Antiviral medications for preventing cytomegalovirus disease in solid organ transplant recipients

Abstract

Background

The risk of cytomegalovirus (CMV) infection in solid organ transplant recipients has resulted in the frequent use of prophylaxis to prevent the clinical syndrome associated with CMV infection. This is an update of a review first published in 2005 and updated in 2008 and 2013.

Objectives

To determine the benefits and harms of antiviral medications to prevent CMV disease and all‐cause death in solid organ transplant recipients.

Search methods

We contacted the information specialist and searched the Cochrane Kidney and Transplant Register of Studies up to 5 February 2024 using search terms relevant to this review. Studies in the Register are identified through searches of CENTRAL, MEDLINE, and EMBASE, conference proceedings, the International Clinical Trials Registry Platform (ICTRP) Search Portal, and ClinicalTrials.gov.

Selection criteria

We included randomised controlled trials (RCTs) and quasi‐RCTs comparing antiviral medications with placebo or no treatment, comparing different antiviral medications or different regimens of the same antiviral medications for CMV prophylaxis in recipients of any solid organ transplant. Studies examining pre‐emptive therapy for CMV infection are studied in a separate review and were excluded from this review.

Data collection and analysis

Two authors independently assessed study eligibility, risk of bias and extracted data. Summary estimates of effect were obtained using a random‐effects model, and results were expressed as risk ratios (RR) and their 95% confidence intervals (CI) for dichotomous outcomes and mean difference (MD) and 95% CI for continuous outcomes. Confidence in the evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach.

Main results

This 2024 update found four new studies, bringing the total number of included studies to 41 (5054 participants). The risk of bias was high or unclear across most studies, with a low risk of bias for sequence generation (12), allocation concealment (12), blinding (11) and selective outcome reporting (9) in fewer studies.

There is high‐certainty evidence that prophylaxis with aciclovir, ganciclovir or valaciclovir compared with placebo or no treatment is more effective in preventing CMV disease (19 studies: RR 0.42, 95% CI 0.34 to 0.52), all‐cause death (17 studies: RR 0.63, 95% CI 0.43 to 0.92), and CMV infection (17 studies: RR 0.61, 95% CI 0.48 to 0.77). There is moderate‐certainty evidence that prophylaxis probably reduces death from CMV disease (7 studies: RR 0.26, 95% CI 0.08 to 0.78). Prophylaxis reduces the risk of herpes simplex and herpes zoster disease, bacterial and protozoal infections but probably makes little to no difference to fungal infection, acute rejection or graft loss. No apparent differences in adverse events with aciclovir, ganciclovir or valaciclovir compared with placebo or no treatment were found.

There is high certainty evidence that ganciclovir, when compared with aciclovir, is more effective in preventing CMV disease (7 studies: RR 0.37, 95% CI 0.23 to 0.60). There may be little to no difference in any outcome between valganciclovir and IV ganciclovir compared with oral ganciclovir (low certainty evidence). The efficacy and adverse effects of valganciclovir or ganciclovir were probably no different to valaciclovir in three studies (moderate certainty evidence). There is moderate certainty evidence that extended duration prophylaxis probably reduces the risk of CMV disease compared with three months of therapy (2 studies: RR 0.20, 95% CI 0.12 to 0.35), with probably little to no difference in rates of adverse events. Low certainty evidence suggests that 450 mg/day valganciclovir compared with 900 mg/day valganciclovir results in little to no difference in all‐cause death, CMV infection, acute rejection, and graft loss (no information on adverse events). Maribavir may increase CMV infection compared with ganciclovir (1 study: RR 1.34, 95% CI: 1.10 to 1.65; moderate certainty evidence); however, little to no difference between the two treatments were found for CMV disease, all‐cause death, acute rejection, and adverse events at six months (low certainty evidence).

Authors' conclusions

Prophylaxis with antiviral medications reduces CMV disease and CMV‐associated death, compared with placebo or no treatment, in solid organ transplant recipients. These data support the continued routine use of antiviral prophylaxis in CMV‐positive recipients and CMV‐negative recipients of CMV‐positive organ transplants.

Plain language summary

What are the benefits of giving regular antiviral medications to prevent cytomegalovirus disease, death and side effects in solid organ transplant recipients

Key messages

‐ In people who have received a solid organ transplant, giving antiviral medications reduces cytomegalovirus (CMV) disease and death from CMV disease, compared with placebo or no treatment.

‐ Longer periods of prophylaxis were found to be more effective than three months of therapy in kidney and lung transplant recipients.

‐ Low‐dose valganciclovir was found to be as effective as the standard dose for preventing CMV in moderate‐risk kidney transplant recipients.

Why use antiviral medication to prevent CMV disease in solid organ transplant recipients?

CMV (a herpes virus) is the most common type of virus in people who have received solid organ transplants(kidney, heart, liver, lung and pancreas). CMV is a major cause of illness and death during the first year after transplantation.

What did we want to find out?

We wanted to look at both the benefits and harms of antiviral medication to prevent CMV disease in people who have received a solid organ transplant.

What did we do?

We searched for all trials that assessed the benefits and harms of randomly allocated antiviral treatment for the prevention of CMV disease in people receiving a solid organ transplant. We compared and summarised the results of the trials and rated our confidence in the information based on factors such as trial methods and sizes.

What did we find?

We included 41 studies involving 5051 people who received a kidney, kidney and pancreas, liver, heart, lung, or heart and lung transplant. We found some antiviral drugs (ganciclovir, valaciclovir and aciclovir) reduced the risk of CMV disease, death due to CMV disease, and clinical disease caused by herpes simplex compared with placebo or no treatment. For CMV disease and death, the benefits of aciclovir, ganciclovir, and valaciclovir were seen across people who received heart, kidney or liver transplants. These benefits occur in both those recipients who had had CMV infection in the past, as well as in those recipients who have not had CMV before but who received a transplant from a donor who had had CMV infection in the past. The benefits occurred at all time points. We found that ganciclovir is more effective than aciclovir and as effective as valganciclovir, which is currently the most commonly used antiviral drug to prevent CMV disease in transplant recipients. Different doses of valganciclovir did not result in a difference in preventing CMV disease.

What are the limitations of the evidence?

Future studies may be required in the seronegative donor‐recipient group depending on the prevalence of CMV disease in this group with newer and more potent immunosuppressive regimens.

More information is required on the efficacy of prophylaxis with different regimens of immunosuppressive regimens used for the prevention and treatment of rejection in different organ transplants.

How up‐to‐date is the evidence?

The evidence is up‐to‐date as of February 2024.

Summary of findings

Summary of findings 1. Antiviral prophylaxis versus placebo or no treatment for preventing cytomegalovirus disease in solid organ transplant recipients.

Antiviral prophylaxis versus placebo or no treatment for preventing cytomegalovirus disease in solid organ transplant recipients
Patient or population: solid organ transplant recipients Settings: tertiary hospitals Intervention: antiviral prophylaxis Comparison: placebo or no treatment
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No. of participants (RCTs)	Quality of the evidence (GRADE)
	Assumed risk	Corresponding risk
	Placebo/no treatment	Antiviral prophylaxis
CMV disease: all symptomatic CMV disease Follow‐up: 3 to 18 months	299 per 1000	126 per 1000 (102 to 156)	RR 0.42 (0.34 to 0.52)	1981 (19)	⊕⊕⊕⊕ high
Death due to CMV disease Follow‐up: 3 to 18 months	23 per 1000	6 per 1000 (2 to 18)	RR 0.26 (0.08 to 0.78)	1300 (7)	⊕⊕⊕⊝ moderate1
All‐cause death Follow‐up: 3 to 18 months	71 per 1000	45 per 1000 (30 to 65)	RR 0.63 (0.43 to 0.92)	1838 (17)	⊕⊕⊕⊕ high
CMV infection Follow‐up: 3 to 18 months	488 per 1000	297 per 1000 (234 to 375)	RR 0.61 (0.48 to 0.77)	1786 (17)	⊕⊕⊕⊕ high
Graft loss Follow‐up: 3 to 18 months	93 per 1000	69 per 1000 (44 to 109)	RR 0.74 (0.47 to 1.17)	825 (10)	⊕⊕⊕⊝ moderate1
Acute rejection Follow‐up: 3 to 18 months	468 per 1000	421 per 1000 (365 to 491)	RR 0.90 (0.78 to 1.05)	1420 (13)	⊕⊕⊕⊕ high
Leucopenia Follow‐up: 4 to 12 months	67 per 1000	69 per 1000 (45 to 107)	RR 1.03 (0.67 to 1.59)	1125 (4)	⊕⊕⊕⊝ moderate1
Kidney dysfunction Follow‐up: 3 to 12 months	72 per 1000	130 per 1000 (63 to 269)	RR 1.81 (0.88 to 3.73)	668 (5)	⊕⊕⊝⊝ low2,3
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI) CI: Confidence interval; RR: Risk ratio; CMV: Cytomegalovirus
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate Very low quality: We are very uncertain about the estimate

¹ Downgraded one level for imprecision given the small event rate and limited amount of studies

² Downgraded one level for inconsistency given the substantial heterogeneity (I² = 40%)

³ Downgraded one level for imprecision given wide CIs around the effect estimate

Summary of findings 2. Ganciclovir versus aciclovir for preventing cytomegalovirus disease in solid organ transplant recipients.

Ganciclovir versus aciclovir for preventing cytomegalovirus disease in solid organ transplant recipients
Patient or population: solid organ transplant recipients Settings: tertiary hospitals Intervention: ganciclovir Comparison: aciclovir
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No. of participants (RCTs)	Quality of the evidence (GRADE)
	Assumed risk	Corresponding risk
	Aciclovir	Ganciclovir
CMV disease Follow‐up: 4 to 27 months	177 per 1000	66 per 1000 (41 to 106)	RR 0.37 (0.23 to 0.6)	1113 (7)	⊕⊕⊕⊕ high
Death due to CMV disease Follow‐up: 4 to 27 months	10 per 1000	3 per 1000 (1 to 15)	RR 0.33 (0.07 to 1.58)	832 (6)	⊕⊕⊕⊝ moderate1
All‐cause death Follow‐up: 4 to 27 months	103 per 1000	117 per 1000 (85 to 163)	RR 1.13 (0.82 to 1.58)	1138 (8)	⊕⊕⊕⊝ moderate1
Acute rejection Follow‐up: 4 to 27 months	491 per 1000	481 per 1000 (427 to 540)	RR 0.98 (0.87 to 1.10)	1009 (6)	⊕⊕⊕⊕ high
Graft loss Follow‐up: 6 to 12 months	148 per 1000	81 per 1000 (40 to 167)	RR 0.55 (0.27 to 1.13)	268 (3)	⊕⊕⊕⊝ moderate1
Opportunistic infections: other viral infections Follow‐up: 4 to 12 months	35 per 1000	28 per 1000 (11 to 70)	RR 0.81 (0.32 to 2.01)	740 (4)	⊕⊕⊕⊝ moderate1
Leucopenia Follow‐up: 4 to 12 months	15 per 1000	48 per 1000 (22 to 106)	RR 3.28 (1.48 to 7.25)	955 (6)	⊕⊕⊝⊝ low2
Kidney dysfunction Follow‐up: 4 to 12 months	425 per 1000	406 per 1000 (351 to 465)	RR 0.96 (0.83 to 1.10)	661 (4)	⊕⊕⊕⊝ moderate1
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI) CI: Confidence interval; RR: Risk ratio; CMV: Cytomegalovirus
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate Very low quality: We are very uncertain about the estimate

¹ Downgraded one level for imprecision given the small event rate and limited amount of studies

² Downgraded two levels for imprecision given the small event rate/limited amount of studies and wide CIs around the effect estimate

Summary of findings 3. Valaciclovir versus ganciclovir or valganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients.

Valaciclovir versus ganciclovir or valganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients
Patient or population: solid organ transplant recipients Settings: tertiary hospitals Intervention: valaciclovir Comparison: ganciclovir or valganciclovir
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No. of participants (RCTs)	Quality of the evidence (GRADE)
	Assumed risk	Corresponding risk
	Ganciclovir or valganciclovir	Valaciclovir
CMV disease Follow‐up: 6 to 36 months	48 per 1000	27 per 1000 (8 to 84)	RR 0.55 (0.17 to 1.74)	331 (4)	⊕⊕⊝⊝ low1
All‐cause death Follow‐up: 6 to 12 months	15 per 1000	29 per 1000 (5 to 182)	RR 1.97 (0.31 to 12.37)	273 (3)	⊕⊕⊝⊝ low1
CMV infection Follow‐up: 6 to 36 months	218 per 1000	286 per 1000 (199 to 415)	RR 1.31 (0.91 to 1.90)	331 (4)	⊕⊕⊝⊝ low1
Acute rejection Follow‐up: 6 to 12 months	200 per 1000	176 per 1000 (62 to 494)	RR 0.91 (0.22 to 3.73)	271 (3)	⊕⊝⊝⊝ very low1,2
Graft loss Follow‐up: 12 months	73 per 1000	44 per 1000 (13 to 147)	RR 0.61 (0.18 to 2.02)	190 (2)	⊕⊕⊝⊝ lowₑ
Leucopenia Follow‐up: 12 months	284 per 1000	210 per 1000 (128 to 352)	RR 0.74 (0.45 to 1.24)	188 (2)	⊕⊕⊕⊝ moderate3
SCr at the end of follow‐up Follow‐up: 6 to 12 months	The mean SCr was 0.12 lower with valaciclovir (0.36 lower to 0.12 lower) compared to ganciclovir or valganciclovir		‐	271 (3)	⊕⊕⊕⊝ moderate3
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI) CI: Confidence interval; RR: Risk ratio; CMV: cytomegalovirus; SCr: Serum creatinine
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate Very low quality: We are very uncertain about the estimate

¹ Downgraded two levels for imprecision given the small event rate/limited amount of studies and the wide CIs around the effect estimate

² Downgraded one level for inconsistency given the substantial heterogeneity (I² = 72%)

³ Downgraded one level for imprecision given the small event rate/limited amount of studies

Summary of findings 4. Extended versus short‐duration valganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients.

Extended versus short‐duration valganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients
Patient or population: solid organ transplant recipients Settings: tertiary hospitals Intervention: extended duration (200 to 365 days) Comparison: short duration (90 to 100 days)
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No. of participants (RCTs)	Quality of the evidence (GRADE)
	Assumed risk	Corresponding risk
	Short‐duration valganciclovir	Extended‐duration valganciclovir
CMV disease Follow‐up: 13 to 24 months	314 per 1000	63 per 1000 (38 to 110)	RR 0.20 (0.12 to 0.35)	454 (2)	⊕⊕⊕⊝ moderate1
All‐cause death Follow‐up: 24 months	31 per 1000	3 per 1000 (0 to 52)	RR 0.09 (0.01 to 1.70)	319 (1)	⊕⊕⊝⊝ low2
CMV syndrome Follow‐up: 13 to 24 months	310 per 1000	121 per 1000 (74 to 198)	RR 0.39 (0.24 to 0.64)	454 (2)	⊕⊕⊕⊝ moderate1
CMV infection Follow‐up: 13 to 24 months	502 per 1000	136 per 1000 (50 to 357)	RR 0.27 (0.10 to 0.71)	454 (2)	⊕⊕⊝⊝ low1,3
Acute rejection Follow‐up: 12 months	218 per 1000	140 per 1000 (94 to 207)	RR 0.64 (0.43 to 0.95)	454 (2)	⊕⊕⊕⊝ moderate1
Opportunistic infections Follow‐up: 13 to 24 months	343 per 1000	244 per 1000 (113 to 539)	RR 0.71 (0.33 to 1.57)	456 (2)	⊕⊕⊝⊝ low1,4
Total treatment‐related adverse effects Follow‐up: 13 to 24 months	426 per 1000	494 per 1000 (413 to 597)	RR 1.16 (0.97 to 1.40)	456 (2)	⊕⊕⊕⊝ moderate1
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). **Risks were calculated from pooled risk differences CI: Confidence interval; RR: Risk ratio; CMV: Cytomegalovirus
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate Very low quality: We are very uncertain about the estimate

¹ Downgraded one level for imprecision given the small event rate/limited amount of studies

² Downgraded two levels for imprecision given the small event rate/limited amount of studies and the wide CIs around the effect estimate

³ Downgraded one level for inconsistency given the substantial heterogeneity (I² = 82%)

⁴ Downgraded one level for inconsistency given the substantial heterogeneity (I² = 82%)

Summary of findings 5. Low versus standard dose valganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients.

Low versus standard dose valganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients
Patient or population: solid organ transplant recipients Settings: tertiary hospitals Intervention: valganciclovir 450 mg/d Comparison: valganciclovir 900 mg/d
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No. of participants (RCTs)	Quality of the evidence (GRADE)
	Assumed risk	Corresponding risk
	Valganciclovir 900 mg/d	Valganciclovir 450 mg/d
All‐cause death Follow‐up: 24 months	20 per 1000	4 per 1000 (2 to 84)	RR 0.20 (0.11 to 4.11)	196 (1)	⊕⊕⊝⊝ low1
CMV infection Follow‐up: 24 months	70 per 1000	26 per 1000 (8 to 86)	RR 0.37 (0.11 to 1.22)	256 (2)	⊕⊕⊝⊝ low1
Acute rejection Follow‐up: 24 months	61 per 1000	20 per 1000 (4 to 99)	RR 0.33 (0.07 to 1.61)	196 (1)	⊕⊕⊝⊝ low1
Graft loss Follow‐up: 24 months	10 per 1000	10 per 1000 (1 to 161)	RR 1.00 (0.06 to 15.76)	196 (1)	⊕⊕⊝⊝ low1
Opportunistic infection	Not reported	Not reported	‐	‐	‐
Adverse events	Not reported	Not reported	‐	‐	‐
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio; CMV: Cytomegalovirus
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate Very low quality: We are very uncertain about the estimate

¹ Downgraded two levels for imprecision given the small event rate/limited amount of studies and the wide CIs around the effect estimate

Summary of findings 6. Maribavir versus ganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients.

Maribavir versus ganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients
Patient or population: solid organ transplant recipients Settings: tertiary hospitals Intervention: maribavir Comparison: ganciclovir
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No. of participants (RCTs)	Quality of the evidence (GRADE)
	Assumed risk	Corresponding risk
	Ganciclovir	Maribavir
CMV disease Follow‐up: 6 months	150 per 1000	195 per 1000 (111 to 344)	RR 1.30 (0.74 to 2.29)	233 (1)	⊕⊕⊝⊝ low1
All‐cause death Follow‐up: 6 months	41 per 1000	20 per 1000 (18 to 127)	RR 1.50 (0.55 to 4.11)	294 (1)	⊕⊕⊝⊝ low1
CMV infection Follow‐up: 6 months	533 per 1000	181 per 1000 (53 to 347)	RR 1.34 (1.10 to 1.65)	233 (1)	⊕⊕⊕⊝ moderate2
Acute rejection Follow‐up: 6 months	147 per 1000	12 per 1000 (69 to 90)	RR 0.92 (0.53 to 1.61)	303 (1)	⊕⊕⊝⊝ low1
Graft loss	Not reported	Not reported	‐	‐	‐
Opportunistic infection	Not reported	Not reported	‐	‐	‐
Adverse events Follow‐up: 6 months	487 per 1000	5 per 1000 (102 to 122)	RR 0.99 (0.79 to 1.25)	303 (1)	⊕⊕⊕⊝ moderate2
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio; CMV: Cytomegalovirus
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate Very low quality: We are very uncertain about the estimate

¹ Downgraded two levels for imprecision given the small event rate/limited amount of studies and the wide CIs around the effect estimate

² Downgraded one level for imprecision given the small event rate/limited amount of studies

Background

Description of the condition

Cytomegalovirus (CMV) is the most common virus pathogen in solid organ transplant recipients, being a major cause of morbidity and death during the first six months post‐transplant (Kotton 2013; Razonable 2013). The overall incidence of symptomatic CMV disease in the transplant population ranges from 30% to 50%, with the incidence and severity being highest among lung transplant recipients (Bate 2010; Linden 2000; Zuhair 2019). Without preventive treatment, CMV‐seropositive transplant recipients have a 10% to 20% risk of developing CMV disease (Pascual 2018; Sommerer 2019). Approximately 50% of deaths following lung transplantation are attributed to infection (Hakimi 2017; Hartmann 2006; Michaels 2000; Sagedal 2004). Like all herpes viruses, CMV has the propensity to establish lifelong latency infection in the host after the initial infection has resolved. Therefore, a solid organ transplant recipient may be infected either by exogenous virus or by reactivation of latent virus if they were CMV‐positive pre‐transplant. When the immune system is compromised, a reactivation of a latent infection might occur. Those at highest risk of symptomatic CMV disease are CMV seronegative patients who receive organs from CMV seropositive donors and CMV seropositive patients on heavily immunosuppressive regimens (Fishman 1998; Rubin 2000). CMV may manifest as a non‐specific illness characterised by fever, mononucleosis, leucopenia and thrombocytopenia or as a variety of clinical syndromes, including pneumonitis, hepatitis, encephalitis and focal gastrointestinal disease. In addition, CMV infection causes morbidity in organ recipients through indirect effects on their immune response (Rubin 1989) and is associated with increased risk of allograft injury and rejection (Grattan 1989; Keenan 1991), opportunistic infections (Fishman 1995; Hadley 1995; Van den Berg 1996) and late‐onset malignancies such as Epstein‐Barr virus lymphoproliferative disease (Basgoz 1995).

Description of the intervention

Two main strategies to prevent CMV disease have been adopted: universal prophylaxis of organ recipients with antiviral agents and/or immunoglobulins, or pre‐emptive therapy of organ recipients who develop evidence of asymptomatic CMV infection during screening (Kotton 2018; Rubin 1989). Antiviral medications may be given intravenously (ganciclovir, aciclovir, immunoglobulins) in acute settings but are now more commonly administered daily orally with the availability of the longer‐acting oral preparations valganciclovir and valaciclovir. Prophylaxis is usually administered for three to six months after transplantation, during the time that patients are most at risk of CMV infection and disease. Pre‐emptive therapy relies upon monitoring for CMV infection by pp65 antigenaemia assay or for CMV DNA using quantitative polymerase chain reaction (PCR) with administration of antiviral therapy when CMV infection is diagnosed (Emery 2000). While antivirals are considered effective for the prevention and treatment of CMV infection and disease, they might come at the price of undesirable hematologic adverse events.

How the intervention might work

This review examines the use of prophylaxis to prevent CMV infection and CMV disease. Prophylaxis is usually administered for the first three to six months after transplant when the recipient is at the highest risk of CMV infection. Prevention of CMV disease should reduce the associated morbidity and death. In addition, prophylaxis may reduce the indirect effects of CMV infection, including opportunistic infections, acute rejection and graft loss. Pre‐emptive therapy is the subject of a different review (Owers 2013).

Why it is important to do this review

There remains a lack of consensus on the merits of the various CMV prophylaxis protocols and guidelines available (Fishman 1998; Humar 2009; Kotton 2018). Universal prophylaxis exposes all solid organ transplant recipients to the adverse effects of medications, particularly haematological effects (leucopenia, neutropenia, increased risk of other infections) with valganciclovir and neurological effects with valaciclovir. Valganciclovir is the most commonly used prophylactic drug, yet valaciclovir might be an alternative because of less bone marrow suppression and lower incidence of acute rejections. However, based on epidemiological studies, many recipients without prophylaxis do not develop disease (Humar 2009). CMV prophylaxis is recommended for all kidney transplant recipients except when the donor and recipient both have negative CMV serologies (KDIGO Transplant Working Group 2009). The Transplantation Society International CMV Consensus Group recommend the use of either prophylactic or pre‐emptive treatment for transplant recipients in CMV‐positive donor/CMV‐negative recipient and CMV‐positive recipient transplantations (Kotton 2018). Prophylaxis may also be associated with an increased risk of late‐onset CMV disease occurring after discontinuation of prophylaxis and with the development of resistant organisms (Humar 2009). Antiviral prophylactic treatment might be beneficial for patients with the highest CMV risk, while patients with lower CMV risk might benefit from pre‐emptive monitoring. Therefore, a patient‐tailored approach to trade‐off the efficacy and safety of the available for personalised CMV prevention is needed (Hellemans 2022). The heterogeneity in treatment approaches highlights the need to reach a consensus on the optimal CMV prevention with the available antiviral medication. A systematic review is therefore required to assess the benefits and harms of antiviral prophylaxis in solid organ transplants.

A meta‐analysis of all possible types of CMV prophylactic treatment versus placebo or no treatment was first published in 1998 (Couchoud 1998a) and later updated in 2005 (Hodson 2005b), 2008 (Hodson 2008), and 2013 (Hodson 2013). A wide variety of preventive antiviral approaches, both prophylactic and pre‐emptive, are included in the recent review of Raval 2020, where the head‐to‐head comparison of prophylactic approaches showed no consensus on the optimum dose, duration and route of administration on CMV outcomes. Other reviews have evaluated pre‐emptive therapy in the detection of CMV viraemia (Strippoli 2006a; Strippoli 2006b) and the use of other agents (immunoglobulins, vaccines, interferon) alone or in combination with antiviral medications (Hodson 2007). Both indirect and direct comparisons of pre‐emptive therapy versus prophylactic strategies show equal effectiveness in preventing CMV disease, death, graft loss, and acute rejections but with a higher risk of leukopenia and neutropenia for patients receiving prophylactic treatment caused by a longer antiviral exposure (Florescu 2014; Owers 2013).

The current update includes new data on the prophylactic treatment of CMV in any newly published studies since the last update. The Cochrane review Pre‐emptive therapy for cytomegalovirus viraemia to prevent cytomegalovirus disease in solid organ transplant recipients (Owers 2013) will be updated concomitantly with this review.

Objectives

This review aimed to assess the benefits and harms of all antiviral medications for preventing symptomatic CMV disease in solid organ transplant recipients of all ages, irrespective of CMV serostatus prior to transplantation. The secondary aims were to evaluate the efficacy of antiviral medications in preventing all CMV infections (symptomatic and asymptomatic where CMV is detected only by laboratory investigation) and in decreasing the incidence of acute rejection, graft loss, death (all‐cause death and death due to CMV disease), opportunistic infections, and to evaluate the harms of each antiviral medication.

The review compared studies of antiviral medications with placebo or no treatment and explored comparisons between two or more antiviral agents and/or two different doses or durations of the same antiviral agent. Thirdly, it has compared the treatment effect of each regimen between different solid organs and finally, among the different risk groups (i.e. pre‐existent CMV serostatus and/or level of immunosuppression).

Methods

Criteria for considering studies for this review

Types of studies

We included all randomised controlled trials (RCTs) and quasi‐RCTs (RCTs in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable method). We did not include single‐arm studies or studies without a control arm, and we included all studies regardless of their publication status or language of publication. If available, data for the first period of cross‐over studies were to be included in meta‐analyses; otherwise, cross‐over studies were reported in the text only.

Types of participants

We included studies with participants of all ages, irrespective of CMV serostatus before transplantation, who have undergone at least one solid organ transplant (kidney, liver, lung, heart, pancreas). Bone marrow and other cellular transplants were excluded.

Types of interventions

Interventions included antiviral medications (e.g. aciclovir, maribavir, ganciclovir, valaciclovir, valganciclovir). Comparisons were made between antiviral medications and placebo or no treatment, two different antiviral medications, or two varying doses or durations of an antiviral medication.

Studies of pre‐emptive treatment (i.e. treatment on detection of CMV viraemia), immunoglobulin alone or with antiviral medications, vaccines or interferon were excluded. Treatment regimens for symptomatic CMV disease were excluded as these are the subject of other reviews (Strippoli 2006a; Strippoli 2006b; Hodson 2007).

Types of outcome measures

We did not use the measurement of the outcomes assessed in this review as an eligibility criterion.

Primary outcomes

The primary outcome measures were:

• CMV disease (documented CMV infection with clinical symptoms)

• All‐cause death.

The study investigators' definition of symptomatic CMV disease was used. This was usually the diagnosis of CMV infection in association with one or more of the following: CMV syndrome (temperature of 38°C or more with no other documented source in association with one or more of atypical lymphocytosis, leucopenia or thrombocytopenia), pneumonitis, focal gastrointestinal disease, liver function abnormality, or encephalitis.

Secondary outcomes

Secondary outcomes included:

• CMV infection (symptomatic and asymptomatic)

• Death due to CMV disease

• Time to CMV disease

• Acute rejection

• Graft loss

• Opportunistic infections

• Kidney function at the end of the study

• Harms (including adverse events, nephrotoxicity, bone marrow suppression, emergence of resistant CMV strains, late onset of CMV disease).

The study investigators' definition of CMV infection was used. This was usually the isolation of CMV from a cultured specimen from any site, positive histopathology or CMV antigen detection in a tissue specimen, the presence of CMV pp65 antigenaemia, or an elevation in CMV viral load as detected by qualitative or quantitative PCR (as defined by the investigator).

Graft loss was defined as the need for dialysis for kidney transplantation or retransplantation for other organs during the follow‐up period of the study. The study investigators' definition of acute rejection was used. This was either biopsy‐proven or clinical, defined by a rise in serum creatinine (SCr) levels with respect to kidney transplants or response to rejection treatment. The study investigators' definition of kidney function at the end of the study was used and could be measured by SCr or estimated glomerular filtration rate (eGFR). All outcomes were recorded as dichotomous (present/absent), except time to the development of CMV disease and kidney function.

Search methods for identification of studies

We performed a comprehensive search with no restrictions on the language of publication or publication status.

Electronic searches

We searched the Cochrane Kidney and Transplant Register of Studies up to 5 February 2024 through contact with the Information Specialist using search terms relevant to this review. The Register contains studies identified from the following sources.

1. Monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL)

2. Weekly searches of MEDLINE OVID SP

3. Handsearching of kidney‐related journals and the proceedings of major kidney and transplant conferences

4. Searching of the current year of EMBASE OVID SP

5. Weekly current awareness alerts for selected kidney journals

6. Searches of the International Clinical Trials Registry Platform (ICTRP) Search Portal and ClinicalTrials.gov.

Studies contained in the Register are identified through searches of CENTRAL, MEDLINE, and EMBASE based on the scope of Cochrane Kidney and Transplant. Details of search strategies, as well as a list of handsearched journals, conference proceedings and current awareness alerts, are available on the Cochrane Kidney and Transplant website under CKT Register of Studies.

See Appendix 1 for search terms used in strategies for this review.

Searching other resources

1. Reference lists of review articles, relevant studies, and clinical practice guidelines.

2. Letters seeking information about unpublished or incomplete trials to investigators known to be involved in previous studies.

Data collection and analysis

Selection of studies

Two authors independently screened titles and abstracts retrieved from the searches and identified those studies that met the inclusion criteria. This process favoured over‐selection in order to include all relevant studies. The full article was retrieved if uncertainty existed or when the abstract was not available. Any disagreement with article selection was resolved through discussion and consultation through a third review author. We documented reasons for excluding studies in the Characteristics of excluded studies tables. We presented a Preferred Reporting Items for Systematic Reviews and Meta‐analyses (PRISMA) flow diagram showing the process of study selection (Liberati 2009).

Data extraction and management

Two authors independently extracted data from eligible studies using standardised data extraction forms. Studies reported in foreign language journals were translated before data extraction. Participant characteristics (number, age, sex, comorbidities), interventions (type of treatment, dose, duration, co‐interventions) and primary and secondary outcome measures were recorded. Any discrepancies in data extraction were resolved via discussion. Where results of a study were published in more than one article, data from the most complete study, with the longest follow‐up, were extracted from all sources and used in the analysis only once.

We extracted the outcome data relevant to this Cochrane review as needed for the calculation of summary statistics and measures of variance. For dichotomous outcomes, we tried to obtain numbers of events and totals for the population of a 2 x 2 table, as well as summary statistics with corresponding measures of variances. For continuous outcomes, we attempted to obtain means and standard deviation or data necessary to calculate these.

We resolved any data extraction disagreements between the two authors by discussion or, if required, by consultation with a third review author. We provided information, including the trial identified, about potentially relevant ongoing studies in the table Characteristics of ongoing studies.

In the event of duplicate publications, companion documents, or multiple articles of a primary study, we maximised the yield of information by mapping all publications to unique studies and collating all data available. We used the most complete dataset aggregated across all known publications with the longest follow‐up.

Assessment of risk of bias in included studies

The following items were independently assessed by two authors using the risk of bias assessment tool (Higgins 2022) (see Appendix 2). All risk of bias domains were judged as 'low risk', 'high risk', or 'unclear risk', and we evaluated individual bias items as described in the Cochrane Handbook for Systematic Reviews of Interviews (Higgins 2022).

• Was there adequate sequence generation (selection bias)?

• Was allocation adequately concealed (selection bias)?

• Was knowledge of the allocated interventions adequately prevented during the study?

  • Participants and personnel (performance bias)

  • Outcome assessors (detection bias)

• Were incomplete outcome data adequately addressed (attrition bias)?

• Are reports of the study free of selective outcome reporting (reporting bias)?

• Was the study apparently free of other problems that could put it at risk of bias?

Measures of treatment effect

Dichotomous outcomes were expressed as risk ratios (RR) with 95% confidence intervals (CI). Risk differences (RD) with 95% CI were calculated for adverse effects. Continuous outcomes were calculated as mean differences (MD) with 95% CI.

Unit of analysis issues

The unit of analysis was the individual participant. We did not identify any cross‐over trials. Trials with more than two intervention groups for inclusion in this Cochrane review were handled in accordance with guidance provided in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022).

Dealing with missing data

Study authors were contacted for information on sequence generation, allocation concealments and missing data. Where missing data were few and not thought likely to influence results, the available data were analysed.

Assessment of heterogeneity

We first assessed the heterogeneity by visual inspection of the forest plot. We then quantified statistical heterogeneity using the I² statistic, which describes the percentage of total variation across studies that is due to heterogeneity rather than sampling error (Higgins 2003). The following was used as a guide to the interpretation of I² values.

• 0% to 40%: might not be important

• 30% to 60%: may represent moderate heterogeneity

• 50% to 90%: may represent substantial heterogeneity

• 75% to 100%: considerable heterogeneity.

The importance of the observed value of I² depends on the magnitude and direction of treatment effects and the strength of evidence for heterogeneity (e.g., a P value from the Chi² test or a CI for I²) (Higgins 2003). We tried to determine possible reasons for heterogeneity by examining individual studies and subgroup characteristics.

Assessment of reporting biases

The study protocols or trial registration of the included studies were assessed for selective outcome reporting.

If we identified 10 or more studies investigating a particular outcome, we assessed small study effects through funnel plots.

Data synthesis

Data were pooled using a random‐effects model to calculate a summary estimate of the effect. For the dichotomous outcomes, we will use the Mantel‐Haenszel method, and for continuous outcomes, the inverse variance method.

Subgroup analysis and investigation of heterogeneity

To explore clinical differences among studies that might be expected to influence the magnitude of the treatment effect for the primary outcomes of CMV disease and all‐cause death, subgroup analysis and univariate meta‐regression were performed using STATA^© software (StataCorp LP, Texas, USA) using restricted maximum‐likelihood to estimate the between‐study variance. The potential sources of variability defined a priori were organ transplanted, antiviral medication used, use of immunosuppressive regimen including antibody therapy, treatment duration, donor/recipient CMV status at transplant, the time from transplant that the outcomes were measured, and methodological quality. Multivariate meta‐regression was performed to investigate whether the results were altered after allowing for the differences in drug used, organ transplanted and recipient CMV serostatus at the time of transplantation.

Sensitivity analysis

We planned to conduct sensitivity analyses by restricting the analyses by taking the risk of bias into account and excluding the 'high risk' or 'unclear risk' studies, yet for the majority of the comparison, this was not possible due to no 'low risk' of studies available.

Summary of findings and assessment of the certainty of the evidence

We presented the main results of the review in the Summary of the findings tables. These tables present key information concerning the quality of the evidence, the magnitude of the effects of the interventions examined, and the sum of the available data for the main outcomes (Schünemann 2022a). The Summary of findings tables also include an overall grading of the evidence related to each main outcome using the GRADE (Grades of Recommendation, Assessment, Development and Evaluation) approach (GRADE 2008; GRADE 2011). The GRADE approach defines the quality of a body of evidence as the extent to which one can be confident that an estimate of effect or association is close to the true quantity of specific interest. The quality of a body of evidence involves consideration of within‐trial risk of bias (methodological quality), directness of evidence, heterogeneity, precision of effect estimates and risk of publication bias (Schünemann 2022b). Standard statements, according to the GRADE approach, were used to communicate findings combining size and certainty of effect (Santesso 2020). We presented the following outcomes in the Summary of findings tables.

• CMV disease

• All‐cause death

• CMV infection

• Acute rejection

• Graft loss

• Opportunistic infections

• Harms.

Results

Description of studies

The following section contains broad descriptions of the studies considered in this review. For further details on each individual study, please see Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification; Characteristics of ongoing studies.

Results of the search

For this update, we searched the Cochrane Kidney and Transplant Register of Studies u to 5 February 2024 and identified 94 new reports. Four new studies (13 reports) were included, 14 new studies (34 reports) were excluded, and one ongoing study (NCT04225923) was identified. Two new studies (3 reports) are awaiting classification; one recently completed (Limaye 2023), and one is an abstract‐only publication (Verghese 2022). We also identified 43 new reports of existing included and excluded.

We reassessed and reclassified 46 studies. One ongoing study was moved to an existing included study. One excluded study was an additional report of an existing included study, and 44 excluded studies were deleted as they were not RCTs, the wrong population or comparator.

A total of 41 studies were included (219 reports, 5054 participants); 34 were excluded, two are awaiting classification, and there is one ongoing study (Figure 1).

1.

Flow diagram show study selection

Included studies

In the 2024 update, four additional studies were included, with 619 new patients (Ali Ibrahim 2020 Kidney; Halim 2016 Kidney; Prabakaran 2020 Kidney; Winston 2012 Liver).

• Halim 2016 Kidney compared valganciclovir 450 mg daily with valganciclovir 900 mg/day for the first six months after kidney transplant.

• Winston 2012 Liver compared maribavir (an oral benzimidazole riboside with potent in vitro activity against CMV) 100 mg twice daily with oral ganciclovir 1000 mg three times/day for 14 weeks after liver transplant.

• Ali Ibrahim 2020 Kidney compared low‐dose valganciclovir (450 mg/day) with valacyclovir regimen (8 g/day) after kidney transplant.

• Prabakaran 2020 Kidney compared low‐dose valganciclovir (900 mg three times/week) with standard dose valganciclovir (900 mg once/day) for recipients of a kidney transplant.

The results of the 2VAL 2010 Kidney study have also been updated, given that additional reports have been available since the 2013 update. One ongoing study has been identified (NCT04225923). A description of the included studies in the previous updates of this review has been noted in Appendix 3.

The 2024 update includes 41 studies (5054 participants). This includes the following comparisons.

1. Prophylaxis versus placebo or no treatment: 19 studies (Balfour 1989 Kidney; Barkholt 1999 Liver; Gavalda 1997 Liver; Kletzmayr 1996 Kidney; Ahsan 1997 Kidney; Brennan 1997 Kidney; Cohen 1993 Liver; Conti 1995 Kidney; Egan 2002 Heart; Gane 1997 Liver; Hibberd 1995 Kidney; Leray 1995 Kidney; Lowance 1999 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Pouteil‐Noble 1996 Kidney; Rondeau 1993 Kidney; Rostaing 1994 Kidney; Saliba 1993 Liver).

2. Ganciclovir versus aciclovir: eight studies (Badley 1997 Liver; Duncan 1993 Lung; Flechner 1998 Kidney; Martin 1994 Liver; Nakazato 1993 Liver; Rubin 2002 All; Winston 1995 Liver; Winston 2003 Liver).

3. Ganciclovir/aciclovir versus ganciclovir: one study (Green 1997 Liver).

4. Valganciclovir versus ganciclovir: one study (Paya 2004 All).

5. Valaciclovir versus ganciclovir or valganciclovir: five studies (2VAL 2010 Kidney; Ali Ibrahim 2020 Kidney; Pavlopoulou 2005 Kidney; Reischig 2005 Kidney).

6. Valganciclovir low dose versus standard dose: two studies (Halim 2016 Kidney; Prabakaran 2020 Kidney).

7. Maribavir versus ganciclovir, including 1 study (Winston 2012 Liver).

8. Prophylaxis with different regimens of ganciclovir, including three studies (Hertz 1998 Heart/lung; Winston 2004 Liver; Nafar 2005 Kidney)

9. Prophylaxis with extended durations of valganciclovir, including two studies (IMPACT 2010 Kidney; Palmer 2010 Lung).

Nineteen studies investigated the antiviral medications in kidney transplant recipients (2VAL 2010 Kidney; Ahsan 1997 Kidney; Ali Ibrahim 2020 Kidney; Balfour 1989 Kidney; Brennan 1997 Kidney; Conti 1995 Kidney; Flechner 1998 Kidney; Halim 2016 Kidney; Hibberd 1995 Kidney; IMPACT 2010 Kidney; Kletzmayr 1996 Kidney; Leray 1995 Kidney; Lowance 1999 Kidney; Nafar 2005 Kidney; Pouteil‐Noble 1996 Kidney; Prabakaran 2020 Kidney; Reischig 2005 Kidney; Rondeau 1993 Kidney; Rostaing 1994 Kidney). Liver transplant recipients are included in 13 studies (Badley 1997 Liver; Barkholt 1999 Liver; Cohen 1993 Liver; Gane 1997 Liver; Gavalda 1997 Liver; Green 1997 Liver; Martin 1994 Liver; Nakazato 1993 Liver; Saliba 1993 Liver; Winston 1995 Liver; Winston 2003 Liver; Winston 2004 Liver; Winston 2012 Liver). Lung transplant recipients are included in two studies (Duncan 1993 Lung; Palmer 2010 Lung), and heart transplant recipients in three studies (Egan 2002 Heart; Macdonald 1995 Heart; Merigan 1992 Heart). Finally, Hertz 1998 Heart/lung included both heart and lung transplant recipients; Paya 2004 All included liver, kidney, heart, and kidney‐pancreas recipients, and Rubin 2002 All investigated antiviral medication in kidney, liver or heart transplant recipients.

Excluded studies

A description of the excluded studies in the previous updates of this review has been noted in Appendix 3. In this 2024 update, there are a total of 34 excluded studies. The reasons for exclusion in the current update were: wrong design (4 studies), wrong population (1 study), wrong intervention (pre‐emptive treatment and/or combination of immunoglobin or monoclonal antibodies) (29 studies).

See Characteristics of excluded studies.

Risk of bias in included studies

Risk of bias ratings for all individual included studies can be found in Characteristics of included studies; Figure 2; Figure 3

2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies

3.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study

Allocation

Random sequence generation

The risk of bias was low for sequence generation in 12 studies (2VAL 2010 Kidney; Ahsan 1997 Kidney; Badley 1997 Liver; Balfour 1989 Kidney; Cohen 1993 Liver; Egan 2002 Heart; Flechner 1998 Kidney; Macdonald 1995 Heart; Martin 1994 Liver; Palmer 2010 Lung; Paya 2004 All; Reischig 2005 Kidney); high in one study (Brennan 1997 Kidney); and unclear in the remaining 28 studies.

Allocation concealment

Overall, the risk of bias for allocation concealment was assessed as low risk in 12 studies (2VAL 2010 Kidney; Badley 1997 Liver; Cohen 1993 Liver; Egan 2002 Heart; Flechner 1998 Kidney; IMPACT 2010 Kidney; Palmer 2010 Lung; Paya 2004 All; Pouteil‐Noble 1996 Kidney; Reischig 2005 Kidney; Rubin 2002 All; Saliba 1993 Liver), high in one study (Brennan 1997 Kidney), and unclear in the remaining 28 studies.

Of 19 studies comparing prophylaxis with placebo or no treatment, the risk of bias was low for allocation concealment in four studies (Cohen 1993 Liver; Egan 2002 Heart; Pouteil‐Noble 1996 Kidney; Saliba 1993 Liver); high in one study (Brennan 1997 Kidney) and the information was unclear in 14 studies. Of the 22 studies comparing different medications, allocation concealment bias was low in eight studies (2VAL 2010 Kidney; Badley 1997 Liver; Flechner 1998 Kidney; IMPACT 2010 Kidney; Palmer 2010 Lung; Paya 2004 All; Reischig 2005 Kidney; Rubin 2002 All); and information was not available for 14 studies.

Blinding

Performance bias

Performance bias was assessed as low risk in 11 studies, including seven studies that compared prophylaxis with placebo (Balfour 1989 Kidney; Barkholt 1999 Liver; Gane 1997 Liver; Lowance 1999 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Pouteil‐Noble 1996 Kidney), one study comparing different antiviral agents (Paya 2004 All), one study comparing ganciclovir versus maribavir (Winston 2012 Liver), and two studies comparing different durations of valganciclovir (IMPACT 2010 Kidney; Palmer 2010 Lung). The risk of bias was unclear for blinding of participants and investigators in three studies (Ali Ibrahim 2020 Kidney; Egan 2002 Heart; Halim 2016 Kidney). The remaining 27 studies were assessed as being at high risk of performance bias.

Detection bias

The risk of detection bias was low in 10 studies (Balfour 1989 Kidney; Barkholt 1999 Liver; Gane 1997 Liver; IMPACT 2010 Kidney; Lowance 1999 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Palmer 2010 Lung; Paya 2004 All; Pouteil‐Noble 1996 Kidney) and unclear in four studies (Ali Ibrahim 2020 Kidney; Egan 2002 Heart; Halim 2016 Kidney; Winston 2012 Liver). The remaining 27 studies were judged to be at high risk of detection bias.

Incomplete outcome data

We identified 37 studies that were considered to be at low risk of attrition bias. Of these, 18 studies compared prophylaxis with placebo or no treatment (Ahsan 1997 Kidney; Balfour 1989 Kidney; Barkholt 1999 Liver; Brennan 1997 Kidney; Cohen 1993 Liver; Conti 1995 Kidney; Egan 2002 Heart; Gane 1997 Liver; Gavalda 1997 Liver; Hibberd 1995 Kidney; Kletzmayr 1996 Kidney; Lowance 1999 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Pouteil‐Noble 1996 Kidney; Rondeau 1993 Kidney; Rostaing 1994 Kidney; Saliba 1993 Liver) and 19 studies compared different antiviral medications or regimens (2VAL 2010 Kidney; Badley 1997 Liver; Duncan 1993 Lung; Flechner 1998 Kidney; Green 1997 Liver; Halim 2016 Kidney; Hertz 1998 Heart/lung; IMPACT 2010 Kidney; Martin 1994 Liver; Nakazato 1993 Liver; Palmer 2010 Lung; Pavlopoulou 2005 Kidney; Paya 2004 All; Prabakaran 2020 Kidney; Reischig 2005 Kidney; Rubin 2002 All; Winston 1995 Liver; Winston 2003 Liver; Winston 2004 Liver). Two studies were considered to be at high risk of attrition bias (Nafar 2005 Kidney; Winston 2012 Liver), and two studies were at unclear risk of attrition bias (Ali Ibrahim 2020 Kidney; Leray 1995 Kidney).

Selective reporting

Studies were considered to be at low risk of bias if they reported all the expected outcomes (CMV disease, CMV infection, acute rejection, graft loss, death, opportunistic infections, adverse effects). Nine studies were considered to be at low risk of bias (2VAL 2010 Kidney; Balfour 1989 Kidney; Barkholt 1999 Liver; Egan 2002 Heart; Gane 1997 Liver; IMPACT 2010 Kidney; Paya 2004 All; Prabakaran 2020 Kidney; Winston 1995 Liver), and six studies were considered to be at unclear risk of bias (Ali Ibrahim 2020 Kidney; Halim 2016 Kidney; Leray 1995 Kidney; Pouteil‐Noble 1996 Kidney; Saliba 1993 Liver; Winston 2012 Liver). The remaining 26 studies were considered to be at high risk of bias because they failed to report adequately on one or more outcomes.

Other potential sources of bias

Six studies were considered at low risk of bias as they reported funding from government or university sources (2VAL 2010 Kidney; Badley 1997 Liver; Balfour 1989 Kidney; Halim 2016 Kidney; Reischig 2005 Kidney; Rondeau 1993 Kidney). Fifteen studies were considered to be at high risk of bias because they reported pharmaceutical sponsorship (Barkholt 1999 Liver; Brennan 1997 Kidney; Egan 2002 Heart; Gane 1997 Liver; Hibberd 1995 Kidney; IMPACT 2010 Kidney; Lowance 1999 Kidney; Merigan 1992 Heart; Nakazato 1993 Liver; Palmer 2010 Lung; Paya 2004 All; Rubin 2002 All; Winston 1995 Liver; Winston 2003 Liver; Winston 2012 Liver). In the remaining 20 studies, it was unclear whether pharmaceutical sponsorship existed or what impact it had on the study's conduct.

Effects of interventions

See: Table 1; Table 2; Table 3; Table 4; Table 5; Table 6

Antiviral medication versus placebo or no treatment

We identified 19 studies (1981 analysed patients) that compared antiviral medications with placebo or no treatment. Six studies administered aciclovir (Balfour 1989 Kidney; Barkholt 1999 Liver; Gavalda 1997 Liver; Kletzmayr 1996 Kidney; Rostaing 1994 Kidney; Saliba 1993 Liver); 11 studies administered ganciclovir (Ahsan 1997 Kidney; Brennan 1997 Kidney; Cohen 1993 Liver; Conti 1995 Kidney; Gane 1997 Liver; Hibberd 1995 Kidney; Leray 1995 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Pouteil‐Noble 1996 Kidney; Rondeau 1993 Kidney), and two studies administered valaciclovir (Egan 2002 Heart; Lowance 1999 Kidney). Using a funnel plot (Figure 4), no significant publication bias could be demonstrated among studies comparing antiviral medications with placebo or no treatment. There were too few studies comparing ganciclovir and aciclovir to subject the data to a funnel plot.

4.

Funnel plot of 19 trials comparing antiviral medications with placebo or no treatment

See Table 1.

CMV disease

The average risk of CMV disease was 30% (range 11% to 72%). There is high certainty evidence to suggest that prophylaxis with all agents reduced the risk for CMV disease (Analysis 1.1.1 (19 studies, 1981 participants): RR 0.42, 95% CI 0.34 to 0.52; I² = 13%), CMV syndrome (Analysis 1.1.2 (11 studies, 1570 participants): RR 0.41, 95% CI 0.29 to 0.57; I² = 0%) and CMV invasive organ disease (Analysis 1.1.3 (12 studies, 1628 participants): RR 0.34, 95% CI 0.21 to 0.55; I² = 35%) compared with placebo or no treatment. No significant heterogeneity between studies was detected in the effect of prophylaxis on CMV disease, syndrome and invasive organ disease.

1.1. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 1: CMV disease

Figure 5 shows the cumulative meta‐analysis demonstrating changes over time for CMV disease. There was a consistent reduction in CMV disease with antiviral prophylaxis from the first study in 1989, with the relative risk remaining stable from 1996 but with a progressive narrowing in CIs.

5.

CMV disease: cumulative meta‐analysis showing change over time

The time to onset of CMV disease was reported in 11 studies. Prophylaxis increased the time from transplant to the onset of CMV disease in nine studies. Different methods of reporting prevented these data from being combined in a meta‐analysis.

Subgroup analyses for CMV disease

Subgroup analyses according to antibody status, antiviral medications, organ transplanted, treatment duration, use of antilymphocyte therapy, time to outcome assessment, study quality and other aspects of study design did not demonstrate any differences in treatment effects. Multivariate meta‐regression showed no difference in CMV disease with high certainty evidence after allowing for potential confounding or effect‐modification by prophylactic drug used, organ transplanted or recipient serostatus in CMV‐positive recipients and CMV‐negative recipients of CMV‐positive donors (See Table 7).

1. Potential sources of variability: CMV disease and all‐cause death.

Variable	CMV disease			All‐cause death
	Studies	RR (95% CI)	P value for interaction	Studies	RR (95% CI)	P value for interaction
Antiviral medication
Aciclovir Ganciclovir Valaciclovir	6 11 2	0.45 (0.29 to 0.69) 0.44 (0.34 to 0.58) 0.30 (0.19 to 0.49)	0.43	5 10 2	0.67 (0.38 to 1.20) 0.69 (0.29 to 1.65) 0.50 (0.22 to 1.15)	0.85
Time to outcome assessment
3 to 6 months 9 to 12 months	11 8	0.46 (0.36 to 0.58) 0.36 (0.22 to 0.58)	0.37	7 10	0.63 (0.40 to 0.97) 0.64 (0.31 to 1.33)	0.83
Recipient CMV status
Positive (donor +ve or ‐ve)a Negative (donor +ve)b	13 10	0.34 (0.24 to 0.50) 0.52 (0.37 to 0.74)	0.12	7 4	0.59 (0.30 to 1.18) 1.42 (0.44 to 4.66)	0.23
Donor CMV statusc
Positive (recipients all +ve) Negative (recipients all +ve)	5 5	0.18 (0.09 to 0.36) 0.33 (0.11 to 0.95)	0.37	No data No data	No data No data	No data
Organ transplanted
Kidney Liver Heart	11 5 3	0.42 (0.31 to 0.57) 0.49 (0.29 to 0.84) 0.39 (0.25 to 0.63)	0.93	10 4 3	0.49 (0.24 to 1.00) 0.64 (0.39 to 1.00) 1.82 (0.39 to 8.51)	0.13
Antibody therapy
Yes No	11 6	0.43 (0.33 to 0.55) 0.47 (0.29 to 0.76)	0.74	10 5	0.81 (0.33 to 2.01) 0.63 (0.39 to 1.00)	0.93
Treatment durationd
6 weeks or less More than 6 weeks	7 4	0.49 (0.36 to 0.68) 0.33 (0.21 to 0.53)	0.72	6 4	0.91 (0.17 to 4.92) 0.62 (0.30 to 1.30)	0.15
Allocation concealment
Adequate Unclear or inadequate	4 15	0.50 (0.31 to 0.79) 0.41 (0.33 to 0.51)	0.64	3 14	0.26 (0.06 to 1.20) 0.67 (0.45 to 0.99)	0.88
Blinding
Yes No	5 14	0.35 (0.25 to 0.48) 0.47 (0.37 to 0.59)	0.18	5 12	0.62 (0.39 to 0.98) 0.65 (0.33 to 1.27)	0.97
Intention to treat
Yes No	10 9	0.38 (0.30 to 0.48) 0.47 (0.33 to 0.68)	0.37	9 8	0.62 (0.40 to 0.98) 0.65 (0.32 to 1.29)	0.57

^aStudies in "positive" group included those in which recipients were positive for CMV with donor positive or negative for CMV
^bStudies in "negative" group included those in which CMV‐negative recipients received CMV‐positive organs
^cStudies in which recipients were CMV positive and the donors CMV positive (positive group) or negative (CMV negative group)
^dGanciclovir studies only

CMV disease in patients stratified by antibody status

Subgroup analysis revealed that treatment efficacy in CMV disease did not vary according to recipient serostatus.

Antiviral medication reduced the risk of CMV disease (Analysis 1.2.1 (13 studies, 1348 participants): RR 0.34, 95% CI 0.24 to 0.50; I² = 24%; high certainty evidence) in CMV‐positive recipients (donor positive or negative).

1.2. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 2: All symptomatic CMV disease stratified by antibody status

Antiviral medication reduced the risk of CMV disease (Analysis 1.2.2 (10 studies, 423 participants): RR 0.52, 95% CI 0.37 to 0.73; I² = 27%; high certainty evidence) in CMV‐negative recipients of CMV‐positive organs.

It is uncertain whether prophylaxis reduces CMV disease in CMV‐negative recipients of CMV‐negative donors because the certainty of this evidence is very low (Analysis 1.2.3 (4 studies, 38 participants): 1 event in each group).

Subgroup analysis showed with high certainty evidence that treatment efficacy did not vary in CMV‐positive recipients if they received a CMV‐positive organ (Analysis 1.2.4 (5 studies, 276 participants): RR 0.19, 95% CI 0.09 to 0.37; I² = 0%) and with moderate certainty (downgraded for imprecision) evidence probably for CMV‐negative organ (Analysis 1.2.5 (5 studies, 160 participants): RR 0.32, 95% CI 0.11 to 0.95; I² = 0%).

CMV disease in all patients stratified by antiviral medication

Subgroup analysis showed with high certainty evidence that the treatment efficacy did not vary according to antiviral medication. When analysed separately aciclovir (Analysis 1.3.1 (6 studies, 421 participants): RR 0.45, 95% CI 0.29 to 0.69; I² = 8%), ganciclovir (Analysis 1.3.2 (11 studies, 917 participants): RR 0.44, 95% CI 0.34 to 0.58; I² = 23%) and valaciclovir (Analysis 1.3.3 (2 studies, 643 participants): RR 0.30, 95% CI 0.19 to 0.49; I² = 0%) reduced the risk for CMV disease compared with placebo or no treatment.

1.3. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 3: CMV disease in all patients by antiviral medication

CMV disease in all patients stratified by transplanted organ

The treatment efficacy on CMV disease did not vary according to organ transplanted. Prophylaxis reduced the risk of CMV disease in kidney transplant recipients (Analysis 1.4.1 (11 studies, 1132 participants): RR 0.42, 95% CI 0.31 to 0.57; I² = 27%; high certainty evidence). Similarly, prophylaxis probably reduces the risk of CMV disease in liver (Analysis 1.4.2 (5 studies, 616 participants): RR 0.49, 95% CI 0.29 to 0.84; I² = 57%; moderate certainty evidence (downgraded for inconsistency)) and heart transplant recipients (Analysis 1.4.3 (3 studies, 232 participants): RR 0.39, 95% CI 0.25 to 0.63; I² = 0%; moderate certainty evidence (downgraded for imprecision)).

1.4. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 4: CMV disease for different organ transplants

CMV disease in ganciclovir‐treated patients stratified by treatment duration

In ganciclovir studies, the duration of treatment was arbitrarily divided into less than six weeks and six weeks or more. There was no difference in treatment efficacy (Analysis 1.5) (Test for subgroup differences: Chi² = 1.99, df = 1 (P = 0.16), I² = 49.7%). The effect of duration could not be assessed for other medications, which were generally administered for three months.

1.5. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 5: CMV disease and ganciclovir duration

CMV disease in patients stratified for the use of antilymphocyte antibody

Subgroup analysis with high certainty evidence showed no difference in treatment efficacy against CMV disease if the immunosuppressive regimen did (Analysis 1.6.1 (11 studies, 666 participants): RR 0.43, 95% CI 0.33 to 0.55; I² = 0%) or did not (Analysis 1.7.1 (6 studies, 649 participants): RR 0.47, 95% CI 0.29 to 0.76; I² = 47%) include an antilymphocyte antibody given during prophylaxis for induction or rejection.

1.6. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 6: CMV disease and ATG therapy and antiviral efficacy

1.7. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 7: CMV disease and immunosuppression without ATG induction and antiviral efficacy

All‐cause death

The average all‐cause death rate reported at one year or less post‐transplant in the placebo or no treatment arms of all studies was 7.1% (range 0% to 37%). Prophylaxis reduced all‐cause death (Analysis 1.8.1 (7 studies, 1838 participants): RR 0.63, 95% CI 0.43 to 0.92; I² = 0%; high certainty evidence).

1.8. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 8: All‐cause death according to antiviral medication

Figure 6 shows the cumulative meta‐analyses demonstrating change over time for all‐cause death. While the relative risk remained stable, the CIs narrowed progressively, with evidence for a significant reduction in all‐cause death becoming evident with the addition of Lowance 1999 Kidney.

6.

All‐cause death cumulative meta‐analysis showing change over time

Subgroup analyses for all‐cause death

Subgroup analyses according to CMV status, antiviral medications, organ transplanted, treatment duration, use of antilymphocyte therapy, time to outcome assessment, study quality and other aspects of study design did not demonstrate any differences in all‐cause death. Multivariate meta‐regression showed no difference in all‐cause death after allowing for potential confounding or effect‐modification by prophylactic drug used, organ transplanted or recipient serostatus in CMV‐positive recipients and CMV‐negative recipients of CMV‐positive donors (See Table 7).

All‐cause death stratified by CMV status

Antiviral prophylaxis probably makes little or no difference to all‐cause death in CMV‐positive recipients (Analysis 1.9.1 (7 studies, 738 participants): RR 0.59, 95% CI 0.30 to 1.18; I² = 2%; moderate certainty evidence (downgraded for imprecision)) or CMV‐negative recipients of CMV‐positive organs (Analysis 1.9.2 (4 studies, 288 participants): RR 1.42, 95% CI 0.44 to 4.66; I² = 0%; low certainty evidence (downgraded twice for imprecision)). Data were not available to determine if the effects of antiviral medications on all‐cause death differed between CMV‐positive recipients of CMV‐negative and CMV‐positive recipients of CMV‐positive organs.

1.9. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 9: All‐cause death according to CMV status

All‐cause death stratified by transplanted organ

All‐cause death was reduced (Analysis 1.10 (17 studies, 1838 participants): RR 0.63, 95% CI 0.43 to 0.92; I² = 0%; high certainty evidence). However, the reduction could not be demonstrated for individual organs because of the small number of events and patients for individual organs.

1.10. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 10: All‐cause death for different organ transplants

All‐cause death in ganciclovir‐treated patients stratified by treatment duration

There may be little or no difference in all‐cause death among studies evaluating ganciclovir for six weeks or less or more than six weeks (Analysis 1.11) (Test for subgroup differences: Chi² = 0.17, df = 1 (P = 0.68), I² = 0%).

1.11. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 11: All‐cause death and ganciclovir duration

All‐cause death in studies stratified according to the use of antilymphocyte therapy

There may be no difference in all‐cause death whether or not antibody therapy was administered (Analysis 1.12.1; Analysis 1.12.2) (for induction) (Test for subgroup differences: Chi² = 0.26, df = 1 (P = 0.61), I² = 0%).

1.12. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 12: All‐cause death with or without ATG therapy and antiviral efficacy

CMV infection

The average risk of CMV infection in the placebo or no treatment arms of all studies was 49% (range 36% to 100%). Prophylaxis probably reduces CMV infection (Analysis 1.13 (17 studies, 1786 participants): RR 0.61, 95% CI 0.48 to 0.77; I² = 76%; moderate certainty evidence (downgraded for inconsistency)). Considerable heterogeneity existed between studies for CMV infection with no explanation apparent, but the summary estimates for individual studies favoured prophylaxis in 15/17 studies.

1.13. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 13: CMV infection

Death due to CMV disease

The average death rates in the placebo or no treatment arms due to CMV disease and non‐CMV causes were 2.3% (range 0.3% to 7.4%) and 5.7% (0% to 15.6%), respectively. Prophylaxis reduced the risk of death due to CMV disease (Analysis 1.14.1 (7 studies, 1300 participants): RR 0.26, 95% CI 0.08 to 0.78; I² = 0%; high certainty evidence) but not the risk from non‐CMV causes (Analysis 1.14.2 (7 studies, 1300 participants): RR 0.71, 95% CI 0.44 to 1.17; I² = 0%; moderate certainty evidence (downgraded for imprecision)).

1.14. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 14: Death due to CMV disease or other causes

Additional outcomes

For graft loss, acute rejection, invasive fungal infection and post‐transplant lymphoproliferative disease (PTLD), there was low certainty evidence showing little to no differences between antiviral prophylaxis and placebo or no treatment (Analysis 1.15.1; Analysis 1.15.2; Analysis 1.15.4; Analysis 1.15.6).

1.15. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 15: Additional outcomes: all medications

Prophylaxis with aciclovir, ganciclovir or valaciclovir reduced the risk for clinical disease caused by herpes simplex and herpes zoster (Analysis 1.15.3 (9 studies, 1483 participants): RR 0.27, 95% CI 0.19 to 0.40; I² = 27%; high certainty evidence). Combining the studies of different medications showed that bacterial (Analysis 1.15.5 (3 studies, 175 participants): RR 0.65, 95% CI 0.44 to 0.96; I² = 0%; moderate certainty evidence (downgraded for imprecision)) and protozoal infections (Analysis 1.15.7 (2 studies, 114 participants): RR 0.31, 95% CI 0.10 to 0.99; I² = 0%; moderate certainty evidence (downgraded for imprecision)) were probably reduced by prophylaxis.

The risk of acute rejection may make little or no difference between studies using biopsy diagnosis (Analysis 1.16.1 (5 studies, 827 participants): RR 0.97, 95% CI 0.71 to 1.32; I² = 62%; low certainty evidence (downgraded for imprecision and inconsistency)) and those using clinical criteria (Analysis 1.16.2 (8 studies, 599 participants): RR 0.91, 95% CI 0.76 to 1.08; I² = 14%; moderate quality evidence (downgraded for imprecision)) (Test for subgroup differences: Chi² = 0.13, df = 1 (P = 0.71), I² = 0%). In one study using valaciclovir with subgroups pre‐specified according to CMV serostatus, prophylaxis reduced the risk of acute rejection in CMV‐negative recipients of CMV‐positive kidneys (Lowance 1999 Kidney) (Analysis 1.17.1 (208 participants): RR 0.51, 95% CI 0.35 to 0.74) compared with CMV‐positive recipients (Analysis 1.17.2 (408 participants): RR 0.84, 95% CI 0.63 to 1.10) (test of interaction Chi² = 4.33; P = 0.04). This difference is responsible for the heterogeneity demonstrated between valaciclovir studies for acute rejection (Analysis 1.17.3 (2 studies, 643 participants): RR 0.81, 95% CI 0.55 to 1.19; I² = 85%; moderate certainty evidence (downgraded for imprecision)).

1.16. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 16: Acute rejection according to method of diagnosis

1.17. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 17: Valaciclovir: additional outcomes

There were 16 studies that reported data on the adverse effects of medications. Except for six placebo‐controlled studies, we could not determine baseline adjusted effects of medications on leucopenia, kidney function and neurological dysfunction as the numbers of patients with these abnormalities were not reported for the no treatment groups. In one placebo‐controlled study, valaciclovir probably increased the risk for hallucinations (8.5% compared with 0.97%) (Analysis 1.18.9 (616 participants): RR 8.78, 95% CI 2.69 to 28.71; low certainty evidence (downgraded twice for imprecision)). There was low certainty evidence for any difference in neurological dysfunction with aciclovir (Analysis 1.18.3). No differences were identified for leucopenia (Analysis 1.18.1; Analysis 1.18.4; Analysis 1.18.7) or reduced kidney function (Analysis 1.18.2; Analysis 1.18.5; Analysis 1.18.8) with any medication (See Table 8).

1.18. Analysis.

Comparison 1: Antiviral prophylaxis versus placebo or no treatment, Outcome 18: Adverse effects

2. Summary of outcomes for antiviral medication versus placebo/no treatment.

Outcome	Aciclovir Studies; RR (95% CI)	Ganciclovir Studies; RR (95% CI)	Valaciclovir Studies; RR (95% CI)	All medications Studies; RR (95% CI)
Acute rejection	4; 1.03 (0.78 to 1.36)	7; 0.92 (0.70 to 1.21)	2; 0.81 (0.51 to 1.28)a	13; 0.90 (0.78 to 1.17)
Graft loss	4; 0.77 (0.35 to 1.68)	6; 0.73 (0.41 to 1.28)	No data	10; 0.74 (0.47 to 1.17)
Herpes simplex or zoster infections	3; 0.30 (0.14 to 0.62)	4; 0.25 (0.08 to 0.78)	2; 0.28 (0.20 to 0.40)	9; 0.27 (0.19 to 0.40)
Post‐transplant lymphoproliferative disease	1; 2.90 (0.12 to 68.2)	1; 0.34 (0.01 to 8.33)	No data	2; 1.01 (0.11 to 9.51)
Bacterial infections	1; 0.67 (0.33 to 1.38)	1; 0.72 (0.44 to 1.17)	1; 0.27 (0.07 to 1.05)	3; 0.65 (0.44 to 0.96)
Fungal infections	1; 1.30 (0.31 to 5.39)	2; 0.28 (0.07 to 1.12)	No data	3; 0.58 (0.19 to 1.73)
Protozoal infections	No data	2; 0.31 (0.01 to 0.99)	No data	2; 0.31 (0.01 to 0.99)
Leucopeniaª	No data	3; 0.99 (0.37 to 2.65)	1; 1.05 (0.62 to 1.78)	‐
Creatinine > 200 µmol/Lb	2; 1.14 (0.27 to 4.70)	3; 2.36 (0.91 to 6.15)	No data	‐
Hallucinationsb	1; 10.6 (0.62 to 183.3)	1; 1.59 (0.98 to 2.58)	1; 8.78 (2.69 to 28.7)	‐

^aHeterogeneity of study results present

^bPlacebo‐controlled RCTs only

Subgroup analyses by methodological quality for CMV disease and all‐cause death

Subgroup analysis, stratifying studies by methodological quality and aspects of study design specified a priori, showed that treatment efficacy to reduce CMV disease and all‐cause death did not change among studies.

• Study quality: Studies were divided according to quality assessment (adequate allocation concealment or other, blinding or no blinding, intention‐to‐treat (ITT) analysis carried out or not). On subgroup analysis, no differences in treatment efficacy for CMV disease or all‐cause death could be detected for allocation concealment (Analysis 2.1; Analysis 3.1), blinding (Analysis 2.2; Analysis 3.2), or ITT analysis (Analysis 2.3; Analysis 3.3).

• Time of outcome assessment: There may be no difference in treatment efficacy for CMV disease and all‐cause death if the outcome was assessed at three to six months or nine to 12 months (Analysis 2.4; Analysis 3.4).

• Study publication date: Studies were arbitrarily divided into those published before 1997 and those published in or after 1997. No difference in treatment efficacy could be demonstrated (Analysis 2.5; Analysis 3.5).

2.1. Analysis.

Comparison 2: Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo or no treatment, Outcome 1: Allocation concealment

3.1. Analysis.

Comparison 3: Effect of methodological quality on all‐cause death in studies of prophylaxis versus placebo or no treatment, Outcome 1: Allocation concealment

2.2. Analysis.

Comparison 2: Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo or no treatment, Outcome 2: Blinding of participants/investigators

3.2. Analysis.

Comparison 3: Effect of methodological quality on all‐cause death in studies of prophylaxis versus placebo or no treatment, Outcome 2: Blinding of participants and investigators

2.3. Analysis.

Comparison 2: Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo or no treatment, Outcome 3: Intention‐to‐treat analysis (ITT)

3.3. Analysis.

Comparison 3: Effect of methodological quality on all‐cause death in studies of prophylaxis versus placebo or no treatment, Outcome 3: Intention‐to‐treat analysis (ITT)

2.4. Analysis.

Comparison 2: Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo or no treatment, Outcome 4: CMV disease by time of outcome assessment

3.4. Analysis.

Comparison 3: Effect of methodological quality on all‐cause death in studies of prophylaxis versus placebo or no treatment, Outcome 4: All‐cause death and time of outcome assessment or trial publication date

2.5. Analysis.

Comparison 2: Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo or no treatment, Outcome 5: CMV disease by trial publication date

3.5. Analysis.

Comparison 3: Effect of methodological quality on all‐cause death in studies of prophylaxis versus placebo or no treatment, Outcome 5: All‐cause death and trial publication date

Publication bias

Publication bias has been assessed for seven outcomes (Analysis 1.1; Analysis 1.2; Analysis 1.3; Analysis 1.4; Analysis 1.6; Analysis 1.10; Analysis 1.13), and no strong conclusion can be drawn (figures not shown).

Ganciclovir versus aciclovir

Eight studies compared ganciclovir with aciclovir (Badley 1997 Liver; Duncan 1993 Lung; Flechner 1998 Kidney; Martin 1994 Liver; Nakazato 1993 Liver; Rubin 2002 All; Winston 1995 Liver; Winston 2003 Liver).

See Table 2.

CMV disease

In head‐to‐head studies, there was high certainty evidence that ganciclovir was more effective than aciclovir in preventing CMV disease (Analysis 4.1.1 (7 studies, 1113 participants) RR 0.37, 95% CI 0.23 to 0.60; I² = 33%) and CMV invasive organ involvement (Analysis 4.1.2 (7 studies (1034 participants): RR 0.28, 95% CI 0.15 to 0.49; I² = 0%), and may reduce CMV syndrome (Analysis 4.1.3 (6 studies, 1009 participants): RR 0.40, 95% CI 0.16 to 1.02. I² = 55%; low certainty evidence (downgraded for imprecision and inconsistency) in all recipients. This was consistent in CMV‐positive recipients (Analysis 4.2.1 (5 studies, 722 participants): RR 0.27, 95% CI 0.13 to 0.55; I² = 7%; high certainty evidence) and in CMV‐negative recipients of CMV‐positive organs (Analysis 4.3.1 (5 studies, 246 participants): RR 0.64, 95% CI 0.41 to 0.99; I² = 0%; high certainty evidence). There was insufficient data in CMV‐negative recipients of CMV‐negative donors to determine if a difference in efficacy exists (Analysis 4.4).

4.1. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 1: CMV disease in all treated patients

4.2. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 2: CMV disease by antibody +ve recipients

4.3. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 3: CMV disease by +ve donors / CMV ‐ve recipients

4.4. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 4: CMV ‐ve donor / CMV ‐ve recipient

On subgroup analysis, differences in efficacy could be demonstrated between studies in which the participants received ganciclovir for three months (Analysis 4.1.5 (4 studies, 703 participants): RR 0.28, 95% CI 0.09 to 0.82; I² = 62%; low certainty evidence (downgraded for imprecision and inconsistency)) and probably in those in which the participants received ganciclovir followed by aciclovir (Analysis 4.1.6 (3 studies, 410 participants): RR 0.38, 95% CI 0.22 to 0.64; I² = 0%; moderate certainty evidence (downgraded for imprecision)). Subgroup analysis demonstrated the efficacy of antiviral medication was not dependent on the organ transplanted for CMV disease (Analysis 4.5.1; Analysis 4.5.2; Analysis 4.5.3).

4.5. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 5: CMV disease and effect of prophylaxis for different transplanted organs

Death

There was probably little or no difference in the risk of death due to CMV disease (Analysis 4.6.1 (6 studies, 832 participants): RR 0.33, 95% CI 0.07 to 1.58; I² = 0%; moderate certainty evidence (downgraded for imprecision)) or all‐cause death (Analysis 4.6.2 (8 studies, 1138 participants): RR 1.13, 95% CI 0.82 to 1.58; I² = 0%; moderate certainty evidence (downgraded for imprecision)).

4.6. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 6: Death

CMV infection

Ganciclovir was probably more effective than aciclovir in reducing CMV infection (Analysis 4.7 (6 studies, 815 participants): RR 0.44, 95% CI 0.28 to 0.67; I² = 73%; moderate certainty evidence (downgraded for inconsistency)) in all recipients and probably in CMV‐positive recipients (Analysis 4.8 (5 studies, 522 participants): RR 0.30, 95% CI 0.16 to 0.58; I² = 70%; moderate certainty evidence (downgraded for inconsistency)) but may make little or no difference in CMV‐negative recipients of CMV‐positive organs (Analysis 4.9 (4 studies, 228 participants): RR 0.63, 95% CI 0.36 to 1.09; I² = 58%; low certainty evidence (downgraded for inconsistency and imprecision)) but there was significant heterogeneity among the studies.

4.7. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 7: CMV infection

4.8. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 8: CMV infection by antibody +ve recipients

4.9. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 9: CMV infection by +ve donors / CMV ‐ve recipients

Ganciclovir was probably more effective than aciclovir in reducing CMV infection in kidney (Analysis 4.10.1 (2 studies, 168 participants): RR 0.20, 0.04 to 0.95; I² = 59%; low certainty evidence (downgraded for inconsistency and imprecision)) and liver transplant recipients (Analysis 4.10.2 (4 studies, 472 participants): RR 0.42, 95% CI 0.25 to 0.73; I² = 75%; low certainty evidence (downgraded for inconsistency and imprecision); but may make little or no difference in heart or lung transplant recipients (Analysis 4.10.3 (2 studies, 75 participants): RR 0.88, 95% CI 0.50 to 1.55; I² = 57%; low certainty evidence (downgraded for inconsistency and imprecision)).

4.10. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 10: CMV infection and effect of prophylaxis for different transplanted organs

Additional outcomes

With moderate certainty evidence, there are probably little to no differences found for acute rejection (Analysis 4.11.1), graft loss (Analysis 4.11.2), other viral infections (Analysis 4.11.3), fungal infections (Analysis 4.11.4), bacterial infections (Analysis 4.11.5), protozoal infections (Analysis 4.11.6), obliterative bronchiolitis in lung transplant recipients (Analysis 4.11.7), kidney dysfunction (Analysis 4.11.9) or neurological dysfunction (Analysis 4.11.10). A maximum of three studies provided outcomes for graft loss, obliterative bronchiolitis, opportunistic infections other than other viral infections and neurological dysfunction.

4.11. Analysis.

Comparison 4: Ganciclovir versus aciclovir, Outcome 11: Additional outcomes

Leucopenia may be more common with ganciclovir compared with aciclovir (Analysis 4.11.8 (6 studies, 955 participants): RR 3.28, 95% CI 1.48 to 7.25; I² = 0%); low certainty evidence (downgraded twice for imprecision).

Ganciclovir then aciclovir versus ganciclovir

Green 1997 Liver compared ganciclovir given for 14 days followed by aciclovir for one year with ganciclovir for 14 days in 48 children who had received liver transplants. Little to no differences in efficacy were demonstrated for CMV disease (Analysis 5.1), all‐cause death (Analysis 5.2), CMV infection (Analysis 5.3) or Epstein‐Barr virus infections (Analysis 5.4) (all low certainty evidence).

5.1. Analysis.

Comparison 5: Ganciclovir then aciclovir versus ganciclovir, Outcome 1: CMV disease in all treated patients

5.2. Analysis.

Comparison 5: Ganciclovir then aciclovir versus ganciclovir, Outcome 2: Death

5.3. Analysis.

Comparison 5: Ganciclovir then aciclovir versus ganciclovir, Outcome 3: CMV infection in all treated patients

5.4. Analysis.

Comparison 5: Ganciclovir then aciclovir versus ganciclovir, Outcome 4: Additional outcomes

Valganciclovir versus ganciclovir

Paya 2004 All compared valganciclovir with ganciclovir in CMV‐negative recipients of CMV‐positive organs and included patients receiving kidney, liver, heart and combined kidney‐pancreas transplants.

CMV disease

Compared to ganciclovir, valganciclovir probably made little to no difference in preventing CMV disease at six months (Analysis 6.1.1) or one‐year post‐transplant (Analysis 6.1.2) (low certainty evidence). Similarly, there was probably little to no difference at six months and one year in the prevention of CMV syndrome (Analysis 6.1.3; Analysis 6.1.4) and CMV invasive organ disease (Analysis 6.1.5; Analysis 6.1.6) (low certainty evidence). Subgroup analysis showed that, at six months, valganciclovir was probably more effective than ganciclovir in kidney transplant recipients (Analysis 6.1.8; 120 participants; RR 0.27, 95% CI 0.01 to 0.75) compared with liver, heart or kidney‐pancreas transplant recipients (Analysis 6.1.7; Analysis 6.1.9; Analysis 6.1.10) (test of interaction Chi² = 6.34; P = 0.01).

6.1. Analysis.

Comparison 6: Valganciclovir versus ganciclovir, Outcome 1: CMV disease in CMV donor +ve / recipient ‐ve

All‐cause death

There may be little or no difference in death due to CMV disease between valganciclovir and ganciclovir (Analysis 6.2.1) (low certainty evidence) and all‐cause death (Analysis 6.2.2) (moderate certainty evidence).

6.2. Analysis.

Comparison 6: Valganciclovir versus ganciclovir, Outcome 2: Death

CMV infection

There may be little or no difference in the prevention of CMV disease at six months and one year between valganciclovir and ganciclovir (Analysis 6.3).

6.3. Analysis.

Comparison 6: Valganciclovir versus ganciclovir, Outcome 3: CMV infection in CMV donor +ve / recipient ‐ve

Additional outcomes

There were probably little or no differences in acute rejection, graft loss and opportunistic infections (Analysis 6.4.1; Analysis 6.4.2; Analysis 6.4.3) (moderate certainty evidence). Neutrophil counts < 1000/mm³ occurred in 13% of patients treated with valganciclovir compared with 8% treated with ganciclovir, but the difference was based on low certainty evidence (Analysis 6.4.7). Similarly, there were probably little to no differences in cessation of medications due to neutropenia, anaemia, thrombocytopenia or tremor (Analysis 6.4.4; Analysis 6.4.5; Analysis 6.4.6; Analysis 6.4.7; Analysis 6.4.8) (moderate certainty evidence).

6.4. Analysis.

Comparison 6: Valganciclovir versus ganciclovir, Outcome 4: Additional outcomes

Valaciclovir versus ganciclovir or valganciclovir

Four studies compared valaciclovir with ganciclovir (Pavlopoulou 2005 Kidney; Reischig 2005 Kidney) or valganciclovir (2VAL 2010 Kidney; Ali Ibrahim 2020 Kidney) in kidney transplant recipients.

See Table 3.

CMV disease

There was probably little or no difference in the risk of CMV disease (Analysis 7.1.1) with valaciclovir compared with ganciclovir or valganciclovir prophylaxis with low certainty evidence. There were probably little or no differences in the risk of CMV disease (Analysis 7.1.2) in CMV‐positive recipients of CMV‐positive or negative transplants on the risk of CMV disease (Analysis 7.1.3) in CMV‐negative recipients of CMV‐positive organs with moderate certainty evidence.

7.1. Analysis.

Comparison 7: Valaciclovir versus ganciclovir or valganciclovir, Outcome 1: CMV disease in all treated patients

All‐cause death

There was probably little or no difference in the risk of all‐cause death; however, the certainty of evidence for this outcome was low (Analysis 7.2.1).

7.2. Analysis.

Comparison 7: Valaciclovir versus ganciclovir or valganciclovir, Outcome 2: Death

CMV infection

There was probably little or no difference in the risk of CMV infection with valaciclovir compared with ganciclovir or valganciclovir prophylaxis with moderate certainty evidence (Analysis 7.3.1). There were probably little or no differences in the risk of CMV infection in CMV‐positive recipients of CMV‐positive or negative transplants (Analysis 7.3.2) or in CMV‐negative recipients of CMV‐positive organs (Analysis 7.3.3) (moderate certainty evidence).

7.3. Analysis.

Comparison 7: Valaciclovir versus ganciclovir or valganciclovir, Outcome 3: CMV infection

Additional outcomes

The risk of acute rejection may not change with valaciclovir compared with ganciclovir; however, the evidence was very uncertain (Analysis 7.4.1 (3 studies, 271 participants): RR 0.88, 95% CI 0.31 to 2.47; I² = 72%; very low certainty evidence (downgraded for inconsistency and imprecision). However, there was heterogeneity among the three studies, with Reischig 2005 Kidney reporting a significantly reduced risk for acute rejection with valaciclovir (seen in participants with delayed graft function), while the other studies showed no differences between valaciclovir compared with ganciclovir. There was probably little to no difference in the risk of graft loss based on moderate certainty evidence (Analysis 7.4.2).

There were probably little to no differences in the risk of leucopenia, thrombocytopenia, anaemia, neurological dysfunction, the need to reduce or cease study medications, other herpes infection, polyomavirus‐associated nephropathy or neutropenia with moderate certainty evidence (Analysis 7.4.3; Analysis 7.4.4; Analysis 7.4.5; Analysis 7.4.6; Analysis 7.4.7; Analysis 7.4.8; Analysis 7.4.11; Analysis 7.4.12). Pavlopoulou 2005 Kidney reported non‐viral infections were probably increased in patients treated with valaciclovir (Analysis 7.4.9 (83 participants): RR 0.59, 95% CI 0.44 to 0.80) due to the increase in urinary tract infections in that group. 2VAL 2010 Kidney reported polyoma viremia increased in patients treated with valganciclovir (Analysis 7.4.10 (119 participants): RR 0.48, 95% CI 0.25 to 0.94) (moderate certainty evidence).

7.4. Analysis.

Comparison 7: Valaciclovir versus ganciclovir or valganciclovir, Outcome 4: Additional outcomes

Kidney function

There was probably little or no difference in kidney function (SCr and calculated GFR) at the end of the study with valaciclovir compared with ganciclovir or valganciclovir based on moderate certainty evidence (Analysis 7.5.1; Analysis 7.5.2) (moderate certainty evidence).

7.5. Analysis.

Comparison 7: Valaciclovir versus ganciclovir or valganciclovir, Outcome 5: Kidney function at end of study

Different regimens of ganciclovir

Hertz 1998 Heart/lung compared daily with three times/week IV ganciclovir in heart‐lung transplant recipients. Winston 2004 Liver and Nafar 2005 Kidney compared oral with IV ganciclovir.

Daily versus three times/week ganciclovir

There were probably little or no differences in CMV disease, CMV syndrome, CMV invasive tissue disease or CMV infection, all‐cause death, and deaths due to CMV disease bacteraemia, bronchiolitis obliterans or leucopenia between the two treatment regimens (Analysis 8.1) (moderate and low certainty evidence).

8.1. Analysis.

Comparison 8: Different ganciclovir regimens, Outcome 1: IV doses given at different frequencies

Oral versus IV ganciclovir

There are probably little to no differences in CMV disease, CMV syndrome, CMV invasive tissue disease or CMV infection, all‐cause death, acute rejection, graft loss, leucopenia and the need to cease medications due to leucopenia (Analysis 8.2) (moderate and low certainty evidence).

8.2. Analysis.

Comparison 8: Different ganciclovir regimens, Outcome 2: Oral versus IV ganciclovir

Extended versus short‐duration valganciclovir

Two studies compared extended durations (definition of extended differed across the studies) of valganciclovir. One study compared 200 days with 100 days in kidney transplant recipients (IMPACT 2010 Kidney), and the other study compared one year with three months in lung transplant recipients (Palmer 2010 Lung). Data included in meta‐analyses from Palmer 2010 Lung were taken from percentages reported in the study as the authors were not able to provide the original data.

See Table 4.

CMV disease

There was high certainty evidence that the risk of CMV disease was reduced at the end of treatment with extended‐duration valbanciclovir (Analysis 9.1.1 (2 studies, 454 participants): RR 0.20, 95% CI 0.12 to 0.35; I² = 0%), at nine months (Analysis 9.1.2 (1 study, 318 participants): RR 0.39, 95% CI 0.25 to 0.60), 12 months (Analysis 9.1.3 (1 study, 318 participants): RR 0.44, 95% CI 0.29 to 0.66) and 24 months (Analysis 9.1.4 (1 study, 318 participants): RR 0.55, 95% CI 0.38 to 0.79).

9.1. Analysis.

Comparison 9: Extended versus short‐duration valganciclovir, Outcome 1: CMV disease

The number of patients with CMV syndrome (Analysis 9.2 (2 studies, 454 participants): RR 0.27, 95% CI 0.10 to 0.71; I² = 12%; high certainty evidence) was also reduced.

9.2. Analysis.

Comparison 9: Extended versus short‐duration valganciclovir, Outcome 2: CMV syndrome

There was little or no difference in the risk for CMV invasive disease at 12 months (Analysis 9.3.1 (2 studies, 454 participants): RR 0.17, 95% CI 0.03 to 1.34; I² = 44%; moderate certainty evidence (downgraded for imprecision)). There were few episodes of CMV invasive disease in kidney transplant recipients, and the numbers probably did not differ at 24 months (Analysis 9.3.2; moderate certainty evidence).

9.3. Analysis.

Comparison 9: Extended versus short‐duration valganciclovir, Outcome 3: CMV invasive disease

All‐cause death

IMPACT 2010 Kidney reported there was probably little to no difference in all‐cause death at either 12 months (Analysis 9.4.1) or 24 months (Analysis 9.4.2) (moderate certainty evidence).

9.4. Analysis.

Comparison 9: Extended versus short‐duration valganciclovir, Outcome 4: All‐cause death

CMV infection

The risk of CMV infection was probably reduced with extended‐duration treatment at the end of treatment (Analysis 9.5.1 (2 studies, 454 participants): RR 0.27, 95% CI 0.10 to 0.71; I² = 82%), at nine months (Analysis 9.5.2 (1 study, 318 participants): RR 0.27, 95% CI 0.10 to 0.71) and at 12 months (Analysis 9.5.3 (1 study, 318 participants): RR 0.73, 95% CI 0.57 to 0.95) (moderate certainty evidence).

9.5. Analysis.

Comparison 9: Extended versus short‐duration valganciclovir, Outcome 5: CMV infection

Other outcomes

There were probably little or no differences in graft loss at 12 months (Analysis 9.6.1; moderate certainty evidence) and 24 months (Analysis 9.6.2; moderate certainty evidence), biopsy‐proven acute rejection at < 100 days (Analysis 9.7.1; moderate certainty evidence), 12 months (Analysis 9.7.2; high certainty evidence) and 24 months (Analysis 9.7.3; high certainty evidence), and in post‐transplant diabetes mellitus (Analysis 9.8.2; moderate certainty evidence).

9.6. Analysis.

Comparison 9: Extended versus short‐duration valganciclovir, Outcome 6: Graft loss

9.7. Analysis.

Comparison 9: Extended versus short‐duration valganciclovir, Outcome 7: Acute rejection

9.8. Analysis.

Comparison 9: Extended versus short‐duration valganciclovir, Outcome 8: Other outcomes

There was considerable heterogeneity (I² = 87%) in the analysis of opportunistic infections (Analysis 9.8.1 (2 studies, 454 participants; RR 0.71, 95% CI 0.33 to 1.57; low certainty evidence). IMPACT 2010 Kidney reported that opportunistic infections were less common among patients treated with extended duration valganciclovir (RR 0.48, 95% CI 0.30 to 0.77), while Palmer 2010 Lung reported no difference (RR 1.02, 95% CI 0.75 to 1.40) (low certainty evidence).

Adverse effects

There was little or no difference between treatment groups for total treatment‐related adverse effects (Analysis 9.9.1) and serious treatment‐related adverse effects (Analysis 9.9.2), based on moderate certainty evidence. Leucopenia was probably more common (Analysis 9.9.3 (1 study, 320 participants): RD 0.12, 95% CI 0.01 to 0.22) and more likely to result in treatment termination (Analysis 9.9.4 (1 study, 320 participants): RD 0.04, 95% CI 0.00 to 0.07) in patients treated for 200 days compared with those treated for 100 days in the IMPACT 2010 Kidney. Termination for any treatment‐related adverse effect did not differ in Palmer 2010 Lung (Analysis 9.9.5). While there was little or no difference in the number of hospitalisations for all adverse effects (Analysis 9.9.7) among treatment groups, there were fewer hospitalisations for CMV disease in patients treated for 200 days (Analysis 9.9.6 (1 study, 418 total hospitalisations): RD ‐0.10, 95% CI ‐0.17 to ‐0.04) in IMPACT 2010 Kidney. There was probably no reduction or increase in CMV mutations, which confer ganciclovir resistance, in participants with positive viral load who were treated for an extended duration compared with those treated for 100 days or three months (Analysis 9.9.8) (moderate certainty evidence).

9.9. Analysis.

Comparison 9: Extended versus short‐duration valganciclovir, Outcome 9: Adverse effects

Low versus standard dose valganciclovir

Two studies compared different valganciclovir regimens: Halim 2016 Kidney compared valganciclovir 450 mg/day with valganciclovir 900 mg/day for CMV prophylaxis in kidney transplant patients for six months, and Prabakaran 2020 Kidney compared low‐dose valganciclovir prophylaxis (900 mg, 3 times/week) versus standard prophylaxis dosing (900 mg/day) for three months in kidney transplant recipients.

See Table 5.

There may be little or no difference in the risks of all‐cause death (Analysis 10.1), CMV infection (Analysis 10.2), acute rejection (Analysis 10.3), or graft loss (Analysis 10.4) (all low certainty evidence).

10.1. Analysis.

Comparison 10: Different valganciclovir regimens, Outcome 1: All‐cause death

10.2. Analysis.

Comparison 10: Different valganciclovir regimens, Outcome 2: CMV infection

10.3. Analysis.

Comparison 10: Different valganciclovir regimens, Outcome 3: Acute rejection

10.4. Analysis.

Comparison 10: Different valganciclovir regimens, Outcome 4: Graft loss

Maribavir versus ganciclovir

Winston 2012 Liver compared 100 mg of maribavir twice/day with 1000 mg oral ganciclovir three times/day for CMV prophylaxis in liver transplant patients for 14 weeks. As maribavir lacks in vitro activity against herpes simplex virus and varicella‐zoster viral, 400 mg of oral acyclovir was given twice/day concomitantly with maribavir.

See Table 6.

There may be little or no differences in efficacy demonstrated for CMV disease (Analysis 11.1), all‐cause death (Analysis 11.2), acute rejection (Analysis 11.4) and adverse events (Analysis 11.5). The event rate of CMV infections may be higher in patients who received maribavir versus ganciclovir (Analysis 11.3 (233 participants): RR 1.34, 95% CI 1.10 to 1.65) (low to moderate certainty evidence).

11.1. Analysis.

Comparison 11: Maribavir versus ganciclovir, Outcome 1: CMV disease

11.2. Analysis.

Comparison 11: Maribavir versus ganciclovir, Outcome 2: All‐cause death

11.4. Analysis.

Comparison 11: Maribavir versus ganciclovir, Outcome 4: Acute rejection

11.5. Analysis.

Comparison 11: Maribavir versus ganciclovir, Outcome 5: Adverse events

11.3. Analysis.

Comparison 11: Maribavir versus ganciclovir, Outcome 3: CMV infection

Discussion

Summary of main results

Antiviral agents compared with placebo or no specific treatment

This systematic review found that the antiviral agents ganciclovir, valaciclovir and aciclovir improve outcomes for solid organ transplant recipients far beyond the primary indication for use. These data support the continued routine use of antiviral prophylaxis in CMV‐positive recipients and in CMV‐negative recipients of CMV‐positive organ transplants. In addition to reducing the risk of CMV disease by 60%, these agents reduced all‐cause death by 40%, predominantly due to reduced death from CMV disease, as well as reducing clinical disease caused by herpes simplex and herpes zoster (70%), bacterial infections (35%), and protozoal infections (70%). The relative benefits of aciclovir, ganciclovir and valaciclovir in relation to CMV disease and death appeared to be consistent among recipients of heart, kidney and liver transplants. These benefits occurred in both CMV‐positive recipients and CMV‐negative recipients of CMV‐positive organs, irrespective of whether immunosuppression included antilymphocyte antibody therapy, and were not dependent on the time of outcome assessment. Although there were no placebo‐controlled RCTs of valganciclovir, Paya 2004 All compared valganciclovir (the prodrug of ganciclovir) and ganciclovir and demonstrated no differences in the risk for CMV disease, all‐cause death and other outcomes, indicating that outcomes demonstrated in this systematic review in placebo or no treatment studies can be extrapolated to valganciclovir.

There was no clear reduction in graft loss or acute rejection, although a small but clinically important benefit has not been excluded. The summary RR for both outcomes favours antiviral agents, but the 95% CIs were relatively wide and consistent, with there being no effect. The exception was in a predefined subgroup in a single study (Lowance 1999 Kidney) in which CMV prophylaxis reduced the risk for biopsy‐proven acute rejection in CMV‐negative recipients of CMV‐positive kidney transplants by 50%.

Based on data from a single large study (Lowance 1999 Kidney), valaciclovir increased the risk for hallucinations. There was no increase in adverse effects with aciclovir or ganciclovir, although the 95% CIs were wide. Very few studies adequately reported harms, so differences in adverse effects between medication and placebo could not be excluded. It is possible that other differences in side effect profiles exist between agents but have not been demonstrated.

Relative efficacy of antiviral medications

Having demonstrated that antiviral agents as a drug class reduced all‐cause death and CMV disease, we then sought to determine which antiviral regimen was the most beneficial. Indirect comparisons demonstrated no difference between antiviral agents administered. In head‐to‐head studies, ganciclovir was more effective than aciclovir in preventing CMV disease, demonstrating the importance of assessing the comparative effects of drugs in direct comparison studies. This difference may be explained by differences in the duration of therapy in the indirect studies. Aciclovir was administered for 84 days or more, but ganciclovir was given for shorter durations (9 to 42 days) in seven of the 11 included ganciclovir studies. Hence, agent and duration were evaluated rather than the agent alone, as in direct comparison studies.

One large study (Paya 2004 All) demonstrated no difference in efficacy between ganciclovir and its prodrug, valganciclovir. Although three small studies demonstrated no difference in efficacy to prevent CMV disease between ganciclovir or valganciclovir and valaciclovir (2VAL 2010 Kidney; Ali Ibrahim 2020 Kidney; Pavlopoulou 2005 Kidney; Reischig 2005 Kidney), the wide CIs of the summary estimate urging moderate certainty evidence. Based on existing study data, aciclovir appears to be inferior to ganciclovir, and no clear superiority has been demonstrated between ganciclovir and valganciclovir or between valaciclovir and ganciclovir/valganciclovir.

Maribavir (100 mg) was safe but not adequate for the prevention of CMV disease in liver transplant recipients at high risk for CMV disease (Winston 2012 Liver). There was no difference between the ganciclovir and maribavir group for the incidence of CMV disease at six months, but the incidence of CMV disease within the period of 100 days after transplantation was lower in the ganciclovir group. A higher dose of maribavir (≥ 400 mg twice/day) was used in a phase 2 trial (Papanicolaou 2018), in which maribavir was active against refractory or resistant CMV infections in transplant recipients. Further studies with higher doses of miribavir may be indicated to see its effect as CMV prophylaxis.

Prophylaxis with low dose valganciclovir

Six months of low‐dose valganciclovir prophylaxis compared to high‐dose valganciclovir for moderate‐risk kidney transplant recipients and high‐dose valganciclovir showed similar effects with probably a better safety profile (Halim 2016 Kidney). However, further studies are needed to confirm this effect.

Prophylaxis with extended durations of valganciclovir

Extended prophylaxis with valganciclovir resulted in reductions in the risks of CMV disease, CMV infection and opportunistic infections with moderate certainty evidence but no differences in other outcomes (acute rejection, all‐cause death, graft loss) with low certainty evidence. Leucopenia was more common with extended duration of prophylaxis, but hospitalisations due to CMV disease were reduced.

Overall completeness and applicability of evidence

Antiviral agents compared with placebo or no specific treatment

Our major findings that CMV antiviral prophylaxis prevents CMV disease and all‐cause death, irrespective of the organ transplanted and CMV serostatus, are strengthened by two features of the data: the consistency of these findings across all studies and the finding that almost all eligible studies reported both major outcomes of interest (lack of outcome reporting bias). We identified 19 eligible studies, and the summary estimate favours antiviral medication for the outcome prevention of CMV disease in 18 studies. Similarly, 17 studies contributed data to the all‐cause death outcome. With fewer events, the play of chance would be expected to be greater, but only two studies (Macdonald 1995 Heart; Merigan 1992 Heart) had point estimates suggesting increased death from CMV prophylaxis. Unlike the outcome of CMV disease, no individual study demonstrated a reduction in all‐cause death with antiviral medication. This was evident only from the meta‐analytic estimate. The overall I² was 12.6% for CMV disease and 0% for all‐cause death, demonstrating very low heterogeneity beyond chance despite the clear differences in patient groups (Characteristics of included studies). Supporting this, as shown in Table 7, no pre‐defined potential source of variability for the effects of antiviral medication was significant, including standard quality items for study conduct and reporting, such as allocation concealment, blinding and intention‐to‐treat. We cannot exclude with any certainty a difference in the magnitude of the effect of antiviral medication in solid organ transplant recipients. However, any difference is likely to be clinically unimportant since data from 19 studies and about 2000 patients were insufficient to demonstrate any difference. In addition, the remarkable consistency in results across all studies suggests any undetected difference would be in magnitude and not direction of effect.

The data were relatively sparse in four areas, and further research is still needed. For the outcome of all‐cause death in heart transplant recipients, there are few relevant studies (2), patients (205) and events (4), making the effects of antiviral medications on heart transplant recipients very uncertain. Both studies had higher death rates in the active arms, but the 95% CIs were very wide, results were consistent with other patient groups (liver and kidney), and the likely pathway for benefit ‐ reduction in CMV disease ‐ is evident in this patient group.

Second, there was very scant data for the CMV‐negative donor to a CMV‐negative recipient group, even though this group is frequently given antiviral agents to prevent CMV disease (Baliga 2004). These patients are almost exclusively not enrolled in studies because of low event rates and low burden of disease. However, there are no studies examining the efficacy of antiviral agents to prevent de novo CMV disease in such CMV‐negative patients.

Third, our conclusions on the other benefits of antiviral medications and the adverse effects of these drugs (Table 8) must be considered more cautiously for reasons of the imprecision of summary estimates and that many eligible studies did not report these outcomes. Therefore, these results may be biased. The direction of bias cannot be determined without obtaining additional data, e.g. individual patient data, from the authors regarding these outcomes. At the same time, based on the included studies, it remains difficult to draw conclusions on drug resistance and dosing strategy.

Fourth, only one study specifically addressed children (Green 1997 Liver). This is despite the fact that children commonly receive prophylaxis with antiviral agents since they are at a high risk of CMV disease. Many are CMV‐negative and receive organs from CMV‐positive donors. Information on the efficacy of prophylaxis with antiviral agents from RCTs of adult transplant recipients has been extrapolated to children. Non‐randomised studies suggest valganciclovir is effective and tolerated in children (Camacho‐Gonzalez 2011).

Relative efficacy of antiviral medications

The data demonstrated that ganciclovir was superior to aciclovir in preventing CMV disease, and aciclovir is no longer used for prophylaxis. A single large study indicated no differences between oral ganciclovir and oral valganciclovir. Clinical practice data from this study have been extrapolated to indicate that oral valganciclovir can substitute for oral ganciclovir, and valganciclovir is now generally the preferred agent for prophylaxis. Oral ganciclovir is no longer marketed.

The data demonstrated little differences between valaciclovir compared with ganciclovir or valganciclovir in terms of CMV disease or infection, all‐cause death, or adverse events based on four studies (2VAL 2010 Kidney; Ali Ibrahim 2020 Kidney; Pavlopoulou 2005 Kidney; Reischig 2005 Kidney). The available studies comparing valganciclovir/ganciclovir with valaciclovir have only enrolled kidney transplant recipients, and it is unclear whether the data can be extrapolated to other transplanted organs.

Prophylaxis with extended durations of valganciclovir

Two studies, one in kidney (318 CMV‐positive recipient/CMV‐negative donor participants) and one in lung transplant recipients (136 CMV‐positive donor/CMV‐negative recipient and donor CMV‐positive or negative donor/CMV‐positive recipient participants), have demonstrated that extended durations of prophylaxis with valganciclovir resulted in a lower risk of CMV disease and infection. Neither study identified an increase in CMV mutations resistant to therapy, but study numbers were likely to be too small to demonstrate any difference. Both studies reported few cases of CMV disease occurring after the end of the extended period of prophylaxis. Further data are required to demonstrate whether the benefits of extended prophylaxis in other organ transplants justify the increased costs and adverse effects.

Quality of the evidence

This review now contains 41 studies. Most studies, including those recently published, did not provide sufficient information to determine whether sequence generation and allocation concealment were at a low risk of bias (Figure 2; Figure 3). It is a matter of concern that there was no blinding of participants, investigators and outcome assessors in almost 75% of studies. The primary outcome of CMV syndrome is a clinical diagnosis supported by a laboratory diagnosis of CMV infection and other information. Therefore, it is possible that CMV syndrome was misdiagnosed in some participants. Studies that lack adequate allocation concealment and blinding may overestimate treatment effects (Moher 1998; Schultz 1995).

The overall certainty of the evidence for studies comparing antiviral medications with placebo or no specific treatment was considered high for some outcomes (CMV disease, all‐cause death, acute rejection, CMV disease in kidney transplant recipients). It was considered moderate for death due to CMV disease, CMV disease in liver or heart transplants and graft loss because of the limited numbers of studies reporting these outcomes (Table 1).

The overall certainty of the evidence for studies comparing ganciclovir and aciclovir was considered high for CMV disease in all patients and for acute rejection. It was considered moderate for all‐cause death, death due to CMV disease and other viral infections, and low for other fungal infections and graft loss because of the limited number of events in the studies in which these outcomes were reported (Table 2).

The overall certainty of the evidence for studies comparing ganciclovir/valganciclovir with aciclovir/valaciclovir was considered low because of the small number of studies with few participants (Table 3).

The overall certainty of the evidence for studies comparing extended duration with three months of therapy was considered high for CMV disease, CMV syndrome, CMV infection and total adverse reactions. It was considered low for invasive CMV disease, acute rejection and opportunistic infections because of the heterogeneity between studies (Table 4).

Potential biases in the review process

Approximately half the studies did not report all important outcomes, so there is a risk of selection bias. In particular, there was limited data on death due to CMV disease, graft loss and other infections.

Agreements and disagreements with other studies or reviews

The results of this review update confirm and expand the findings of five previous systematic reviews (Couchoud 1998a and Couchoud 1998b; Fiddian 2002; Gourishankar 2001; Kalil 2005; Raval 2020), which included 12, 10, 9, 11, and 12 studies, respectively, which compared antiviral medications with placebo or no treatment for preventing of CMV disease. All found that prophylaxis reduced the risk of CMV disease in solid organ transplant recipients. One review (Couchoud 1998a; Couchoud 1998b) found no effect on death (10 studies: RR 0.69, 95% CI 0.41 to 1.18) and a second (Fiddian 2002), which included two studies using immunoglobulin and antiviral agents found that prophylaxis with aciclovir or valaciclovir significantly reduced all‐cause death (1321 patients: OR 0.60, 95% CI 0.40 to 0.90). Similarly, Kalil 2005, including 11 studies, found that prophylaxis with antiviral medications compared with placebo or no specific treatment significantly reduced CMV disease, all‐cause death and opportunistic infections with similar degrees of benefit to those found in our review, although inclusion criteria differed between these two reviews. Eight studies of prophylaxis included in our review were excluded from the analyses of universal prophylaxis in the review by Kalil 2005. The two reviews differed in that our review showed no significant reduction of acute rejection with antiviral prophylaxis, but Kalil 2005 identified a significant reduction in acute rejection with treatment (OR 0.72, 95% CI 0.57 to 0.91) using a fixed‐effect model for the analysis. However, there was some heterogeneity in the analyses of acute rejection in both reviews. Further analyses using a random‐effects model identified that both reviews found no significant differences in the risk of acute rejection between antiviral therapy and placebo or no specific treatment. Both reviews found a significant reduction in acute rejection using a fixed‐effect model. In a more recent systematic review (Raval 2020), the efficacy, safety, and costs of prophylactic CMV treatment options are compared with both no prophylactic treatment and pre‐emptive treatment. Prophylactic treatment showed a reduction in CMV infection incidence irrespective of heterogeneity in the dose, duration, or route of administration of treatment. No differences between treatment arms were found for acute rejection and graft loss, which is similar to the results of our review.

Our systematic review differs from previous reviews in that comparisons of different antiviral medications were included so that conclusions on the comparative effects of agents can be made. In addition, our review included a detailed exploration of potential heterogeneity. The finding of a reduction in all‐cause death is largely explained by a reduced death due to CMV disease, although a reduction in death due to other causes cannot be totally excluded. The latter is biologically plausible because CMV disease leads to an increase in other opportunistic infections in heart and liver transplant recipients (George 1997; Valentine 1999). This is suggestive of a mechanism whereby the prevention of CMV disease may prevent other infective complications that contribute to overall death.

Both prophylaxis and pre‐emptive therapy significantly reduce CMV disease compared with placebo or no specific therapy in solid organ transplant recipients. However, the available evidence base for the prevention of CMV disease with prophylaxis compared with placebo or no specific therapy (19 studies, 1981 participants) is large and of high certainty (GRADE) compared with the low‐certainty data (6 studies, 288 participants) supporting pre‐emptive therapy (Owers 2013). Further studies are required to determine the relative efficacies, adverse effects and costs of pre‐emptive therapy and prophylaxis because currently available data (7 studies, 753 participants), while showing no differences in efficacy though a lower risk of leucopenia with pre‐emptive therapy, demonstrated considerable heterogeneity among studies thus limiting the applicability of these data to patient management (Owers 2013)

Authors' conclusions

Implications for practice.

This systematic review has shown that prophylaxis of CMV‐positive recipients and CMV‐negative recipients of CMV‐positive organs with antiviral medications given for three months post solid organ transplantation reduces the risk of CMV disease and all‐cause death and may reduce the risk of other opportunistic infections. What are the implications of this study for clinical practice? Previous treatment guidelines (Jassal 1998; Van der Bij 2001) recommended CMV prophylaxis for all recipients of solid organ transplants who received immunosuppression with antilymphocyte antibody products and for CMV‐negative recipients of CMV‐positive organs. In liver and heart transplant recipients, prophylaxis was also recommended for all CMV‐positive recipients of solid organ transplants because of the higher risk for CMV disease. Our data suggested that these recommendations for use were too narrow because the benefits for patient survival and the constant relative benefits for preventing CMV disease, irrespective of CMV serostatus, had not been recognised previously.

Recent guidelines recommend that all kidney transplant recipients except donor‐negative/recipient‐negative recipients should receive antiviral prophylaxis for at least three months post‐transplant (KDIGO 2009). Similarly, guidelines from the AST Infectious Diseases Community of Practice (Humar 2009) recommend antiviral prophylaxis for both CMV‐positive recipients and for CMV‐negative recipients of CMV‐positive donors of any solid organ transplant. Consensus guidelines from the Infectious Disease Section of the Transplantation Society (Razonable 2019) recommended antiviral prophylaxis for CMV‐negative recipients of CMV‐positive donor organs. These guidelines considered that either prophylaxis or pre‐emptive therapies could be used in CMV‐positive recipients but noted the lack of data on pre‐emptive therapy in subpopulations, including lung and small bowel transplants. If a pre‐emptive approach is non‐inferior compared with universal prophylaxis, the former might be preferred by clinicians and patients, yet further studies are required to establish the safety and efficacy for all subgroups of patients.

The absolute effects of antiviral medications on the prevention of CMV disease and all‐cause death are shown quantitatively in groups of patients at different baseline risks for these outcomes (Table 9). The primary determinants for CMV disease are the organ transplanted and serostatus, whereas organ transplanted is the most important determinant for all‐cause death. Table 9 shows that benefit exceeds harm for all but the lowest risk groups, assuming equal importance of the outcomes. However, given that the clinical importance of all‐cause death and CMV disease are significantly greater than the adverse effects of medications, most patients and clinicians, when provided with this information, are likely to use CMV prophylaxis with antiviral medications across all risk categories, except in the seronegative donor and recipient groups for whom there are few data. The decision to give CMV prophylaxis for all risk categories could be favoured by one study (Halim 2016 Kidney) added in this update, which showed comparable effectiveness of low‐dose valganciclovir prophylaxis as high‐dose valganciclovir for six months, for moderate risk kidney transplant recipients (CMV‐positive donor/CMV‐positive recipient or CMV‐negative donor/CMV‐positive recipient) with a better safety profile.

Two RCTs (IMPACT 2010 Kidney; Palmer 2010 Lung) have now demonstrated that extended‐duration prophylaxis with valganciclovir in CMV‐positive donor/CMV‐negative recipients of kidney and lung transplants and CMV‐positive recipients of lung transplants reduces the risk of CMV disease compared with three months of therapy, suggesting that extended‐duration prophylaxis should be considered in patients at higher risk of CMV disease (Humar 2009).

3. Effects of antiviral medication on CMV disease and all‐cause death.

Recipient group	Without prophylaxisa	With prophylaxisb	Number prevented	Number with harmsc
CMV disease
Kidneyd Kidneyd; livere; heartd Liver, heartd; alle, antibody therapy included in immunosuppressive regimen	7/100 28/100 59/100	3/100 12/100 25/100	4/100 16/100 39/100	7/100 7/100 7/100
All‐cause death
Kidney Liver Heart or lung	6/100 20/100 24/100	4/100 13/100 15/100	2/100 7/100 9/100	7/100 7/100 7/100

^aData from references
^bCalculated from summary estimates of RR (0.42 for prevention of CMV disease, 0.63 for all‐cause death)
^cBased on proportion of patients, treated with valaciclovir, who developed hallucinations
^dDonor positive or negative for CMV; recipient negative
^eDonor positive recipient negative for CMV

Implications for research.

There are no data from RCTs on the efficacy of prophylaxis compared with placebo in lung transplants and few data for heart transplants. However, such studies are no longer ethical based on the demonstration of efficacy in other organ transplants. Future studies may be required in the seronegative donor‐recipient group depending on the prevalence of CMV disease in this group with newer and more potent immunosuppressive regimens. Further studies are required to determine the optimum duration and dosage of medications in different organ transplants. Currently, valganciclovir is most commonly used for prophylaxis. It remains possible that smaller doses than currently recommended may be effective for prophylaxis, as demonstrated for IV ganciclovir (Hertz 1998 Heart/lung) and valganciclovir (Halim 2016 Kidney) 450 mg/day versus valganciclovir 900 mg/day for CMV prophylaxis in kidney transplant patients for six months.

Further studies are required to evaluate the comparative effects, including harms, of antiviral medications in clinical use at present or in the future. More information is also required on the efficacy of prophylaxis with different immunosuppressive regimens used for the prevention and treatment of rejection in different organ transplants.

Overall, prophylaxis did not significantly reduce the risk of acute rejection or graft loss. Further information is required to determine whether prophylaxis can reduce the risk for rejection in particular groups of patients, whether it affects the number or severity of rejection episodes, and whether it reduces graft loss at time periods beyond one year.

Adequately powered and well‐designed RCTs are required to determine the relative efficacies, adverse effects and costs of universal prophylaxis compared to pre‐emptive therapies, particularly in transplant populations at lower risk of CMV disease. There is one ongoing study identified that investigates prophylactic treatments (NCT04225923). Consequently, we will monitor and update the search for this review regularly; however, we do not expect that the findings of this review will change on a short‐term basis.

What's new

Date	Event	Description
3 May 2024	New citation required and conclusions have changed	New studies included
3 May 2024	New search has been performed	GRADE quality of evidence assessment included

History

Protocol first published: Issue 3, 2002
Review first published: Issue 4, 2005

Date	Event	Description
3 January 2013	New citation required and conclusions have changed	Risk of bias assessment incorporated
3 January 2013	New search has been performed	New studies included
18 March 2010	Amended	Contact details updated.
13 May 2009	Amended	Contact details updated.
13 August 2008	Amended	Converted to new review format.
7 January 2008	New citation required and conclusions have changed	Substantive amendment, 6 additional publications identified, 2 new studies included
16 October 2004	Amended	Title changed. Background, methods edited to reflect limitation of review to prophylaxis with antiviral medication. Quality assessment criteria added.

Appendices

Appendix 1. Electronic search strategies

Database	Search terms
CENTRAL	MeSH descriptor Cytomegalovirus, this term only in MeSH products MeSH descriptor Cytomegalovirus Infections explode all trees in MeSH products MeSH descriptor Cytomegalovirus Vaccines explode all trees cytomegalovirus* in All Fields in CENTRAL cmv* in All Fields in CENTRAL (#1 OR #2 OR #3 OR #4 OR #5) (organ or renal or kidney or heart or lung or liver or pancreas) adj transplant in All Fields in all products MeSH descriptor Organ Transplantation, this term only MeSH descriptor Heart Transplantation explode all trees MeSH descriptor Lung Transplantation explode all trees MeSH descriptor Kidney Transplantation, this term only MeSH descriptor Liver Transplantation, this term only MeSH descriptor Pancreas Transplantation, this term only (#7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13) (#6 AND #14)
MEDLINE (OVID SP)	Cytomegalovirus/ exp Cytomegalovirus Infections/ Cytomegalovirus Vaccines/ cytomegalovirus.tw. cmv.tw. or/1‐5 Organ Transplantation/ exp Heart Transplantation/ exp Lung Transplantation/ Kidney Transplantation/ Liver Transplantation/ Pancreas Transplantation/ ((organ or renal or kidney or heart or lung or liver or pancreas) adj transplant$).tw. or/8‐13 6 and 14
EMBASE (OVID SP)	exp CYTOMEGALOVIRUS/ Cytomegalovirus Infection/ Cytomegalovirus Antibody/ Cytomegalovirus Vaccine/ cytomegalovirus.tw. CMV.tw. or/1‐6 exp organ transplantation/ ((organ or renal or kidney or heart or lung or liver or pancreas) adj transplant$).tw. or/8‐9 7 and 10

Appendix 2. Risk of bias assessment tool

Potential source of bias	Assessment criteria
Random sequence generation Selection bias (biased allocation to interventions) due to inadequate generation of a randomised sequence	Low risk of bias: Random number table; computer random number generator; coin tossing; shuffling cards or envelopes; throwing dice; drawing of lots; minimization (minimization may be implemented without a random element, and this is considered to be equivalent to being random).
	High risk of bias: Sequence generated by odd or even date of birth; date (or day) of admission; sequence generated by hospital or clinic record number; allocation by judgement of the clinician; by preference of the participant; based on the results of a laboratory test or a series of tests; by availability of the intervention.
	Unclear: Insufficient information about the sequence generation process to permit judgement.
Allocation concealment Selection bias (biased allocation to interventions) due to inadequate concealment of allocations prior to assignment	Low risk of bias: Randomisation method described that would not allow investigator/participant to know or influence intervention group before eligible participant entered in the study (e.g. central allocation, including telephone, web‐based, and pharmacy‐controlled, randomisation; sequentially numbered drug containers of identical appearance; sequentially numbered, opaque, sealed envelopes).
	High risk of bias: Using an open random allocation schedule (e.g. a list of random numbers); assignment envelopes were used without appropriate safeguards (e.g. if envelopes were unsealed or non‐opaque or not sequentially numbered); alternation or rotation; date of birth; case record number; any other explicitly unconcealed procedure.
	Unclear: Randomisation stated but no information on method used is available.
Blinding of participants and personnel Performance bias due to knowledge of the allocated interventions by participants and personnel during the study	Low risk of bias: No blinding or incomplete blinding, but the review authors judge that the outcome is not likely to be influenced by lack of blinding; blinding of participants and key study personnel ensured, and unlikely that the blinding could have been broken.
	High risk of bias: No blinding or incomplete blinding, and the outcome is likely to be influenced by lack of blinding; blinding of key study participants and personnel attempted, but likely that the blinding could have been broken, and the outcome is likely to be influenced by lack of blinding.
	Unclear: Insufficient information to permit judgement
Blinding of outcome assessment Detection bias due to knowledge of the allocated interventions by outcome assessors.	Low risk of bias: No blinding of outcome assessment, but the review authors judge that the outcome measurement is not likely to be influenced by lack of blinding; blinding of outcome assessment ensured, and unlikely that the blinding could have been broken.
	High risk of bias: No blinding of outcome assessment, and the outcome measurement is likely to be influenced by lack of blinding; blinding of outcome assessment, but likely that the blinding could have been broken, and the outcome measurement is likely to be influenced by lack of blinding.
	Unclear: Insufficient information to permit judgement
Incomplete outcome data Attrition bias due to amount, nature or handling of incomplete outcome data.	Low risk of bias: No missing outcome data; reasons for missing outcome data unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias); missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes not enough to have a clinically relevant impact on observed effect size; missing data have been imputed using appropriate methods.
	High risk of bias: Reason for missing outcome data likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk enough to induce clinically relevant bias in intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes enough to induce clinically relevant bias in observed effect size; ‘as‐treated’ analysis done with substantial departure of the intervention received from that assigned at randomisation; potentially inappropriate application of simple imputation.
	Unclear: Insufficient information to permit judgement
Selective reporting Reporting bias due to selective outcome reporting	Low risk of bias: The study protocol is available and all of the study’s pre‐specified (primary and secondary) outcomes that are of interest in the review have been reported in the pre‐specified way; the study protocol is not available but it is clear that the published reports include all expected outcomes, including those that were pre‐specified (convincing text of this nature may be uncommon).
	High risk of bias: Not all of the study’s pre‐specified primary outcomes have been reported; one or more primary outcomes is reported using measurements, analysis methods or subsets of the data (e.g. subscales) that were not pre‐specified; one or more reported primary outcomes were not pre‐specified (unless clear justification for their reporting is provided, such as an unexpected adverse effect); one or more outcomes of interest in the review are reported incompletely so that they cannot be entered in a meta‐analysis; the study report fails to include results for a key outcome that would be expected to have been reported for such a study.
	Unclear: Insufficient information to permit judgement
Other bias Bias due to problems not covered elsewhere in the table	Low risk of bias: The study appears to be free of other sources of bias.
	High risk of bias: Had a potential source of bias related to the specific study design used; stopped early due to some data‐dependent process (including a formal‐stopping rule); had extreme baseline imbalance; has been claimed to have been fraudulent; had some other problem.
	Unclear: Insufficient information to assess whether an important risk of bias exists; insufficient rationale or evidence that an identified problem will introduce bias.

Appendix 3. Selection procedure in previous updates

Included studies

In the original review published in 2005, 19 studies compared aciclovir (Balfour 1989 Kidney; Barkholt 1999 Liver; Gavalda 1997 Liver; Kletzmayr 1996 Kidney; Rostaing 1994 Kidney; Saliba 1993 Liver), ganciclovir (Ahsan 1997 Kidney; Brennan 1997 Kidney; Cohen 1993 Liver; Conti 1995 Kidney; Gane 1997 Liver; Hibberd 1995 Kidney; Leray 1995 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Pouteil‐Noble 1996 Kidney; Rondeau 1993 Kidney) or valaciclovir (Egan 2002 Heart; Lowance 1999 Kidney) with placebo or no treatment. Fifteen of these 19 studies excluded CMV‐negative recipients of CMV‐negative donors. Eleven studies compared different antiviral medications (Badley 1997 Liver, Duncan 1993 Lung, Flechner 1998 Kidney, Green 1997 Liver, Martin 1994 Liver; Nakazato 1993 Liver; Paya 2004 All; Reischig 2005 Kidney; Rubin 2002 All; Winston 2003 Liver; Winston 1995 Liver); and two studies (Hertz 1998 Heart/lung; Winston 2004 Liver) compared different regimens of ganciclovir administration. Recipients of transplants other than heart, kidney and liver were not included in studies comparing treatment with placebo or no treatment and were investigated in only three comparison studies. Green 1997 Liver specifically included children; the inclusion criteria for the Paya 2004 All and Rubin 2002 All studies indicated that children aged over 12 years could be included; however, the youngest participant in the Rubin 2002 All study was aged 20 years, and the average participant age in the Paya 2004 All study was 45 years. All identified studies were published in English language. The 2005 review included 32 studies (3737 participants) (Figure 1).

In the 2008 update, five additional publications were included. These were an abstract of an included study (Ahsan 1997 Kidney); one publication reported the full results of an included study, and an additional publication assessed one outcome from that study (Reischig 2005 Kidney); and two new studies (Nafar 2005 Kidney; Pavlopoulou 2005 Kidney). Pavlopoulou 2005 Kidney compared valaciclovir with ganciclovir and Nafar 2005 Kidney compared oral with IV ganciclovir. The 2008 update included 34 studies (3850 participants).

In the 2013 update, three additional studies were included (2VAL 2010 Kidney; IMPACT 2010 Kidney; Palmer 2010 Lung). 2VAL 2010 Kidney compared valaciclovir with valganciclovir, but only preliminary results at four months were available; IMPACT 2010 Kidney compared 200 days of oral valganciclovir with 100 days in kidney transplant recipients; and Palmer 2010 Lung compared 12 months of oral valganciclovir with three months in lung transplant recipients. The 2013 update included 37 studies (4342 participants).

Excluded studies

In the 2005 review, we excluded 47 studies after full text review: four were systematic reviews; 10 were narrative reviews; 12 involved ineligible interventions; and 21 were not RCTs.

In the 2008 update, one study was excluded because it compared pre‐emptive therapy with prophylaxis (Khoury 2006 Kidney).

In the 2013 update, 19 additional studies (34 reports) were excluded after reviewing abstracts: six were not RCTs and 13 studies involved an ineligible intervention. We excluded two studies after full text review: Pescovitz 2009 was a pharmacokinetic study and Said 2007 was a sequential study. We also identified four additional reports of three studies that had previously been excluded.

Data and analyses

Comparison 1. Antiviral prophylaxis versus placebo or no treatment.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 CMV disease	19		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.1.1 All symptomatic CMV disease	19	1981	Risk Ratio (M‐H, Random, 95% CI)	0.42 [0.34, 0.52]
1.1.2 CMV syndrome	11	1570	Risk Ratio (M‐H, Random, 95% CI)	0.41 [0.29, 0.57]
1.1.3 CMV organ involvement	12	1628	Risk Ratio (M‐H, Random, 95% CI)	0.34 [0.21, 0.55]
1.2 All symptomatic CMV disease stratified by antibody status	17		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.2.1 CMV antibody +ve recipients	13	1348	Risk Ratio (M‐H, Random, 95% CI)	0.34 [0.24, 0.50]
1.2.2 CMV +ve donor / CMV ‐ve recipient	10	423	Risk Ratio (M‐H, Random, 95% CI)	0.52 [0.37, 0.73]
1.2.3 CMV ‐ve donor / CMV ‐ve recipient	4	38	Risk Ratio (M‐H, Random, 95% CI)	1.00 [0.09, 11.03]
1.2.4 CMV +ve donor / CMV +ve recipient	5	276	Risk Ratio (M‐H, Random, 95% CI)	0.19 [0.09, 0.37]
1.2.5 CMV ‐ve donor / CMV +ve recipient	5	160	Risk Ratio (M‐H, Random, 95% CI)	0.32 [0.11, 0.95]
1.3 CMV disease in all patients by antiviral medication	19	1981	Risk Ratio (M‐H, Random, 95% CI)	0.42 [0.34, 0.52]
1.3.1 Aciclovir	6	421	Risk Ratio (M‐H, Random, 95% CI)	0.45 [0.29, 0.69]
1.3.2 Ganciclovir	11	917	Risk Ratio (M‐H, Random, 95% CI)	0.44 [0.34, 0.58]
1.3.3 Valaciclovir	2	643	Risk Ratio (M‐H, Random, 95% CI)	0.30 [0.19, 0.49]
1.4 CMV disease for different organ transplants	19	1980	Risk Ratio (M‐H, Random, 95% CI)	0.44 [0.35, 0.55]
1.4.1 Kidney transplant recipients	11	1132	Risk Ratio (M‐H, Random, 95% CI)	0.42 [0.31, 0.57]
1.4.2 Liver transplant recipients	5	616	Risk Ratio (M‐H, Random, 95% CI)	0.49 [0.29, 0.84]
1.4.3 Heart transplant recipients	3	232	Risk Ratio (M‐H, Random, 95% CI)	0.39 [0.25, 0.63]
1.5 CMV disease and ganciclovir duration	11		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.5.1 Six weeks or less	7	478	Risk Ratio (M‐H, Random, 95% CI)	0.49 [0.36, 0.68]
1.5.2 More than 6 weeks	4	439	Risk Ratio (M‐H, Random, 95% CI)	0.33 [0.21, 0.53]
1.6 CMV disease and ATG therapy and antiviral efficacy	11		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.6.1 CMV disease in all treated patients	11	666	Risk Ratio (M‐H, Random, 95% CI)	0.43 [0.33, 0.55]
1.7 CMV disease and immunosuppression without ATG induction and antiviral efficacy	6		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.7.1 CMV disease in all treated patients	6	649	Risk Ratio (M‐H, Random, 95% CI)	0.47 [0.29, 0.76]
1.8 All‐cause death according to antiviral medication	17	1838	Risk Ratio (M‐H, Random, 95% CI)	0.63 [0.43, 0.92]
1.8.1 Aciclovir	5	301	Risk Ratio (M‐H, Random, 95% CI)	0.67 [0.38, 1.20]
1.8.2 Ganciclovir	10	894	Risk Ratio (M‐H, Random, 95% CI)	0.69 [0.29, 1.65]
1.8.3 Valaciclovir	2	643	Risk Ratio (M‐H, Random, 95% CI)	0.50 [0.22, 1.15]
1.9 All‐cause death according to CMV status	9	1026	Risk Ratio (M‐H, Random, 95% CI)	0.74 [0.41, 1.32]
1.9.1 CMV +ve recipients	7	738	Risk Ratio (M‐H, Random, 95% CI)	0.59 [0.30, 1.18]
1.9.2 CMV ‐ve recipients of CMV +ve organs	4	288	Risk Ratio (M‐H, Random, 95% CI)	1.42 [0.44, 4.66]
1.10 All‐cause death for different organ transplants	17	1838	Risk Ratio (M‐H, Random, 95% CI)	0.63 [0.43, 0.92]
1.10.1 Kidney transplant recipients	10	1109	Risk Ratio (M‐H, Random, 95% CI)	0.49 [0.24, 1.00]
1.10.2 Liver transplant patients	4	497	Risk Ratio (M‐H, Random, 95% CI)	0.63 [0.39, 1.00]
1.10.3 Heart transplant recipients	3	232	Risk Ratio (M‐H, Random, 95% CI)	1.82 [0.39, 8.51]
1.11 All‐cause death and ganciclovir duration	10		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.11.1 Six weeks or less	6	455	Risk Ratio (M‐H, Random, 95% CI)	0.91 [0.17, 4.92]
1.11.2 More than 6 weeks	4	439	Risk Ratio (M‐H, Random, 95% CI)	0.62 [0.30, 1.30]
1.12 All‐cause death with or without ATG therapy and antiviral efficacy	15		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.12.2 With ATG therapy	10	643	Risk Ratio (M‐H, Random, 95% CI)	0.82 [0.33, 2.02]
1.12.3 Without ATG therapy	5	529	Risk Ratio (M‐H, Random, 95% CI)	0.63 [0.39, 1.00]
1.13 CMV infection	17		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.13.1 Total CMV infection	17	1786	Risk Ratio (M‐H, Random, 95% CI)	0.61 [0.48, 0.77]
1.14 Death due to CMV disease or other causes	7		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.14.1 CMV disease	7	1300	Risk Ratio (M‐H, Random, 95% CI)	0.26 [0.08, 0.78]
1.14.2 Other causes	7	1300	Risk Ratio (M‐H, Random, 95% CI)	0.71 [0.44, 1.17]
1.15 Additional outcomes: all medications	16		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.15.1 Graft loss	10	825	Risk Ratio (M‐H, Random, 95% CI)	0.74 [0.47, 1.17]
1.15.2 Acute rejection	13	1420	Risk Ratio (M‐H, Random, 95% CI)	0.90 [0.78, 1.05]
1.15.3 Herpes simplex and H. zoster infection	9	1483	Risk Ratio (M‐H, Random, 95% CI)	0.27 [0.19, 0.40]
1.15.4 Invasive fungal infection	3	189	Risk Ratio (M‐H, Random, 95% CI)	0.58 [0.19, 1.73]
1.15.5 Bacterial infection	3	174	Risk Ratio (M‐H, Random, 95% CI)	0.65 [0.44, 0.96]
1.15.6 EBV‐associated PTLD	2	359	Risk Ratio (M‐H, Random, 95% CI)	1.01 [0.11, 9.51]
1.15.7 Protozoal infections	2	114	Risk Ratio (M‐H, Random, 95% CI)	0.31 [0.10, 0.99]
1.16 Acute rejection according to method of diagnosis	13	1420	Risk Ratio (M‐H, Random, 95% CI)	0.90 [0.78, 1.05]
1.16.1 Biopsy‐proven acute rejection	5	821	Risk Ratio (M‐H, Random, 95% CI)	0.97 [0.71, 1.32]
1.16.2 Clinical diagnosis of acute rejection or method not stated	8	599	Risk Ratio (M‐H, Random, 95% CI)	0.91 [0.76, 1.08]
1.17 Valaciclovir: additional outcomes	2		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.17.1 Acute rejection in donor CMV +ve / recipient CMV ‐ve grafts	1	208	Risk Ratio (M‐H, Random, 95% CI)	0.51 [0.35, 0.74]
1.17.2 Acute rejection in CMV +ve recipients	1	408	Risk Ratio (M‐H, Random, 95% CI)	0.84 [0.63, 1.10]
1.17.3 Total with acute rejection	2	643	Risk Ratio (M‐H, Random, 95% CI)	0.81 [0.55, 1.19]
1.18 Adverse effects	6		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.18.1 Kidney dysfunction with aciclovir	2	159	Risk Ratio (M‐H, Random, 95% CI)	1.14 [0.27, 4.70]
1.18.2 Neurological dysfunction with aciclovir	1	55	Risk Ratio (M‐H, Random, 95% CI)	10.62 [0.62, 183.26]
1.18.3 Leucopenia with ganciclovir	3	509	Risk Ratio (M‐H, Random, 95% CI)	0.99 [0.37, 2.65]
1.18.4 Kidney dysfunction with ganciclovir	3	509	Risk Ratio (M‐H, Random, 95% CI)	2.36 [0.91, 6.15]
1.18.5 Neurological dysfunction with ganciclovir	3	509	Risk Ratio (M‐H, Random, 95% CI)	1.59 [0.98, 2.58]
1.18.6 Leucopenia with valaciclovir	1	616	Risk Ratio (M‐H, Random, 95% CI)	1.05 [0.62, 1.78]
1.18.7 Neurological dysfunction with valaciclovir	1	616	Risk Ratio (M‐H, Random, 95% CI)	8.78 [2.69, 28.71]
1.18.8 Leucopenia for all medications	4	1125	Risk Ratio (M‐H, Random, 95% CI)	1.03 [0.67, 1.59]
1.18.9 Kidney dysfunction for all medications	5	668	Risk Ratio (M‐H, Random, 95% CI)	1.81 [0.88, 3.73]

Comparison 2. Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo or no treatment.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Allocation concealment	19		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
2.1.1 Adequate	4	262	Risk Ratio (M‐H, Random, 95% CI)	0.50 [0.31, 0.79]
2.1.2 Inadequate/unclear	15	1719	Risk Ratio (M‐H, Random, 95% CI)	0.41 [0.33, 0.51]
2.2 Blinding of participants/investigators	19		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
2.2.1 Blinding	5	1135	Risk Ratio (M‐H, Random, 95% CI)	0.35 [0.25, 0.48]
2.2.2 No blinding	14	846	Risk Ratio (M‐H, Random, 95% CI)	0.47 [0.37, 0.59]
2.3 Intention‐to‐treat analysis (ITT)	19		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
2.3.1 ITT undertaken	10	1569	Risk Ratio (M‐H, Random, 95% CI)	0.38 [0.30, 0.48]
2.3.2 ITT not undertaken	9	412	Risk Ratio (M‐H, Random, 95% CI)	0.47 [0.33, 0.68]
2.4 CMV disease by time of outcome assessment	19		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
2.4.1 Outcome at 3‐6 months	11	704	Risk Ratio (M‐H, Random, 95% CI)	0.46 [0.36, 0.58]
2.4.2 Outcome at 9‐12 months	8	1277	Risk Ratio (M‐H, Random, 95% CI)	0.36 [0.22, 0.58]
2.5 CMV disease by trial publication date	19		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
2.5.1 Trials published before 1997	12	821	Risk Ratio (M‐H, Random, 95% CI)	0.48 [0.37, 0.63]
2.5.2 Trials published in 1997 and later	7	1160	Risk Ratio (M‐H, Random, 95% CI)	0.32 [0.24, 0.44]

Comparison 3. Effect of methodological quality on all‐cause death in studies of prophylaxis versus placebo or no treatment.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
3.1 Allocation concealment	17		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
3.1.1 Adequate	3	142	Risk Ratio (M‐H, Random, 95% CI)	0.26 [0.06, 1.20]
3.1.2 Inadequate/unclear	14	1695	Risk Ratio (M‐H, Random, 95% CI)	0.67 [0.45, 0.99]
3.2 Blinding of participants and investigators	17		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
3.2.1 Blinding	5	1135	Risk Ratio (M‐H, Random, 95% CI)	0.62 [0.39, 0.98]
3.2.2 No blinding	12	702	Risk Ratio (M‐H, Random, 95% CI)	0.65 [0.33, 1.27]
3.3 Intention‐to‐treat analysis (ITT)	17		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
3.3.1 ITT undertaken	9	1448	Risk Ratio (M‐H, Random, 95% CI)	0.62 [0.40, 0.98]
3.3.2 ITT not undertaken	8	389	Risk Ratio (M‐H, Random, 95% CI)	0.65 [0.32, 1.29]
3.4 All‐cause death and time of outcome assessment or trial publication date	17		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
3.4.1 Outcome at 4‐6 months	7	468	Risk Ratio (M‐H, Random, 95% CI)	0.64 [0.31, 1.33]
3.4.2 Outcome at 9‐12 months	10	1370	Risk Ratio (M‐H, Random, 95% CI)	0.63 [0.40, 0.97]
3.5 All‐cause death and trial publication date	17		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
3.5.1 Outcome in trials published before 1997	10	678	Risk Ratio (M‐H, Random, 95% CI)	0.71 [0.25, 2.08]
3.5.2 Outcome in trials published in 1997 or later	7	1160	Risk Ratio (M‐H, Random, 95% CI)	0.62 [0.41, 0.94]

Comparison 4. Ganciclovir versus aciclovir.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
4.1 CMV disease in all treated patients	8		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.1.1 CMV disease in all patients	7	1113	Risk Ratio (M‐H, Random, 95% CI)	0.37 [0.23, 0.60]
4.1.2 CMV organ involvement	7	1034	Risk Ratio (M‐H, Random, 95% CI)	0.28 [0.15, 0.49]
4.1.3 CMV syndrome	6	1009	Risk Ratio (M‐H, Random, 95% CI)	0.40 [0.16, 1.02]
4.1.5 CMV disease in patients treated with ganciclovir for 3 months	4	703	Risk Ratio (M‐H, Random, 95% CI)	0.28 [0.09, 0.82]
4.1.6 CMV disease in patients treated with ganciclovir for 2‐4 weeks then aciclovir	3	410	Risk Ratio (M‐H, Random, 95% CI)	0.38 [0.22, 0.64]
4.2 CMV disease by antibody +ve recipients	5		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.2.1 All symptomatic CMV disease	5	722	Risk Ratio (M‐H, Random, 95% CI)	0.27 [0.13, 0.55]
4.3 CMV disease by +ve donors / CMV ‐ve recipients	5		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.3.1 All symptomatic CMV disease	5	246	Risk Ratio (M‐H, Random, 95% CI)	0.64 [0.41, 0.99]
4.4 CMV ‐ve donor / CMV ‐ve recipient	3		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.4.1 CMV disease	3	41	Risk Ratio (M‐H, Random, 95% CI)	0.45 [0.07, 3.07]
4.5 CMV disease and effect of prophylaxis for different transplanted organs	7		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.5.1 CMV disease in kidney transplant patients	2	168	Risk Ratio (M‐H, Random, 95% CI)	0.30 [0.07, 1.35]
4.5.2 CMV disease in liver transplant patients	5	791	Risk Ratio (M‐H, Random, 95% CI)	0.37 [0.23, 0.59]
4.5.3 CMV disease in heart or lung transplant patients	2	75	Risk Ratio (M‐H, Random, 95% CI)	0.55 [0.10, 3.00]
4.6 Death	8		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.6.1 Death associated with CMV disease	6	832	Risk Ratio (M‐H, Random, 95% CI)	0.33 [0.07, 1.58]
4.6.2 All‐cause death	8	1138	Risk Ratio (M‐H, Random, 95% CI)	1.13 [0.82, 1.58]
4.7 CMV infection	6		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.7.1 CMV infection	6	815	Risk Ratio (M‐H, Random, 95% CI)	0.44 [0.28, 0.67]
4.8 CMV infection by antibody +ve recipients	5		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.8.1 CMV infection	5	522	Risk Ratio (M‐H, Random, 95% CI)	0.30 [0.16, 0.58]
4.9 CMV infection by +ve donors / CMV ‐ve recipients	4		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.9.1 CMV infection	4	228	Risk Ratio (M‐H, Random, 95% CI)	0.63 [0.36, 1.09]
4.10 CMV infection and effect of prophylaxis for different transplanted organs	6		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.10.1 CMV infection in kidney transplant patients	2	168	Risk Ratio (M‐H, Random, 95% CI)	0.20 [0.04, 0.95]
4.10.2 CMV infection in liver transplant patients	4	572	Risk Ratio (M‐H, Random, 95% CI)	0.42 [0.25, 0.73]
4.10.3 CMV infection in heart or lung transplant patients	2	75	Risk Ratio (M‐H, Random, 95% CI)	0.88 [0.50, 1.55]
4.11 Additional outcomes	8		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
4.11.1 Acute rejection	6	1009	Risk Ratio (M‐H, Random, 95% CI)	0.98 [0.87, 1.10]
4.11.2 Graft loss	3	268	Risk Ratio (M‐H, Random, 95% CI)	0.55 [0.27, 1.13]
4.11.3 Other viral infections	4	740	Risk Ratio (M‐H, Random, 95% CI)	0.81 [0.32, 2.01]
4.11.4 Invasive fungal infections	3	401	Risk Ratio (M‐H, Random, 95% CI)	0.67 [0.40, 1.10]
4.11.5 Bacterial infections	1	167	Risk Ratio (M‐H, Random, 95% CI)	1.10 [0.78, 1.53]
4.11.6 Protozoal infections	1	167	Risk Ratio (M‐H, Random, 95% CI)	0.34 [0.01, 8.16]
4.11.7 Obliterative bronchiolitis in lung transplant recipients	1	25	Risk Ratio (M‐H, Random, 95% CI)	0.81 [0.42, 1.54]
4.11.8 Leucopenia	6	955	Risk Ratio (M‐H, Random, 95% CI)	3.28 [1.48, 7.25]
4.11.9 Kidney dysfunction	4	661	Risk Ratio (M‐H, Random, 95% CI)	0.96 [0.83, 1.10]
4.11.10 Neurological dysfunction	2	306	Risk Ratio (M‐H, Random, 95% CI)	1.01 [0.24, 4.15]

Comparison 5. Ganciclovir then aciclovir versus ganciclovir.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
5.1 CMV disease in all treated patients	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
5.1.1 CMV disease	1	48	Risk Ratio (M‐H, Random, 95% CI)	3.50 [0.81, 15.16]
5.2 Death	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
5.2.1 All‐cause death	1	48	Risk Ratio (M‐H, Random, 95% CI)	5.00 [0.25, 98.96]
5.3 CMV infection in all treated patients	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
5.3.1 CMV infection	1	29	Risk Ratio (M‐H, Random, 95% CI)	2.85 [0.57, 14.36]
5.4 Additional outcomes	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
5.4.1 EBV infection	1	48	Risk Ratio (M‐H, Random, 95% CI)	1.60 [0.61, 4.19]

Comparison 6. Valganciclovir versus ganciclovir.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
6.1 CMV disease in CMV donor +ve / recipient ‐ve	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
6.1.1 CMV disease by 6 months	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.80 [0.47, 1.37]
6.1.2 CMV disease by 1 year	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.93 [0.59, 1.48]
6.1.3 CMV syndrome by 6 months	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.48 [0.23, 1.03]
6.1.4 CMV syndrome by 1 year	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.76 [0.39, 1.50]
6.1.5 Tissue invasive CMV disease by 6 months	1	364	Risk Ratio (M‐H, Random, 95% CI)	1.48 [0.60, 3.66]
6.1.6 Tissue invasive CMV disease by 1 year	1	364	Risk Ratio (M‐H, Random, 95% CI)	1.44 [0.66, 3.14]
6.1.7 CMV disease in liver transplant recipients by 6 months	1	177	Risk Ratio (M‐H, Random, 95% CI)	1.57 [0.71, 3.47]
6.1.8 CMV disease in renal transplant recipients by 6 months	1	120	Risk Ratio (M‐H, Random, 95% CI)	0.27 [0.10, 0.74]
6.1.9 CMV disease in heart transplant recipients by 6 months	1	56	Risk Ratio (M‐H, Random, 95% CI)	0.60 [0.09, 3.95]
6.1.10 CMV disease in renal‐pancreas transplant recipients by 6 months	1	11	Risk Ratio (M‐H, Random, 95% CI)	0.39 [0.02, 7.88]
6.1.11 CMV infection by 6 months	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.92 [0.71, 1.19]
6.1.12 CMV infection by 1 year	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.99 [0.80, 1.24]
6.2 Death	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
6.2.1 Death due to CMV disease	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.52 [0.03, 8.29]
6.2.2 All‐cause death	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.98 [0.43, 2.25]
6.3 CMV infection in CMV donor +ve / recipient ‐ve	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
6.3.1 CMV infection by 6 months	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.92 [0.71, 1.19]
6.3.2 CMV infection by 1 year	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.99 [0.80, 1.24]
6.4 Additional outcomes	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
6.4.1 Acute rejection in all recipients	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.91 [0.67, 1.22]
6.4.2 Graft loss	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.78 [0.13, 4.63]
6.4.3 Opportunistic infections	1	364	Risk Ratio (M‐H, Random, 95% CI)	0.86 [0.42, 1.76]
6.4.4 Neutrophil count < 1000/mm³	1	370	Risk Ratio (M‐H, Random, 95% CI)	1.60 [0.81, 3.16]
6.4.5 Medications ceased because of neutropenia	1	370	Risk Ratio (M‐H, Random, 95% CI)	0.86 [0.21, 3.54]
6.4.6 Anaemia (< 80 g/L)	1	370	Risk Ratio (M‐H, Random, 95% CI)	1.55 [0.68, 3.55]
6.4.7 Thrombocytopenia	1	370	Risk Ratio (M‐H, Random, 95% CI)	1.33 [0.88, 2.03]
6.4.8 Tremor	1	370	Risk Ratio (M‐H, Random, 95% CI)	1.10 [0.76, 1.57]

Comparison 7. Valaciclovir versus ganciclovir or valganciclovir.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
7.1 CMV disease in all treated patients	4		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
7.1.1 CMV disease	4	331	Risk Ratio (M‐H, Random, 95% CI)	0.55 [0.17, 1.74]
7.1.2 CMV disease in donor +ve or ‐ve/recipient +ve	1	63	Risk Ratio (M‐H, Random, 95% CI)	Not estimable
7.1.3 CMV disease in donor +ve/recipient ‐ve	1	12	Risk Ratio (M‐H, Random, 95% CI)	0.33 [0.02, 6.86]
7.2 Death	3		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
7.2.1 All‐cause death	3	273	Risk Ratio (M‐H, Random, 95% CI)	1.97 [0.31, 12.37]
7.3 CMV infection	4		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
7.3.1 CMV infection	4	331	Risk Ratio (M‐H, Random, 95% CI)	1.31 [0.91, 1.90]
7.3.2 CMV infection in donor +ve or ‐ve/recipient +ve	2	171	Risk Ratio (M‐H, Random, 95% CI)	1.04 [0.39, 2.82]
7.3.3 CMV infection in donor +ve/recipient ‐ve	2	23	Risk Ratio (M‐H, Random, 95% CI)	1.65 [0.85, 3.19]
7.4 Additional outcomes	3		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
7.4.1 Acute rejection	3	271	Risk Ratio (M‐H, Random, 95% CI)	0.88 [0.31, 2.47]
7.4.2 Graft loss	2	190	Risk Ratio (M‐H, Random, 95% CI)	0.61 [0.18, 2.02]
7.4.3 Leucopenia	2	188	Risk Ratio (M‐H, Random, 95% CI)	0.75 [0.45, 1.24]
7.4.4 Thrombocytopenia	2	188	Risk Ratio (M‐H, Random, 95% CI)	0.87 [0.45, 1.69]
7.4.5 Anaemia	2	187	Risk Ratio (M‐H, Random, 95% CI)	0.86 [0.31, 2.35]
7.4.6 Neurological dysfunction	1	69	Risk Ratio (M‐H, Random, 95% CI)	1.54 [0.62, 3.87]
7.4.7 Dose reduction or cessation for adverse effects	2	188	Risk Ratio (M‐H, Random, 95% CI)	0.54 [0.29, 1.02]
7.4.8 Other herpes virus infections	1	83	Risk Ratio (M‐H, Random, 95% CI)	1.86 [0.18, 19.73]
7.4.9 Non‐viral infections	1	83	Risk Ratio (M‐H, Random, 95% CI)	0.59 [0.44, 0.80]
7.4.10 Polyoma viremia	1	119	Risk Ratio (M‐H, Random, 95% CI)	0.48 [0.25, 0.94]
7.4.11 Polyomavirus associated nephropathy	1	119	Risk Ratio (M‐H, Random, 95% CI)	0.34 [0.07, 1.61]
7.4.12 Neutropenia	1	119	Risk Ratio (M‐H, Random, 95% CI)	0.66 [0.36, 1.20]
7.5 Kidney function at end of study	3		Std. Mean Difference (IV, Random, 95% CI)	Subtotals only
7.5.1 Serum creatinine	3	271	Std. Mean Difference (IV, Random, 95% CI)	‐0.12 [‐0.36, 0.12]
7.5.2 Calculated GFR	2	188	Std. Mean Difference (IV, Random, 95% CI)	0.21 [‐0.09, 0.52]

Comparison 8. Different ganciclovir regimens.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
8.1 IV doses given at different frequencies	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
8.1.1 CMV disease	1	72	Risk Ratio (M‐H, Random, 95% CI)	0.58 [0.32, 1.04]
8.1.2 CMV syndrome	1	72	Risk Ratio (M‐H, Random, 95% CI)	0.47 [0.09, 2.42]
8.1.3 Invasive CMV disease	1	72	Risk Ratio (M‐H, Random, 95% CI)	0.61 [0.30, 1.22]
8.1.4 CMV infection	1	72	Risk Ratio (M‐H, Random, 95% CI)	0.64 [0.45, 0.92]
8.1.5 All‐cause death	1	72	Risk Ratio (M‐H, Random, 95% CI)	4.26 [0.99, 18.34]
8.1.6 Death due to CMV disease	1	72	Risk Ratio (M‐H, Random, 95% CI)	0.19 [0.01, 3.81]
8.1.7 Bacteraemia	1	72	Risk Ratio (M‐H, Random, 95% CI)	0.95 [0.34, 2.66]
8.1.8 Bronchiolitis obliterans syndrome	1	72	Risk Ratio (M‐H, Random, 95% CI)	0.63 [0.25, 1.59]
8.1.9 Leucopenia	1	72	Risk Ratio (M‐H, Random, 95% CI)	4.74 [0.24, 95.33]
8.2 Oral versus IV ganciclovir	2		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
8.2.1 CMV disease	2	94	Risk Ratio (M‐H, Random, 95% CI)	0.57 [0.16, 2.05]
8.2.2 CMV syndrome	2	94	Risk Ratio (M‐H, Random, 95% CI)	0.48 [0.11, 2.11]
8.2.3 CMV invasive organ disease	1	64	Risk Ratio (M‐H, Random, 95% CI)	1.00 [0.07, 15.30]
8.2.4 CMV infection	1	30	Risk Ratio (M‐H, Random, 95% CI)	1.05 [0.41, 2.70]
8.2.5 All‐cause death	1	64	Risk Ratio (M‐H, Random, 95% CI)	5.00 [0.62, 40.44]
8.2.6 Acute rejection	2	94	Risk Ratio (M‐H, Random, 95% CI)	0.85 [0.45, 1.59]
8.2.7 Graft loss	1	34	Risk Ratio (M‐H, Random, 95% CI)	1.00 [0.07, 14.72]
8.2.8 Leucopenia due to ganciclovir	1	64	Risk Ratio (M‐H, Random, 95% CI)	0.69 [0.35, 1.39]
8.2.9 Medications ceased due to leucopenia	1	64	Risk Ratio (M‐H, Random, 95% CI)	1.00 [0.27, 3.66]

Comparison 9. Extended versus short‐duration valganciclovir.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
9.1 CMV disease	2		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
9.1.1 CMV disease at end of treatment	2	454	Risk Ratio (M‐H, Random, 95% CI)	0.20 [0.12, 0.35]
9.1.2 CMV disease at 9 months	1	310	Risk Ratio (M‐H, Random, 95% CI)	0.39 [0.25, 0.60]
9.1.3 CMV disease at 12 months	1	318	Risk Ratio (M‐H, Random, 95% CI)	0.44 [0.29, 0.66]
9.1.4 CMV disease at 24 months	1	318	Risk Ratio (M‐H, Random, 95% CI)	0.55 [0.38, 0.79]
9.2 CMV syndrome	2	454	Risk Ratio (M‐H, Random, 95% CI)	0.39 [0.24, 0.64]
9.3 CMV invasive disease	2		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
9.3.1 Number at 12 months	2	454	Risk Ratio (M‐H, Random, 95% CI)	0.17 [0.02, 1.34]
9.3.2 Number at 24 months	1	318	Risk Ratio (M‐H, Random, 95% CI)	0.70 [0.12, 4.14]
9.4 All‐cause death	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
9.4.1 Number at 12 months	1	319	Risk Ratio (M‐H, Random, 95% CI)	0.15 [0.01, 2.87]
9.4.2 Number at 2 years	1	319	Risk Ratio (M‐H, Random, 95% CI)	0.09 [0.01, 1.70]
9.5 CMV infection	2		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
9.5.1 CMV infection at end of treatment	2	454	Risk Ratio (M‐H, Random, 95% CI)	0.27 [0.10, 0.71]
9.5.2 CMV infection at 9 months	1	318	Risk Ratio (M‐H, Random, 95% CI)	0.72 [0.56, 0.94]
9.5.3 CMV infection at 12 months	1	318	Risk Ratio (M‐H, Random, 95% CI)	0.73 [0.57, 0.95]
9.6 Graft loss	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
9.6.1 Number at 12 months	1	318	Risk Ratio (M‐H, Random, 95% CI)	1.05 [0.22, 5.13]
9.6.2 Number at 24 months	1	318	Risk Ratio (M‐H, Random, 95% CI)	0.45 [0.12, 1.71]
9.7 Acute rejection	2		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
9.7.1 Biopsy‐proved acute rejection < 100 days	1	318	Risk Ratio (M‐H, Random, 95% CI)	0.70 [0.32, 1.51]
9.7.2 Biopsy‐proven acute rejection at 12 months	2	454	Risk Ratio (M‐H, Random, 95% CI)	0.64 [0.43, 0.95]
9.7.3 Biopsy‐proven acute rejection at 24 months	1	318	Risk Ratio (M‐H, Random, 95% CI)	0.62 [0.35, 1.08]
9.8 Other outcomes	2		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
9.8.1 Opportunistic infections	2	456	Risk Ratio (M‐H, Random, 95% CI)	0.71 [0.33, 1.57]
9.8.2 Post‐transplant diabetes mellitus	1	244	Risk Ratio (M‐H, Random, 95% CI)	1.17 [0.58, 2.36]
9.9 Adverse effects	2		Risk Difference (M‐H, Random, 95% CI)	Subtotals only
9.9.1 Total treatment‐related adverse effects	2	456	Risk Difference (M‐H, Random, 95% CI)	0.08 [‐0.01, 0.16]
9.9.2 Treatment‐related serious adverse effects	2	456	Risk Difference (M‐H, Random, 95% CI)	0.02 [‐0.02, 0.07]
9.9.3 Leukopenia	1	320	Risk Difference (M‐H, Random, 95% CI)	0.12 [0.01, 0.22]
9.9.4 Leucopenia leading to VGCV cessation	1	320	Risk Difference (M‐H, Random, 95% CI)	0.04 [0.00, 0.07]
9.9.5 Termination due to treatment related adverse effects	1	136	Risk Difference (M‐H, Random, 95% CI)	0.07 [‐0.04, 0.18]
9.9.6 Hospitalisations due to CMV disease	1	418	Risk Difference (M‐H, Random, 95% CI)	‐0.10 [‐0.17, ‐0.04]
9.9.7 Hospitalisations due to adverse effects	1	418	Risk Difference (M‐H, Random, 95% CI)	0.04 [‐0.05, 0.13]
9.9.8 CMV mutations known to confer ganciclovir resistance	2	208	Risk Difference (M‐H, Random, 95% CI)	0.02 [‐0.08, 0.11]

Comparison 10. Different valganciclovir regimens.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
10.1 All‐cause death	1	196	Risk Ratio (M‐H, Random, 95% CI)	0.20 [0.01, 4.11]
10.2 CMV infection	2	256	Risk Ratio (M‐H, Random, 95% CI)	0.37 [0.11, 1.22]
10.3 Acute rejection	1	196	Risk Ratio (M‐H, Random, 95% CI)	0.33 [0.07, 1.61]
10.4 Graft loss	1	196	Risk Ratio (M‐H, Random, 95% CI)	1.00 [0.06, 15.76]

Comparison 11. Maribavir versus ganciclovir.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
11.1 CMV disease	1	233	Risk Ratio (M‐H, Random, 95% CI)	1.30 [0.74, 2.29]
11.2 All‐cause death	1	294	Risk Ratio (M‐H, Random, 95% CI)	1.50 [0.55, 4.11]
11.3 CMV infection	1	233	Risk Ratio (M‐H, Random, 95% CI)	1.34 [1.10, 1.65]
11.4 Acute rejection	1	303	Risk Ratio (M‐H, Random, 95% CI)	0.92 [0.53, 1.61]
11.5 Adverse events	1	303	Risk Ratio (M‐H, Random, 95% CI)	0.99 [0.79, 1.25]

Characteristics of studies

Characteristics of included studies [ordered by study ID]

2VAL 2010 Kidney.

Study characteristics
Methods	Study design Parallel R CT Time frame: November 2007 to April 2012 Follow‐up period: 12 months Loss to follow‐up: 6.7%
Participants	Study characteristics Country: Czech Republic Setting: tertiary single centre Inclusion criteria: all adult kidney transplant recipients with recipient and/or D+ for CMV serology Exclusion criteria: D‐/R‐ serostatus; allergy to (val)ganciclovir or (val)acyclovir; severe leukopenia or thrombocytopenia; participation in another clinical trial; inability to provide informed consent Treatment group 1: valganciclovir Number: 60 Mean age ± SD: 48 ± 13 years Sex (M/F): 47/13 Treatment group 2: valacyclovir Number: 59 Mean age ± SD: 50 ± 11 years Sex (M/F): 37/22
Interventions	Treatment group 1 Valganciclovir: 900 mg orally/d for 12 weeks Treatment group 2 Valacyclovir: 2000 mg 4 times/d for 12 weeks Co‐interventions CSA, TAC, MMF, prednisone, ALG
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV DNA by PCR Graft loss Acute rejection Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: 4/123 (3.3%) Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Patients were randomized by the transplant physician using a random number table at a 1:1 ratio to valganciclovir or valacyclovir prophylaxis" Random number table is considered an adequate methodology for randomising patients
Allocation concealment (selection bias)	Low risk	Quote: "Sequentially numbered sealed envelopes were used for allocation concealment." Sequentially numbered sealed envelopes is considered an adequate methodology for allocation concealment
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "This was an open‐label, single‐center, randomized study." Open‐label study. Lack of blinding could influence clinical assessment of symptoms of possible CMV disease
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "This was an open‐label, single‐center, randomized study." Open‐label study. Lack of blinding could influence clinical assessment of symptoms of possible CMV disease
Incomplete outcome data (attrition bias) All outcomes	Low risk	Four patients excluded but reason unlikely to be related to true outcome
Selective reporting (reporting bias)	Low risk	Quote: "The trial is registered at Australian New Zealand Clinical Trials Registry: ACTRN1260000016033." Trial registration included the same outcomes as the final article.
Other bias	Low risk	Grants from Ministry of Health

Ahsan 1997 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: March 1995 to December 1995 Follow‐up period: 9 months Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: idney transplant recipients; D/R+, D+/R‐, D‐/R‐; if diabetic or receiving OKT‐3 Exclusion criteria: not reported Treatment group Number: 22 Mean age ± SEM: 50.4 ± 2.3 years Sex (M/F): 10/11 CD/LD: 18/3 Control group Number: 22 Mean age ± SEM: 47.6 ± 2.1 years Sex (M/F): 12/11 CD/LD: 7/15
Interventions	Treatment group Ganciclovir: 750 mg orally twice/d for 12 weeks starting day 1 Control group No treatment Co‐interventions CSA, AZA, prednisone, OKT‐3 (CD recipients)
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV culture, IgM3 All‐cause death Death due to CMV disease Acute rejection Graft loss Opportunistic infections
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computerised generated code with 4 patients in each block
Allocation concealment (selection bias)	Unclear risk	Randomisation stated but no information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open‐label study. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Open‐label study. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	One patient excluded but reason unlikely to be related to true outcome
Selective reporting (reporting bias)	High risk	Incomplete reporting of adverse effects
Other bias	Unclear risk	No information about pharmaceutical sponsorship

Ali Ibrahim 2020 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: not reported Follow‐up period: 36 months Loss to follow‐up: not reported
Participants	Study characteristics Country: Egypt Setting: not reported Inclusion criteria: kidney transplant recipients Exclusion criteria: not reported Treatment group Number: 30 Mean age: not reported Sex (M/F): not reported CD/LD: not reported Control group Number: 30 Mean age: not reported Sex (M/F): not reported CD/LD: not reported
Interventions	Treatment group Valaciclovir: 4 g/d Control group Valganciclovir: 450 mg/d Co‐interventions Not reported
Outcomes	Outcomes relevant to this review CMV disease CMV infection All‐cause death Graft loss Leucopenia
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: not reported Stop or endpoint: not reported Abstract‐only publication
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No description on the random sequence generation
Allocation concealment (selection bias)	Unclear risk	No description on the allocation concealment
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	No description on the blinding of participants or personnel
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No description on the blinding of outcome assessment
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	No description on missing data or patients lost to follow‐up
Selective reporting (reporting bias)	Unclear risk	No protocol or trial registry identified
Other bias	Unclear risk	No information about pharmaceutical sponsorship

Badley 1997 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: January 1991 to June 1994 Follow‐up period: 1 year Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary multicentre (2 sites) Inclusion criteria: first liver transplant Exclusion criteria: allergy to ganciclovir/aciclovir; Cr > 3 mg/dL or GFR < 10; stage 3/4 coma post‐transplant; existing CMV infection Treatment group Number: 83 Age range: 16 to 68 years Sex (M/F): 50/33 Control group Number: 84 Age range: 16 to 68 years Sex (M/F): 46/38
Interventions	Treatment group Ganciclovir: 5 mg/kg IV twice/d for 14 days starting first day post‐transplant Aciclovir: 800 mg orally 4 times/d to 120 days Control group Aciclovir: 800 mg orally 4 times/d to 120 days Co‐interventions CSA, AZA (one centre), prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture All‐cause death Acute rejection Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: 3 excluded Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Block randomisation scheme was used to generate a series of 150 randomly selected treatment assignments for each transplant centre"
Allocation concealment (selection bias)	Low risk	Patient randomisation and all statistical analyses were performed at coordinating centre
Blinding of participants and personnel (performance bias) All outcomes	High risk	Medications schedules differ between intervention groups. Assessment of primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Medications schedules differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	Three patients excluded but exclusions unlikely to be related to outcomes
Selective reporting (reporting bias)	High risk	No graft loss reported
Other bias	Low risk	Study carried out under NIH contracts

Balfour 1989 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: August 1985 to May 1988 Follow‐up period: 1 year Loss to follow‐up: 6% at 1 year, 0% at 6 months
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: cadaveric kidney transplant recipients > 10 years Exclusion criteria: intolerance of aciclovir Treatment group Number: 53 Median age (range): 43 years (15 to 67) Sex (M/F): 36/17 Control group Number: 51 Median age (range): 42 years (17 to 68) Sex (M/F): 34/17
Interventions	Treatment group Aciclovir: 800 mg orally 4 times/d for 12 weeks starting day of transplant Control group Placebo: 1 tablet 4 times/d for 12 weeks starting day of transplant Co‐interventions CSA, AZA, prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture, rising CMV antibody All‐cause death Death due to CMV disease Acute rejection Graft loss Opportunistic infections Adverse events
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none reported Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation scheme generated by computer program
Allocation concealment (selection bias)	Unclear risk	No information provided
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Placebo‐controlled study. Placebo tablets identical in appearance to acyclovir
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Placebo‐controlled study. Placebo tablets identical in appearance to acyclovir
Incomplete outcome data (attrition bias) All outcomes	Low risk	14 patients (6 intervention, 8 placebo) excluded but reasons unlikely to be related to true outcome
Selective reporting (reporting bias)	Low risk	All expected outcomes reported
Other bias	Low risk	Report partial support from NIH, Minnesota Medical Foundation and Burroughs Wellcome

Barkholt 1999 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: May 1993 to December 1994 Follow‐up period: 3 months Loss to follow‐up: 0%
Participants	Study characteristics Country: Sweden Setting: tertiary single centre Inclusion criteria: liver transplant recipients; all CMV serostatus Exclusion criteria: < 6 years; HIV infection; CMV therapy in previous 4 weeks Treatment group Number: 28 Mean age ± SD: 41 ± 17 years Sex (M/F): 16/12 Control group Number: 27 Mean age ± SD: 47 ± 15 years Sex (M/F): 12/15
Interventions	Treatment group Aciclovir: 800 mg (1 tablet) orally 4 times/d for 12 weeks starting 6 hours pre‐transplant Control group Placebo: 1 tablet orally 4 times/d for 12 weeks starting 6 hours pre‐transplant Co‐interventions Cyclosporin, cyclosporin, prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV culture, CMV DNA, IgM All‐cause death Death due to CMV disease Acute rejection Graft loss Opportunistic infections Adverse reactions
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: 5 Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided
Allocation concealment (selection bias)	Unclear risk	No information provided
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Placebo‐controlled study
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Placebo‐controlled study; patients with verified CMV infection were withdrawn from study drug without breaking the code
Incomplete outcome data (attrition bias) All outcomes	Low risk	5 excluded (3 given acyclovir outside study; 2 under 6 years) but reasons unlikely to be related to true outcome
Selective reporting (reporting bias)	Low risk	All expected outcomes reported
Other bias	High risk	Supported by Wellcome Research Laboratories

Brennan 1997 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: not reported Follow‐up period: 6 months Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: kidney transplant recipients; D/R+, D+/R‐ recipients Exclusion criteria: D‐/R‐ recipients Treatment group Number: 19 Mean age ± SEM: 50.6 ± 2.8 years Sex (M/F): 13/6 Control group Number: 23 Mean age ± SEM: 44.2 ± 3.0 years Sex (M/F): 5/18
Interventions	Treatment group Ganciclovir: 1000 mg orally 3 times/d for 12 weeks starting at transplant Control group No treatment except ganciclovir low dose to prevent Herpes simplex Co‐interventions CSA, AZA, prednisone, ATG
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV DNA All‐cause death Acute rejection Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Odd and even numbers according to last digit of medical record number. Information obtained from authors
Allocation concealment (selection bias)	High risk	Odd and even numbers according to last digit of medical record number. Information obtained from authors
Blinding of participants and personnel (performance bias) All outcomes	High risk	Medications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Medications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing data on primary outcome
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No report of graft loss
Other bias	High risk	Hoffman‐La Roche Laboratory pharmaceutical sponsorship

Cohen 1993 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: not reported Follow‐up period: 18 months Loss to follow‐up: 0%
Participants	Study characteristics Country: UK Setting: tertiary single centre Inclusion criteria: liver transplant recipients; D/R+, D+/R‐ Exclusion criteria: AKI; multiple organ system failure; D‐/R‐ recipients Treatment group Number: 33 Mean age: 42.4 years Sex (M/F): 15/18 Control group Number: 32 Mean age: 46.3 years Sex (M/F): 16/16
Interventions	Treatment group Ganciclovir: 5 mg/kg IV twice/d for 14 days starting on day 14 Control group No treatment Co‐interventions CSA, AZA, prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture, IgM All‐cause death Death due to CMV disease Acute rejection Graft loss Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "65 patients were randomised in a distribution determined by random numbers"
Allocation concealment (selection bias)	Low risk	Information obtained from authors that method used would not allow investigator/participant to know allocation before participant entered study
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open‐label study; primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Open‐label study; primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients completed follow‐up
Selective reporting (reporting bias)	High risk	Incomplete reporting of outcomes. No or limited report on other infections or adverse effects
Other bias	Unclear risk	No report on pharmaceutical sponsorship

Conti 1995 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: January 1992 to January 1994 Follow‐up period: 12 months Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: kidney transplant recipients; D/R+; receiving ALG for induction or rejection Exclusion criteria: not reported Treatment group Number: 22 Mean age: 43 years Sex (M/F): 11/11 Control group Number: 18 Mean age: 45 years Sex (M/F): 12/6
Interventions	Treatment group Ganciclovir: 5 mg/kg/d IV during ALG therapy (median 10 days) starting on first day of ALG Control group No treatment Co‐interventions CSA, AZA, prednisone, ALG
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease All‐cause death Acute rejection Graft loss Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Patients were randomly assigned" but method of sequence generation not stated
Allocation concealment (selection bias)	Unclear risk	Quote: "Patients were randomly assigned" but no information provided on method used
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants in control group received no specific intervention. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Participants in control group received no specific intervention. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients evaluated
Selective reporting (reporting bias)	High risk	Incomplete reporting of outcomes. No report or limited reporting of CMV infection/adverse effects
Other bias	Unclear risk	Supported in part by grant from National Kidney Foundation. No report on pharmaceutical sponsorship

Duncan 1993 Lung.

Study characteristics
Methods	Study design Parallel RCT Time frame: not reported Follow‐up period: 1 year Loss to follow‐up: 0%
Participants	Study design Country: USA Setting: tertiary single centre Inclusion criteria: lung transplant recipients; D/R+, D+/R‐; neutrophils > 1000/mm³, creatinine > 2.5 mg/dL Exclusion criteria: D‐/R‐ Treatment group Number: 13 Mean age ± SD: 41.8 ± 9.6 years Sex (M/F): 9/4 Control group Number: 12 Mean age ± SD: 45.6 ± 8.4 years Sex (M/F): 7/5
Interventions	Treatment group Ganciclovir: 5 mg/kg 4 times/d IV for 14 days starting day 7; 5 mg/kg/d IV for days 21 to 28; 5 mg/kg IV 5 times/wk to day 90 Control group Ganciclovir: 5 mg/kg 4 times/d IV for 14 days starting day 7; 5 mg/kg/d IV for days 21 to 28 Aciclovir: 800 mg orally 4 times/d to day 90 Co‐interventions CSA, AZA
Outcomes	Outcomes relevant to this review CMV tissue invasive disease CMV infection: CMV culture of bronchial lavage All‐cause death Death due to CMV disease Obliterative bronchiolitis Graft loss Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided other than that patients were stratified according to CMV serostatus and type of transplant
Allocation concealment (selection bias)	Unclear risk	Said to be "randomly assigned" but no other information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Medications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Medications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	Consecutive lung transplant recipients randomised. Results from all reported
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No or limited reporting of CMV disease, acute rejection, opportunistic infections
Other bias	Unclear risk	No report of pharmaceutical sponsorship

Egan 2002 Heart.

Study characteristics
Methods	Study design Parallel RCT Time frame: September 1994 to February 1998 Follow‐up period: 6 months Loss to follow‐up: 0%
Participants	Study characteristics Country: UK Setting: tertiary single centre Inclusion criteria: heart transplant recipients; D/R+ Exclusion criteria: active herpes infection; required other antiviral agents Treatment group Number: 14 Mean age (range): 51.6 years (39 to 63) Sex (M/F): 11/1 Control group Number: 13 Mean age (range): 50.4 years (31 to 62) Sex (M/F): 10/3
Interventions	Treatment group Valaciclovir: 2000 mg orally 4 times/d for 90 days starting within 72 hours of transplant Control group Aciclovir: 200 mg orally 4 times/d for 90 days starting within 72 hours of transplant for herpes simplex Co‐interventions CSA, AZA, prednisone, ATG
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV antigenaemia, culture All‐cause death Death due to CMV disease Acute rejection Graft loss Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Computer generated randomization schedule (block size 4)"
Allocation concealment (selection bias)	Low risk	Quote: "Allocation by opening sealed envelopes corresponding to patient number in sequence"
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Control group given low dose acyclovir to "maintain double blind by effective prophylaxis of herpes simplex outbreaks" but no information that acyclovir and valacyclovir tablets were indistinguishable
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Control group given low dose acyclovir to "maintain double blind by effective prophylaxis of herpes simplex outbreaks" but no information that acyclovir and valacyclovir tablets were indistinguishable
Incomplete outcome data (attrition bias) All outcomes	Low risk	All enrolled patients were included in the analysis including 2 patients randomised in error
Selective reporting (reporting bias)	Low risk	All expected outcomes reported
Other bias	High risk	Funding provided by Glaxo Wellcome Research and Development

Flechner 1998 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: April 1996 to December 1997 Follow‐up period: 6 to 27 months Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: kidney transplant recipients > 18 years and < 101 kg; D/R+, D+/R‐ Exclusion criteria: D‐/R‐; allergy to ganciclovir/aciclovir; AIDS; WBC < 3000; platelets < 100,000; previous viral hepatitis Treatment group Number: 40 Mean age: 47.9 years Sex (M/F): 30/10 Control group Number: 39 Mean age: 50.2 years Sex (M/F): 31/8
Interventions	Treatment group Ganciclovir: 1000 mg orally 3 times/d for 84 days starting on day 1 Control group Aciclovir: 800 mg orally 4 times/d for 84 days starting on day 1 Co‐interventions CMV IgG given to D+/R‐ recipients in each group; CSA, AZA (⅓), MMF (⅔), OKT‐3
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture All‐cause death Death due to CMV disease Acute rejection Opportunistic infections
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer generated list. Information provided by authors
Allocation concealment (selection bias)	Low risk	Central research coordinator
Blinding of participants and personnel (performance bias) All outcomes	High risk	Medications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Medications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All participants were followed to death/graft loss or June 1998
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No report of graft loss
Other bias	Unclear risk	No information provided about pharmaceutical sponsorship

Gane 1997 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: December 1993 to April 1995 Follow‐up period: 1 year Loss to follow‐up: 0%
Participants	Study characteristics Country: USA, Europe Setting: tertiary multicentre Primary liver transplant recipients aged > 18 years; D/R+, D+/R‐ Exclusion criteria: multiple organ transplant; D‐/R‐ (2 patients inadvertently randomised and included in analysis); unable to take oral medications; neutrophils < 1000; platelets < 25,000; Cr > 300 Treatment group Number: 150 Mean age ± SD: 46.8 ± 11.6 years Sex (M/F): 92/58 Control group Number: 154 Mean age ± SD: 48.1 ± 10.9 years Sex (M/F): 82/72
Interventions	Treatment group Ganciclovir: 1000 mg (4 tablets) orally 3 times/d until day 98 starting within 10 days of transplant Control group Matching placebo: 4 tablets orally 3 times/d until day 98 starting within 10 days of transplant Co‐interventions CSA, TAC (52 patients), ALG (61 patients)
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV antigenaemia, IgM, CMV culture All‐cause death Death due to CMV disease Acute rejection Graft loss Opportunistic infection Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or end point: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided
Allocation concealment (selection bias)	Unclear risk	"Randomised trial" but no further information provided
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Matching placebo capsules
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Matching placebo capsules
Incomplete outcome data (attrition bias) All outcomes	Low risk	Complete 12 month data available on all participants
Selective reporting (reporting bias)	Low risk	All expected outcomes reported
Other bias	High risk	Grant support from Roche Global Development

Gavalda 1997 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: June 1991 to November 1993 Follow‐up period: 12 months Loss to follow‐up: 0%
Participants	Study characteristics Country: Spain Setting: tertiary single centre Inclusion criteria: primary liver transplant recipient; D/R+ Exclusion criteria: second transplant recipients Treatment group Number: 37 Median age (range): 57 years (34 to 66) Sex (M/F): 25/12 Control group Number: 36 Median age (range): 54 years (20 to 65) Sex (M/F): 23/13
Interventions	Treatment group Aciclovir: 400 mg orally 5 times/d for 16 weeks starting 3 to 30 days (median 7 days) post‐transplant Control group No treatment Co‐interventions CSA, prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture All‐cause death Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided
Allocation concealment (selection bias)	Unclear risk	"Randomized study" but no other information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Control group received no medication. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Control group received no medication. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	Consecutive adult recipients enrolled. 7 did not complete study. All included in analysis
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No or limited reporting of acute rejection, adverse effects
Other bias	Unclear risk	No information provided on pharmaceutical sponsorship

Green 1997 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: July 1992 to March 1994 Follow‐up period: 1 year Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: first liver transplant recipients aged < 18 years Exclusion criteria: multi‐organ recipients Treatment group Number: 24 Mean age: 4.9 years Sex (M/F): not reported Control group Number: 24 Mean age: 4.3 years Sex (M/F): not reported
Interventions	Treatment group Ganciclovir: 5 mg/kg twice/d IV for 14 days starting day 1 Aciclovir: 800 mg/m² orally 4 times/d to 1 year Control group Ganciclovir: 5 mg/kg twice/d IV for 14 days starting day 1 Co‐interventions TAC, prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive tissue disease CMV infection: CMV culture All‐cause death Opportunistic infections
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Stratified according to donor/recipient serostatus. Method not reported
Allocation concealment (selection bias)	Unclear risk	"A randomized trial" but no further information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Control group received no medication after initial two weeks of ganciclovir therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Control group received no medication after initial two weeks of ganciclovir therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients enrolled in study were included in analysis
Selective reporting (reporting bias)	High risk	Incomplete reporting of outcomes. No or limited reporting of acute rejection, graft loss, adverse effects
Other bias	Unclear risk	Study ended following interim analysis which showed no benefit of prolonged course of acyclovir and families requesting that their children receive acyclovir rather than enter trial. No information provided on pharmaceutical sponsorship

Halim 2016 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: 2010 to 2013 Follow‐up period: 2 years Loss to follow‐up: 5 patients (2%)
Participants	Study characteristics Country: Kuwait Setting: single centre Inclusion criteria: adult kidney transplant recipients (> 18 years) who could tolerate oral valganciclovir within 1 week post‐transplant Exclusion criteria: high‐risk patients (D+/R‐) were excluded from the analysis to stratify the patients into a homogenous group of kidney transplant recipients with intermediate risk for CMV infection (D‐/R+ or D+/R+) Treatment group Number: 98 Mean age ± SD: 43.2 ± 14.2 years Sex (M/F): 62/36 Control group Number: 98 Mean age ± SD: 42 ± 17.5 years Sex (M/F): 57/41
Interventions	Treatment group Valganciclovir: 450 mg/d for 6 months Control group Valganciclovir: 900 mg/d for 6 months Co‐interventions None
Outcomes	Outcomes relevant to this review All‐cause death CMV infection Acute rejection Graft loss
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Patients were sequentially randomized (1:1) using the simple randomization method" No clear description of the random sequence generation has been provided
Allocation concealment (selection bias)	Unclear risk	No clear description of the allocation concealment has been provided
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	No clear description of the blinding of participants and personnel has been provided
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No clear description of the blinding of participants and personnel has been provided
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "Our intent‐to‐treat population included 201 patients. Of these, 196 were included in the final analysis as intermediate‐risk patients (98 in each group) after exclusion of high‐risk patients (n = 5 patients)." All patients were included in the intention‐to‐treat analysis and very little missing patients
Selective reporting (reporting bias)	Unclear risk	No registration or protocol identified to assess selective reporting
Other bias	Low risk	No other sources of bias detected

Hertz 1998 Heart/lung.

Study characteristics
Methods	Study design Parallel RCT Time frame: January 1993 to January 1996 Follow‐up period: 1 year Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: lung or heart/lung transplant recipients; D/R+; D+/R‐ Exclusion criteria: D‐/R‐ Treatment group Number: 35 Mean age ± SD: 46.4 ± 11.4 years Sex (M/F): 15/20 Control group Number: 37 Mean age ± SD: 49.1 ± 8.7 years Sex (M/F): 14/23
Interventions	Treatment group Ganciclovir: 5 mg/kg twice/d IV on days 8 to 21; 5 mg/kg IV 3 times/wk to 90 days Control group Ganciclovir: 5 mg/kg twice/d IV on days 8 to 21; 5 mg/kg IV daily to 90 days Co‐interventions CSA, AZA, prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV tissue invasive disease CMV infection: CMV culture of bronchial lavage All‐cause death Death due to CMV disease Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided
Allocation concealment (selection bias)	Unclear risk	"Randomized trial" in title but no information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	One patient unable to complete therapy but included in analyses
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No or limited reporting of graft loss, adverse effects
Other bias	Unclear risk	No information provided about pharmaceutical sponsorship

Hibberd 1995 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: November 1990 to September 1992 Follow‐up period: 6 months Loss to follow‐up: 1.8% (2 lost at 32 days and 78 days)
Participants	Study characteristics Country: USA Setting: tertiary multicentre (6 sites) Inclusion criteria: kidney transplant recipients receiving ALG preparations for induction or treatment of rejection; D/R+ Exclusion criteria: < 20 years; pregnant; multi‐organ recipient; treatment with other antiviral agent Treatment group Number: 64 Mean age ± SEM: 44.2 ± 1.62 years Sex (M/F): 36/28 Control group Number: 49 Mean age ± SEM: 42.8 ± 1.99 years Sex (M/F): 33/16
Interventions	Treatment group Ganciclovir: 2.5 mg/kg/d IV during ALG therapy (median duration 9 days) starting within 24 hours of first dose of ALG Control group No treatment Co‐interventions CSA, AZA, prednisone, ALG or OKT‐3
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture All‐cause death Death due to CMV disease Graft loss Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Separate randomization lists for each center" but no other information available
Allocation concealment (selection bias)	Unclear risk	Quote: "Patients were randomly assigned" but no other information available
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "Investigators at each site knew which patients received the study drug" Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "Investigators at each site knew which patients received the study drug" Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All participants included in the analyses
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No or limited reporting of acute rejection, adverse effects
Other bias	High risk	Supported in part by a grant from Ortho Pharmaceutical Corporation. Ganciclovir provided by Syntex Laboratories Inc

IMPACT 2010 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: March 2006 to August 2008 (final data collection date for primary outcome measure) Follow‐up period: 24 months Loss to follow‐up: 6/326 did not receive experimental therapy; 103 subsequently withdrew from treatment, but all who received at least one dose of medication and underwent post‐randomisation safety assessment were included in ITT analysis for safety. All who received at least one dose of therapy and were D+/R‐ were included in the efficacy study
Participants	Study characteristics Countries: 13 countries Setting: tertiary multicentre (65 sites) Inclusion criteria: kidney transplant recipients Exclusion criteria: CMV disease; HIV; hepatitis B; hepatitis C at enrolment; received CMV IgG in previous 1 month; multi‐organ transplant Treatment group 1 Number: 156 Mean age ± SD: 47 ± 13.5 years Sex (M/F): 116/40 Treatment group 2 Number: 164 Mean age ± SD: 48.5 ± 13.8 years Sex (M/F): 119/45
Interventions	Treatment group 1 Valganciclovir: 900 mg/d orally for 200 days, started as soon as able to tolerate oral medications and by 10 days post‐transplant Treatment group 2 Valganciclovir: 900 mg/d orally for 100 days, started as soon as able to tolerate oral medications and by 10 days post‐transplant, followed by placebo orally for 100 days Co‐interventions Induction therapy with ATG (52, 52) or IL2Ra (79, 72)
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV DNA by PCR, CMV antigenaemia All‐cause death Acute rejection Graft loss Opportunistic infections Adverse effects Death due to CMV disease Ganciclovir resistant mutations
Notes	Additional information Further Information sought from the authors on sequence generation and allocation concealment but no response obtained
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Patients randomized sequentially in a 1:1 ratio at each study centre in the order in which they were enrolled" No other information provided
Allocation concealment (selection bias)	Low risk	Central randomisation
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double blind. Placebo and active drug "were indistinguishable"
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "Study investigators, site staff and sponsors were fully blinded to treatment allocation until after analysis of the primary endpoint"
Incomplete outcome data (attrition bias) All outcomes	Low risk	ITT analysis. Patients excluded who did not receive at least one dose of medication but only 8 patients excluded and numbers unlikely to influence true outcome
Selective reporting (reporting bias)	Low risk	All expected outcomes reported
Other bias	High risk	Funded by F Hoffman‐La‐Roche. Medical writers funded by sponsors. "There is an agreement between the Principal Investigators and the Sponsor that restricts the principal investigators' rights to discuss or publish trial results after the trial is completed"

Kletzmayr 1996 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: not reported Follow‐up period: 1 year Loss to follow‐up: 5.6%
Participants	Study characteristics Country: Austria Setting: tertiary single centre Inclusion criteria: kidney transplant recipients; D+/R‐ Exclusion criteria: not reported Treatment group Number: 22 Mean age ± SD: 46 ± 14 years Sex (M/F): 17/5 Control group Number: 10 Mean age ± SD: 44 ± 13 years Sex (M/F): 7/3
Interventions	Treatment group Aciclovir: 800 mg 3 times/d orally for 3 months starting first post‐op day Control group No treatment Co‐interventions CSA, AZA, prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV antigenaemia, CMV culture, IgM All‐cause death Acute rejection Graft loss
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Patients randomized... in a 2:1 ratio" No information on sequence generation provided
Allocation concealment (selection bias)	Unclear risk	Quote: "Patients were randomly assigned" No information provided on method
Blinding of participants and personnel (performance bias) All outcomes	High risk	Control group received no specific treatment. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Control group received no specific treatment. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	4/36 excluded from analysis
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No or limited reporting of opportunistic infections/adverse effects
Other bias	Unclear risk	No information provided on pharmaceutical sponsorship

Leray 1995 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: January 1991 to July 1994 Follow‐up period: unclear Loss to follow‐up: 0%
Participants	Study characteristics Country: France Setting: tertiary single centre Inclusion criteria: kidney transplant recipients; D+/R‐ Exclusion criteria: not reported Treatment group Number: 13 Age: not reported Sex (M/F): not reported Control group Number: 10 Age: not reported Sex (M/F): not reported
Interventions	Treatment group Ganciclovir: 5 mg/kg IV twice/d for 14 days starting 14 days post‐transplant Control group No treatment Co‐interventions CSA, AZA, prednisone, ALG
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV antigenaemia, CMV culture, IgM Acute rejection Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none reported Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided
Allocation concealment (selection bias)	Unclear risk	Quote: "On day 14 patients were randomized". No other information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Control group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Control group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Unclear if any patients were excluded from analysis
Selective reporting (reporting bias)	Unclear risk	Abstract only available
Other bias	Unclear risk	No information provided on sponsorship

Lowance 1999 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: July 1992 to December 1996 Follow‐up period: 12 months Loss to follow‐up: 0%
Participants	Study characteristics Country: USA/Europe Setting: tertiary multicentre (27 sites) Inclusion criteria: kidney transplant recipients; D/R+, D+/R‐ Exclusion criteria: D‐/R‐; active herpes infection; antiviral therapy in previous 2 months Treatment group Number: 306; D/R+ (204); D+/R‐ (102) Mean age ± SD: D/R+ (43.6 ± 13.1 years); D+/R‐ (40.3 ± 14.2 years) Sex (M/F): D/R+ 153/51; D+/R‐ 60/42 Control group Number: 310; D/R+ (204); D+/R‐ (106) Mean age ± SD: D/R+ (45.1 ± 13 years); D+/R‐ (45.6 ± 13.5 years) Sex (M/F): D/R+ 124/80; D+/R‐ 65/41
Interventions	Treatment group Valaciclovir: 2000 mg orally 4 times/d for 90 days starting within 3 days of transplant Control group Placebo: orally 4 times/d for 90 days starting within 3 days of transplant Co‐interventions CSA, AZA, TAC (6), MMF (7), ATG or ALG (251), OKT‐3 (102)
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture All‐cause death Death due to CMV disease Acute rejection Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Randomly assigned in 1:1 ratio according to study site". No other information provided
Allocation concealment (selection bias)	Unclear risk	"Randomly assigned" but method of allocation unstated
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Matching placebo tablets
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Matching placebo tablets
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients included in intention to treat analysis
Selective reporting (reporting bias)	High risk	Not all expected outcomes reported. No graft loss data reported
Other bias	High risk	Supported by Glaxo Wellcome. Employees included as authors

Macdonald 1995 Heart.

Study characteristics
Methods	Study design Parallel RCT Time frame: not reported Follow‐up period: 12 months Loss to follow‐up: 0%
Participants	Study characteristics Country: Australia Setting: tertiary single centre Inclusion criteria: heart transplant recipients; D/R+, D+/R‐ Exclusion criteria: D‐/R‐ Treatment group Number: 28 Mean age ± SD: 48 ± 15 years Sex (M/F): 24/4 Control group Number: 28 Mean age ± SD: 45 ± 15 years Sex (M/F): 25/3
Interventions	Treatment group Ganciclovir: 5 mg/kg IV 3 times/wk for 6 weeks starting pre‐transplant Control group Placebo: IV 3 times/wk for 6 weeks starting pre‐transplant Co‐interventions CSA, AZA, prednisone, ATG
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture All‐cause death Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Table of random numbers. Separate randomisation sequences were used according to serostatus
Allocation concealment (selection bias)	Unclear risk	Method of allocation not stated
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Matching placebo administered to control group
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Matching placebo administered to control group
Incomplete outcome data (attrition bias) All outcomes	Low risk	Consecutive patients enrolled and all included in analysis
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No report of graft loss
Other bias	Unclear risk	No report on pharmaceutical sponsorship

Martin 1994 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: February 1991 to August 1991 Follow‐up period: 24 weeks Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: liver transplant recipients aged > 18 years Exclusion criteria: fulminant hepatic failure; stage 3/4 hepatic coma; hepatic malignancies with pre‐operative chemotherapy Treatment group Number: 68 Mean age ± SD: 48.1 ± 13.2 years Sex (M/F): 43/25 Control group Number: 71 Mean age ± SD: 47 ± 12.9 years Sex (M/F): 35/36
Interventions	Treatment group Ganciclovir: 5 mg/kg twice/d IV for 14 days starting 2 days post‐transplant Aciclovir: 800 mg orally 4 times/d to 10 weeks Control group Aciclovir: 800 mg orally 4 times/d for 10 weeks starting 2 days post‐transplant Co‐interventions TAC
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive tissue disease CMV infection: CMV culture, IgM All‐cause death Acute rejection Graft loss Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported Four excluded after randomisation (active CMV (1), death from sepsis (2), unable to take medication (1)) and one randomised to ganciclovir given acyclovir and analysed in acyclovir group
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Fixed block randomization scheme (block size = 4)"
Allocation concealment (selection bias)	Unclear risk	No information provided on allocation
Blinding of participants and personnel (performance bias) All outcomes	High risk	Groups received different medications by different routes. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Groups received different medications by different routes. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	4/143. Missing outcome data unlikely to be related to true outcome
Selective reporting (reporting bias)	High risk	Did not report opportunistic infections
Other bias	Unclear risk	No information on pharmaceutical sponsorship provided

Merigan 1992 Heart.

Study characteristics
Methods	Study design Parallel RCT Time frame: not reported Follow‐up period: 120 days Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary multicentre (4 sites) Inclusion criteria: heart transplant recipients; D/R+, D+/R‐ Exclusion criteria: D‐/R‐; combined heart‐lung transplant recipients; antiviral agents in previous 7 days; WBC < 1500; platelets < 50,000; GFR < 10 or > 400 Treatment group Number: 76 Mean age ± SEM: 47.1 ± 1.55 years Sex (M/F): 68/8 Control group Number: 73 Mean age ± SEM: 47.6 ± 1.4 years Sex (M/F): 63/10
Interventions	Treatment group Ganciclovir: 5 mg/kg IV twice/d for 14 days starting on day 1 post‐transplant but delay for 2 to 7 days in 21% Control group Placebo: IV twice/d for 14 days starting on day 1 post‐transplant but delay for 2 to 7 days in 23% Co‐interventions CSA, AZA, prednisone, OKT‐3
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture All‐cause death Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: study stopped after interim assessment after 80 patients enrolled when the difference between treatment groups evident
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Stratified at randomization according to their CMV serostatus" Otherwise no information provided
Allocation concealment (selection bias)	Unclear risk	Patients were "... randomly assigned". No information provided on allocation
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Patients in control group received infusions of placebo medication
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Patients in control group received infusions of placebo medication
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients included in analysis
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No report of graft loss
Other bias	High risk	Supported by Public Health Service grant and by grant from Syntex Corporation (employees included as authors)

Nafar 2005 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: September 2001 to November 2001 Follow‐up period: 12 months Loss to follow‐up: 0%
Participants	Study characteristics Country: Iran Setting: tertiary single centre Inclusion criteria: kidney transplant recipients; D+/R+; ATG required for rejection; second transplant; deceased donor transplant Mean age ± SD: 37.8 ± 9.8 years Exclusion criteria: not reported Treatment group Number: 16 (17 entered the study) Age: not reported Sex (M/F): 11/5 Control group Number: 14 (17 entered study) Age: not reported Sex (M/F): 9/5
Interventions	Treatment group Ganciclovir: 1000 mg oral 3 times/d for 3 months Control group Ganciclovir: 5 mg/kg/d IV for 2 weeks Co‐interventions ATG for induction or rejection; other immunosuppression not reported
Outcomes	Outcomes relevant to this review CMV disease CMV viraemia: CMV antigenaemia Acute rejection Adverse effects Kidney function at 12 months
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none One patient from treatment group excluded following graft loss; 3 excluded from control group (graft loss 1, pre‐existing CMV antigenaemia, refusal to be followed) Stop or endpoint: not repoerted Additional data requested from authors: none
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information available
Allocation concealment (selection bias)	Unclear risk	"Randomized prospective trial" in title but no other information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	High risk	4/34 excluded. 3 excluded from IV ganciclovir arm
Selective reporting (reporting bias)	High risk	Drug toxicity and side effects not reported
Other bias	Unclear risk	No information provided on pharmaceutical sponsorship

Nakazato 1993 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: August 1990 to November 1991 Follow‐up period: 1 year Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: liver transplant recipients Exclusion criteria: not reported Treatment group Number: 52 Mean age ± SD: 38.7 ± 21.5 years Sex (M/F): not reported Control group Number: 52 Mean age ± SD: 34.9 ± 22.8 years Sex (M/F): not reported
Interventions	Treatment group Ganciclovir: 5 mg/kg/d IV during inpatient periods in first 3 months post‐transplant Aciclovir: 5 mg/kg/d oral to 3 months Control group Aciclovir: 5 mg/kg/d IV during inpatient periods in first 3 months post‐transplant Aciclovir: 5 mg/kg/d oral to 3 months Co‐interventions IgG IV 200 mg/kg/d during inpatient periods in first 3 months post‐transplant; CSA (81), TAC (23), prednisone
Outcomes	Outcomes relevant to this review CMV disease: CMV culture/histopathology and symptoms All‐cause death Acute rejection Graft loss Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided
Allocation concealment (selection bias)	Unclear risk	Quote: "Preliminary report of a randomized trial..." in title. Otherwise no information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients included in analyses
Selective reporting (reporting bias)	High risk	No CMV infection or adverse effects reported
Other bias	High risk	Supported in part by Sandoz Pharmaceuticals

Palmer 2010 Lung.

Study characteristics
Methods	Study design Parallel RCT Time frame: July 2003 to January 2007 Follow‐up period: 13 months Loss to follow‐up: 45/136 withdrawn but all included in analysis
Participants	Study characteristics Country: USA Setting: tertiary multicentre (11 sites) Inclusion criteria: single or double first lung transplant recipient; aged ≥ 18 years; adequate haematological, kidney and liver function; D/R+, D+/R‐; received IV ganciclovir for 2 weeks post‐transplant; able to tolerate oral medications; negative PCR/bronchoscopy for CMV at baseline and at day 75 when randomisation occurred Exclusion criteria: re‐transplant, on ventilator; current/previous ganciclovir outside study; invasive fungal disease; using disallowed medications; previous severe reaction to ganciclovir; diarrhoea; malabsorption; liver/kidney/haematological dysfunction Treatment group 1 Number: 70 Mean age (IQR): 56 (45 to 62) years Sex (M/F): 29/41 Treatment group 2 Number: 66 Mean age (IQR): 55 (42 to 61) years Sex (M/F): 38/28
Interventions	Treatment group 1 12 months group IV ganciclovir: for 2 weeks starting within 24 hours of transplant Oral valganciclovir: 900 mg/d for 3 months Oral valganciclovir: 900 mg/d for 9 months Treatment group 2 3 months group IV ganciclovir: for 2 weeks starting within 24 hours of transplant Oral valganciclovir: 900 mg/d for 3 months Placebo: for 9 months Co‐interventions TAC (50/70 and 46/66), ALG (23/70 and 21/66)
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV‐DNA by PCR on blood and/or broncholavage All‐cause death (data from 1 centre) Acute rejection Opportunistic infections Adverse reactions
Notes	Additional information Information on absolute numbers with outcomes requested from investigators. Response received but information not available
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomised 1.1 stratified by site at 3 months. Computer‐generated randomised list managed centrally
Allocation concealment (selection bias)	Low risk	Randomised at 3 months. Independent pharmacist dispensed medically centrally
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Placebo‐controlled study
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Bronchoscopies performed by investigators blinded to treatment group
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients included in analysis
Selective reporting (reporting bias)	High risk	Incomplete reporting of outcomes. Reports of deaths only available for one institution
Other bias	High risk	Funded by Roche Pharmaceuticals. All data analyses performed at Duke Clinical Research Institute

Pavlopoulou 2005 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: April 1999 to September 2000 Follow‐up period: 6 months Loss to follow‐up: 0%
Participants	Study characteristics Country: Greece Setting: tertiary single centre Inclusion criteria: kidney transplant recipient; D/R+, D+/R‐ Exclusion criteria: active herpes viral infection; antiviral therapy in previous 14 days Treatment group Number: 43 Mean age ± SD: 40.7 ± 12 years Sex (M/F): 34/9 Control group Number: 40 Mean age ± SD: 43.1 ± 15 years Sex (M/F): 29/11
Interventions	Treatment group Valaciclovir: 2000 mg oral 4 times/d starting within 72 hours of transplant for 3 months Control group Ganciclovir: 1000 mg oral 3 times/d starting within 72 hours of transplant for 3 months Co‐interventions CSA or TAC, sirolimus (11), IL2R a (treatment: 23; control: 25), ATG 4 (treatment: 4; control: 2)
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV‐DNA All‐cause death Acute rejection Opportunistic infections Adverse reactions Kidney function at 6 months
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: not reported Stop or end point: not reported Additional data requested from authors: none
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Assigned randomly in 1:1 ratio but no other information provided
Allocation concealment (selection bias)	Unclear risk	No information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open‐label study. Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Open‐label study. Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients included in analyses
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. Limited reporting of adverse effects
Other bias	Unclear risk	No information provided on pharmaceutical sponsorship

Paya 2004 All.

Study characteristics
Methods	Study design Parallel RCT Time frame: April 2000 to August 2001 Follow‐up period: 12 months Loss to follow‐up: 0%
Participants	Study characteristics Countries: USA, Europe, Canada, Australia Setting: tertiary multicentre (57 sites) Inclusion criteria: solid organ transplant recipient aged >12 years (liver, kidney, heart, kidney‐pancreas); D+/R‐; first transplant; adequate liver and kidney function Exclusion criteria: retransplant; history of CMV infection/disease; CMV therapy in previous 30 days; severe uncontrolled diarrhoea; malabsorption Treatment group Number: 245 Mean age: 45.7 years Sex (M/F): 179/66 Control group Number: 127 Mean age: 45.3 years Sex (M/F): 95/32
Interventions	Treatment group Valganciclovir: 900 mg oral/d starting within 10 days of transplant for 100 days Control group Ganciclovir: 1000 mg oral 3 times/d starting within 10 days of transplant for 100 days Co‐interventions Immunosuppression according to protocol of centre
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV tissue invasive disease CMV infection: CMV‐DNA; infection confirmed in central lab All‐cause death Death due to CMV disease Acute rejection Graft loss Opportunistic infections Adverse reactions
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: 2 excluded from safety analysis as did not receive medication, 8 excluded from primary outcome analysis as not D+/R‐
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Stratified according to organ transplanted and assigned in 2:1 ratio at each centre
Allocation concealment (selection bias)	Low risk	Quote: "Treatment randomization numbers were assigned by telephone via a central randomization center"
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐dummy. Placebo tablets given to both groups
Blinding of outcome assessment (detection bias) All outcomes	Low risk	End points adjudicated by independent (of sponsor and study) blinded Endpoint Committee
Incomplete outcome data (attrition bias) All outcomes	Low risk	ITT population included 364/372 patients. Safety 370/372. Reasons for missing outcomes data unlikely to be related to true outcome
Selective reporting (reporting bias)	Low risk	Expected outcomes all reported
Other bias	High risk	Study funded by Hoffman‐La Roche

Pouteil‐Noble 1996 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: not reported Follow‐up period: 6 months Loss to follow‐up: 0%
Participants	Study characteristics Country: France Setting: tertiary single centre Inclusion criteria: kidney transplant recipients; all CMV serostatus Exclusion criteria: not reported Treatment group Number: 24 Age: not reported Sex (M/F): not reported Control group Number: 26 Age: not reported Sex (M/F): not reported
Interventions	Treatment group Ganciclovir: 5 mg/kg/d IV for 14 days starting on day of transplant Aciclovir: 800 mg oral 3 times/d from day 14 to 3 months Control group Placebo: given as for treatment arm Co‐interventions Not reported
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV culture, IgM All‐cause death
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided except stratification for CMV serostatus
Allocation concealment (selection bias)	Low risk	Adequate allocation (information received from authors)
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Control group received placebo
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Control group received placebo
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients included in analyses
Selective reporting (reporting bias)	Unclear risk	Abstract only
Other bias	Unclear risk	Work supported by Wellcome Laboratories and Hospices Civils de Lyon

Prabakaran 2020 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: February 2018 to December 2018 Follow‐up period: 6 months Loss to follow‐up: not reported
Participants	Study characteristics Country: India Setting: single centre Inclusion criteria: kidney transplant recipients; D+/R+ Exclusion criteria: not reported Treatment group Number: 30 Mean age: 37 years Sex (M/F): not reported Control group Number: 30 Mean age: 37 years Sex (M/F): not reported
Interventions	Treatment group Valganciclovir: 900 mg once/d for 3 months Control group Low‐dose valganciclovir prophylaxis: 900 mg 3 times/wk for 3 months Co‐interventions Not reported
Outcomes	Outcomes relevant to this review CMV infection Acute rejection Adverse reactions
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Abstract‐only publication
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information reported on randomisation
Allocation concealment (selection bias)	Unclear risk	No information reported on allocation concealment
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open‐label study
Blinding of outcome assessment (detection bias) All outcomes	High risk	Open‐label study
Incomplete outcome data (attrition bias) All outcomes	Low risk	No patients lost to follow‐up
Selective reporting (reporting bias)	Low risk	No differences identified between trial registry and manuscript
Other bias	Unclear risk	No information on funding reported

Reischig 2005 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: April 1999 to December 2000; January 2001 to January 2003 Follow‐up period: 12 months Loss to follow‐up: 0%
Participants	Study characteristics Country: Czech Republic Setting: tertiary single centre Inclusion criteria: kidney transplant recipients; D/R+, D+/R‐ Exclusion criteria: D‐/R‐; unknown CMV status; active CMV infection; treatment with antiviral agents; WBC < 4000; platelets < 150,000; allergy to study drugs Treatment group Number: 35 Mean age ± SD: 45 ± 12 years Sex (M/F): 26/9 Control group Number: 36 Mean age ± SD: 48 ± 11 years Sex (M/F): 25/11
Interventions	Treatment group Valaciclovir: 2000 mg oral 4 times/d starting within 3 days of transplant for 3 months Control group Ganciclovir: 1000 mg oral 3 times/d starting within 3 days of transplant for 3 months Co‐interventions Aciclovir low dose to prevent herpes simplex; CSA, MMF, prednisone, ATG or OKT‐3 (9), IL2Ra/sirolimus (6)
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV‐DNA, CMV antigenaemia, CMV culture All‐cause death Acute rejection Graft loss Adverse reactions
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported Additional data requested from authors: data on quality assessment and results obtained
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random number generator used. (Information from authors)
Allocation concealment (selection bias)	Low risk	Adequate allocation based on information from authors
Blinding of participants and personnel (performance bias) All outcomes	High risk	Different medication schedules in each group. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Different medication schedules in each group. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	Randomised consecutive patients. All patients included in analyses
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No data of opportunistic infections
Other bias	Low risk	Quote: "The study was independent and not funded by any commercial sources"

Rondeau 1993 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: January 1990 to July 1992 Follow‐up period: 3 months Loss to follow‐up: 0%
Participants	Study characteristics Country: France Setting: tertiary multicentre (2 sites) Inclusion criteria: kidney transplant recipients; D+/R‐ Exclusion criteria: living‐related donor transplant recipients; WBC < 1500; platelets < 50,000; treatment with another antiviral agent Treatment group Number: 17 Mean age ± SEM: 43.8 ± 2.9 years Sex (M/F): 13/4 Control group Number: 15 Mean age ± SEM: 43.5 ± 3.3 years Sex (M/F): 6/9
Interventions	Treatment group Ganciclovir: 5 mg/kg IV twice/d for 14 days starting day 14 post‐transplant Control group No treatment Co‐interventions Not reported
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture, IgM All‐cause death Acute rejection Graft loss
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none reported Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information available
Allocation concealment (selection bias)	Unclear risk	Quote: "On day 14 after transplantation, patients were randomized...". No further information available
Blinding of participants and personnel (performance bias) All outcomes	High risk	Control group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Control group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients included in analyses
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No or limited reporting of opportunistic infections/adverse effects
Other bias	Low risk	Work supported in part by grants from non‐pharmaceutical sources

Rostaing 1994 Kidney.

Study characteristics
Methods	Study design Parallel RCT Time frame: April 1992 to February 1993 Follow‐up period: mean 12 months Loss to follow‐up: 0%
Participants	Study characteristics Country: France Setting: tertiary single centre Inclusion criteria: kidney transplant recipients; D/R+ Exclusion criteria: D+/R‐; D‐/R‐ recipients Treatment group Number: 19 Mean age ± SD: 50.4 ± 11.3 years Sex (M/F): 13/6 Control group Number: 18 Mean age ± SD: 45.1 ± 11.1 years Sex (M/F): 14/4
Interventions	Treatment group Aciclovir: 6 mg/kg/d IV for 3 days starting day 1 then aciclovir 800 mg oral 4 times/d for 3 months Control group No treatment Co‐interventions CSA, AZA, prednisone, ATG
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture All‐cause death Acute rejection Graft loss Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none reported Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information available
Allocation concealment (selection bias)	Unclear risk	Quote: "The patients were randomized to receive either acyclovir or nothing...", no other information available
Blinding of participants and personnel (performance bias) All outcomes	High risk	Control group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Control group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients included in analysis
Selective reporting (reporting bias)	High risk	No data on opportunistic infections or adverse reactions
Other bias	Unclear risk	No information provided about pharmaceutical sponsorship

Rubin 2002 All.

Study characteristics
Methods	Study design Parallel RCT Time frame: November 1996 to January 1999 Follow‐up period: 12 months Loss to follow‐up: 0% of evaluated patients
Participants	Study characteristics Country: USA Setting: tertiary multicentre (13 sites) Inclusion criteria: first kidney, liver or heart transplant recipients aged >12 years; D+/R‐ Exclusion criteria: D/R+; D‐/R‐ Treatment group Number: 77 Mean age ± SD: 46 ± 13 years Sex (M/F): 60/17 Control group Number: 78 Mean age ± SD: 45 ± 12 years Sex (M/F): 61/17
Interventions	Treatment group Ganciclovir: 5 mg/kg/d IV for 5 to 10 days starting within 72 hours of transplant, then ganciclovir 1000 mg oral 3 times/d to 12 weeks Control group Ganciclovir: 5 mg/kg/d IV for 5 to 10 days starting within 72 hours of transplant then aciclovir 400 mg oral 3 times/d to 12 weeks Co‐interventions CSA (141), TAC (27), AZA (57), MMF (101), antibody therapy (56)
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV antigenaemia, CMV culture All‐cause death Acute rejection Opportunistic infections Adverse effects Time to CMV disease
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported 11 (5 acyclovir, 6 ganciclovir) were deemed unable to be evaluated: 7 did not qualify for protocol, 1 died, 3 lost to follow‐up
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Stratification for organ transplanted. Central randomisation. Otherwise no information available
Allocation concealment (selection bias)	Low risk	Central randomisation
Blinding of participants and personnel (performance bias) All outcomes	High risk	Patients received different oral medications. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Patients received different oral medications. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	11/166 excluded from analyses. Reasons for missing data unlikely to be related to true outcome
Selective reporting (reporting bias)	High risk	Incomplete reporting of outcomes. No report of graft loss
Other bias	High risk	Funded in part by a grant from F. Hoffman‐LaRoche

Saliba 1993 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: February 1990 to February 1991 Follow‐up period: 3 months Loss to follow‐up: 0%
Participants	Study characteristics Country: France Setting: tertiary single centre Inclusion criteria: liver transplant recipients; D/R+ Exclusion criteria: D+/R‐; D‐/R‐ recipients Treatment group Number: 60 Mean age ± SD: 45.3 ± 12 years Sex (M/F): 36/24 Control group Number: 60 Mean age ± SD: 44.5 ± 13 years Sex (M/F): 35/35
Interventions	Treatment group Aciclovir: 500 mg/m²/d IV for 10 days, then 800 mg oral 4 times/d to 3 months Control group No treatment Co‐interventions CSA, AZA, prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV infection: CMV culture Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information available
Allocation concealment (selection bias)	Low risk	Adequate allocation concealment (information from authors)
Blinding of participants and personnel (performance bias) All outcomes	High risk	Control group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Control group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	Consecutive recruitment. All patients included in analyses
Selective reporting (reporting bias)	Unclear risk	Abstract only
Other bias	Unclear risk	No information provided on pharmaceutical sponsorship

Winston 1995 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: not reported Follow‐up period: 4 months Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: first liver transplant recipients aged > 12 years; all serologies Exclusion criteria: second transplants Treatment group Number: 124 Mean age (range): 52 years (20 to 72) Sex (M/F): 67/57 Control group Number: 126 Mean age (range): 47 years (20 to 74) Sex (M/F): 67/59
Interventions	Treatment group Ganciclovir: 6 mg/kg/d IV to day 30; ganciclovir 6 mg/kg/d IV Monday to Friday to day 100 Control group Aciclovir: 10 mg/kg IV 8 hourly until discharge; aciclovir 800 mg oral 4 times/d to day 100 Co‐interventions CSA, TAC (38), AZA, prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV invasive organ disease CMV infection: CMV culture, isolation from any site All‐cause death Death due to CMV disease Acute rejection Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomisation stratified according to CMV status but no other information provided
Allocation concealment (selection bias)	Unclear risk	No information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Patients given different medications by different routes. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Patients given different medications by different routes. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients included in analyses
Selective reporting (reporting bias)	Low risk	All expected outcomes reported
Other bias	High risk	Supported in part by non‐pharmaceutical grants. Ganciclovir from Syntex Research

Winston 2003 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: not reported Follow‐up period: 12 months Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: liver transplant recipients; D/R+ Exclusion criteria: D+/R‐; D‐/R‐ recipients Treatment group Number: 110 Mean age (range): 51 years (7 to 78) Sex (M/F): 58/52 Control group Number: 109 Mean age (range): 51 years (7 to 71) Sex (M/F): 58/51
Interventions	Treatment group Ganciclovir: 6 mg/kg/d IV to day 14 starting day of transplant; ganciclovir 1000 mg oral 3 times/d to day 100 Control group Ganciclovir: 6 mg/kg/d IV to day 14 starting day of transplant; aciclovir 800 mg oral 4 times/d to day 100 Co‐interventions CSA (58), TAC (164), AZA (128), MMF (85), prednisone
Outcomes	Outcomes relevant to this review CMV disease: CMV DNA, CMV culture CMV syndrome CMV tissue invasive disease All‐cause death Death due to CMV disease Acute rejection Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: unclear Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided
Allocation concealment (selection bias)	Unclear risk	"Patients were assigned randomly" but no other information available
Blinding of participants and personnel (performance bias) All outcomes	High risk	Different interventions given to groups with different dose frequency. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Different interventions given to groups with different dose frequency. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients included in analyses
Selective reporting (reporting bias)	High risk	Incomplete reporting of outcomes. No report of CMV infection and graft loss
Other bias	High risk	Supported in part by a research grant from Roche Laboratories

Winston 2004 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: June 1997 to April 2000 Follow‐up period: 1 year Loss to follow‐up: 0%
Participants	Study characteristics Country: USA Setting: tertiary single centre Inclusion criteria: liver transplant recipients; D+/R‐ Exclusion criteria: D/R+; D‐/R‐ Treatment group Number: 32 Mean age (range): 49 years (13 to 67) Sex (M/F): 24/8 Control group Number: 32 Mean age (range): 46 years (6 to 73) Sex (M/F): 23/9
Interventions	Treatment group Ganciclovir: 6 mg/kg IV daily days 1 to 14; ganciclovir 1000 mg oral 3 times/d on days 15 to 86 Control group Ganciclovir: 6 mg/kg IV daily days 1 to 14; ganciclovir 6 mg/kg IV Monday to Friday from days 15 to 86 Co‐interventions CSA (10), TAC (54), MMF (29), AZA (3), prednisone
Outcomes	Outcomes relevant to this review CMV disease CMV syndrome CMV tissue invasive disease All‐cause death Opportunistic infections Adverse effects
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information available
Allocation concealment (selection bias)	Unclear risk	"Randomized controlled trial" in title but no other information provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients followed for 1 year or until death
Selective reporting (reporting bias)	High risk	Incomplete outcome reporting. No report of CMV infection, graft loss
Other bias	Unclear risk	Supported in part by research grant from Roche Laboratories

Winston 2012 Liver.

Study characteristics
Methods	Study design Parallel RCT Time frame: July 2007 to February 2009 Follow‐up period: 6 months Loss to follow‐up: 70 patients (23%)
Participants	Study characteristics Country: USA and Europe Setting: multicentre (70 sites) Inclusion criteria: orthotopic liver transplants; ≥ 18 years; CMV‐seronegative with a CMV‐seropositive donor (D+/R‐) Exclusion criteria: history of CMV organ disease within 6 months prior to enrolment; treatment for CMV at time of enrolment; estimated CrCl < 10 mL/min; need for dialysis; infection with HIV; mechanical ventilation or other serious illness precluding study compliance; undergoing repeat liver transplantation or multiorgan transplantation Treatment group Number: 156 Mean age: 53 (range: 19 to 72) years Sex (M/F): 119/37 Control group Number: 147 Mean age: 55 (range: 19 to 71) years Sex (M/F):120/27
Interventions	Treatment group Maribavir (oral): 100 mg twice/d for 14 weeks Acyclovir (oral): 400 mg twice/d for 14 weeks Control group Ganciclovir (oral): 1000 mg 3 times/d for 14 weeks Co‐interventions None
Outcomes	Outcomes relevant to this review CMV disease All‐cause death CMV infection Acute rejection Harms (nephrotoxicity, bone marrow suppression, emergence of resistant CMV strains, late onset of CMV disease)
Notes	Additional information Exclusions post‐randomisation but pre‐intervention: none Stop or endpoint: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Eligible patients were randomly assigned (1:1) to receive either oral maribavir (100 mg twice daily) with oral acyclovir (400 mg twice daily) or oral ganciclovir alone (1000 mg three times daily)." No clear description of the methods for the random sequence generation
Allocation concealment (selection bias)	Unclear risk	No clear description of the methods for the allocation concealment
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "Allocation was concealed by matching placebo pills for either maribavir and acyclovir or ganciclovir" Identical placebo pills are considered adequate method for blinding participants and personnel
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No clear description of the blinding of the outcome assessment
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote: "Seventy patients (36 ganciclovir, 34 maribavir) did not meet the criteria for inclusion in the ITT‐M population" 23% of the patient data missing in the intervention arm and control arm for the primary outcome
Selective reporting (reporting bias)	Unclear risk	No protocol or registration identified to assess selective reporting
Other bias	High risk	Quote: "All investigators received research funding from ViroPharma to conduct the trial" High risk of bias due to funding of pharmaceutical industry

AIDS: acquired immunodeficiency syndrome; ALG: antilymphocyte globulin; AKI: acute kidney injury; ATG: antithymocyte globulin; AZA: azathioprine; CD/LD: cadaveric donor/living donor; CMV: cytomegalovirus; CMVIgG: cytomegalovirus gamma G immunoglobulin; CSA: cyclosporin; D/R+: donor CMV positive or negative/recipient CMV positive; D+/R‐: donor CMV positive/recipient CMV negative; D‐/R‐: donor CMV negative/recipient CMV negative; Cr: creatinine; CrCl: creatinine clearance; DNA: deoxyribonucleic acid; GFR: glomerular filtration rate; HIV: human immunovirus; IgG: immunoglobulin G; IgM: immunoglobulin M; IgM 3: immunoglobulin M 3; IL2Ra: interleukin 2 receptor alpha; IQR: interquartile range; ITT: intention‐to‐treat; IV: intravenous; M/F: male/female; MMF: mycophenolate mofetil; OKT‐3: monoclonal anti CD3 antibody; PCR: polymerase chain reaction; RCT: randomised controlled trial; SD: stardard deviation; SEM: standard error of the mean; TAC: tacrolimus; WBC: white blood cell

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Basic‐Jukic 2019	RCT comparing different formulations of the same medication (valganciclovir)
Brennan 1997a Kidney	Wrong intervention: pre‐emptive study
CYTOCOR 2019	Wrong intervention: pre‐emptive treatment for 6 months (after 6 months prophylaxis) with immunoguided prophylaxis for 9 months (after 3 months prophylaxis)
First 1993a	Wrong study design: pharmacokinetic study
Gerna 2008 Liver	Wrong intervention: pre‐emptive treatment compared with prophylaxis
Griffiths 2016	Wrong intervention: pre‐emptive treatment
Ishida 2017	Wrong intervention: combination of 2 monoclonal antibodies
Jarque 2020	Wrong intervention: pre‐emptive treatment
Jung 2001 Kidney	Wrong intervention: pre‐emptive treatment
Khoury 2006 Kidney	Wrong intervention: pre‐emptive treatment
Kiser 2012	Wrong study design: pharmacokinetic study
Kliem 2008 Kidney	Wrong intervention: pre‐emptive treatment
Koetz 2001 kidney/Liver	Wrong intervention: pre‐emptive treatment
Marker 1980	Wrong intervention: not prophylaxis of CMV disease
Mattes 2004	Wrong intervention: 2 pre‐emptive treatments
Murray 1997	Wrong intervention: pre‐emptive treatment
NCT00566072	Wrong intervention: compares different methods to encourage compliance
Padulles 2016	Wrong intervention: pre‐emptive treatment
Papanicolaou 2018	Wrong population: hemopoietic stem cell transplant patients
Paya 2002 Liver	Wrong intervention: pre‐emptive treatment
Pescovitz 2009	Wrong study design: pharmacokinetic study
PROTECT 2010	Wrong intervention: pre‐emptive treatment versus prophylaxis
Queiroga 2003 Kidney	Wrong intervention: pre‐emptive treatment
Rayes 2001 Liver	Wrong intervention: pre‐emptive treatment
Reischig 2008 Kidney	Wrong intervention: pre‐emptive treatment
Sagedal 2003 Kidney	Wrong intervention: pre‐emptive treatment
Scott 2011 Liver	Wrong intervention: pre‐emptive treatment
Singh 1994 Liver	Wrong intervention: pre‐emptive treatment
Singh 2000 Liver	Wrong intervention: pre‐emptive treatment
Singh 2019 Liver	Wrong intervention: prophylaxis versus pre‐emptive treatment
VICTOR 2007	Wrong intervention: not prophylaxis
Westall 2018	Wrong intervention: diagnostic tests (CMV specific immune monitoring)
Witzke 2012 Kidney	Wrong intervention: pre‐emptive treatment
Yang 1998 Kidney	Wrong intervention: pre‐emptive treatment

RCT: randomised controlled trial

Characteristics of studies awaiting classification [ordered by study ID]

Limaye 2023.

Methods	Phase III, randomised, double‐blind, active comparator‐controlled study
Participants	Inclusion criteria Documented negative serostatus for CMV within 180 days prior to randomisation Anticipate receiving a primary or secondary allograft kidney from a CMV IgG D+ donor at the time of screening AND have received a primary or secondary allograft kidney from a documented D+ donor at the time of randomization. Be within 0 to 7 days (inclusive) post‐kidney transplant at the time of randomisation Males: agree to use contraception during the treatment period and for at least 90 days after the last dose of study treatment and refrain from donating sperm during this period Females: not pregnant or breastfeeding, and is not a woman of childbearing potential or agrees to follow the contraception guidance during the treatment period and for at least 90 days after the last dose of study treatment Exclusion criteria Received a previous non‐kidney solid organ transplant or hematopoietic stem cell transplant Multi‐organ transplant recipient (e.g. kidney‐pancreas) History of CMV disease or suspected CMV disease within 6 months prior to randomisation Suspected or known hypersensitivity to active or inactive ingredients of letermovir formulations, valganciclovir, ganciclovir, and/or acyclovir formulations On dialysis or plasmapheresis at the time of randomisation Child‐Pugh Class C severe hepatic insufficiency at screening Moderate hepatic insufficiency AND moderate‐to‐severe renal insufficiency at screening Has any uncontrolled infection on the day of randomisation Documented positive HIV‐Ab at any time or HVC‐ab or HBsAg test within 90 days prior to randomisation Requires mechanical ventilation or is hemodynamically unstable at the time of randomisation History of malignancy ≤ 5 years prior to signing informed consent Pregnant or expecting to conceive, is breastfeeding, or plans to breastfeed from the time of consent through at least 90 days following cessation of study therapy, is expecting to donate eggs or sperm starting from the time of consent through at least 90 days following cessation of study therapy Received within 30 days prior to randomisation or plans to receive during the study any of the following anti‐CMV IgG antibody treatment or anti‐CMV drug therapy Has received within 7 days prior to randomisation or plans to receive during the study any anti‐CMV drug therapy User of recreational or illicit drugs or has had a recent history (within the last year) of drug or alcohol abuse or dependence Currently participating or has participated in a study with an unapproved investigational compound or device within 28 days, or 5 × half‐life of the investigational compound, whichever is longer, of initial dosing on this study Previously participated in this study or any other study involving letermovir Previously participated or is currently participating in any study involving the administration of a CMV vaccine or another CMV investigational agent, or is planning to participate in a study of a CMV vaccine or another CMV investigational agent during the course of this study
Interventions	Treatment group 1 Letermovir: 480 mg (or 240 mg when administered concomitantly with CSA) tablet orally once/d Placebo to valganciclovir tablet orally once/d Acyclovir: 400 mg capsule of acyclovir orally every 12 hours for 28 weeks Treatment group 1 Valganciclovir: 900 mg tablet orally, once/d Placebo to letermovir tablet orally once/d Placebo to acyclovir orally every 12 hours for 28 weeks
Outcomes	Planned outcomes Percentage of participants with adjudicated CMV disease through 52 weeks post‐transplant Percentage of participants with adjudicated CMV disease through 28 weeks post‐transplant Time to onset of adjudicated CMV disease through 52 weeks post‐transplant Percentage of participants with any adverse event up to 52 weeks Percentage of participants with any drug‐related serious adverse event up to 52 weeks
Notes	Actual study start date: 3 May 2018 Actual primary completion date: 5 April 2022 Actual study completion date: 5 April 2022

Verghese 2022.

Methods	Double blinded placebo controlled RCT
Participants	Inclusion criteria Approved as a kidney transplant donor with a recipient who has never undergone a previous transplantation Kidney transplant donor ≥ 18 years old The kidney transplant donor must be positive for CMV IgG and / or EBV IgG The donor must be to a recipient that is discordantly seronegative for the virus for which the donor is seropositive (D+ R‐) They must have provided signed informed consent The potential donors must be willing to contribute samples of blood and oral washings at regular intervals The potential donor must state willingness to use effective contraception during treatment and 30 days following receiving the study drug/placebo All females must have a negative pregnancy test Person must have estimated creatinine clearance (Cockcroft and Gault method) >= 60 ml/min Person must have absolute neutrophil count >= 1000 cells/uL Person must have platelets >= 100,000/uL Person must have Hemoglobin >= 9.5 g/dL Exclusion criteria Any potential kidney transplant donor who is seronegative for both CMV & EBV IgG Any potential kidney transplant donor who is receiving or have received anti‐herpes medication in the past week Any potential kidney transplant donor to a recipient who has received a previous solid organ transplant Any potential kidney transplant donor who is immunosuppressed due to medical disease and/or immunosuppressive or immunomodulating medications Any potential kidney transplant donor who is breast feeding during the study Any potential kidney transplant donor who is on corticosteroids
Interventions	Treatment group Kidney transplant donors randomised to receive 450 mg of valganciclovir twice/d for 14 days prior to the transplant date Control group Kidney transplant donors randomised to receive 1 placebo in the morning and 1 in the evening for 14 days prior to transplant date
Outcomes	Planned outcomes Incidence of EBV or CMV related disease in transplant recipient
Notes	Study start date: June 2011 Actual primary completion date: April 2014 Actual study completion date: April 2014 Abstract‐only publication Study was unblinded 2 months following the enrollment of the final donor, when a recipient developed post‐transplant lymphoproliferative disorder with evidence of EBV infection at the single cell level by detection of EBV encoded small nuclear RNA

CMV: cytomegalovirus; EBV: Epstein‐Barr virus

Characteristics of ongoing studies [ordered by study ID]

NCT04225923.

Study name	A study for kidney transplant recipients at high‐risk of cytomegalovirus infection
Methods	Study design Parallel RCT Time frame: June 2020 to February 2023 Follow‐up period: not yet known Loss to follow‐up: not yet known
Participants	Study characteristics Country: USA and Japan Setting: not yet known Inclusion criteria: adult kidney transplant recipients CMV seronegative pre‐transplant and scheduled to receive or have received (within 7 days prior to first study drug administration) a first kidney transplant from a CMV seropositive donor. Planned enrolment: 87 participants Exclusion criteria Received a previous solid organ transplantation or hematopoietic stem cell transplantation or a multi‐organ transplant. CMV disease or CMV viremia at screening. Positive donor‐specific antibody within 90 days prior to randomisation confirmed via medical records Body weight > 120 kg at screening Have received the following anti‐CMV therapy within 7 days prior to randomisation and/or plan to receive the following anti‐CMV therapy during the study: Anti‐CMV agents (eg, foscarnet, ganciclovir, valganciclovir, letermovir, high dose acyclovir, high dose valacyclovir, high dose famciclovir, or cidofovir). Patients who have received the following therapy within 28 days prior to randomisation and/or plan to receive the following anti‐CMV therapy during the study: CMV hyperimmune globulin (eg, CytoGam) Intravenous immunoglobulin Plasmapheresis (receipt prior to first study drug administration is acceptable) History of a serious drug allergy to proteins, immunoglobulins, transfusions, or vaccines or any excipient of the NPC‐21 formulation Severe hepatic insufficiency at Screening (eg, Child‐Pugh Class C) Active and untreated hepatitis B virus or hepatitis C virus, as documented as part of the pre‐transplant screening Known HIV infection, based on medical records serology Uncontrolled infection at randomisation or a history of serious and uncontrolled infection within 6 months prior to randomisation Pregnant or lactating History of malignancy within 5 years prior to randomisation other than curatively treated in situ cervical carcinoma, cutaneous basal cell carcinoma, or cutaneous squamous cell carcinoma. History of alcohol or drug abuse or dependence within 1 year prior to randomisation that, in the opinion of the Investigator, would preclude study participation Previously participated in this study or any other study involving NPC‐21 Previously participated or are currently participating in any study involving the administration of a CMV vaccine or another CMV investigational agent Participated in another interventional clinical study and received another investigational product (ie, not approved by the Food and Drug Administration in the United States or the Ministry of Health, Labour and Welfare in Japan) within 90 days before randomisation Unable or unwilling, in the opinion of the Investigator, to comply with the protocol
Interventions	Treatment group 1 NPC‐21 low dose Treatment group 2 NPC‐21 high dose Control group NPC‐21 placebo
Outcomes	Planned outcomes Incidence of CMV disease or CMV viremia Time to detectable CMV disease or CMV viremia Amount of CMV DNA Incidence and duration of anti‐CMV therapy during the rescue phase Changes in EQ‐5D‐5L score from baseline
Starting date	1 June 2020
Contact information	Responsible party: Nobelpharma
Notes	Actual primary completion date: 2 November 2022 Actual study completion date: 8 February 2023

CMV: cytomegalovirus; HIV: human immunodeficiency virus; RCT: randomised controlled trial

Differences between protocol and review

The previous version of this review was published in 2013 (Hodson 2013). Therefore, to allign this update with the current methodological standards of Cochrane reviews, substantial changes since the last version has been made, including the use of GRADE to assess the certainty of evidence per outcome and reporting of the results / conclusions accordingly. Specific issues are highlighted below.

• Two new comparisons added: 1) Comparison 10: Different valganciclovir regimens; 2) Comparison 11: Maribavir versus ganciclovir. Including new SoFs tables.

• GRADE: We have reported the results / concluding statements according to the GRADE approach for all results.

• Kidney function added as outcome.

Contributions of authors

2024 review update

• RV, MM, and EMH contributed to the data extraction, quality assessment, data analysis and rewriting of the review update.

• ACW, ML, JCC, and GFMS contributed to editing the draft update and revision of the manuscript.

2013 review update

• EMH, ML, ACW and JCC contributed to the data extraction, quality assessment, data analysis and rewriting of the review update.

2008 review update

• EMH, ACW, JCC, and GFMS contributed to the data extraction, quality assessment, data analysis and rewriting of the review update.

2005 review

• EMH identified and extracted data from included studies, contacted authors, analysed and interpreted the results and wrote the manuscript.

• CAJ conceived, designed and developed the protocol and search strategy for the review, identified and extracted data from included studies and participated in revising the manuscript.

• ACW analysed and interpreted the results and participated in revising the manuscript.

• GFMS checked the analysis and interpretation of the results and participated in revising the manuscript.

• PGB and KK identified and extracted data from included studies and participated in revising the manuscript.

• DV developed the protocol and search strategy for the review.

• JCC conceived, designed and developed the protocol, analysed and interpreted the results and edited the drafting and revision of the manuscript.

Sources of support

Internal sources

• No sources of support provided

External sources

• No sources of support provided

Declarations of interest

• Robin Vernooij: no relevant interests were disclosed

• Mini Michael: no relevant interests were disclosed

• Maleeka Ladhani: no relevant interests were disclosed

• Angela Webster: no relevant interests were disclosed

• Giovanni Strippoli: no relevant interests were disclosed

• Jonathan Craig: no relevant interests were disclosed

• Elisabeth Hodson: no relevant interests were disclosed

New search for studies and content updated (conclusions changed)