Delayed initiation or reduced initial dose of calcineurin‐inhibitors for kidney transplant recipients at high risk of delayed graft function

Abstract

Background

Kidney transplantation is the preferred therapy for many patients with kidney failure. Delayed graft function (DGF) is more common in donors after cardiac death (DCD), especially those with older age, longer cold ischemia time, or higher creatinine levels. Currently, there is no agreement on the optimal immunosuppressive approach for patients at increased risk of DGF. Strategies include delaying the introduction of calcineurin inhibitors (CNI) or using an initial low dose of CNI.

Objectives

To evaluate the benefits and harms of delayed initiation of CNI or reduced CNI dose as initial immunosuppression therapy for kidney transplant recipients at high risk of DGF.

Search methods

The Cochrane Kidney and Transplant Register of Studies was searched up to 11 December 2024 through contact with the Information Specialist 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

All randomised controlled trials (RCTs) and quasi‐RCTs evaluating delayed versus early initiation of CNI or reduced versus standard initial dose of CNI in kidney transplant recipients at high risk of DGF.

Data collection and analysis

Three authors independently assessed study eligibility, and two assessed the risk of bias, certainty of evidence, extracted the data, and performed the analysis. Results were reported as risk ratios (RR) with 95% confidence intervals (CI) for dichotomous outcomes and as mean difference (MD) with 95% CI for continuous outcomes. Statistical analysis was performed using the random‐effects model. Risk of bias was assessed with the Cochrane risk of bias assessment tool 1.0, and the certainty of the evidence according to the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) methods, which are presented in the summary of findings tables.

Main results

We included 12 studies (2230 randomised participants). All studies were performed in Europe. Around 60% of the participants were males, reflecting the expected proportion in the population on kidney replacement therapy in Europe. Most studies had insufficient information to judge adequate random sequence generation and, or allocation concealment. All studies were unblinded, and judged as high risk of bias for DGF if the definition was based on need for dialysis, and for acute rejection if the diagnosis did not require a biopsy. Overall, the level of certainty was low, and reasons to downgrade were mainly due to risk of bias and imprecision.

Delayed versus early initiation of CNI

There may be little or no difference in DGF between the groups (6 studies, 905 recipients: RR 0.92, 95% CI 0.76 to 1.12; low certainty evidence) or in acute rejection (8 studies, 1295 recipients: RR 1.02, 95% CI 0.75 to 1.40; low certainty evidence).

Delaying the initiation of CNI probably makes little or no difference to eGFR (6 studies, 851 recipients: MD ‐0.81 mL/min, 95% CI ‐3.33 to 1.72; moderate certainty evidence).

Delaying the initiation of CNI may make little or no difference to graft loss censored for death (8 studies, 1295 recipients: RR 1.58, 95% CI 0.68 to 3.65; very low certainty evidence) or to all‐cause death (8 studies, 907 recipients: RR 1.08, 95% CI 0.54 to 2.14; very low certainty evidence) although the evidence is very uncertain.

There is probably little or no difference in all infections between the groups (6 studies, 1226 recipients: RR 1.10, 95% CI 0.97 to 1.25; moderate certainty evidence).

Low versus standard initial dose of CNI

There may be little or no difference to DGF between the groups (5 studies, 983 recipients: RR 1.16, 95% CI 0.90 to 1.50; low certainty evidence) or in acute rejection (5 studies, 947 recipients: RR 0.83, 95% CI 0.52 to 1.30; low certainty evidence).

Starting CNI at a lower dose may make little or no difference to eGFR (5 studies, 935 recipients: MD 4.06 mL/min, 95% CI ‐1.36 to 9.48, low certainty evidence).

Starting CNI at a lower dose may make little or no difference to graft loss censored for death, although the evidence is very uncertain (5 studies, 983 recipients: RR 1.05, 95% CI 0.64 to 1.71; very low certainty evidence), or to all‐cause death (4 studies, 521 recipients: RR 1.01, 95% CI 0.41 to 2.47; low certainty evidence).

There is probably little or no difference in all infections between the groups (4 studies, 828 recipients: RR 0.87, 95% CI 0.71 to 1.07; moderate certainty evidence).

Authors' conclusions

There may be little or no difference in DGF or acute rejection when delaying the start of CNI or when starting it at a lower dose in kidney transplant recipients at high risk of DGF. The available data are of low certainty.

Plain language summary

Does delaying the initiation or starting calcineurin inhibitors at a lower dose reduce the risk of delayed recovery of a transplanted kidney?

Key messages

• After kidney transplantation (a surgical procedure where a kidney is transferred from one person ‐ the donor ‐ to another person ‐ the recipient), it may take a few days for the kidney to start working, especially if the kidney donor was older or the kidney was without blood flow for a long time.

• The best treatment for people at high risk of this delayed recovery is still unclear. Current options include delaying the use of a medication to prevent rejection, called a calcineurin inhibitor, or starting it at a lower dose.

What is the treatment for kidney failure?

Kidney transplantation (a surgical procedure where a kidney is transferred from one person ‐ the donor ‐ to another person ‐ the recipient) is the preferred treatment for most people with kidney failure because it improves quality of life and eliminates the need for dialysis (a procedure where a machine is used to remove waste products and excess salts and fluid from the blood when the kidneys stop working properly). After transplantation, it may take a few days for the transplanted kidney to start working. This condition is called delayed graft function. This delay is more likely when the donor kidney comes from someone older (usually older than 50 years of age) or who died from a heart condition, or when the kidney has been stored for a long time without blood flow. The best treatment for people at high risk of this delayed recovery is not yet known. Some approaches include delaying the start of calcineurin inhibitors (an immunosuppressive medicine that dampens down the body's immune system and prevents it from attacking the transplanted kidney) or using them at a lower dose.

What did we want to find out?

We wanted to know if delaying the initiation or starting calcineurin inhibitors at a lower dose can reduce the risk of delayed recovery of the kidney.

What did we do?

We searched for all studies that compared the benefits and harms of randomly allocating the delayed start of calcineurin inhibitors or using them at a lower dose in people who have received a kidney transplant and are at risk of delayed kidney function immediately after receiving the transplanted kidney.

We compared and summarised the studies' results and rated our confidence in the information based on factors such as study methods and sizes.

What did we find?

We reviewed 12 studies, with 2230 randomised participants, which compared delayed versus early initiation or reduced versus standard initial dose of calcineurin inhibitors for kidney transplant recipients at high risk of delayed recovery of the kidney.

All studies were performed in Europe. Around 60% of the participants were males, reflecting the expected proportion in the population on kidney replacement therapy in Europe. There may be little to no difference in delayed recovery of the kidney or acute rejection risk between any of the evaluated strategies. There may also be little or no differences found in kidney function, risk of death, or infection between the groups.

What are the limitations of the evidence?

We have little confidence in the evidence because studies used different types of additional treatments, and most studies were published more than 10 years ago.

How up to date is this evidence?

The review is current to 11 December 2024.

Summary of findings

Summary of findings 1. Delayed versus early calcineurin inhibitor (CNI) initiation for kidney transplant recipients at high risk of delayed graft function (DGF).

Delayed versus early calcineurin inhibitor (CNI) initiation for kidney transplant recipients at high risk of delayed graft function (DGF)
Patient or population: kidney transplant recipients at high risk of DGF Setting: kidney transplant units Intervention: delayed CNI (usually defined after day 3 post‐transplant or following evidence of improved graft function) Comparison: early CNI (within first 3 days)
Outcomes	Anticipated absolute effects (95% CI)		Relative effect (95% CI)	Nº of participants (studies)	Quality of evidence	Commentary
	Risk with early CNI	Risk with delayed CNI
DGF Follow‐up: 1 year	309 per 1000	334 per 1000	RR 0.92 (0.76 to 1.12)	905 (6)	⊕⊕⊝⊝ low1	Delayed initiation of CNI may make little or no difference to DGF.
Acute rejection (clinical) Follow‐up: 1 year	267 per 1000	256 per 1000	RR 1.02 (0.75 to 1.40)	1295 (8)	⊕⊕⊝⊝ low2	Delayed initiation of CNI may make little or no difference to acute rejection.
eGFR (mL/min) Follow‐up: 1 year	The mean eGFR in the control group ranged from 45 to 66.5	Difference: MD 0.81 lower (3.33 lower to 1.72 higher)	‐‐	851 (6)	⊕⊕⊕⊝ moderate3	Delayed initiation of CNI probably makes little or no difference to eGFR.
Graft loss censored for death Follow‐up: 1 year	47 per 1000	71 per 1000	RR 1.58 (0.68 to 3.65)	1295 (8)	⊕⊝⊝⊝ very low3,4	Delayed initiation of CNI may make little or no difference to graft loss censored for death, but the evidence is very uncertain.
Death (all causes) Follow‐up: 1 year	30 per 1000	27 per 1000	RR 1.08 (0.54 to 2.14)	1280 (8)	⊕⊝⊝⊝ very low3,4	Delayed initiation of CNI may make little or no difference to death (all causes), but the evidence is very uncertain.
Infection (all) Follow‐up: 1 year	658 per 1000	606 per 1000	RR 1.10 (0.97 to 1.25)	1226 (6)	⊕⊕⊕⊝ moderate3	Delayed initiation of CNI probably makes no difference to infection (all).

*The risk in the intervention group (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, CMV: cytomegalovirus, eGFR: estimated glomerular filtration rate, MD: mean difference, RR: risk ratio

_____________________________________

GRADE Working Group grades of evidence

High certainty: We are very confident that the true effect lies close to that of the estimate of the effect

Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different

Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect.

Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

_____________________________________

¹ Due to very serious risk of bias (downgraded 2 levels): All studies were unblinded and only one based the DGF definition based on creatinine levels. Most of them did not report details about random sequence generation or allocation concealment.

² Due to very serious risk of bias (downgraded 2 levels): All studies were unblinded. Most of them did not report details about random sequence generation or allocation concealment.

³ Due to serious risk of bias (downgraded 1 level): most of them did not report details about random sequence generation or allocation concealment.

⁴ Due to very serious risk of imprecision (downgraded 2 levels): total number of events fewer than 100, wide confidence interval.

Summary of findings 2. Low versus standard calcineurin inhibitors (CNI) initial dose for kidney transplant recipients at high risk of delayed graft function (DGF).

Low versus standard calcineurin inhibitors (CNI) initial dose for kidney transplant recipients at high risk of delayed graft function (DGF)
Patient or population: kidney transplant recipients at high risk of DGF Setting: kidney transplant units Intervention: low CNI dose (as defined by the authors) Comparison: standard CNI dose (as defined by the authors)
Outcomes	Anticipated absolute effects (95% CI)		Relative effect (95% CI)	Nº of participants (studies)	Quality of evidence	Commentary
	Risk with standard CNI dose	Risk with low CNI dose
DGF Follow‐up: 2 years	217 per 1000	233 per 1000	1.16 (0.90 to 1.50)	983 (5)	⊕⊕⊝⊝ low1,2	Low CNI dose may make little or no difference to DGF.
Acute rejection (clinical) Follow‐up: 2 years	175 per 1000	141 per 1000	0.83 (0.52 to 1.30)	947(5)	⊕⊕⊝⊝ low1,2	Low CNI dose may make little or no difference to acute rejection (clinical).
eGFR (mL/min) Follow‐up: 2 years	The mean score in the control group ranged from 42.4 to 72.6	Difference: MD was 4.06 higher (1.36 lower to 9.48 higher)	‐‐	935 (5)	⊕⊕⊝⊝ low3,4	Low CNI dose may make little or no difference to eGFR.
Graft loss censored for death Follow‐up: 2 years	60 per 1000	70 per 1000	1.05 (0.64 to 1.71)	983 (5)	⊕⊝⊝⊝ very low3,5	Low CNI dose may make little or no difference to graft loss censored for death, but the evidence is very uncertain.
Death (all causes) Follow‐up: 2 years	44 per 1000	34 per 1000	1.01 (0.41 to 2.47)	521 (4)	⊕⊕⊝⊝ low5	Low CNI dose may make little or no difference to death.
Infection (all) Follow‐up: 2 years	582 per 1000	457 per 1000	0.87 (0.71 to 1.07)	828(4)	⊕⊕⊕⊝ moderate3	Low CNI dose probably makes little or no difference to infection (all).

*The risk in the intervention group (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, CMV: cytomegalovirus, eGFR: estimated glomerular filtration rate, RR: risk ratio, MD: mean difference.

_____________________________________

GRADE Working Group grades of evidence

High certainty: We are very confident that the true effect lies close to that of the estimate of the effect

Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different

Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect.

Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect.

_____________________________________

¹ Due to serious risk of bias (downgraded 1 level): All studies were unblinded, and two of them did not report details about random sequence generation or allocation concealment.

² Due to serious risk of imprecision (downgraded 1 level): total number of events fewer than 300.

³ Due to serious risk of bias (downgraded 1 level): two studies did not report details about random sequence generation or allocation concealment. Two studies were industry sponsored.

⁴ Due to serious risk of inconsistency (downgraded 1 level): I² = 81%.

⁵ Due to very serious risk of imprecision (downgraded 2 levels): total number of events fewer than 100, wide confidence interval.

Background

Description of the condition

The number of people requiring kidney replacement therapy (KRT) worldwide in 2017 was 3.9 million. This number is expected to reach 5.4 million by 2030 (Huijben 2023; Liyanage 2015). The preferred mode of KRT is kidney transplantation; however, the number of people needing a kidney transplant exceeds the number of available organs. To increase organ availability, transplant modalities have expanded to include living donation, paired kidney exchange programs, the inclusion of expanded criteria donors (ECD), and donors after controlled circulatory death (DCD) (Johansen 2021). ECDs are defined as donors who are either (1) ≥ 60 years old, or (2) 50 to 59 years old with two of the following criteria: hypertension, terminal serum creatinine (SCr) > 1.5 mg/dL or death from cerebrovascular accident (Kawakita 2020).

Delayed graft function (DGF) is an important clinical issue in the kidney transplantation setting and can vary between 2% and 50%, depending on whether it was a donor after brain dead (DBD) or DCD organ (Bahl 2019). The most common definitions of DGF are the need for postoperative dialysis during the first week after kidney transplantation or failure of SCr to decrease by at least 10% daily on three consecutive days during the first week post‐transplant (Montagud‐Marrahi 2021). Recipients at greater risk of DGF include those receiving a DCD‐kidney transplantation, especially ECD kidneys, and those with longer cold ischaemia time (CIT), greater donor body mass index (BMI), or longer duration of pre‐transplant dialysis (Irish 2010; Kawakita 2020; Weber 2002). DGF impacts both short‐ and long‐term outcomes by increasing rejection rates and treatment costs, and is associated with residual graft dysfunction (Bahl 2019).

The standard immunosuppressive therapy for de novo kidney transplantation recipients includes a combination of mycophenolic acid (MPA), calcineurin inhibitors (CNI), corticosteroids, and different induction treatments based on the stratification of immune rejection risk (Ekberg 2007; Halloran 2004; KDIGO 2009). However, for patients at increased risk of DGF, there is no consensus on the optimal immunosuppressive approach (Lebranchu 2002; Ruiz‐Martínez 2019). There are several meta‐analyses that evaluate the effect of delaying the start of CNI or starting at a lower dose (Karpe 2017; Moore 2009; Sharif 2011), but not in our population of interest: kidney transplant recipients at high risk of DGF.

Description of the intervention

As the first ever used CNI, cyclosporin (CsA) changed the standard of care with a substantial increase in graft survival one year after kidney transplantation by reducing the incidence of acute transplant rejection and decreasing early graft loss. However, the use of CsA is associated with vasoconstriction and transient reduction of renal plasma flow, causing more frequent and longer DGF and, in some cases, a never‐functioning graft (Lusco 2017). Tacrolimus (TAC) is increasingly used in kidney transplant recipients, and conversely, its effect on renal haemodynamics is mild (Klein 2002; Radermacher 1998).

There are no Food and Drug Administration‐approved therapies to mitigate the ischaemic injury that occurs after DGF. In cases of increased risk of DGF, delayed introduction of CNI in the immediate post‐kidney transplant period has been used as a strategy to minimise its potential nephrotoxic effect with subsequent appearance of acute tubular necrosis (Bentata 2020; Fioretto 2011; Naesens 2009; Nankivell 2016). In these cases, induction therapy with thymoglobulin is usually applied to avoid rejection due to insufficient immunosuppression. However, since the use of thymoglobulin (ATG) is related to an increased risk of complications due to over‐immunosuppression, the benefits of delayed introduction of CNI to avoid subsequent nephrotoxicity remain controversial (Gonwa 2002).

In recipients with a high immunological risk, the effects of delayed introduction or reduced initial CNI dose strategy to prevent DGF should be balanced against the risks of acute rejection and chronic antibody‐mediated rejection (Charpentier 2002).

How the intervention might work

Acute and chronic CNI nephrotoxicity appears in both kidney and non‐kidney transplantation (Naesens 2009), and it is mediated by different mechanisms. Vasoconstriction of the afferent arterioles has been shown with the use of CsA (Dieperink 1985; Murray 1985), and this effect contributes to an impairment of blood flow (Laskow 1990). Although TAC has a different chemical structure, its inhibition of the calcineurin pathway shows that its nephrotoxic properties are similar to those of CsA (Mihatsch 1998). This effect on renal vascular resistance and blood flow in the arterioles of a newly transplanted kidney graft contributes to an acute functional injury which is associated with pathologic contributors derived from the donor (ischaemic injury, inflammatory signalling) and the recipient (reperfusion injury, the innate immune response and the adaptive immune response) (Siedlecki 2011). Altogether, these factors lead to a more difficult tubular regeneration and, consequently, a progression to the clinical diagnosis of DGF.

These detrimental effects of initial CNI use have prompted many studies to investigate delayed initiation (Andres 2009; Lebranchu 2002; SENIOR 2009) or reduced initial CNI dose in the early post‐kidney transplant period (Kyllonen 2002). These comparisons have been made with different induction strategies (ATG or interleukin‐2 receptor antagonists).

Why it is important to do this review

This meta‐analysis will synthesise evidence from available studies of the benefits and harms of delayed introduction or reduced initial CNI dose on DGF and graft outcomes in kidney transplant recipients at risk of DGF. The results would be of value to clinicians, patients, and health policy‐makers regarding a possible strategy to reduce DGF and improve transplant outcomes.

Objectives

To evaluate the benefits and harms of delayed initiation of CNI or reduced CNI dose as initial immunosuppression therapy for kidney transplant recipients at high risk of DGF.

Methods

Criteria for considering studies for this review

Types of studies

We included 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 methods) that compared delayed versus early initiation of CNI, or low versus standard CNI initial dose.

Types of participants

Inclusion criteria

We included de novo kidney transplant recipients, regardless of age or gender, who received a transplant from a first or subsequent deceased donor, were treated with CNI as the primary immunosuppressive therapy, and were at high risk of DGF. High risk of DGF was defined by one or more of the following criteria.

• CIT ≥ 18 hours

• Donors after cardiac death

• ECD: defined as either (1) > 50 years old with two of the following characteristics: a history of high blood pressure, SCr > 1.5 mg/dL, or death resulting from a stroke, or (2) ≥ 60 years old

• Mean BMI ≥ 30 kg/m²

• Long duration of pre‐transplant dialysis (≥ 2 years)

• Increased SCr (donors with any degree of acute kidney injury).

Exclusion criteria

We excluded patients receiving multiorgan transplants (including simultaneous pancreas‐kidney transplants) or a living donor kidney transplant.

Types of interventions

• Delayed CNI introduction (usually defined after day three post‐transplant or following evidence of improved graft function) versus early CNI initiation.

• Reduced versus standard dose of CNIs (as defined by the authors).

Types of outcome measures

This review did not exclude studies based on non‐reporting of the outcomes of interest.

The outcomes selected include the relevant SONG core outcome sets as specified by the Standardised Outcomes in Nephrology initiative (SONG 2017).

Primary outcomes

• DGF (as defined by the authors): incidence and duration (days).

• Acute graft rejection, both clinical rejection and biopsy‐proven: incidence (primary time point of interest: years 1 and 2).

Secondary outcomes

• Graft function measured as an absolute value or change in SCr (µmol/L), glomerular filtration rate (GFR), or creatinine clearance (CrCl), with measurements taken at days 3, 7, 14, 30, months 2, 3, 6, and years 1, 2, 3 and 5 (primary time point of interest: years 1 and 2)

• Graft loss (censored and not censored for death): incidence and time to graft loss (months) (primary time point of interest: years 1 and 2)

• Death: incidence of and time to death (any cause) and incidence of and time to cause‐specific death (months) (primary time point of interest: years 1 and 2)

• Malignancy rates (at any time point)

• Cardiovascular events (at any time point)

• Incidence of infection: any cause, viral, bacterial, or fungal (at any time point)

• Incidence of anaemia or leukopenia, as defined by the authors (at any time point).

Search methods for identification of studies

Electronic searches

The Cochrane Kidney and Transplant (CKT) Register of Studies was searched by the CKT Information Specialist using search terms relevant to this review up to 11 December 2024. The Register contains studies identified from the following sources.

1. Cochrane Central Register of Controlled Trials (CENTRAL)

   • CENTRAL includes records from EMBASE, Clinicaltrials.gov and the International Clinical Trials Registry Platform (ICTRP) Search Portal.

2. Weekly searches of MEDLINE OVID SP using CKT's scope.

3. Handsearched records from kidney‐related journals up to 2021.

Studies contained in the Register are identified through search strategies for CENTRAL, MEDLINE, and EMBASE based on the scope of CKT. Details of these strategies, as well as a list of handsearched journals, conference proceedings and current awareness alerts, are available in the "Specialised Register" section of information about CKT (Cochrane Kidney and Transplant Register of Studies).

Additional search strategies have been designed by the Information Specialist for MEDLINE, EMBASE and CENTRAL Please see Appendix 1.

Searching other resources

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

2. Contacting relevant individuals/organisations seeking information about unpublished or incomplete studies

3. Grey literature sources (e.g. abstracts, dissertations and theses), as well as those already included in the CKT Register of Studies.

Data collection and analysis

Selection of studies

Three authors (LO, SC, EM) independently screened titles and abstracts and discarded studies that were not applicable. The three authors then assessed the full text of the retained studies to determine if they satisfied the inclusion criteria. Disagreements were resolved in consultation with a fourth author (NM).

Data extraction and management

Data extraction was carried out independently by two authors (LO, NM) using standard data extraction forms. Disagreements were resolved in consultation with a third author (EM). Studies reported in non‐English language journals were translated before being assessed. If more than one publication of a study existed, reports were grouped, and the publication with the most complete data was used in the analyses. Where relevant outcomes were only published in earlier versions, these data were used.

Assessment of risk of bias in included studies

The following items were assessed independently by two authors (LO, NM) using the Cochrane risk of bias assessment tool 1 (Higgins 2024) (see Appendix 2).

• 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 suggestion 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

Studies were grouped and analysed according to the following comparisons.

1. Delayed versus early initiation of CNI

2. Low versus standard initial dose of CNI.

For dichotomous outcomes (e.g. acute rejection, DGF, graft loss, death), results were expressed as risk ratios (RR) with 95% confidence intervals (CI). For continuous outcomes, the mean difference (MD) or standardised mean difference (SMD) if different scales were used. For the time‐to‐death analysis, we aimed to include direct estimates of the log hazard ratio and its standard error and perform a generic inverse variance meta‐analysis. However, all the included studies reported death only as a dichotomous outcome. In the case of missing standard deviations, they were obtained from standard errors and CIs for group means (Higgins 2024); however, imputation techniques were avoided.

Unit of analysis issues

Multiple intervention group studies were analysed with different methods: (1) using only the groups with the intervention of interest to create a single pairwise comparison (e.g. if there were three groups including different induction therapies, only one induction therapy was included) and (2) each pairwise comparison was included separately, but with shared intervention groups divided evenly amongst the comparisons. In the case of dichotomous outcomes, both the number of events and the total number of patients were divided. For continuous outcomes, only the total number of participants was divided, and the means and SDs were left unchanged (Higgins 2024).

Dealing with missing data

Missing information or data required from the original author was requested by electronic correspondence and any relevant information obtained in this manner was included in the review.

Evaluation of important numerical data such as screened, randomised patients, as well as intention‐to‐treat, as‐treated, and per‐protocol population were carefully performed. Attrition rates, for example, drop‐outs, losses to follow‐up, and withdrawals were investigated. Issues of missing data and imputation methods (e.g. last‐observation‐carried‐forward) were critically appraised (Higgins 2024).

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). A guide to the interpretation of I² values will be as follows.

• 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. P value from the Chi² test or a CI for I²) (Higgins 2024).

Assessment of reporting biases

We only investigated for publication bias using funnel plots if sufficient studies (10 studies) for each comparison were available (Higgins 2024).

Data synthesis

Data were summarised using the random‐effects model. The fixed‐effect model was also used to ensure the robustness of the model chosen.

Subgroup analysis and investigation of heterogeneity

Subgroup analysis by type of CNI (CsA or TAC) was performed to explore possible sources of heterogeneity, using the P value for subgroup differences provided by RevMan 2024. When evidence of subgroup differences was found, the results were summarised separately for each group in the results section.

Heterogeneity amongst participants could be related to age or sensitisation grade, considering all the definitions used by the authors (e.g. complement‐dependent cytotoxicity cross‐match assays, FlowPRA, Luminex, number of HLA mismatches, or first or retransplantation). Heterogeneity in treatments could be related to prior induction treatment used as well as the drug type (CsA or TAC), dose, and duration of therapy.

If a meta‐analysis of effect estimates was not possible, we planned to summarise effect estimates or combine P values.

Sensitivity analysis

Where possible, we performed sensitivity analyses to evaluate the effect on the overall result of removing studies with low methodological quality.

Summary of findings and assessment of the certainty of the evidence

The main results of the review are presented in the summary of findings (SoF) tables. These tables present key information concerning the certainty of the evidence, the magnitude of the effects of the interventions examined, and the sum of the available data for the main outcomes (Schunemann 2024a). The SoF tables also include an overall grading of the evidence related to each of the main outcomes using the Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) approach (GRADE 2008; GRADE 2011). The GRADE approach defines the certainty 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. This was assessed by two authors (LO, NM). The certainty of a body of evidence involves consideration of the within‐trial risk of bias (methodological quality), directness of the evidence, heterogeneity, the precision of effect estimates, and the risk of publication bias (Schunemann 2024b). We presented the following outcomes in the SoF tables at one year of follow‐up for the delayed versus early CNI, and up to two years of follow‐up for the low versus standard dose of CNI.

• DGF incidence

• Acute rejection (clinical)

• Graft function

• Graft loss censored for death

• Death (all causes)

• Incidence of infections (all).

Results

Description of studies

The following section contains brief 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 ongoing studies.

Results of the search

After searching the CKT’s Specialised Register (up to December 2024), we identified 769 reports, with one additional record identified through handsearching. After removing duplicates, 92 studies (568 records) remained. After title and abstract screening, 44 studies were excluded (wrong population (19), wrong intervention (25)). Following full‐text assessment, a further 35 studies were excluded (wrong population (23), wrong intervention (12)). One study is still ongoing. Twelve studies were included in the qualitative and quantitative synthesis (Figure 1).

1.

Figure 1. Study flow diagram

Included studies

See Characteristics of included studies.

We included 12 studies (53 reports) randomising 2230 kidney transplant recipients at high risk of DGF. Eight studies compared delayed versus early initiation of CNI (three with CsA and five with TAC), and five studies compared low versus standard initial dose of CNI (three with CsA and two with TAC). One study (Andres 2009) included three interventions (early low versus early standard versus delayed standard), allowing for the two comparisons. OSAKA 2013 included a heterogeneous population of kidney transplant recipients, but one report (Albano 2020) separated results between standard‐criteria donors and ECDs. We included only data from ECDs in the meta‐analysis. All studies were performed in Europe. Around 60% of the participants were males, reflecting the expected proportion in the population on KRT in Europe.

In studies comparing delayed versus early initiation of CNI, the starting day in the delayed group was defined by the day after transplantation (ranging from three to 10 days) or when a certain value of SCr was obtained (from less than 250 µmol/L to less than 350 µmol/L). Donors were mostly DBD (except Wilson 2005 who were all DCD) but with CIT longer than 18 hours or older age. Follow‐up time was between six and 12 months, except for one study that was three months (Wilson 2005). Concomitant medication varied between studies. In two studies, recipients in both groups received the same concomitant immunosuppression (Andres 2009; Mourad 2005). The remaining 11 studies had important differences in induction regimens for each group (basiliximab (BSX) or daclizumab versus ATG, alemtuzumab versus ATG, no induction versus BSX/daclizumab). Moreover, SENIOR 2009 had early steroid discontinuation (day 8) in the delayed group and later discontinuation (day 91) in the early group, and Charpentier 2002 received TAC as monotherapy after ATG in the delayed group.

In studies comparing low versus standard CNI initial dose, the definition of low or standard dose varied amongst studies. In CsA, the low dose ranged from 75 to 175 ng/mL, and the standard dose from 150 to 300 ng/mL. In TAC, the low dose was expressed as 0.2 mg/kg/d or a concentration of 3 to 7 ng/mL, and the standard dose was 0.3 mg/kg/d or a concentration of 10 to 15 ng/mL. Donors were mostly DBD but with CIT longer than 18 hours. There were also differences in the concomitant medication. Kyllonen 2002 used induction (ATG or BSX) in the low CsA group, while there was no induction in the standard dose group. Hernandez 2007 used BSX with low CsA and ATG with standard CsA. Most studies used mycophenolate mofetil (MMF) or azathioprine as concomitant immunosuppression, except for Bechstein‐193 2013, which used sirolimus. Follow‐up time was between six months and five years.

Excluded studies

See Characteristics of excluded studies.

In total, we excluded 79 studies (514 records). We excluded 44 studies during the screening phase and 35 studies after assessing the remaining full‐text articles for eligibility. In total, 42 studies were excluded because kidney transplant recipients were not at high risk of DGF, and 37 due to wrong intervention, mainly because they either 1) compared avoidance of CNI to standard dose; or 2) the intervention (delayed or reduced CNI dose) was instigated several weeks or months after the kidney transplant.

Ongoing studies

See Characteristics of studies awaiting classification.

One ongoing study was identified that met our inclusion criteria but has not yet published any results (OPTIMIZE 2021).

Risk of bias in included studies

The risk of bias is summarised for all studies in Figure 2 and Figure 3.

2.

Figure 3. Review authors' judgements about each risk of bias item presented as percentages across all included studies.

3.

Figure 3. Risk of bias summary: review authors' judgements about each risk of bias item for each included study. Blank cells in the table indicate outcomes that were not measured in studies.

Allocation

Three studies were at low risk (Hernandez 2007; Kyllonen 2002; OSAKA 2013) and one at high risk (Wilson 2005) for sequence generation and allocation concealment. The remaining eight studies had insufficient information to judge adequate random sequence generation, allocation concealment, or both.

Blinding

All the included studies were open‐label.

Performance bias

For the DGF outcome, only studies that used a DGF definition based on creatinine were judged as low risk of bias. However, definitions based on the need for dialysis can be biased by lack of blinding. For acute rejection outcomes, eight studies were considered to be at high risk of bias because some acute rejection episodes were not biopsy‐proven. Only four were judged as low bias because they were diagnosed with a biopsy (Hernandez 2007; Kyllonen 2002; Noel 2009; OSAKA 2013).

For eGFR and death, all studies were judged as low risk of bias because they were unlikely to be biased despite the lack of blinding. However, for infections, anaemia, leukopenia, and graft loss, all studies were judged as unclear risk of bias because lack of blinding could have affected the reporting of the outcome.

Detection bias

For the DGF outcome, all studies were judged as low risk of bias, and for acute rejection outcome, eight studies (Andres 2009; Bechstein‐193 2013; Charpentier 2002; Lebranchu 2002; Mourad 2005; SENIOR 2009; Thomas 2007; Wilson 2005) were at high risk because acute rejection episodes were not all biopsy‐proven. For the remaining outcomes, all studies were judged as low risk because they were unlikely to be biased despite the lack of blinding.

Incomplete outcome data

One study was considered to be at high risk of attrition bias (Bechstein‐193 2013) and another as unclear risk of bias due to insufficient information to permit judgement (Mourad 2005). The remaining 10 studies were considered as low risk of attrition bias.

Selective reporting

Nine studies provided insufficient information to permit judgement and remain unclear for their potential for reporting bias. The remaining three were considered at low risk (Kyllonen 2002; Noel 2009; SENIOR 2009).

Other potential sources of bias

Four studies were industry‐funded and assessed as high risk of bias (Bechstein‐193 2013; Charpentier 2002; OSAKA 2013; Wilson 2005). Five studies had insufficient information to judge other potential sources of bias, and the remaining three were considered low risk (Hernandez 2007; Kyllonen 2002; Noel 2009).

Effects of interventions

See: Table 1; Table 2

All outcomes were assessed using GRADE as shown in the results below, but only the six most important outcomes (bold) are included in Table 1 and Table 2.

Delayed versus early initiation of calcineurin inhibitors

Primary outcomes

Up to one year post kidney transplant, delayed initiation of CNI compared with early initiation:

• May make little or no difference to DGF (Analysis 1.1 (6 studies, 905 recipients): RR 0.92, 95% CI 0.76 to 1.12; I² = 18%) (low certainty evidence).

• Probably makes little or no difference to DGF duration (Analysis 1.2 (3 studies, 405 recipients): MD ‐0.71 days, 95% CI ‐5.48 to 4.05; I² = 0%) (moderate certainty evidence).

• May make little or no difference to all acute rejection (Analysis 1.3 (8 studies, 1295 recipients): RR 1.02, 95% CI 0.75 to 1.40; I² = 54%) (low certainty evidence). Early initiation of TAC may have a higher risk of biopsy‐proven acute rejection (Analysis 1.4, subgroup 1.4.2 (3 studies, 852 recipients): RR 0.68, 95% CI 0.48 to 0.97; I² = 37%) (low certainty evidence), while those differences were not detected in the CsA group. However, two of the studies in the TAC group had significant differences in induction therapy between the groups (Charpentier 2002; Noel 2009): early TAC was paired with induction with daclizumab or no induction, while delayed TAC was paired with ATG. The decreased frequency of acute rejection seen in the delayed TAC group could be affected by the type of induction therapy. The evidence is very uncertain about the effect of the intervention on acute rejection resistant to steroid treatment (Analysis 1.5 (6 studies, 1295 recipients): RR 0.82, 95% CI 0.38 to 1.78; I² = 51%) (very low certainty evidence).

1.1. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 1: Delayed graft function

1.2. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 2: Duration of delayed graft function [days]

1.3. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 3: Acute rejection (clinical) < 1 year

1.4. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 4: Acute rejection (biopsy proven) < 1 year

1.5. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 5: Acute rejection resistant to steroid treatment < 1 year

Secondary outcomes

Delayed initiation of CNI compared with early initiation:

• Probably makes little or no difference to eGFR (Analysis 1.6 (6 studies, 851 recipients): MD ‐0.81 mL/min, 95% CI ‐3.33 to 1.72; I² = 0%) (moderate certainty evidence).

• May make little or no difference to graft loss censored for death (Analysis 1.7 (8 studies, 1295 recipients): RR 1.58, 95% CI 0.68 to 3.65; I² = 52%) (very low certainty evidence), death with functioning graft (Analysis 1.8 (3 studies, 674 recipients): RR 0.75, 95% CI 0.26 to 2.14; I² = 0%), graft loss due to acute rejection (Analysis 1.9 (3 studies, 674 recipients): RR 0.75, 95% CI 0.31 to 1.82; I² = 0%) or graft loss due to surgical reasons/primary non‐function (Analysis 1.10 (4 studies, 725 recipients): RR 1.51, 95% CI 0.62 to 3.67; I² = 0%), although the evidence is very uncertain (very low certainty evidence).

• May make little or no difference to all‐cause death (Analysis 1.11 (8 studies, 907 recipients): RR 1.08, 95% CI 0.54 to 2.14; I² = 0%) (very low certainty evidence) or death due to infection (Analysis 1.12 (2 studies, 598 recipients): RR 0.53, 95% CI 0.03 to 9.76; I² = 58%), although the evidence is very uncertain (very low certainty evidence).

• Probably makes little or no difference to all infections (Analysis 1.13 (6 studies, 1226 recipients): RR 1.10, 95% CI 0.97 to 1.25; I² = 53%) (moderate certainty evidence), bacterial infection (Analysis 1.14 (5 studies, 1125 recipients): RR 1.01, 95% CI 0.88 to 1.16; I² = 0%), fungal infection (Analysis 1.15 (3 studies, 674 recipients): RR 1.07, 95% CI 0.57 to 2.02; I² = 7%), or CMV infection (Analysis 1.16 (7 studies, 1262 recipients): RR 1.47, 95% CI 0.94 to 2.29; I² = 52%) (moderate certainty of evidence).

• Probably makes little or no difference to the risk of anaemia (Analysis 1.17 (5 studies, 999 recipients): RR 1.17, 95% CI 0.85 to 1.61; I² = 45%) (moderate certainty evidence).

• May be associated with an increased risk of leukopenia (Analysis 1.18 (5 studies, 894 recipients): RR 2.72, 95% CI 1.42 to 5.22; I² = 71%) (low certainty evidence). However, there are differences in concomitant medication between groups that could considerably affect the risk of leukopenia, such as CMV prophylaxis or treatment with valganciclovir, MMF, or ATG. Most studies used ATG in the delayed introduction of CNI group, while BSX/daclizumab was used in the standard introduction approach. A sensitivity analysis was performed to determine the cause of the high heterogeneity. Mourad 2005, with a weight of around 50% in the CsA subgroup, used BSX/daclizumab as induction in both treatment arms and showed no differences in leukopenia incidence. When excluding this study from the analysis, the heterogeneity dropped to 0%.

1.6. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 6: eGFR < 1 year [mL/min]

1.7. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 7: Graft loss censored for death < 1 year

1.8. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 8: Graft loss: death with a functioning graft < 1 year

1.9. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 9: Graft loss: acute rejection < 1 year

1.10. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 10: Graft loss: technical/surgical/primary non‐function < 1 year

1.11. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 11: Death (all causes)

1.12. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 12: Death due to infection

1.13. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 13: Infection (all)

1.14. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 14: Infection: bacterial

1.15. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 15: Infection: fungal

1.16. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 16: Infection: CMV (all definitions)

1.17. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 17: Anaemia

1.18. Analysis.

Comparison 1: Delayed versus early calcineurin inhibitors (CNI), Outcome 18: Leukopenia

None of the included studies reported malignancy or cardiovascular outcomes.

Low versus standard initial dose of calcineurin inhibitors

Primary outcomes

Up to two years post kidney transplant, initial low dose CNI compared with standard dose CNI:

• May make little or no difference to DGF (Analysis 2.1 (5 studies, 983 recipients): RR 1.16, 95% CI 0.90 to 1.50; I² = 9%) (low certainty evidence). The duration of DGF was not reported by any study.

• May make little or no difference to clinical acute rejection (Analysis 2.2 (5 studies, 947 recipients): RR 0.83, 95% CI 0.52 to 1.30; I² = 9%) (low certainty evidence), biopsy‐proven acute rejection (Analysis 2.3 (5 studies, 947 recipients): RR 0.83, 95% CI 0.53 to 1.31; I² = 40%) or acute rejection resistant to steroid treatment (Analysis 2.4 (3 studies, 393 recipients): RR 1.32, 95% CI 0.44 to 3.97; I² = 0%) (low certainty evidence).

2.1. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 1: Delayed graft function

2.2. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 2: Acute rejection (clinical) < 2 years

2.3. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 3: Acute rejection (biopsy proven) < 2 years

2.4. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 4: Acute rejection resistant to steroid treatment < 2 years (CsA)

Secondary outcomes

Initial low‐dose CNI compared with standard‐dose CNI:

• May make little or no difference to eGFR (Analysis 2.5 (5 studies, 935 recipients): MD 4.06 mL/min, 95% CI ‐1.36 to 9.48; I² = 81%) (low certainty evidence).

• May make little or no difference to graft loss censored for death (Analysis 2.6 (5 studies, 983 recipients): RR 1.05, 95% CI 0.64 to 1.71; I² = 0%) (very low certainty evidence), graft loss not censored for death (Analysis 2.7 (2 studies, 238 recipients): RR 1.74, 95% CI 0.47 to 6.47; I² = 0%), graft loss due to acute rejection (Analysis 2.8 (2 studies, 238 recipients): RR 0.68, 95% CI 0.13 to 3.56; I² = 15%) or graft loss due to surgical reasons/primary non‐function (Analysis 2.9 (3 studies, 393 recipients): RR 1.05, 95% CI 0.27 to 4.04; I² = 3%), although the evidence is very uncertain (very low certainty evidence).

• May make little or no difference to all‐cause death (Analysis 2.10 (4 studies, 521 recipients): RR 1.01, 95% CI 0.41 to 2.47; I² = 5%) (low certainty evidence), death due to cardiovascular cause (Analysis 2.11 (2 studies, 315 recipients): RR 0.40, 95% CI 0.07 to 2.39; I² = 0%), death due to infection (Analysis 2.12 (2 studies, 315 recipients): RR 1.67, 95% CI 0.25 to 11.22; I² = 0%) (low certainty evidence), or death due to malignancy (Analysis 2.13 (2 studies, 315 recipients): RR 0.88, 95% CI 0.14 to 5.29; I² = 0%).

• Probably makes little or no difference to all infections (Analysis 2.14 (4 studies, 828 recipients): RR 0.87, 95% CI 0.71 to 1.07; I² = 62%) (moderate certainty evidence).

• Low‐dose TAC probably reduces bacterial infection risk (Analysis 2.15.2 (2 studies, 590 recipients): RR 0.73, 95% CI 0.56 to 0.94; I² = 0%) (moderate certainty evidence). However, those differences were not detected when the CNI was CsA.

• Low‐dose CNI may make little or no difference to fungal infection (Analysis 2.16 (2 studies, 206 recipients): RR 0.52, 95% CI 0.13 to 2.02; I² = 0%) (low certainty evidence). Probably makes little or no difference to CMV infection (Analysis 2.17 (5 studies, 983 recipients): RR 0.76, 95% CI 0.51 to 1.12; I² = 29%) (moderate certainty evidence).

• Probably makes little or no difference to anaemia (Analysis 2.18 (3 studies, 668 recipients): RR 1.04, 95% CI 0.84 to 1.28; I² = 0%) (moderate certainty evidence).

• It is uncertain whether low‐dose CNI has an effect on leukopenia (Analysis 2.19 (2 studies, 206 recipients): RR 1.11, 95% CI 0.57 to 2.19; I² = 0%) because the evidence is very uncertain.

2.5. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 5: eGFR < 2 years [mL/min]

2.6. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 6: Graft loss censored for death < 2 years

2.7. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 7: Graft loss: death with functioning graft < 2 years (CsA)

2.8. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 8: Graft loss: acute rejection < 2 years (CsA)

2.9. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 9: Graft loss: technical/surgical/primary non‐function < 2 years (CsA)

2.10. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 10: Death (all causes) < 2 years

2.11. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 11: Death due to cardiovascular cause < 2 years (CsA)

2.12. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 12: Death due to infection < 2 years (CsA)

2.13. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 13: Death due to malignancy < 2 years (CsA)

2.14. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 14: Infection (all)

2.15. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 15: Infection: bacterial

2.16. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 16: Infection: fungal

2.17. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 17: Infection: CMV (all definitions)

2.18. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 18: Anaemia

2.19. Analysis.

Comparison 2: Low versus standard dose calcineurin inhibitors (CNI), Outcome 19: Leukopenia

Discussion

Summary of main results

The results of this meta‐analysis show that delayed initiation or lower CNI initial dose in de novo kidney transplant recipients at high risk of DGF may make little or no difference to the risk of DGF or acute rejection incidence. There may be little or no difference to graft function, graft loss, death or infection in any of the evaluated strategies.

Subgroup analysis by CNI type showed no differences in most outcomes except for biopsy‐proven acute rejection and bacterial infections. Firstly, delayed TAC may have a lower risk of biopsy‐proven acute rejection (but not delayed CsA). However, in two of the three studies in this comparison, those who received delayed TAC were treated with ATG, while those who received early TAC were treated with daclizumab or no induction. The decreased frequency of acute rejection seen in the delayed TAC group is probably related to the type of induction therapy used. Secondly, there is probably a lower risk of bacterial infections in recipients treated with low doses of TAC. Both studies included in this intervention had similar concomitant immunosuppression. However, the results are mainly driven by OSAKA 2013 (around 70% of the weight).

Overall completeness and applicability of evidence

Most studies did not follow up beyond the first year, limiting the possibility of assessing long‐term death or graft loss. Outcomes such as infections, anaemia or leukopenia were underreported compared to other outcomes such as DGF or acute rejection. Also, none of the included studies reported pretransplant or new donor‐specific antibodies, which could impact graft survival. The included studies may not have been powered enough to detect differences in DGF risk, as DGF was not the primary outcome for most of them, and the definition of high risk of DGF also varied among the studies.

All studies were performed in Europe. Around 60% of the participants were males, reflecting the expected proportion in the population on KRT in Europe.

There are several factors that limit the applicability of the evidence to our current practice. Most studies included in this meta‐analysis are from the 2000s, when the use of CsA was more common than TAC, most deceased donors were DBD, with higher CNI therapy levels, differences in induction therapy and generally less expanded criteria donors.

Quality of the evidence

The overall quality of the evidence of the included studies was low, ranging from moderate to very low. Factors downgraded on GRADE were primarily a severe risk of bias and inconsistency. Over 50% of the included studies did not adequately report the random sequence generation or the allocation concealment, and all studies were unblinded. Most studies had unclear reporting bias and were judged as unclear or high risk of other biases because they were industry‐funded. Due to few events and wide CIs, several outcomes were downgraded for imprecision.

Potential biases in the review process

There are some limitations in this review. One potential source of bias is that there were differences between studies when defining low or standard dose: some studies used CNI concentration while others used weight‐based dosing. Our review classified low or standard doses based on each author's definition, which may introduce bias. Due to the low number of included studies, the effect of different concomitant immunosuppression therapies could not be evaluated. Also, the definition of DGF varied among studies. In unblinded studies, creatinine‐based definitions for DGF can be a source of risk of bias.

Although there are some limitations, there are also several strengths. There is a pre‐published peer‐reviewed protocol. A comprehensive literature search was performed of the CKT Specialised Register by the CKT Information Specialist, reducing the probability of omitting published studies. Finally, at least two authors independently participated in study selection, data extraction, analysis, and assessment of the risk of bias and certainty, thus enhancing the accuracy of the data collected during the entire review process.

Agreements and disagreements with other studies or reviews

Several reviews have evaluated different CNI‐sparing regimens. However, they did not limit the population to people with a high risk of DGF and did not analyse separately by type of CNI (CsA versus TAC).

Moore 2009 studied different CNI‐sparing regimens, including de novo CNI minimisation, but did not restrict the population to those at high risk of DGF. The results on graft loss, death, and biopsy‐proven acute rejection were similar to our review, showing no differences between low or standard CNI dose. However, they detected a slightly better graft function in the low CNI dose.

Sharif 2011 also studied different strategies to reduce CNI exposure, including delayed introduction of CNI and minimisation of CNI dose as our review, although not limiting the population to high risk of DGF. They reported similar results on death, graft failure, and biopsy‐proven acute rejection. Similar to Moore 2009, a slightly higher graft function was detected in the low TAC group as well as the delayed TAC group.

Karpe 2017 compared different CNI‐sparing regiments, with and without mammalian target of rapamycin inhibitors, including low versus standard CNI. Similar to the other previous reviews, they did not limit the population to a high risk of DGF. Their conclusions on graft loss, acute rejection, death, and CMV infection are similar to ours, but they also reported a slightly higher eGFR in the low CNI dose group.

Authors' conclusions

Implications for practice.

Delayed or reduced CNI dose initiation has been proposed as a strategy to improve DGF in recipients at high risk of DGF. Current evidence suggests that these strategies may have little or no impact on DGF or the risk of acute rejection. However, the use of reduced TAC instead of the standard dose in de‐novo KT probably reduced the risk of bacterial infections.

The medium to long‐term effect of these strategies remains unknown. Evidence from registry data or cohort studies could be integrated into our current understanding of benefits and harms when informing clinical decision‐making.

Implications for research.

Immunosuppression treatments have undergone changes in dosage and drug types over time. Therefore, it is important to conduct studies that represent current clinical practice to assess the impact of the type of immunosuppression on the risk of DGF. Additionally, the demographics of kidney transplant recipients are also changing, with a growing need for well‐conducted studies on optimal immunosuppression for elderly recipients at high risk of DGF.

History

Protocol first published: Issue 11, 2022

Appendices

Appendix 1. Electronic search strategies

Database	Search terms
CENTRAL	(calcineurin next (inhibit* or block*)):ti,ab,kw CNI.tw. (tacrolimus or TAC):ti,ab,kw prograf*:ti,ab,kw (fk506 or "fk 506"):ti,ab,kw cyclosporin*:ti,ab,kw ciclosporin:ti,ab,kw neoral:ti,ab,kw sandimmun:ti,ab,kw CSA:ti,ab,kw {or #1‐#10} ((kidney or renal) next (transplant* or graft* or allograft*)):ti,ab,kw #11 and #12 (delay* or reduce* or reduction):ti,ab,kw (cadaveric or deceased or high‐risk):ti,ab,kw ((donor* near/2 brain death) or (donat* near/2 brain death)):ti,ab ((donor* near/2 circulatory death) or (donat* near/2 circulatory death)):ti,ab ((donor* near/2 cardiac death) or (donat* near/2 cardiac death)):ti,ab ("extended criteria" near/2 (donor* or donat* or cadaveric or deceased)):ti,ab {or #14‐#19} #13 and #20 in Trials
MEDLINE	exp Calcineurin Inhibitors/ (calcineurin adj (inhibit* or block*)).tw. CNI.tw. (tacrolimus or TAC).tw. prograf*.tw. fk 506.tw. fk506.tw. cyclosporin*.tw. ciclosporin.tw. neoral.tw. sandimmun*.tw. CSA.tw. or/1‐12 Kidney Transplantation/ ((kidney or renal) adj (transplant* or graft* or allograft*)).tw. or/14‐15 and/13,16 (delay* or reduce* or reduction).tw. (cadaveric or deceased or high‐risk).tw. ((donor* adj2 brain death) or (donat* adj2 brain death)).tw. ((donor* adj2 circulatory death) or (donat* adj2 circulatory death)).tw. ((donor* adj2 cardiac death) or (donat* adj2 cardiac death)).tw. (extended criteria adj2 (donor* or donat* or cadaveric or deceased)).tw. or/18‐23 and/17,24 MEDLINE search terms have been combined with the Cochrane highly sensitive search terms for RCTS (Cochrane Handbook Chapter 4).
EMBASE	exp calcineurin inhibitor/ (calcineurin adj (inhibit* or block*)).tw. CNI.tw. (tacrolimus or TAC).tw. prograf*.tw. fk 506.tw. fk506.tw. cyclosporin*.tw. ciclosporin.tw. neoral.tw. sandimmun*.tw. CSA.tw. or/1‐12 exp kidney transplantation/ ((kidney or renal) adj (transplant* or graft* or allograft*)).tw. or/14‐15 and/13,16 (delay* or reduce* or reduction).tw. (cadaveric or deceased or high‐risk).tw. ((donor* adj2 brain death) or (donat* adj2 brain death)).tw. ((donor* adj2 circulatory death) or (donat* adj2 circulatory death)).tw. ((donor* adj2 cardiac death) or (donat* adj2 cardiac death)).tw. (extended criteria adj2 (donor* or donat* or cadaveric or deceased)).tw. or/18‐23 and/17,24 EMBASE search terms have been combined with the Cochrane highly sensitive search terms for RCTS (Cochrane Handbook Chapter 4)

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; minimisation (minimisation 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 standardised 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. sub‐scales) 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.

Data and analyses

Comparison 1. Delayed versus early calcineurin inhibitors (CNI).

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Delayed graft function	6	905	Risk Ratio (M‐H, Random, 95% CI)	0.92 [0.76, 1.12]
1.1.1 CsA	3	373	Risk Ratio (M‐H, Random, 95% CI)	1.00 [0.72, 1.40]
1.1.2 TAC	3	532	Risk Ratio (M‐H, Random, 95% CI)	0.90 [0.65, 1.24]
1.2 Duration of delayed graft function [days]	3	405	Mean Difference (IV, Random, 95% CI)	‐0.71 [‐5.48, 4.05]
1.2.1 CsA	1	100	Mean Difference (IV, Random, 95% CI)	1.40 [‐6.27, 9.07]
1.2.2 TAC	2	305	Mean Difference (IV, Random, 95% CI)	‐2.04 [‐8.11, 4.04]
1.3 Acute rejection (clinical) < 1 year	8	1295	Risk Ratio (M‐H, Random, 95% CI)	1.02 [0.75, 1.40]
1.3.1 CsA	3	373	Risk Ratio (M‐H, Random, 95% CI)	1.47 [0.97, 2.21]
1.3.2 TAC	5	922	Risk Ratio (M‐H, Random, 95% CI)	0.88 [0.60, 1.29]
1.4 Acute rejection (biopsy proven) < 1 year	6	1225	Risk Ratio (M‐H, Random, 95% CI)	0.86 [0.58, 1.27]
1.4.1 CsA	3	373	Risk Ratio (M‐H, Random, 95% CI)	1.42 [0.88, 2.31]
1.4.2 TAC	3	852	Risk Ratio (M‐H, Random, 95% CI)	0.68 [0.48, 0.97]
1.5 Acute rejection resistant to steroid treatment < 1 year	6	1225	Risk Ratio (M‐H, Random, 95% CI)	0.82 [0.38, 1.78]
1.5.1 CsA	3	373	Risk Ratio (M‐H, Random, 95% CI)	2.06 [0.76, 5.60]
1.5.2 TAC	3	852	Risk Ratio (M‐H, Random, 95% CI)	0.54 [0.21, 1.39]
1.6 eGFR < 1 year [mL/min]	6	851	Mean Difference (IV, Random, 95% CI)	‐0.81 [‐3.33, 1.72]
1.6.1 CsA	3	365	Mean Difference (IV, Random, 95% CI)	‐1.06 [‐6.42, 4.30]
1.6.2 TAC	3	486	Mean Difference (IV, Random, 95% CI)	‐0.73 [‐3.91, 2.45]
1.7 Graft loss censored for death < 1 year	8	1295	Risk Ratio (M‐H, Random, 95% CI)	1.58 [0.68, 3.65]
1.7.1 CsA	3	373	Risk Ratio (M‐H, Random, 95% CI)	2.75 [0.37, 20.16]
1.7.2 TAC	5	922	Risk Ratio (M‐H, Random, 95% CI)	1.39 [0.55, 3.56]
1.8 Graft loss: death with a functioning graft < 1 year	3	674	Risk Ratio (M‐H, Random, 95% CI)	0.75 [0.26, 2.14]
1.8.1 CsA	1	76	Risk Ratio (M‐H, Random, 95% CI)	Not estimable
1.8.2 TAC	2	598	Risk Ratio (M‐H, Random, 95% CI)	0.75 [0.26, 2.14]
1.9 Graft loss: acute rejection < 1 year	3	674	Risk Ratio (M‐H, Random, 95% CI)	0.75 [0.31, 1.82]
1.9.1 CsA	1	76	Risk Ratio (M‐H, Random, 95% CI)	0.89 [0.26, 3.06]
1.9.2 TAC	2	598	Risk Ratio (M‐H, Random, 95% CI)	0.62 [0.16, 2.31]
1.10 Graft loss: technical/surgical/primary non‐function < 1 year	4	725	Risk Ratio (M‐H, Random, 95% CI)	1.51 [0.62, 3.67]
1.10.1 CsA	1	76	Risk Ratio (M‐H, Random, 95% CI)	9.97 [0.56, 179.04]
1.10.2 TAC	3	649	Risk Ratio (M‐H, Random, 95% CI)	1.24 [0.49, 3.15]
1.11 Death (all causes)	8	1280	Risk Ratio (M‐H, Random, 95% CI)	1.08 [0.54, 2.14]
1.11.1 CsA	3	373	Risk Ratio (M‐H, Random, 95% CI)	0.52 [0.14, 1.93]
1.11.2 TAC	5	907	Risk Ratio (M‐H, Random, 95% CI)	1.42 [0.63, 3.19]
1.12 Death due to infection	2	598	Risk Ratio (M‐H, Random, 95% CI)	0.53 [0.03, 9.76]
1.12.1 TAC	2	598	Risk Ratio (M‐H, Random, 95% CI)	0.53 [0.03, 9.76]
1.13 Infection (all)	6	1226	Risk Ratio (M‐H, Random, 95% CI)	1.10 [0.97, 1.25]
1.13.1 CsA	3	374	Risk Ratio (M‐H, Random, 95% CI)	1.06 [0.77, 1.47]
1.13.2 TAC	3	852	Risk Ratio (M‐H, Random, 95% CI)	1.13 [1.01, 1.25]
1.14 Infection: bacterial	5	1125	Risk Ratio (M‐H, Random, 95% CI)	1.01 [0.88, 1.16]
1.14.1 CsA	2	273	Risk Ratio (M‐H, Random, 95% CI)	0.85 [0.65, 1.11]
1.14.2 TAC	3	852	Risk Ratio (M‐H, Random, 95% CI)	1.07 [0.91, 1.26]
1.15 Infection: fungal	3	674	Risk Ratio (M‐H, Random, 95% CI)	1.07 [0.57, 2.02]
1.15.1 CsA	1	76	Risk Ratio (M‐H, Random, 95% CI)	2.78 [0.57, 13.44]
1.15.2 TAC	2	598	Risk Ratio (M‐H, Random, 95% CI)	0.89 [0.46, 1.72]
1.16 Infection: CMV (all definitions)	7	1262	Risk Ratio (M‐H, Random, 95% CI)	1.47 [0.94, 2.29]
1.16.1 CsA	3	374	Risk Ratio (M‐H, Random, 95% CI)	1.24 [0.36, 4.24]
1.16.2 TAC	4	888	Risk Ratio (M‐H, Random, 95% CI)	1.62 [1.17, 2.24]
1.17 Anaemia	5	999	Risk Ratio (M‐H, Random, 95% CI)	1.17 [0.85, 1.61]
1.17.1 CsA	3	374	Risk Ratio (M‐H, Random, 95% CI)	1.03 [0.56, 1.88]
1.17.2 TAC	2	625	Risk Ratio (M‐H, Random, 95% CI)	1.33 [1.01, 1.75]
1.18 Leukopenia	5	999	Risk Ratio (M‐H, Random, 95% CI)	2.72 [1.42, 5.22]
1.18.1 CsA	3	374	Risk Ratio (M‐H, Random, 95% CI)	1.94 [0.75, 5.03]
1.18.2 TAC	2	625	Risk Ratio (M‐H, Random, 95% CI)	3.93 [2.64, 5.86]

Comparison 2. Low versus standard dose calcineurin inhibitors (CNI).

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Delayed graft function	5	983	Risk Ratio (M‐H, Random, 95% CI)	1.16 [0.90, 1.50]
2.1.1 CsA	3	393	Risk Ratio (M‐H, Random, 95% CI)	1.03 [0.73, 1.44]
2.1.2 TAC	2	590	Risk Ratio (M‐H, Random, 95% CI)	1.31 [0.75, 2.29]
2.2 Acute rejection (clinical) < 2 years	5	947	Risk Ratio (M‐H, Random, 95% CI)	0.83 [0.52, 1.30]
2.2.1 CsA	3	393	Risk Ratio (M‐H, Random, 95% CI)	0.67 [0.40, 1.12]
2.2.2 TAC	2	554	Risk Ratio (M‐H, Random, 95% CI)	1.13 [0.41, 3.13]
2.3 Acute rejection (biopsy proven) < 2 years	5	947	Risk Ratio (M‐H, Random, 95% CI)	0.83 [0.53, 1.31]
2.3.1 CsA	3	393	Risk Ratio (M‐H, Random, 95% CI)	0.69 [0.41, 1.15]
2.3.2 TAC	2	554	Risk Ratio (M‐H, Random, 95% CI)	1.20 [0.37, 3.94]
2.4 Acute rejection resistant to steroid treatment < 2 years (CsA)	3	393	Risk Ratio (M‐H, Random, 95% CI)	1.32 [0.44, 3.97]
2.5 eGFR < 2 years [mL/min]	5	935	Mean Difference (IV, Random, 95% CI)	4.06 [‐1.36, 9.48]
2.5.1 CsA	3	383	Mean Difference (IV, Random, 95% CI)	1.50 [‐5.49, 8.48]
2.5.2 TAC	2	552	Mean Difference (IV, Random, 95% CI)	8.48 [‐4.63, 21.58]
2.6 Graft loss censored for death < 2 years	5	983	Risk Ratio (M‐H, Random, 95% CI)	1.05 [0.64, 1.71]
2.6.1 CsA	3	393	Risk Ratio (M‐H, Random, 95% CI)	1.01 [0.38, 2.66]
2.6.2 TAC	2	590	Risk Ratio (M‐H, Random, 95% CI)	1.36 [0.40, 4.64]
2.7 Graft loss: death with functioning graft < 2 years (CsA)	2	238	Risk Ratio (M‐H, Random, 95% CI)	1.74 [0.47, 6.47]
2.8 Graft loss: acute rejection < 2 years (CsA)	2	238	Risk Ratio (M‐H, Random, 95% CI)	0.68 [0.13, 3.56]
2.9 Graft loss: technical/surgical/primary non‐function < 2 years (CsA)	3	393	Risk Ratio (M‐H, Random, 95% CI)	1.05 [0.27, 4.04]
2.10 Death (all causes) < 2 years	4	521	Risk Ratio (M‐H, Random, 95% CI)	1.01 [0.41, 2.47]
2.10.1 CsA	3	393	Risk Ratio (M‐H, Random, 95% CI)	0.87 [0.22, 3.54]
2.10.2 TAC	1	128	Risk Ratio (M‐H, Random, 95% CI)	1.03 [0.22, 4.92]
2.11 Death due to cardiovascular cause < 2 years (CsA)	2	315	Risk Ratio (M‐H, Random, 95% CI)	0.40 [0.07, 2.39]
2.12 Death due to infection < 2 years (CsA)	2	315	Risk Ratio (M‐H, Random, 95% CI)	1.67 [0.25, 11.22]
2.13 Death due to malignancy < 2 years (CsA)	2	315	Risk Ratio (M‐H, Random, 95% CI)	0.88 [0.14, 5.29]
2.14 Infection (all)	4	828	Risk Ratio (M‐H, Random, 95% CI)	0.87 [0.71, 1.07]
2.14.1 CsA	2	238	Risk Ratio (M‐H, Random, 95% CI)	0.97 [0.74, 1.27]
2.14.2 TAC	2	590	Risk Ratio (M‐H, Random, 95% CI)	0.74 [0.61, 0.89]
2.15 Infection: bacterial	4	828	Risk Ratio (M‐H, Random, 95% CI)	0.96 [0.70, 1.30]
2.15.1 CsA	2	238	Risk Ratio (M‐H, Random, 95% CI)	1.17 [0.88, 1.56]
2.15.2 TAC	2	590	Risk Ratio (M‐H, Random, 95% CI)	0.73 [0.56, 0.94]
2.16 Infection: fungal	2	206	Risk Ratio (M‐H, Random, 95% CI)	0.52 [0.13, 2.02]
2.16.1 CsA	1	78	Risk Ratio (M‐H, Random, 95% CI)	0.53 [0.05, 5.57]
2.16.2 TAC	1	128	Risk Ratio (M‐H, Random, 95% CI)	0.52 [0.10, 2.72]
2.17 Infection: CMV (all definitions)	5	983	Risk Ratio (M‐H, Random, 95% CI)	0.76 [0.51, 1.12]
2.17.1 CsA	3	393	Risk Ratio (M‐H, Random, 95% CI)	0.86 [0.39, 1.89]
2.17.2 TAC	2	590	Risk Ratio (M‐H, Random, 95% CI)	0.77 [0.50, 1.20]
2.18 Anaemia	3	668	Risk Ratio (M‐H, Random, 95% CI)	1.04 [0.84, 1.28]
2.18.1 CsA	1	78	Risk Ratio (M‐H, Random, 95% CI)	1.13 [0.62, 2.09]
2.18.2 TAC	2	590	Risk Ratio (M‐H, Random, 95% CI)	1.03 [0.82, 1.28]
2.19 Leukopenia	2	206	Risk Ratio (M‐H, Random, 95% CI)	1.11 [0.57, 2.19]
2.19.1 CsA	1	78	Risk Ratio (M‐H, Random, 95% CI)	1.05 [0.33, 3.35]
2.19.2 TAC	1	128	Risk Ratio (M‐H, Random, 95% CI)	1.15 [0.50, 2.63]

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Andres 2009.

Study characteristics
Methods	Study design Parallel RCT Study duration: 2002 to 2003 Mean duration of follow‐up: 6 months Country: 17 centres in Spain Definition of DGF: need for perioperative dialysis within the first week after transplantation Definition of eGFR: at 3 and 6 months (Nankivell formula)
Participants	Recipient characteristics Number: 114; CsA early low (38); CsA early (40); delayed CsA (36) Mean age ± SD (years): CsA early low (56.4 ± 9.5); CsA early (55.7 ±9.5); CsA delayed (57.7 ± 12.3) Sex (M/F): CsA early low (23/15); CsA early (26/14); delayed (24/12) Recipients considered to be at high risk of developing DGF: donor > 60 years; SCr > 2 mg/dL; DCD; death from cerebrovascular event or CIT > 24 hours Donor characteristics Type of donor: deceased donors (> 97% DBD) Mean age ± SD (years): 57.6±12.5 CsA early low, 56.3±10.7 CsA early, 58.9±11.3 CsA delayed Other information Mean CIT ± SD (hours): CsA early low (20.6 ± 5.4); CsA early (19.4 ± 4.6); CsA delayed (19.2 ± 5.4) Reason for high risk of DGF: mean CIT > 18h (with approximately 20% of them CIT between 24‐40 h)
Interventions	Group 1: CsA early low CsA < 24 hours: 3 mg/kg/d Group 2: CsA early CsA < 24 hours: 5 mg/kg/d Group 3: CsA delayed CsA day 7‐10: 5mg/kg/d Co‐interventions (all groups) Basiliximab, steroids, MMF
Outcomes	Primary outcome eGFR at 3 and 6 months Secondary outcomes Incidence of DGF, BPAR SCr level Patient and graft survival
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information about the sequence generation process to permit judgement
Allocation concealment (selection bias)	Unclear risk	Randomisation stated, but no complete information on method used is available
Blinding of participants and personnel (performance bias) DGF	High risk	Not blind (DGF definition based on dialysis indication)
Blinding of participants and personnel (performance bias) Acute rejection	High risk	Not blinded
Blinding of participants and personnel (performance bias) eGFR	Low risk	No blinding, but outcome not likely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	No blinding
Blinding of participants and personnel (performance bias) Death	Low risk	No blinding, but outcome not likely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Infections	Unclear risk	No blinding, could affect reporting
Blinding of participants and personnel (performance bias) Cytopenias	Unclear risk	No blinding, could affect reporting
Blinding of outcome assessment (detection bias) DGF	High risk	No blinding, but DGF measurement is not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Acute rejection	High risk	No blinding
Blinding of outcome assessment (detection bias) eGFR	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Infections	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Cytopenias	Low risk	Unlikely to be affected by lack of blinding
Incomplete outcome data (attrition bias) DGF	Low risk	Missing data balanced between groups, reasons for missing outcome stated and similar between groups
Incomplete outcome data (attrition bias) Acute rejection	Low risk	Missing data balanced between groups, reasons for missing outcome stated and similar between groups
Incomplete outcome data (attrition bias) eGFR	Low risk	Missing data balanced between groups, reasons for missing outcome stated and similar between groups
Incomplete outcome data (attrition bias) Graft loss	Low risk	Missing data balanced between groups, reasons for missing outcome stated and similar between groups
Incomplete outcome data (attrition bias) Death	Low risk	Missing data balanced between groups, reasons for missing outcome stated and similar between groups
Incomplete outcome data (attrition bias) Infections	Low risk	Missing data balanced between groups, reasons for missing outcome stated and similar between groups
Incomplete outcome data (attrition bias) Cytopenias	Low risk	Missing data balanced between groups, reasons for missing outcome stated and similar between groups
Selective reporting (reporting bias)	Unclear risk	Insufficient information to permit judgement
Other bias	Unclear risk	Insufficient information to assess whether an important risk of bias exists

Bechstein‐193 2013.

Study characteristics
Methods	Study design Parallel RCT Study duration: not reported Country: multicentre (13 European sites) Mean duration of follow‐up: 6 months Definition of DGF: definition not reported Definition of eGFR: at 6 months (Nankivell formula)
Participants	Recipient characteristics Number: 128; reduced TAC (63); standard TAC (65) Mean age ± SD (years): reduced TAC (47.9 ± 13.3); standard TAC (44.6 ± 14.8) Sex (M/F): reduced TAC (45/18); standard TAC (38/27) Donor characteristics Type of donor: deceased donors (all DBD) Mean age ± SD (years): not reported Other information Mean CIT ± SD (hours): reduced TAC (17.6 ± 6.6); standard TAC (17.9 ± 6.9) Reason for high risk of DGF: mean CIT 17.6 and 17.9 hours, which rounds to 18 hours when rounded to the nearest whole hour
Interventions	Group 1: reduced TAC + SRL TAC initiated within 7 days: levels 3 to 7 ng/mL throughout the study period SRL initiated on day 1: levels 10 to 20 ng/mL (month 1), 10 to 15 ng/mL (months 2 to 3), then 8 to 15 ng/mL Group 2: standard TAC + SRL TAC initiated within 7 days: levels 10 to 15 ng/mL (month 1), then 8 to 12 ng/mL SRL initiated on day 1: levels 5 to 10 ng/mL throughout the study period Co‐interventions (all groups) Steroid tapering No antibody induction
Outcomes	Primary outcome CrCl Secondary outcomes SCr AR Patient and graft survival Incidence of infection NODAT
Notes	Additional information Study sponsored by Wyeth Pharmaceuticals TAC and SRL doses increased (amended protocol) in the reduced TAC group because of increased incidence of AR and target levels not achieved. However, target TAC and SRL concentrations were not changed.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information about the sequence generation process to permit judgement
Allocation concealment (selection bias)	Unclear risk	Randomisation stated but no information on method used is available
Blinding of participants and personnel (performance bias) DGF	Unclear risk	Unblinded, no clear definition of DGF
Blinding of participants and personnel (performance bias) Acute rejection	High risk	No blinding
Blinding of participants and personnel (performance bias) eGFR	Low risk	Unlikely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Death	Low risk	Unlikely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Infections	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Cytopenias	Unclear risk	Unblinded, could affect reporting
Blinding of outcome assessment (detection bias) DGF	Unclear risk	Unblinded, no clear definition of DGF
Blinding of outcome assessment (detection bias) Acute rejection	High risk	No blinding
Blinding of outcome assessment (detection bias) eGFR	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Infections	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Cytopenias	Low risk	Unlikely to be affected by lack of blinding
Incomplete outcome data (attrition bias) DGF	High risk	Imbalance in reasons for missing data across intervention groups
Incomplete outcome data (attrition bias) Acute rejection	High risk	Imbalance in reasons for missing data across intervention groups
Incomplete outcome data (attrition bias) eGFR	High risk	Imbalance in reasons for missing data across intervention groups
Incomplete outcome data (attrition bias) Graft loss	High risk	Imbalance in reasons for missing data across intervention groups
Incomplete outcome data (attrition bias) Death	High risk	Imbalance in reasons for missing data across intervention groups
Incomplete outcome data (attrition bias) Infections	High risk	Imbalance in reasons for missing data across intervention groups
Incomplete outcome data (attrition bias) Cytopenias	High risk	Imbalance in reasons for missing data across intervention groups
Selective reporting (reporting bias)	Unclear risk	Insufficient information to permit judgement
Other bias	High risk	Study sponsored by Wyeth Pharmaceuticals

Charpentier 2002.

Study characteristics
Methods	Study design Parallel RCT Study duration: not reported Mean duration of follow‐up: 6 months Country: multicentre (30 centres in France, Belgium, Switzerland, and Italy) Definition of DGF: not reported Definition of eGFR: not reported
Participants	Recipient characteristics Number: 555; TAC early (185); ATG+TAC delayed (186); ATG+CsA delayed (184) Mean age ± SD (years): TAC early (44.5 ± 11); ATG+TAC delayed (44.7 ± 11.4); ATG+CsA delayed (43.6 ± 10.9) Sex (M/F): TAC early (121/64); ATG+TAC delayed (118/68); ATG+CsA delayed (116/68) Donor characteristics Type of donor: deceased donors Mean age ± SD (years): not reported Other information Mean CIT ± SD (hours): TAC early (19.7); ATG+TAC delayed (20.4); ATG+CsA delayed (21) Reason for high risk of DGF: mean CIT > 18 hours
Interventions	Group 1: early TAC + no induction TAC 0.3 mg/kg/d: levels 10 to 20 ng/mL from day 1 to 42, and 5 to 15 ng/mL from day 43 to 180 Group 2: ATG + delayed TAC (day 9) TAC 0.3 mg/kg/d: levels 10 to 20 ng/mL from day 1 to 42, and 5 to 15 ng/mL from day 43 to 180 Group 3: ATG + delayed CsA (day 9)* CsA 8 mg/kg/d Co‐interventions (all groups) AZA, steroids *intervention not included in the meta‐analysis
Outcomes	Primary outcome Incidence and time to first BPAR Secondary outcomes Incidence of clinically apparent rejection Number of corticosteroid‐resistant rejections Severity of histopathology Number of days on dialysis
Notes	Study supported by Fujisawa GmbH
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information about the sequence generation process to permit judgement
Allocation concealment (selection bias)	Low risk	"Patients were allocated the consecutive center specific patient number"; "sealed envelope"
Blinding of participants and personnel (performance bias) Acute rejection	High risk	Not blinded
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Not blinded, could affect reporting
Blinding of participants and personnel (performance bias) Death	Low risk	Unlikely to be affected by lack of blinding
Blinding of participants and personnel (performance bias) Infections	Unclear risk	Not blinded, could affect reporting
Blinding of participants and personnel (performance bias) Cytopenias	Unclear risk	Not blinded, could affect reporting
Blinding of outcome assessment (detection bias) Acute rejection	High risk	Not blinded
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Infections	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Cytopenias	Low risk	Unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias) DGF	Low risk	Reasons for missing outcome with similar reasons between groups
Incomplete outcome data (attrition bias) Acute rejection	Low risk	Reasons for missing outcome with similar reasons between groups
Incomplete outcome data (attrition bias) Graft loss	Low risk	Reasons for missing outcome with similar reasons between groups
Incomplete outcome data (attrition bias) Death	Low risk	Reasons for missing outcome with similar reasons between groups
Incomplete outcome data (attrition bias) Infections	Low risk	Reasons for missing outcome with similar reasons between groups
Incomplete outcome data (attrition bias) Cytopenias	Low risk	Reasons for missing outcome with similar reasons between groups
Selective reporting (reporting bias)	Unclear risk	Insufficient information to permit judgement
Other bias	High risk	Supported by Fujisawa GmbH

Hernandez 2007.

Study characteristics
Methods	Study design Parallel RCT Study duration: 1999 to 2004 Mean duration of follow‐up: 24 months Country: Spain (single centre) Definition of DGF: need for dialysis in the first week after transplant, SCr that did not decrease by 20% or urine output < 40 mL/h within 24 hours of transplantation Definition of eGFR (included in the meta‐analysis): at 2 years, Cockroft‐Gault equation
Participants	Recipient characteristics Number: 240; ATG+standard CsA (80); BSX+reduced CsA (80); BSX+reduced TAC (80) Mean age ± SD (years): ATG+standard CsA (47 ± 12); BSX+reduced CsA (48 ± 14); BSX+reduced TAC (47 ± 11) Sex (M/F): ATG+standard CsA (59/21); BSX+reduced CsA (50/30); BSX+reduced TAC (44/36) Donor characteristics: Type of donor: deceased donors (all DBD) Mean age ± SD (years): ATG+standard CsA (45 ± 16); BSX+reduced CsA (42 ± 15); BSX+reduced TAC (44 ± 17) Other information Mean CIT ± SD (hours): ATG+standard CsA (20.3 ± 4); BSX+reduced CsA (21.0 ± 4); BSX+reduced TAC (21.0 ± 4) Reason for high risk of DGF: mean CIT > 18 hours
Interventions	Group 1: ATG + standard CsA + AZA CsA 4 mg/kg twice/day: target level 175 to 300 ng/mL (3 months), then 150 to 200 ng/mL Group 2: BSX + reduced CsA + MMF CsA 2 mg/kg twice/day: target level 125 to 175 ng/mL throughout the study Group 3: BSX + reduced TAC + MMF* TAC 0.05 mg/kg twice/day: target level 7 to 10 ng/mL throughout the study Co‐interventions (all groups) Steroid tapering *Intervention not included in the meta‐analysis
Outcomes	Primary outcome CrCl Secondary outcomes BPAR Patient and graft survival Safety
Notes	In group ATG+standard CsA, in case of DGF CsA dose was reduced to 2mg/kg twice daily until SCr < 3 mg/dL
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation sequence was computer generated
Allocation concealment (selection bias)	Low risk	Sequentially numbered sealed envelopes
Blinding of participants and personnel (performance bias) DGF	Unclear risk	Unblinded, DGF definition based on dialysis indication (potentially affected by lack of blinding) or creatinine (unlikely to be affected by lack of blinding)
Blinding of participants and personnel (performance bias) Acute rejection	Low risk	Defined by biopsy, not likely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) eGFR	Low risk	Unlikely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Death	Low risk	Unlikely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Infections	Unclear risk	Unblinded, could affect reporting
Blinding of outcome assessment (detection bias) DGF	Unclear risk	Unblinded, DGF definition based on dialysis indication (potentially affected by lack of blinding) or creatinine (unlikely to be affected by lack of blinding)
Blinding of outcome assessment (detection bias) Acute rejection	Low risk	Defined by biopsy, not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) eGFR	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Infections	Low risk	Not likely to be influenced by lack of blinding
Incomplete outcome data (attrition bias) DGF	Low risk	Similar reasons for missing data across groups
Incomplete outcome data (attrition bias) Acute rejection	Low risk	Similar reasons for missing data across groups
Incomplete outcome data (attrition bias) eGFR	Low risk	Similar reasons for missing data across groups
Incomplete outcome data (attrition bias) Graft loss	Low risk	Similar reasons for missing data across groups
Incomplete outcome data (attrition bias) Death	Low risk	Similar reasons for missing data across groups
Incomplete outcome data (attrition bias) Infections	Low risk	Similar reasons for missing data across groups
Selective reporting (reporting bias)	Unclear risk	Insufficient information to permit judgement
Other bias	Low risk	The study appears to be free of other sources of bias

Kyllonen 2002.

Study characteristics
Methods	Study design Parallel RCT Study duration: 1999 to 2001 Mean duration of follow‐up: 5 years Country: Finland Definition of DGF: plasma creatinine concentration higher than 500umol/L throughout the first posttransplant week, or need of more than one dialysis session in the first week, or oliguria <1 L/24 hr for more than 2 days. Definition of eGFR (included in the meta‐analysis): at 2 years, Cockroft‐Gault equation.
Participants	Recipient characteristics Number: 168; ATG+low CsA (53); BSX+low CsA (58); standard CsA (45) Mean age ± SD (years): ATG+low CsA (44.4 ± 32.1); BSX+low CsA (44.1 ± 32.7); standard CsA (46.4 ± 27.6) Sex (M/F): ATG+low CsA (14/39); BSX+low CsA (27/31); standard CsA (15/29) Donor characteristics Type of donor: deceased donors (100% DBD) Mean age ± SD (years): ATG+low CsA (40.6); BSX+low CsA (42.0); standard CsA (40.0) Other information Mean CIT ± SD (hours): ATG+low CsA (21.8); BSX+low CsA (22.7); standard CsA (22.1) Reason for high risk of DGF: mean CIT > 18 hours
Interventions	Group 1: ATG + low CsA CsA 2.5 mg/kg pre‐op, continued at 5 mg/kg after surgery: target level at day 7 200 to 300 ug/L Group 2: BSX + low CsA CsA 2.5 mg/kg pre‐op, continued at 5 mg/kg after surgery: target level at day 7 200 to 300 ug/L Group 3: no induction + standard CsA CsA 5 mg/kg pre‐op, continued at 10 mg/kg after surgery: target level at day 7 200 to 300 ug/L Co‐interventions (all groups) AZA, steroids
Outcomes	Primary outcome DGF AR Secondary outcomes Adverse events Graft survival
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated numbered randomisation
Allocation concealment (selection bias)	Low risk	Consecutively numbered sealed envelopes
Blinding of participants and personnel (performance bias) DGF	Unclear risk	Unblinded, DGF definition based on dialysis indication (potentially affected by lack of blinding) or creatinine (unlikely to be affected by lack of blinding)
Blinding of participants and personnel (performance bias) Acute rejection	Low risk	All rejection episodes were biopsy proven, unlikely to be affected by lack of blinding
Blinding of participants and personnel (performance bias) eGFR	Low risk	Unlikely to be affected by lack of blinding
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Death	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) DGF	Unclear risk	Unblinded, DGF definition based on dialysis indication (potentially affected by lack of blinding) or creatinine (unlikely to be affected by lack of blinding)
Blinding of outcome assessment (detection bias) Acute rejection	Low risk	All rejection episodes were biopsy proven, unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) eGFR	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias) DGF	Low risk	Low number missing, balanced between groups
Incomplete outcome data (attrition bias) Acute rejection	Low risk	Low number missing, balanced between groups
Incomplete outcome data (attrition bias) eGFR	Low risk	Low number missing, balanced between groups
Incomplete outcome data (attrition bias) Graft loss	Low risk	Low number missing, balanced between groups
Incomplete outcome data (attrition bias) Death	Low risk	Low number missing, balanced between groups
Selective reporting (reporting bias)	Low risk	Protocol
Other bias	Low risk	The study appears to be free from other sources of bias

Lebranchu 2002.

Study characteristics
Methods	Study design Parallel RCT Study duration: 1998 to 2000 Country: France (9 transplant centres) Mean duration of follow‐up: 6 months Definition of DGF: defined as the need for regular postoperative dialysis, graftectomy, or retransplantation Definition of eGFR: at 1 year, Cockroft formula
Participants	Recipient characteristics Number: 100; BSX+early CsA (50); ATG+delayed CsA (50) Mean age ± SD (years): BSX+early CsA (44.1 ± 11.5); ATG+delayed CsA (45.8 ± 10.8) Sex (M/F): BSX+early CsA (36/14); ATG+delayed CsA (32/18) Donor characteristics Type of donor: deceased donors Mean age ± SD (years): BSX+early CsA (41.1 ± 12.1); ATG+delayed CsA (41.1 ± 12.1) Other information Mean CIT ± SD (hours): BSX+early CsA (19.3 ± 6.5); ATG+delayed CsA (20.4 ± 6.7) Reason for high risk of DGF: mean CIT > 18 hours
Interventions	Group 1: BSX + early CsA Start of CsA during the first 24 hours post‐transplant; dose 6 to 8 mg/kg: levels 150 to 250 ng/mL (days 0 to 14), 150 to 200 ng/mL (day 15 to week 12), and 125 to 175 ng/mL (weeks 13 to 24) Group 2: ATG + delayed CsA Start of CsA when SCr < 250 µmol/L, dose 6 to 8 mg/kg Co‐interventions (all groups) MMF, steroids
Outcomes	Reported outcomes First AR First BPAR First AR treated with antibody therapy or TAC Graft loss Treatment failure Death Kidney function
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information about the sequence generation process to permit judgement
Allocation concealment (selection bias)	Unclear risk	Randomisation stated but no information on method used is available
Blinding of participants and personnel (performance bias) DGF	Unclear risk	Not blinded, but definition of DGF can be either based on creatinine (unlikely to be affected by lack of blinding) or dialysis (potentially affected by lack of blinding)
Blinding of participants and personnel (performance bias) Acute rejection	High risk	Not blinded
Blinding of participants and personnel (performance bias) eGFR	Low risk	Not likely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Death	Low risk	Not likely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Infections	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Cytopenias	Unclear risk	Unblinded, could affect reporting
Blinding of outcome assessment (detection bias) DGF	Unclear risk	Not blinded, but definition of DGF can be either based on creatinine (unlikely to be affected by lack of blinding) or dialysis (potentially affected by lack of blinding)
Blinding of outcome assessment (detection bias) Acute rejection	High risk	Not blinded
Blinding of outcome assessment (detection bias) eGFR	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Infections	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Cytopenias	Low risk	Unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias) DGF	Low risk	Few missing, balanced between groups
Incomplete outcome data (attrition bias) Acute rejection	Low risk	Few missing, balanced between groups
Incomplete outcome data (attrition bias) eGFR	Low risk	Few missing, balanced between groups
Incomplete outcome data (attrition bias) Graft loss	Low risk	Few missing, balanced between groups
Incomplete outcome data (attrition bias) Death	Low risk	Few missing, balanced between groups
Incomplete outcome data (attrition bias) Infections	Low risk	Few missing, balanced between groups
Incomplete outcome data (attrition bias) Cytopenias	Low risk	Few missing, balanced between groups
Selective reporting (reporting bias)	Unclear risk	Insufficient information to permit judgement
Other bias	Unclear risk	Insufficient information to permit judgement

Mourad 2005.

Study characteristics
Methods	Study design Parallel RCT Study duration: 2002 to 2004 Country: France (18 centres) Mean duration of follow‐up: 12 months Definition of DGF: SCr > 264 µmol/L at day 5 Definition of eGFR: at 1 year, Cockroft formula
Participants	Recipient characteristics Number 197; early CsA (97); delayed CsA (100) Mean age ± SD (years): early CsA (47 ± 13); delayed CsA (48 ± 13) Sex (M/F): early CsA (66/31); delayed CsA (62/38) Donor characteristics Type of donor: deceased donors (> 95%) Age > 50 years: early CsA (37%); delayed CsA (39%) Other information Mean CIT ± SD (hours): early CsA (19 ± 8); delayed CsA (21 ± 8) Reason for high risk of DGF: mean CIT > 18h.
Interventions	Group 1: early CsA CsA started at day 0, 8 mg/kg/d: C2 levels 1300 ng/mL (weeks 0 to 6); 1150 ng/mL (weeks 6 to 12), 900 ng/mL (months 4 to 6), and then 700 ng/mL. C0 levels not recorded Group 2: delayed CsA CsA started at day 6, 8 mg/kg/d: C2 levels: 1300 ng/mL (weeks 0 to 6), 1150 ng/mL (weeks 6 to 12), 900 ng/mL (months 4 to 6), and then 700 ng/mL. C0 levels not recorded Co‐interventions (all groups) MPA, steroid tapering, BSX/daclizumab
Outcomes	Primary outcome CrCl Secondary outcomes AR DGF Adverse events
Notes	Funded by a grant from Novartis Pharma AG
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated randomisation sequence
Allocation concealment (selection bias)	Unclear risk	Randomisation stated but no information on method use is available
Blinding of participants and personnel (performance bias) DGF	Low risk	DGF outcome not likely to be influenced by lack of blinding (based on creatinine definition)
Blinding of participants and personnel (performance bias) Acute rejection	High risk	Not blinded. Not all acute rejections were biopsy‐proven
Blinding of participants and personnel (performance bias) eGFR	Low risk	Not likely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Death	Low risk	Not likely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Infections	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Cytopenias	Unclear risk	Unblinded, could affect reporting
Blinding of outcome assessment (detection bias) DGF	Low risk	DGF mesurement not likely to be influenced by lack of blinding (based on creatinine definition)
Blinding of outcome assessment (detection bias) Acute rejection	High risk	Not blinded. Not all acute rejections were biopsy‐proven
Blinding of outcome assessment (detection bias) eGFR	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Infections	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Cytopenias	Low risk	Not likely to be influenced by lack of blinding
Incomplete outcome data (attrition bias) DGF	Unclear risk	Insufficient information to permit judgement
Incomplete outcome data (attrition bias) Acute rejection	Unclear risk	Insufficient information to permit judgement
Incomplete outcome data (attrition bias) eGFR	Unclear risk	Insufficient information to permit judgement
Incomplete outcome data (attrition bias) Graft loss	Unclear risk	Insufficient information to permit judgement
Incomplete outcome data (attrition bias) Death	Unclear risk	Insufficient information to permit judgement
Incomplete outcome data (attrition bias) Infections	Unclear risk	Insufficient information to permit judgement
Incomplete outcome data (attrition bias) Cytopenias	Unclear risk	Insufficient information to permit judgement
Selective reporting (reporting bias)	Unclear risk	Insufficient information to permit judgement
Other bias	Unclear risk	Funded by a grant from Novartis Pharma AG

Noel 2009.

Study characteristics
Methods	Study design Parallel RCT Study duration: 2001 to 2005 Country: France and Belgium (19 sites) Mean duration of follow‐up: 12 months Definition of DGF: need for dialysis within the first week after transplantation Definition of eGFR: 1 year, MDRD formula
Participants	Recipient characteristics Number: 227; ATG+delayed TAC (113); daclizumab+early TAC (114) Mean age ± SD (years): ATG+delayed TAC (45.4 ± 10.3); daclizumab+early TAC (46.9 ± 9.0) Sex (M/F): ATG+delayed TAC (52/61); daclizumab+early TAC (59/55) Donor characteristics: Type of donor: deceased donors (all DBD) Mean age ± SD (years): ATG+delayed TAC (44.3 ± 13.8); daclizumab+early TAC (44.6 ± 14.8) Other information Mean CIT ± SD (hours): ATG+delayed TAC (24.0 ± 7.9); daclizumab+early TAC (22.7±6.8) Reason for high risk of DGF: mean CIT >18 hours
Interventions	Group 1: ATG + delayed TAC TAC started on day 2 or day 5 if no spontaneous decrease in SCr, 0.2 mg/kg: target levels 10 to 15 ng/mL (3 months), then 8 to 12 ng/mL Group 2: daclizumab + early TAC TAC started before transplant, 0.2 mg/kg: target levels 10 to 15 ng/mL (3 months), then 8 to 12 ng/mL Co‐interventions (all groups) MMF, steroid tapering
Outcomes	Primary outcome BPAR Secondary outcomes DGF Kidney function Patient and graft survival
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Central randomisation procedure
Allocation concealment (selection bias)	Unclear risk	Randomisation stated but no information on method used is available
Blinding of participants and personnel (performance bias) DGF	High risk	Not blinded (DGF definition based on dialysis indication)
Blinding of participants and personnel (performance bias) Acute rejection	Low risk	Primary endpoint BPAR, unlikely to be affected by lack of blinding
Blinding of participants and personnel (performance bias) eGFR	Low risk	Unlikely to be affected by lack of blinding
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Death	Low risk	Unlikely to be affected by lack of blinding
Blinding of participants and personnel (performance bias) Infections	Unclear risk	Unblinded, could affect reporting
Blinding of outcome assessment (detection bias) DGF	High risk	Not blinded (DGF definition based on dialysis indication)
Blinding of outcome assessment (detection bias) Acute rejection	Low risk	Primary endpoint BPAR, unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) eGFR	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Infections	Low risk	Unlikely to be affected by lack of blinding
Incomplete outcome data (attrition bias) DGF	High risk	Not blinded (DGF definition based on dialysis indication)
Incomplete outcome data (attrition bias) Acute rejection	Low risk	Balanced causes between groups
Incomplete outcome data (attrition bias) eGFR	Low risk	Balanced causes between groups
Incomplete outcome data (attrition bias) Graft loss	Low risk	Balanced causes between groups
Incomplete outcome data (attrition bias) Death	Low risk	Balanced causes between groups
Incomplete outcome data (attrition bias) Infections	Low risk	Balanced causes between groups
Selective reporting (reporting bias)	Low risk	Study protocol available
Other bias	Low risk	The study appears to be free of other sources of bias

OSAKA 2013.

Study characteristics
Methods	Study design Parallel RCT Study duration: 2008 to 2010 Country: international (110 sites) Mean duration of follow‐up: 168 days Definition of DGF: dialysis for more than 1 day within the first week after transplant Definition of eGFR: at 168 days, MDRD formula
Participants	Recipient characteristics Number: 1198; IR reduced TAC (154); PR reduced TAC (155); PR standard TAC (153); PR reduced TAC+BSX (158) Mean age ± SD (years): IR reduced TAC (56.5 ± 11.4); PR reduced TAC (56.3 ± 11.3); PR standard TAC (55.6 ± 11.9); PR reduced TAC+BSX (54.4 ± 12.3) Sex (M/F): IR reduced TAC (110/44); PR reduced TAC (106/49); PR standard TAC (104/49); PR reduced TAC+BSX (109/49) Donor characteristics ECD donors: 620 Type of donor: extended criteria donors (deceased donors > 95%) Mean age ± SD (years): IR reduced TAC (61.6 ± 9.4); PR reduced TAC (62.1 ± 8.7); PR standard TAC (61.8 ± 8.3); PR reduced TAC+BSX (61.1 ± 8.6) Other information Mean CIT ± SD (hours): IR reduced TAC (14.3 ± 5.5); PR reduced TAC (15.4 ± 5.8); PR standard TAC (15.1 ± 5.6); PR reduced TAC+BSX (15.5 ± 5.8) Reason for high risk of DGF: ECD
Interventions	Group 1: IR reduced TAC + steroid taperingº IR TAC: 0.2 mg/kg/d Group 2: PR reduced TAC + steroid taperingº PR TAC: 0.2 mg/kg/d Group 3: PR standard TAC + steroid tapering PR TAC: 0.3 mg/kg/d Group 4: PR reduced TAC + BSX (ST only initial bolus)* PR TAC at 0.2 mg/kg/d Co‐interventions (all groups) MMF Groups 1, 2 and 3 received no induction ºinterventions included in the "low TAC" group *intervention not included in the meta‐analysis
Outcomes	Primary outcome Composite of graft loss, BPAR or graft dysfunction Secondary outcomes eGFR BPAR Graft and patient survival NODAT
Notes	Study funded by Astellas Pharma Global Development
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation using an interactive voice response system
Allocation concealment (selection bias)	Low risk	Allocation using an interactive voice response system
Blinding of participants and personnel (performance bias) DGF	High risk	Unblinded, DGF definition based on dialysis indication
Blinding of participants and personnel (performance bias) Acute rejection	Low risk	Acute rejection diagnosed with biopsy, unlikely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) eGFR	Low risk	Unlikely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Infections	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Cytopenias	Unclear risk	Unblinded, could affect reporting
Blinding of outcome assessment (detection bias) DGF	High risk	Unblinded, DGF definition based on dialysis indication
Blinding of outcome assessment (detection bias) Acute rejection	Low risk	Acute rejection diagnosed with biopsy, unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) eGFR	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Infections	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Cytopenias	Low risk	Unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias) DGF	Low risk	Missing data and causes balanced between groups
Incomplete outcome data (attrition bias) Acute rejection	Low risk	Missing data and causes balanced between groups
Incomplete outcome data (attrition bias) eGFR	Low risk	Missing data and causes balanced between groups
Incomplete outcome data (attrition bias) Graft loss	Low risk	Missing data and causes balanced between groups
Incomplete outcome data (attrition bias) Infections	Low risk	Missing data and causes balanced between groups
Incomplete outcome data (attrition bias) Cytopenias	Low risk	Missing data and causes balanced between groups
Selective reporting (reporting bias)	Unclear risk	Insufficient information to permit judgement
Other bias	High risk	Study funded by Astellas Pharma Global Development

SENIOR 2009.

Study characteristics
Methods	Study design Parallel RCT Study duration: 2004 to 2006 Country: multicentre (32 sites from 7 European countries) Mean duration of follow‐up: 6 months Definition of DGF: dialysis for more than 1 day within the first week after transplant Definition of eGFR: at 6 months, Cockroft‐Gault formula
Participants	Recipient characteristics Number: 254; early TAC (122); delayed TAC (132) Mean age ± SD (years): early TAC (65.5 ± 3.8); delayed TAC (66.4 ± 4) Sex (M/F): early TAC (74/48); delayed TAC (83/49) Donor characteristics Type of donor: > 97% deceased donors Mean age ± SD (years): early TAC (63.2 ± 13.7); delayed TAC (63.2 ± 13.3) Other information Mean CIT ± SD (hours): early TAC (14.2 ± 6.1); early TAC (14.1 ± 6.3) Reason for high risk of DGF: donors > 60 years
Interventions	Group 1: early TAC TAC started on the day of the transplant, initial dose 0.15 mg/kg twice/d: target concentrations 10 to 15 ng/mL (days 0 to 14), 5 to 12 ng/mL (days 15 to 42), then 5 to 10 ng/mL Group 2: delayed TAC + BSX induction TAC started on day 7, initial dose 0.15 mg/kg twice/d: target concentrations 10 to 15 ng/mL (days 0 to 14), 8 to 12 ng/mL (days 15 to 42), then 5 to 10 ng/mL Co‐interventions (all groups) MMF and steroid tapering Early TAC did not receive induction
Outcomes	Primary outcome Kidney function Secondary outcomes BPAR Graft and patient survival Adverse events DGF NODAT
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated randomised list
Allocation concealment (selection bias)	Unclear risk	Randomisation stated but no information on method used is available
Blinding of participants and personnel (performance bias) DGF	High risk	Not blinded (DGF definition based on dialysis indication)
Blinding of participants and personnel (performance bias) Acute rejection	High risk	Not blinded
Blinding of participants and personnel (performance bias) eGFR	Low risk	Not likely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Death	Low risk	Not likely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Infections	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Cytopenias	Unclear risk	Unblinded, could affect reporting
Blinding of outcome assessment (detection bias) DGF	High risk	Not blinded (DGF definition based on dialysis indication)
Blinding of outcome assessment (detection bias) Acute rejection	High risk	Not blinded
Blinding of outcome assessment (detection bias) eGFR	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Infections	Low risk	Not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Cytopenias	Low risk	Not likely to be influenced by lack of blinding
Incomplete outcome data (attrition bias) DGF	Low risk	Balanced causes between groups
Incomplete outcome data (attrition bias) Acute rejection	Low risk	Balanced causes between groups
Incomplete outcome data (attrition bias) eGFR	Low risk	Balanced causes between groups
Incomplete outcome data (attrition bias) Graft loss	Low risk	Balanced causes between groups
Incomplete outcome data (attrition bias) Death	Low risk	Balanced causes between groups
Incomplete outcome data (attrition bias) Infections	Low risk	Balanced causes between groups
Incomplete outcome data (attrition bias) Cytopenias	Low risk	Balanced causes between groups
Selective reporting (reporting bias)	Low risk	Protocol published
Other bias	Unclear risk	Editing assistance received financial remuneration from study sponsor

Thomas 2007.

Study characteristics
Methods	Study design Parallel RCT Study duration: 2005 to 2006 Country: USA Median follow‐up (IQR range): 377 days (127 to 611) Definition of DGF: not reported Definition of eGFR: not reported
Participants	Recipient characteristics Number: 21; early TAC (11); delayed TAC (8) Mean age ± SD (years): early TAC (43.5 ± 4.1); delayed TAC (47.1 ± 4.2) Sex (M/F):early TAC (6/5); delayed TAC (2/6) Donor characteristics Type of donor: deceased donors Mean age ± SD (years): not reported Other information Mean CIT ± SD (hours): early TAC (34.0 ± 4.2); delayed TAC (23.7 ± 4.6) Reason for high risk of DGF: mean CIT > 18 hours
Interventions	Group 1: early TAC + alemtuzumab TAC started on first post‐op day: target level 10 ng/mL Group 2: delayed TAC + ATG TAC started when SCr < 3 mg/dL or on post‐op day 3: target level 10 ng/mL Co‐interventions (all groups) MMF, steroid tapering
Outcomes	Primary outcome Graft function Secondary outcomes AR Safety
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information about the sequence generation process to permit judgement
Allocation concealment (selection bias)	Unclear risk	Randomisation stated but no information on method used is available
Blinding of participants and personnel (performance bias) Acute rejection	High risk	Not blinded
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Death	Low risk	Unlikely to be affected by lack of blinding
Blinding of outcome assessment (detection bias) Acute rejection	High risk	Not blinded
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Unlikely to be affected by lack of blinding
Incomplete outcome data (attrition bias) Acute rejection	Low risk	No missing data
Incomplete outcome data (attrition bias) Graft loss	Low risk	No missing data
Incomplete outcome data (attrition bias) Death	Low risk	No missing data.
Selective reporting (reporting bias)	Unclear risk	Insufficient information to permit judgement
Other bias	Unclear risk	Low number included from eligible patients

Wilson 2005.

Study characteristics
Methods	Study design Parallel RCT Study duration: 2000 to 2003 Country: UK (2 sites) Mean duration of follow‐up: 3 months Definition of DGF: requirement for more than one dialysis session in the first week after transplantation, with subsequent evidence of graft function Definition of eGFR: at 3 months, Cockcroft‐Gault formula
Participants	Recipient characteristics Number: 51; early TAC (26); delayed TAC (25) Mean age ± SD (years): early TAC (47 ± 12); delayed TAC (53 ± 14) Sex (M/F): early TAC (14/12); delayed TAC (17/8) Donor characteristics Type of donor: deceased donors (all DCD) Mean age ± SD (years): early TAC (44 ± 15); delayed TAC (47 ± 15) Other information Mean CIT ± SD (hours): early TAC (21 ± 6); delayed TAC (23 ± 8) Reason for high risk of DGF: DCD donors, mean CIT > 18 hours
Interventions	Group 1: early TAC TAC started the day after transplantation at 0.1mg/kg twice/d: target level 8 to 12 ng/mL No induction Group 2: delayed TAC+daclizumab TAC started when SCr < 350 µmol/L or evidence of AR, at 0.1 mg/kg twice/d: target level 8 to 12 ng/mL Induction with daclizumab Co‐interventions (all groups) MMF, steroid tapering
Outcomes	Primary outcome Graft function Secondary outcomes AR DGF Safety
Notes	Funded by Fujisawa and Roche
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Predesignated sequence overseen by investigators
Allocation concealment (selection bias)	High risk	Predesignated sequence overseen by investigators
Blinding of participants and personnel (performance bias) DGF	Unclear risk	Not blinded, unclear definition of DGF
Blinding of participants and personnel (performance bias) Acute rejection	High risk	Not blinded
Blinding of participants and personnel (performance bias) eGFR	Low risk	Unlikely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Graft loss	Unclear risk	Unblinded, could affect reporting
Blinding of participants and personnel (performance bias) Death	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) DGF	Unclear risk	Not blinded, unclear definition of DGF
Blinding of outcome assessment (detection bias) Acute rejection	High risk	Not blinded
Blinding of outcome assessment (detection bias) eGFR	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Graft loss	Low risk	Unlikely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Death	Low risk	Unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias) DGF	Low risk	Missing and causes balanced between groups
Incomplete outcome data (attrition bias) Acute rejection	Low risk	Missing and causes balanced between groups
Incomplete outcome data (attrition bias) eGFR	Low risk	Missing and causes balanced between groups
Incomplete outcome data (attrition bias) Graft loss	Low risk	Missing and causes balanced between groups
Incomplete outcome data (attrition bias) Death	Low risk	Missing and causes balanced between groups
Selective reporting (reporting bias)	Unclear risk	Insufficient information to permit judgement
Other bias	High risk	Study funded by Fujisawa and Roche Baseline characteristics not balanced between groups

AR: acute rejection; ATG: anti‐thymocyte globulin; AZA: azathioprine; BPAR: biopsy‐proven BSX: basiliximab; AR; CrCl: creatinine clearance; CsA: cyclosporin A; CIT: cold ischaemic time; DBD: donation after brain death; DCD: donation after circulatory death; DGF: delayed graft function; eGFR: estimated GFR; GFR: glomerular filtration rate; IR: immediate release; MDRD: Modification of Diet in Renal Disease; MMF: mycophenolate mofetil; MPA: mycophenolic acid; NODAT: new‐onset diabetes after transplant; NODM: new‐onset diabetes mellitus; pre‐op: pre‐operation; RCT: randomised controlled trial; SCr: serum creatinine; SD: standard deviation; SRL: sirolimus; TAC: tacrolimus

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
3C 2014	Wrong population: none at high risk of DGF
Alexander 2006	Wrong population: none at high risk of DGF
ALLEGRO 2018	Wrong intervention: intervention 6 months after transplantation, not initial immunosuppression
APOLLO 2015	Wrong intervention: conversion to mTORi
ASCERTAIN 2011	Wrong intervention: conversion to mTORi
Asher 2014	Wrong intervention: mTORi vs initial CNI
Asia‐OPTIMIZE 2019	Wrong intervention: reduced or standard dose after 4 weeks, no initial immunosuppression
ASSET 2012	Wrong intervention: reduced dose at 4 weeks, mTORi
ATHENA 2016	Wrong intervention: mTORi
Azevedo 2016	Wrong population: none at high risk of DGF
Baboolal 2003	Wrong intervention: intervention not in initial immunosuppression
Bang 2020	Wrong population: none at high risk of DGF
Bertoni 2011	Wrong population: none at high risk of DGF
BEST 2020	Wrong intervention: CNI avoidance
Bromberg 2011	Wrong population: none at high risk of DGF
Budde 2007	Wrong intervention: low vs standard after 1 month, not as initial immunosuppression
Burkhalter 2012	Wrong intervention: intervention at 3 months, not initial immunosuppression
CAESAR 2007	Wrong population: none at high risk of DGF
Cai 2014	Wrong population: none at high risk of DGF
CAMPASIA 2005	Wrong population: none at high risk of DGF
Chadban 2013	Wrong population: none at high risk of DGF
Chan 2008	Wrong population: none at high risk of DGF
Chan 2012	Wrong population: none at high risk of DGF
Cibrik 2007	Wrong population: none at high risk of DGF
CIS 2014	Wrong population: none at high risk of DGF
Cockfield 2019	Wrong population: none at high risk of DGF
CRADLE 2018	Wrong intervention: conversion to mTORi, at 4 to 6 weeks
David‐Neto 2014	Wrong intervention: conversion to mTORi at 1 month
de Sevaux 2001	Wrong population: approximately 30% of living donors; no separate results for deceased donors
DICAM 2010	Wrong intervention: intervention after 2 years, not initial immunosuppression
EVEREST 2009	Wrong population: none at high risk of DGF
Fangmann 2010	Wrong population: none at high risk of DGF
Ferguson 2006	Wrong population: none at high risk of DGF
Fleming 2016	Wrong intervention: mTORi conversion at 6 to 24 weeks
Fricke 2009	Wrong intervention: intervention at 3 months after transplantation, not initial immunosuppression
Ghadimi 2018	Wrong population: none at high risk of DGF
Ghafari 2007	Wrong population: none at high risk of DGF
HERAKLES 2018	Wrong intervention: intervention at 3 months, not initial immunosuppression
IRCT201604253043N11	Wrong population: none at high risk of DGF
Jindal 2002	Wrong intervention: intervention at 3 months after transplantation, not initial immunosuppression
Kamar 2012	Wrong intervention: intervention at 3 months after transplantation, not initial immunosuppression
Lee 2017	Wrong population: none at high risk of DGF
Liu 2007b	Wrong intervention: no delayed/early CNI, no low/standard dose as intervention
Lo 2003	Wrong intervention: CNI avoidance
Maamoun 2010	Wrong population: none at high risk of DGF
Maiorano 2006	Wrong population: none at high risk of DGF
Margreiter 2008	Wrong population: none at high risk of DGF
MODIFY 2012	Wrong population: none at high risk of DGF
Muhlbacher 2014	Wrong intervention: intervention at 1month, not initial immunosuppression
Nashan 2004	Wrong population: none at high risk of DGF
NCT00470665	Wrong intervention: CsA vs TAC
NCT00681213	Wrong intervention: mTORi
NCT01895049	Wrong intervention: mTORi vs MMF
NCT03644485	Wrong intervention: conversion from Prograf to Advagraf
NEVERWOUND 2020	Wrong intervention: mTORi
Novoa 2011	Wrong intervention: CNI elimination
Oh 2015	Wrong intervention: intervention at 1 week after transplantation, no initial immunosuppression
OPTICEPT 2009	Wrong population: none at high risk of DGF
OPTIMA‐TX 2008	Wrong intervention: not initial immunosuppression
OPTIMUM 2018	Wrong intervention: not initial immunosuppression
Pankewycz 2011	Wrong intervention: intervention at 3 months after transplantation, not initial immunosuppression
Paoletti 2012	Wrong population: none at high risk of DGF
Pascual 2003	Wrong intervention: intervention 1 year after transplantation
Qazi 2017	Wrong population: none at high risk of DGF
REFERENCE 2006	Wrong intervention: intervention 1 year after transplantation, not initial immunosuppression
Russ 2003	Wrong population: none at high risk of DGF
Salvadori 2007	Wrong intervention: intervention 1 year after transplantation, not initial immunosuppression
Saunders 2003	Wrong intervention: mTORi
Shetty 2015	Wrong population: none at high risk of DGF
SLOW & LOW 2023	Wrong population: none at high risk of DGF
SOCRATES 2014	Wrong intervention: intervention 2 weeks after transplantation, not initial immunosuppression
Stallone 2005	Wrong population: none at high risk of DGF
Stevens 2008	Wrong intervention: intervention after 6 months of transplantation, not initial immunosuppression
SYMPHONY 2007	Wrong population: none at high risk of DGF
Takahashi 2013a	Wrong population: none at high risk of DGF
Tang 2006	Wrong population: none at high risk of DGF
Tedesco‐Silva 2010	Wrong population: none at high risk of DGF
TRANSFORM 2013	Wrong population: none at high risk of DGF
Zheng 2022	Wrong intervention: intervention after 4 weeks of transplantation, not initial immunosuppression

CNI: calcineurin inhibitor; DGF: delayed graft function; MMF: mycophenolate mofetil; mTORi: mammalian target of rapamycin inhibitor

Characteristics of ongoing studies [ordered by study ID]

OPTIMIZE 2021.

Study name	RCT comparing Immunosuppressive regimens in elderly renal transplant recipients (OPTIMIZE)
Methods	Study design Parallel RCT Planned duration:2019 to 2026 Mean duration of follow‐up: 24 months Country: Belgium and the Netherlands (7 sites) Recruitment status: active, not recruiting
Participants	Study characteristics Inclusion criteria: written informed consent; male or female subject ≥ 65 years; randomised within 24 hours of completion of transplant surgery Stratum A: recipient of a primary (or secondary, if first graft is not lost due to immunological reasons) kidney transplant from a deceased donor aged 65 years or older Stratum B: recipient of a primary (or secondary, if first graft is not lost due to immunological reasons) kidney transplant from a deceased donor aged below 65 years or a living donor of any age Exclusion criteria: multi‐organ transplant recipient; blood group ABO incompatible allograft or CDC cross‐match positive transplant; at high immunological risk for rejection as determined by local practice for assessment of anti‐donor reactivity; cold ischaemia time > 24 hours; HLA‐identical related living donor; known intolerability for one or more of the study drugs; HIV positive; hepatitis B surface antigen and/or a HCV positive subject with evidence of elevated liver function tests (ALT/AST levels ≥ 2.5 times ULN). Viral serology results obtained within 6 months prior to randomisation are acceptable; recipient of a kidney from a donor who tests positive for HIV, hepatitis B surface antigen or anti‐hepatitis C virus; severe systemic infections, current or within the two weeks prior to randomisation; severe restrictive or obstructive pulmonary disorders; severe hypercholesterolaemia or hypertriglyceridemia that cannot be controlled; white blood cell count ≤ 2,000/mm3 or with platelet count ≤ 50,000/mm3 Planned enrolment: 374
Interventions	Group 1: low TAC + EVE Group 2: standard TAC + MMF Co‐interventions (all groups) Basiliximab induction, steroid tapering
Outcomes	Primary outcome Successful transplantation (absence of graft or patient loss with eGFR > 30) Secondary outcomes Death Graft loss Acute rejection eGFR Adverse events
Starting date	Actual: July 2019 Date of registration: 9 January 2019
Contact information	S. E. de Boer s.e.de.boer@umcg.nl
Notes	Co‐funding by Chiesi Pharmaceuticals

Differences between protocol and review

Although we initially planned to include cohort studies due to an expected reduced number of RCTs, we decided before data extraction to only include RCTs to improve the evidence quality.

Contributions of authors

1. Draft the protocol: LO, SC, EM, JMC, AC, AF, AM, NL, CC, PLV, NM

2. Study selection: LO, EM, SC

3. Data extraction: LO, NM

4. Enter data into RevMan: LO, NM

5. Conduct the analysis: LO, AR, NM

6. Interpret the analysis: LO, NM

7. Draft the final review: LO, SC, EM, AR, JMC, AC, AF, AM, NL, CC, PLV, NM

8. Disagreement resolution: NM

9. Update the review: LO, NM

Sources of support

Internal sources

• No internal sources of support, Other

  N/A

External sources

• Instituto de Salud Carlos III, Spain

  EM, AM, JMC, and NM have received funding: ISCIII‐FEDER RD21/0005/0001

Declarations of interest

• Laia Oliveras: no relevant interests were disclosed

• Pamela López‐Vargas: no relevant interests were disclosed

• Edoardo Melilli: no relevant interests were disclosed

• Sergi Codina: no relevant interests were disclosed

• Ana Royuela: no relevant interests were disclosed

• Ana Coloma López: no relevant interests were disclosed

• Alexandre Favà: no relevant interests were disclosed

• Anna Manonelles: no relevant interests were disclosed

• Núria Lloberas: no relevant interests were disclosed

• Carlos Couceiro: no relevant interests were disclosed

• Josep M Cruzado: no relevant interests were disclosed

• Nuria Montero: no relevant interests were disclosed

New