Background:
The impact and outcomes of postnatal cytomegalovirus (CMV) infection are not entirely clear. We aimed to determine the associations between treatment outcomes of postnatal CMV infection and its antiviral treatment.
Methods:
Retrospective study in a tertiary center. Infants of < 29 weeks gestational age who were tested for postnatal CMV infection were included. CMV-infected infants were compared to uninfected infants (control group). CMV-infected infants were either treated with ganciclovir and/or valganciclovir (CMVPT group) or not (CMVPNT group). Demographic, clinical, laboratory, treatment, and outcome data were collected. Primary outcomes were the length of stay, death before discharge and hearing impairment, cognitive and motor development as assessed by the Denver Developmental Screening Test II, and neurologic impairment at the corrected age of 1.5–2 years.
Results:
We included 103 extremely premature infants. The Median (interquartile range [IQR]) length of stay was 94 (69–112) days in control, 85 (70–102) days in CMVPNT, and 100 (88–137) days in the CMVPT group. Mortality before discharge was 6% in control, 3.8% in CMVPNT, and 3.7% in the CMVPT group. Normal hearing at follow-up was found in 30/37 infants in control (81.1%), 13/13 infants in CMVPNT (100%), and 17/20 infants in the CMVPT group (85%). Denver Developmental Screening Test II results did not differ among the three groups. Neurologic impairment was found in 21/37 infants (56.8%) in control, 9/13 infants in CMVPNT (69.2%), and 14/20 infants in CMVPT group (70%).
Conclusions:
The associations between antiviral treatment of postnatal CMV infection and better treatment outcomes were nonsignificant.
Keywords: cytomegalovirus, neonatology, hearing impairment, neurologic impairment
INTRODUCTION
Human cytomegalovirus (CMV) infection is a common viral infection in infants.1 Despite meagre evidence, there are indications that postnatal CMV infection might increase the risk for cognitive and motor developmental deficit and hearing impairment.2–6 Some studies found better cognitive and motor development in premature infants without postnatal CMV infection compared to postnatally-infected premature infants,3–5 while others did not.7–9
Despite extensive literature search, we did not find any clinical trials on safety and efficacy of antiviral treatment of postnatal CMV infection, only case reports and smaller retrospective cohort studies,10–17 and there are no evidence-based recommendations on which infants with postnatal CMV infection to treat and when to initiate antiviral treatment. The main problems in deciding on antiviral treatment for postnatal CMV infection are the unknown scope of infection-associated morbidity and uncertainties regarding the effect of antiviral treatment in improving treatment outcomes.1,16,18 If a decision is made to treat postnatal CMV infection, studies suggest oral valganciclovir or intravenous ganciclovir but data on their use in children is limited.1,16–19 Guidelines for diagnosis and treatment of congenital CMV infection are well-established in Slovenia20–24 and the world,25–29 but not for postnatal CMV infection.
We aimed to assess short-term and long-term outcomes of postnatal CMV infection in extremely premature infants with or without antiviral therapy and to use this data as treatment recommendations.
MATERIALS AND METHODS
Participants
Extremely premature infants of gestational age less than 29 weeks, who were tested for CMV infection on the 21st day of life or later were included in our 12-year retrospective study. The infants were hospitalized between January 2009 and September 2020 in the Neonatal Intensive Care Unit (NICU) at the Division of Gynaecology and Obstetrics, University Medical Center Ljubljana. The participants were followed-up at the corrected age of 1.5–2 years. Data was collected between July 2020 and August 2021 and obtained from maternity and delivery summaries and infants’ medical charts and discharge summaries.
Infants with postnatal CMV infection were compared to similar-age infants who had negative test results for CMV. Infants were divided into three groups depending on the presence or absence of postnatal CMV infection and antiviral therapy: CMV-negative infants (control group), CMV-positive infants who received antiviral therapy (CMVPT group), and CMV-positive infants who did not receive antiviral therapy (CMVPNT group). Fifty included infants were uninfected and 53 had postnatal CMV infection.
Exclusion criteria were: missing medical records from the intensive care unit, age < 21 days of life at microbiologic sample collection, only positive serologic tests, and diagnosis of congenital CMV infection.
CMV Infection
In all infants, CMV was detected in urine using real-time polymerase chain reaction (PCR) or viral cell culture. In some infants, CMV was also detected in blood samples (in addition to detection in urine). Nucleic acids were automatically isolated using MagNA pure Compact Nucleic Acid Isolation Kit I reagents and MagNA Pure Compact according to the manufacturer’s instructions (Roche Diagnostics, Mannheim, Germany). CMV DNA was detected using real-time PCR and CMV HHV6,7,8 R-GENE® reagent set according to the manufacturer’s instructions (Bioemerieux, France). For the viral cell culture method, CMV was isolated on MRC-5 culture (shell vial, human lung fibroblasts).
At our institution newborns were not routinely tested for CMV infection, therefore congenital infection could not be definitively excluded. However, CMV infection was considered probably postnatal if infants had positive CMV PCR or viral cell culture results on the 21st day of life or later without prior positive results. The most common indications for CMV testing were hepatosplenomegaly, pathologic liver function tests, worsening respiratory failure, and thrombocytopenia.
Maternal Characteristics
Maternal characteristics we assessed were mode of delivery, age at delivery, glucocorticoid application before or during labor, and presence of chorioamnionitis, preeclampsia, and gestational diabetes.
Infant Characteristics
Infant characteristics we assessed were demographic characteristics (sex, gestational age, birth weight, and its percentile), Apgar score at 1 and 5 minutes after birth, small for gestational age (SGA) status, age at CMV infection diagnosis, viral load in urine and/or plasma, presence of hepatosplenomegaly, presence of jaundice, minimal and maximal fraction of inspired oxygen (FiO2). We also assessed laboratory parameters including levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma-glutamyl transferase (γ-GT), direct and total bilirubin, and C-reactive protein (CRP), and platelet, leukocyte, and neutrophils counts. Presence of hepatosplenomegaly and jaundice, minimal and maximal FiO2, and laboratory parameters were assessed at microbiologic sample collection and the end of antiviral therapy course or 3 weeks from sample collection in the case of control and CMVPNT groups. We allowed for a 1-week deviation from this time frame, though data collected still reflected the infant’s condition post antiviral treatment.
Antiviral therapy of infants was characterized in terms of drugs received (ganciclovir and/or valganciclovir) and therapy duration. Antiviral treatment was initiated after consultation with a pediatric infectious disease specialist, with the most common reason for initiation of therapy being worsening respiratory failure. Other medications also documented were intravenous immunoglobulins, dexamethasone, and surfactant.
At follow-up, at the corrected age of 1.5–2 years, infants were assessed for hearing impairment, cognitive and motor development using Denver Developmental Screening Test II (DDST II), neurologic impairment, need for breathing support, and anthropometric measurements (body weight, height, and head circumference).
Definitions of Infant Characteristics
SGA was defined as birth weight lower than the 5th percentile for gestational age based on UK-WHO growth charts. Hepatosplenomegaly was defined as the liver extending 2 cm below the right costal margin and the spleen extending below the left costal margin. Jaundice was defined as yellow pigmentation of the skin. The presence of hepatosplenomegaly and jaundice was discerned from daily physical examination notes.
Primary Outcomes
Length of stay (LOS) was defined as number of days of hospitalization before discharge from the NICU. The second primary short-term outcome was death before discharge.
Sensorineural hearing loss was determined through auditory brainstem response at the corrected age of 1.5–2 years. Cognitive and motor development was evaluated using DDST II with four domains (fine motor skills, gross motor skills, socialization, and speech) which were performed by a pediatrician at follow-up at corrected age 1.5–2 years. Neurologic impairment was defined as the presence of paresis, paralysis, or muscle tone abnormalities.
Secondary Outcomes
Bronchopulmonary dysplasia (BPD) was defined as need for oxygen supplementation at 36 weeks gestational age. Intraventricular hemorrhage (IVH) was classified according to Papile30 and the presence of 3rd or 4th-degree IVH was considered as a secondary outcome. Other secondary outcomes included the presence of necrotizing enterocolitis, periventricular leukomalacia (PVL), retinopathy of prematurity (ROP), and sepsis as diagnosed by the attending physician, as well as results of hearing impairment screening with transient evoked otoacoustic emissions (TEOAE) before discharge.
Ethical Considerations
This study was approved by the National Medical Ethics Committee of the Republic of Slovenia (ref. no. 0120-373/2018/7 and ref. no. 0120/373/2018/16). The research is observational and all data was anonymized, therefore informed consent was not obtained.
Statistical Methods
Categorical variables were presented with frequencies and percentages; normally distributed numerical variables were presented with mean and standard deviation, and non-normally distributed numerical variables were presented with median and interquartile range. Associations between variables and presence of postnatal CMV infection and antiviral treatment were assessed using univariate multinominal logistical regression with the control group as a reference group. Likelihood ratio test was used if any of the cells in the contingency table contained 0 and Mann–Whitney U test when wide 95% confidence interval (CI) for odds ratio (OR) for numerical variables was present. Patients with missing values were included in the analysis, where data was present. No multivariate analysis, controlling for possible confounding factors, was applied, due to the small number of infants in the CMVPNT and CMVPT groups. Statistical analysis was performed at level of significance α = 0.05 in SPSS v. 27.0.
RESULTS
Participants
We initially included 110 infants, but 7 (5 postnatally CMV infected and 2 uninfected) were excluded for having met exclusion criteria (Figure 1). The final sample included 103 extremely premature infants of < 29 weeks gestational age who were tested for CMV infection on the 21st day of life or later and admitted to our NICU between January 2009 and September 2020. All infants were tested for CMV infection using urine samples, and 21 infants (20 CMV positive and 1 CMV negative) were tested for CMV infection using blood samples as well. No results were discordant. Postnatal CMV infection was confirmed in 53/103 infants (51%) who were compared to 50/103 (49%) postnatally uninfected infants (control group). 26/53 (49%) were postnatally CMV-infected and did not receive antiviral therapy (CMVPNT group) while 27/53 postnatally CMV-infected infants (51%) did (CMVPT group).
FIGURE 1.
Flowchart of our cohort.
Thirty-three infants were lost to follow-up at the corrected age of 1.5–2 years; 13/50 (26%) from the control group, 7/27 (25.9%) from the CMVPT group, and 13/26 (50%) from the CMVPNT group. Reasons for loss to follow-up were death before follow-up, not showing up to the appointment, and age < 1.5 years at the time of data collection.
Of 27 infants in the CMVPT group, 5 (18%) received ganciclovir only, 7 (26%) received valganciclovir only and 15 (56%) received both. The Median (IQR) duration of therapy was 19.5 (12–22) days. 7 infants (26 %) also received intravenous immunoglobulins.
Maternal Characteristics
No associations between maternal characteristics and presence of postnatal CMV infection or antiviral treatment were found.
Infant Characteristics
Median (IQR) birth weight was 700 (630–855) grams in the control group compared to 690 (610–800) grams in the CMVPNT group and 660 (600–750) grams in the CMVPT group.
Median (IQR) gestational ages were 25 (24–27) weeks in the control group, 24.5 (23–26) weeks in the CMVPNT group, and 24 (24–27) weeks in the CMVPT group. Infants in the CMVPNT group had significantly lower gestational age compared to the control group (P = 0.014) and were less likely to be SGA (0 infants in the CMVPNT group compared to 7 [14%] in the control group, P = 0.013).
Associations between clinical characteristics and laboratory parameters, and the presence of postnatal CMV infection and antiviral treatment at microbiologic sample collection or the end of antiviral therapy course or three weeks after sample collection are shown in Table 1 and Table 2, respectively.
TABLE 1.
Associations of Infants’ Clinical Characteristics at Sample Collection and Presence of Postnatal CMV Infection and Antiviral Treatment
| CMVPNT vs. NEG | CMVPT vs. NEG | ||||||
|---|---|---|---|---|---|---|---|
| NEG (n = 50) | CMVPNT (n = 26) | CMVPT (n = 27) | OR (95% CI) | P | OR (95% CI) | P | |
| Me (IQR) age at sample collection [days] | 42 (35–55) | 61.5 (54–72) | 57 (48–65) | 1.06 (1.03–1.1) | <0.001 | 1.04 (1.01–1.07) | 0.011 |
| viral load >100,000 copies/ml | |||||||
| Blood [n (%)] | 0 (0) | 0/3 (0) | 4/10 (40) | – | – | – | – |
| Urine [n (%)] | 0 (0) | 19/24 (79.2) | 16/25 (64) | – | – | - | - |
| Hepatosplenomegaly [n (%)] | 16/49 (32.7) | 20 (76.9) | 16/26 (61.5) | 6.87 (2.31–20.45) | <0.001 | 3.3 (1.23–8.88) | 0.018 |
| Jaundice [n (%)] | 12/49 (24,5) | 2 (7,7) | 11/26 (44) | 0.26 (0.05–1.25) | 0.092 | 2.42 (0.87–6.74) | 0.09 |
| Me (IQR; n) AST [μkat/l] | 1.67 (0.65–3.62; 26) | 0.82 (0.51–1.12; 12) | 1.26 (0.96–3.25; 24) | 0.65 (0.35–1.19) | 0.162 | 1.07 (0.86–1.33) | 0.535 |
| Me (IQR; n) ALT [μkat/l] | 0.35 (0.27–0.99; 25) | 0.4 (0.29–0.68; 12) | 0.51 (0.36–0.99; 24) | 0.72 (0.26–1.99) | 0.528 | 1.2 (0.76–1.88) | 0.435 |
| Me (IQR; n) γ–GT [μkat/l] | 1.8 (1.04–5.96; 23) | 1.65 (1.26–2.32; 9) | 2.62 (1.46–5.16; 22) | 0.8 (0.55–1.17) | 0.245 | 0.96 (0.78–1.19) | 0.712 |
| Me (IQR; n) direct bilirubin [μmol/l] | 39 (0–85; 19) | 0 (0–6; 9) | 62 (0.5–115.5; 16) | 0.95 (0.9–1) | 0.067 | 1 (0.99–1.01) | 0.716 |
| Me (IQR; n) total bilirubin [μmol/l] | 64.5 (25.5–114; 20) | 9 (6–32; 11) | 102.65 (10.5–143; 16) | 0.96 (0.93–1) | 0.03 | 1 (0.99–1.01) | 0.654 |
| Me (IQR; n) platelet count [×109/l] | 225 (114–294; 33) | 153.5 (126.5–244; 24) | 80 (66–159; 23) | 1 (1–1) | 0.977 | 0.98 (0.98–0.99) | 0.001 |
| Me (IQR; n) leukocyte count [×109/l] | 15.9 (11.3–20.6; 34) | 10.7 (8.7–11.9; 23) | 12.3 (9.3–17; 23) | 0.8 (0.7–0.92) | 0.001 | 0.9 (0.81–0.99) | 0.038 |
| Me (IQR; n) neutrophils count [×109/l] | 5.95 (3.85–7.95; 32) | 1.75 (1.1–3.95; 16) | 4.2 (2.8–7.1; 21) | 0.58 (0.41–0.82) | 0.002 | 0.91 (0.8–1.04) | 0.187 |
| Me (IQR; n) CRP [mg/l] | 8 (5–22; 35) | 8 (5–19; 18) | 19 (7–44; 25) | 0.98 (0.95–1.02) | 0.331 | 1.02 (1–1.04) | 0.081 |
| Me (IQR) minimal FiO2 | 0.3 (0.23–0.45) | 0.25 (0.21–0.3) | 0.4 (0.35–0.55) | 0.02 (0–1.1) | 0.056 | 8.3 (0.93–74.22) | 0.058 |
| Me (IQR) maximal FiO2 | 0.35 (0.25–0.55) | 0.27 (0.22–0.4) | 0.41 (0.35–0.6) | 0.02 (0–0.67) | 0.03 | 4.04 (0.59–27.84) | 0.157 |
CI, confidence interval; CMVPNT, CMV positive without antiviral treatment group; CMVPT, CMV positive on antiviral therapy group; IQR, interquartile range; Me, median; n, number of subjects; NEG, CMV negative group (control group); OR, odds ratio.
TABLE 2.
Associations of Infants’ Clinical Characteristics After End of Treatment Course or 3 Weeks After Sample Collection and Presence of Postnatal CMV Infection and Antiviral Treatment
| CMVPNT vs. NEG | CMVPT vs. NEG | ||||||
|---|---|---|---|---|---|---|---|
| NEG (n = 50) | CMVPNT (n = 26) | CMVPT (n = 27) | OR (95% CI) | P | OR (95% CI) | P | |
| Hepatosplenomegaly [n (%)] | 6/40 (15) | 0/13 (0) | 9/20 (45) | – | 0.057a | 4.64 (1.35–15.97) | 0.015 |
| Jaundice [n (%)] | 12/40 (30) | 0/13 (0) | 5/21 (23.8) | – | 0.005a | 0.73 (0.22–2.45) | 0.609 |
| Me (IQR; n) AST [μkat/l] | 3.65 (2.47–4.11; 10) | 2.3 (1.46–3.13; 2) | 2.79 (0.92–3.6; 12) | 0.58 (0.19–1.75) | 0.336 | 0.67 (0.36–1.25) | 0.207 |
| Me (IQR; n) ALT [μkat/l] | 1.21 (0.69–1.87; 10) | 1.2 (0.69–1.71; 2) | 0.87 (0.58–1.31; 12) | 0.91 (0.09–9.19) | 0.935 | 0.48 (0.12–1.93) | 0.301 |
| Me (IQR; n) γ–GT [μkat/l] | 1.94 (1.79–5.9; 10) | 4.3 (4.19–4.41; 2) | 3.43 (2.12–6.5; 11) | 1.16 (0.62–2.16) | 0.646 | 1.17 (0.81–1.7) | 0.408 |
| Me (IQR; n) direct bilirubin [μmol/l] | 49 (42–79; 9) | 25.5 (0–51; 2) | 66.5 (32–154; 8) | 0.97 (0.91–1.03) | 0.321 | 1.01 (0.99–1.03) | 0.332 |
| Me (IQR; n) total bilirubin [μmol/l] | 72 (69–121; 9) | 42 (15–69; 2) | 91.5 (46–183.5; 8) | 0.97 (0.92–1.03) | 0.282 | 1.01 (0.99–1.02) | 0.495 |
| Me (IQR; n) platelet count [×109/l] | 260 (192–391; 19) | 223 (157–335; 11) | 168 (120–253; 21) | 1 (0.99–1) | 0.384 | 0.99 (0.98–1) | 0.012 |
| Me (IQR; n) leukocyte count [×109/l] | 9.1 (7.75–15.3; 16) | 7.8 (5.6–11.8; 11) | 7.2 (5–11.9; 21) | 0.92 (0.8–1.06) | 0.273 | 0.95 (0.86–1.04) | 0.273 |
| Me (IQR; n) neutrophils count [×109/l] | 4.9 (3.1–8.1; 9) | 1.9 (1.3–3.2; 10) | 1.5 (0.9–3.1; 15) | 0.79 (0.54–1.17) | 0.24 | 0.96 (0.85–1.09) | 0.569 |
| Me (IQR; n) CRP [mg/l] | 5 (3–10; 11) | 5 (5–14; 9) | 14 (5–29; 14) | 0.98 (0.93–1.04) | 0.55 | 1 (0.98–1.03) | 0.721 |
| Me (IQR) minimal FiO2 | 0.26 (0.22–0.4) | 0.22 (0.22–0.24) | 0.3 (0.24–0.7) | 0 (0–2.94) | 0.077 | 8.53 (0.81–89.4) | 0.074 |
| Me (IQR) maximal FiO2 | 0.3 (0.23–0.45) | 0.23 (0.22–0.27) | 0.34 (0.25–0.7) | 0 (0–1.62) | 0.067 | 6.04 (0.67–54.8) | 0.11 |
CI, confidence interval; CMVPNT, CMV positive without antiviral treatment group; CMVPT, CMV positive on antiviral therapy group; IQR, interquartile range; Me, median; n, number of subjects; NEG, CMV negative group (control group); OR, odds ratio.
Odds ratio test.
For more detailed infant characteristics (viral load, gestational age, birth weight, presence of hepatosplenomegaly or jaundice, and laboratory parameters) see STable, Supplemental Digital Content 1 http://links.lww.com/INF/E846.
Outcomes
Table 3 shows associations between primary outcomes and postnatal CMV infection and antiviral treatment. Median (IQR) LOS was 94 (69–112) days in the control group compared to 85 (70–102) days in the CMVPNT group and 100 (88–137) days in the CMVPT group. In the control group, 3/50 (6%) infants died before discharge compared to 1/26 (3.8%) in the CMVPNT group and 1/27 (3.7%) in the CMVPT group. Normal hearing at corrected age of 1.5–2 years in 30/37 infants in control (81.1%), 13/13 infants in CMVPNT (100%), and 17/20 infants in CMVPT group (85%) with statistical significance reached in CMVPNT group in comparison to the control group (P = 0.032). No associations were found between presence of postnatal CMV infection and antiviral treatment and DDST II scores. Neurologic impairment was found in 21/37 infants (56.8%) in control, 9/13 infants in CMVPNT (69.2%), and 14/20 infants in CMVPT groups (70%). We found no significant associations between secondary outcomes and postnatal CMV infection (Table 4).
TABLE 3.
Associations of Primary Short-Term and Long-Term Outcomes With Presence of Postnatal CMV Infection and Antiviral Treatment
| CMVPNT vs. NEG | CMVPT vs. NEG | ||||||
|---|---|---|---|---|---|---|---|
| NEG (n = 50) | CMVPNT (n = 26) | CMVPT (n = 27) | OR (95% CI) | P | OR (95% CI) | P | |
| Me (IQR) Length of stay [days] | 94 (69–112) | 85 (70–102) | 100 (88–137) | 0.99 (0.98–1.01) | 0.457 | 1.01 (1–1.02) | 0.103 |
| Death before discharge [n (%)] | 3 (6) | 1 (3.8) | 1 (3.7) | 0.63 (0.06–6.34) | 0.692 | 0.6 (0.06–6.09) | 0.668 |
| Hearing at follow-up | |||||||
| Normal hearing [n (%)] | 30 (81.1) | 13 (100) | 17 (85) | – | 0.032 a | 1.32 (0.3–5.79) | 0.711 |
| Sensorineural hearing loss [n (%)] | 7 (18.9) | 0 (0) | 3 (15) | ||||
| Denver developmental screening test II | |||||||
| Me (IQR) Gross motor domain score | 0.94 (0.8–1) | 0.98 (0.88–1) | 1 (0.96–1) | – | 0.635b | – | 0,115b |
| Me (IQR) Fine motor domain score | 1 (0.92–1) | 1.06 (0.98–1.23) | 1 (0.88–1) | – | 0.077b | – | 0.694b |
| Me (IQR) Speech domain score | 0.93 (0.83–1) | 1 (0.9–1) | 0.92 (0.88–1) | 2.89 (0.11–72.82) | 0.519 | 3.97 (0.2–79.6) | 0.368 |
| Me (IQR) Socialization domain score | 1 (1–1) | 1 (0.96–1) | 1 (0.85–1) | 0.77 (0.01–77.65) | 0.91 | 1.26 (0.02–104.03) | 0.918 |
| Neurologic status | |||||||
| Normal [n (%)] | 21 (56.8) | 9 (69.2) | 14 (70) | 1.71 (0.45–6.58) | 0.432 | 1.78 (0.56–5.65) | 0.33 |
| Abnormal [n (%)] | 16 (43.2) | 4 (30.8) | 6 (30) | ||||
CI, confidence interval; CMVPNT, CMV positive without antiviral treatment group; CMVPT, CMV positive on antiviral therapy group; IQR, interquartile range; Me, median; n, number of subjects; NEG, CMV negative group (control group); OR, odds ratio.
Odds ratio test.
Mann–Whitney U test.
TABLE 4.
Associations of Secondary Outcomes With Presence of Postnatal CMV Infection and Antiviral Treatment
| CMVPNT vs. NEG | CMVPT vs. NEG | ||||||
|---|---|---|---|---|---|---|---|
| NEG (n = 50) | CMVPNT (n = 26) | CMVPT (n = 27) | OR (95% CI) | P | OR (95% CI) | P | |
| Bronchopulmonary dysplasia [n (%)] | 34 (68) | 16 (61.5) | 23 (85.2) | 0.75 (0.28–2.02) | 0.574 | 2.71 (0.8–9.14) | 0.109 |
| Intraventricular hemorrhage 3rd or 4th degree [n (%)] | 6 (12) | 1 (3.8) | 1 (3.7) | 0.51 (0.1–2.66) | 0.426 | 0.77 (0.18–3.25) | 0.72 |
| Necrotizing enterocolitis [n (%)] | 7 (14) | 1 (3.8) | 1 (3.7) | 0.25 (0.03–2.11) | 0.201 | 0.24 (0.03–2.03) | 0.189 |
| Periventricular leukomalacia [n (%)] | 0 (0) | 1 (3.8) | 0 (0) | – | 0.141a | – | 1a |
| Retinopathy of prematurtiy [n (%)] | 15 (30) | 12 (46.2) | 13 (48.1) | 2 (0.75–5.33) | 0.166 | 2.17 (0.82–5.7) | 0.117 |
| Sepsis [n (%)] | 8 (16) | 6 (23.1) | 4 (14.8) | 1.57 (0.48–5.15) | 0.452 | 0.91 (0.25–3.36) | 0.891 |
| Transient evoked otoacoustic emissions | |||||||
| Elicited [n (%)] | 28 (68.3) | 15 (65.2) | 14 (63.6) | 0.87 (0.3–2.57) | 0.802 | 0.81 (0.27–2.42) | 0.709 |
| Not elicited (unilateral or bilateral) [n (%)] | 13 (31.7) | 8 (34.8) | 8 (36.4) | ||||
| Me (IQR) weight at follow-up [g] | 10,000 (9,700–11,500) | 10,500 (8,950–10,700) | 10,200 (9,450–11,300) | 1 (1–1) | 0.468 | 1 (1–1) | 0.942 |
| Me (IQR) height at follow-up [cm] | 85 (84–88) | 88 (82–89) | 87 (82–88) | 0.98 (0.83–1.15) | 0.778 | 0.99 (0.86–1.13) | 0.833 |
| Me (IQR) head circumference at follow-up [cm] | 48 (47–49) | 48.3 (47.3–49.5) | 46.8 (46–48) | 1.16 (0.7–1.9) | 0.564 | 0.75 (0.51–1.09) | 0.133 |
| Breathing at follow-up | – | 0.191a | – | 0.171a | |||
| Independently [n (%)] | 24 (96) | 8 (88.9) | 11 (78.6) | ||||
| Addition of oxygen needed [n (%)] | 0 (0) | 1 (11.1) | 1 (7.1) | ||||
| Addition of oxygen and mechanical ventilation needed [n (%)] | 1 (4) | 0 (0) | 2 (14.3) | ||||
CI, confidence interval; CMVPNT, CMV positive without antiviral treatment group; CMVPT, CMV positive on antiviral therapy group; IQR, interquartile range; Me, median; n, number of subjects; NEG, CMV negative group (control group); OR, odds ratio.
Odds ratio test.
CMVPT group had statistically significantly higher odds for longer LOS in comparison to the CMVPNT group (OR [95% CI: 1.02 [1–1.04]), while the odds for death were not significantly reduced in this group (OR [95% CI: 0.96 [0.06–16.22]). No statistically significant differences between the two groups in hearing loss (P = 0.074), fine motor skills (p = 0.203), gross motor skills (P = 0.237), socialization (P = 0.778), speech (P = 0.717), or neurologic status (P = 0.963) was found.
DISCUSSION
The results of our study showed that primary short-term (LOS and death before discharge) and long-term (cognitive and motor development and sensorineural hearing loss at corrected age of 1.5–2 years) outcomes of infants did not differ between the three groups. Moreover, our results showed that the odds for secondary outcomes (BPD, IVH 3rd or 4th degree according to Papile, necrotizing enterocolitis, periventricular leukomalacia, ROP, and sepsis) did not differ among the three groups.
Some studies of premature infants with postnatal CMV infection showed an association between postnatal CMV infection and longer LOS2,31 while others did not.7,32 We did not find an association between postnatal CMV infection and longer LOS in our study. Earlier studies did not show an increase in mortality before discharge in extremely premature infants with postnatal CMV infection7,31–33 which is consistent with our results.
Kelly et al. found an increased risk of developing BPD in premature infants with postnatal CMV infection,33 a finding that was supported by later studies by Weimer et al.2 and Mukhopadhyay et al.34 Contrary to these studies, we did not find an association between the risk of developing BPD and postnatal CMV infection in our study.
Moreover, Weimer et al found an increased risk of hearing impairment at postmenstrual age of 34 weeks in very low birth weight (<1.500 g) premature infants with postnatal CMV infection which they attributed to the infection.2 Other earlier studies did not show an association between postnatal CMV infection and hearing impairment5,8,35 which is consistent with our results. In fact, in our study infected infants in CMVPNT groups were more likely to be without sensorineural hearing loss at follow-up.
Gunkel et al studied the long-term sequelae of postnatal CMV infection in extremely premature infants (<28 weeks of gestation) at corrected ages of 16 months, 24–30 months, and 6 years. They did not find any significant differences in neurodevelopment between infected and uninfected infants at the ages of 16 months and 24–30 months but showed lower scores in infected infants at age of 6 years, with statistical significance attained only for verbal IQ.8 Vollmer et al evaluated anthropometric measurements, hearing impairment, and neurologic status in the same cohort at the age of 2–4.5 years. Their results were consistent with ours since they did not find any differences in outcomes between the control group and the postnatally CMV-infected group.9 Same authors evaluated the same cohort at the age of 6–8 years and 11–17 and found worse cognitive and motor development in postnatally infected children, while the two groups did not differ in other outcomes.3,4 Negative effects of postnatal CMV infection on long-term outcomes have been reported by the same authors in two additional studies.5,6 Therefore, specific effects of early postnatal CMV infection might only be detectable later in life and only using an extensive neuropsychologic evaluation.4
There is little evidence on safety and efficacy of treatment of postnatal CMV infection with ganciclovir or valganciclovir. Existing studies suggest that 4–6 weeks long course of ganciclovir or valganciclovir improves the clinical picture, especially hematologic abnormalities.11–14 Hu et al. found an improvement in hematologic abnormalities after concluding the treatment course with ganciclovir but could not conclude whether this improvement was related to antiviral treatment or spontaneous resolution of the infection which coincided with the end of the treatment course.13 Comparing average neutrophil counts between Table 1 and Table 2 it seems some of the infants in the CMVPT group have developed neutropenia after finishing the treatment course. Indeed 4 infants (14.8%) in the CMVPT group developed neutropenia after the conclusion of antiviral treatment. Furthermore, 4 infants (14.8%) in the CMVPT group maintained thrombocytopenia, which had been present before initiation of antiviral treatment, after treatment conclusion. Also, 5 infants (18.5%) in the CMVPT group maintained elevated transaminases after antiviral treatment conclusion, and in 2 of those, the values were higher than at initiation of antiviral treatment.
The main limitation of our study is retrospecitve nature which makes it difficult to evaluate the impact of antiviral treatment on outcomes of postnatal CMV infection and we could not systematically monitor laboratory parameters. Therefore, included data was collected at different time points in the infants’ hospital stay and some of the data was missing. Another important limitation is confounding. Infants at our institution were treated after consulting a pediatric infectious disease specialist and based on the perceived worse clinical course (sepsis-like syndrome, hepatitis, pneumonitis, or bone marrow suppression). Since it is impossible to exclude that more severely ill postnatally CMV infected infants were treated with antivirals, the ability of our study to detect the benefit of antiviral treatment is reduced, but we can speculate that we managed to prevent worse outcomes with antiviral treatment in the group of more severely ill infants. Nonsignificant trends suggesting higher frequency of jaundice and BPD, and higher FiO2 needed before initiation of treatment in CMVPT group could possibly reflect worse clinical course of infants in this group compared to the CMVPNT group. It is also impossible to exclude, that infants in the control group were more severely affected by prematurity, since SGA was more common in that group, although significant only when compared to infants in the CMVPNT group. This would reduce the ability to detect negative outcomes of postnatal CMV infection and differences in outcomes between groups in our study, but it is also possible that antiviral treatment offered no benefit regarding treatment outcomes.
Infants included in our study were not routinely screened for congenital CMV infection, but prevalence of congenital CMV infection in Slovenia is low (0.14%),36 and it is not expected that many infants had congenital CMV infection. The usefulness of DDST II is limited since it is a screening test and cannot precisely assess the infant’s cognitive and motor developmental deficit. Relatively large loss to follow-up may present a selection bias, with infants lost to follow-up perhaps having worse outcomes. The COVID-19 pandemic also contributed to the loss of follow-up due to the interruption of nonessential medical services.
In conclusion, to answer the question of whether antiviral treatment of postnatal CMV infection affects short-term and long-term outcomes of the infection in extremely premature infants, larger prospective randomized studies are needed.
Supplementary Material
Footnotes
There are no conflicts of interest.
Rok Košiček and Borut Kristan contributed equally to this work.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (www.pidj.com).
Contributor Information
Vanja Erčulj, Email: vanja.erculj@rosigma.si.
Lilijana Kornhauser Cerar, Email: lilijana.kornhauser@kclj.si.
Miroslav Petrovec, Email: mirc.petrovec@mf.uni-lj.si.
Marko Pokorn, Email: marko.pokorn@kclj.si.
Ana Spirovska, Email: a.spirovska74@gmail.com.
Tina Uršič, Email: tina.Ursic@mf.uni-lj.si.
REFERENCES
- 1.Kadambari S, Whittaker E, Lyall H. Postnatally acquired cytomegalovirus infection in extremely premature infants: how best to manage? Arch Dis Child Fetal Neonatal Ed. 2020;105:334–339. [DOI] [PubMed] [Google Scholar]
- 2.Weimer KED, Kelly MS, Permar SR, et al. Association of adverse hearing, growth, and discharge age outcomes with postnatal cytomegalovirus infection in infants with very low birth weight. JAMA Pediatr. 2020;174:133–140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bevot A, Hamprecht K, Krägeloh-Mann I, et al. Long-term outcome in preterm children with human cytomegalovirus infection transmitted via breast milk. Acta Paediatr. 2012;101:e167–e172. [DOI] [PubMed] [Google Scholar]
- 4.Brecht KF, Goelz R, Bevot A, et al. Postnatal human cytomegalovirus infection in preterm infants has long-term neuropsychological sequelae. J Pediatr. 2015;166:834–9.e1. [DOI] [PubMed] [Google Scholar]
- 5.Goelz R, Meisner C, Bevot A, et al. Long-term cognitive and neurological outcome of preterm infants with postnatally acquired CMV infection through breast milk. Arch Dis Child Fetal Neonatal Ed. 2013;98:F430–F433. [DOI] [PubMed] [Google Scholar]
- 6.Dorn M, Lidzba K, Bevot A, et al. Long-term neurobiological consequences of early postnatal hCMV-infection in former preterms: a functional MRI study. Hum Brain Mapp. 2014;35:2594–2606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Jim WT, Chiu NC, Ho CS, et al. Outcome of preterm infants with postnatal cytomegalovirus infection via breast milk: a two-year prospective follow-up study. Medicine (Baltimore). 2015;94:e1835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gunkel J, de Vries LS, Jongmans M, et al. Outcome of preterm infants with postnatal cytomegalovirus infection. Pediatrics. 2018;141:e20170635. [DOI] [PubMed] [Google Scholar]
- 9.Vollmer B, Seibold-Weiger K, Schmitz-Salue C, et al. Postnatally acquired cytomegalovirus infection via breast milk: effects on hearing and development in preterm infants. Pediatr Infect Dis J. 2004;23:322–327. [DOI] [PubMed] [Google Scholar]
- 10.Okulu E, Akin IM, Atasay B, et al. Severe postnatal cytomegalovirus infection with multisystem involvement in an extremely low birth weight infant. J Perinatol. 2012;32:72–74. [DOI] [PubMed] [Google Scholar]
- 11.Meier J, Lienicke U, Tschirch E, et al. Human cytomegalovirus reactivation during lactation and mother-to-child transmission in preterm infants. J Clin Microbiol. 2005;43:1318–1324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mehler K, Oberthuer A, Lang-Roth R, et al. High rate of symptomatic cytomegalovirus infection in extremely low gestational age preterm infants of 22-24 weeks’ gestation after transmission via breast milk. Neonatology. 2014;105:27–32. [DOI] [PubMed] [Google Scholar]
- 13.Hu H, Cheng Y, Peng Q, et al. Clinical features, treatment courses, and distribution of cytomegalovirus genotypes among thrombocytopenia patients aged younger than 12 months. Am J Perinatol. 2020;97:339–345. [DOI] [PubMed] [Google Scholar]
- 14.Fischer C, Meylan P, Bickle Graz M, et al. Severe postnatally acquired cytomegalovirus infection presenting with colitis, pneumonitis and sepsis-like syndrome in an extremely low birthweight infant. Neonatology. 2010;97:339–345. [DOI] [PubMed] [Google Scholar]
- 15.Takahashi R, Tagawa M, Sanjo M, et al. Severe postnatal cytomegalovirus infection in a very premature infant. Neonatology. 2007;92:236–239. [DOI] [PubMed] [Google Scholar]
- 16.Schuster K, Goelz R, Speckmann C, et al. Symptomatic cytomegalovirus infections in the first year of life: when is antiviral therapy conceived to be justified? Pediatr Infect Dis J. 2017;36:224–227. [DOI] [PubMed] [Google Scholar]
- 17.Luck S, Lovering A, Griffiths P, et al. Ganciclovir treatment in children: evidence of subtherapeutic levels. Int J Antimicrob Agents. 2011;37:445–448. [DOI] [PubMed] [Google Scholar]
- 18.Luck S, Sharland M. Postnatal cytomegalovirus: innocent bystander or hidden problem? Arch Dis Child Fetal Neonatal Ed. 2009;94:F58–F64. [DOI] [PubMed] [Google Scholar]
- 19.Sunada M, Kinoshita D, Furukawa N, et al. Therapeutic drug monitoring of ganciclovir for postnatal cytomegalovirus infection in an extremely low birth weight infant: a case report. BMC Pediatr. 2016;16:141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Paro-Panjan D. Prirojena okužba z virusom citomegalije [Congenital cytomegalovirus infection]. Paro-Panjan D, ed. In: Neonatalne okužbe in imunski odziv pri novorojenčkih: mednarodni simpozij 2013. Univerzitetni klinični center Ljubljana; 2013:117–127. [Google Scholar]
- 21.Rednak-Paradiž K, Mlaker M, Paro-Panjan D, et al. Congenital cytomegalovirus infection [Prirojena okužba s citomegalovirusom]. Zdrav Vestn. 2006;75:727–732. [Google Scholar]
- 22.Mrvič T, Plankar Srovin T. Klinična slika prirojene in pridobljene okužbe s humanim virusom citomegalije pri otrocih. Beović B, Strle F, Čižman M, eds. In: Infektološki Simpozij 2008. Slovensko zdravniško društvo; 2008:201–211. [Google Scholar]
- 23.Konjajev Z. Izvirni izsledki o intrauterini okužbi novorojencev s citomegalovirusom [PhD thesis]. University of Ljubljana. 1973. [Google Scholar]
- 24.Arnež M, Bizjak S. Zdravljenje otrok s prirojeno okužbo s humanim virusom citomegalije. Beović B, Strle F, Čižman M, eds. In: Infektološki Simpozij 2008. Slovensko zdravniško društvo; 2008:231–241. [Google Scholar]
- 25.Kadambari S, Williams EJ, Luck S, et al. Evidence based management guidelines for the detection and treatment of congenital CMV. Early Hum Dev. 2011;87:723–728. [DOI] [PubMed] [Google Scholar]
- 26.Oliver SE, Cloud GA, Sánchez PJ, et al. ; National Institute of Allergy, Infectious Diseases Collaborative Antiviral Study Group. Neurodevelopmental outcomes following ganciclovir therapy in symptomatic congenital cytomegalovirus infections involving the central nervous system. J Clin Virol. 2009;46(suppl 4):S22–S26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kimberlin DW, Lin CY, Sánchez PJ, et al. ; National Institute of Allergy and Infectious Diseases Collaborative Antiviral Study Group. Effect of ganciclovir therapy on hearing in symptomatic congenital cytomegalovirus disease involving the central nervous system: a randomized, controlled trial. J Pediatr. 2003;143:16–25. [DOI] [PubMed] [Google Scholar]
- 28.Kimberlin DW, Jester PM, Sánchez PJ, et al. ; National Institute of Allergy and Infectious Diseases Collaborative Antiviral Study Group. Valganciclovir for symptomatic congenital cytomegalovirus disease. N Engl J Med. 2015;372:933–943. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Luck SE, Wieringa JW, Blázquez-Gamero D, et al. ; ESPID Congenital CMV Group Meeting, Leipzig 2015. Congenital cytomegalovirus: a European expert consensus statement on diagnosis and management. Pediatr Infect Dis J. 2017;36:1205–1213. [DOI] [PubMed] [Google Scholar]
- 30.Papile LA, Burstein J, Burstein R, et al. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr. 1978;92:529–534. [DOI] [PubMed] [Google Scholar]
- 31.Tran C, Bennett MV, Gould JB, et al. Cytomegalovirus infection among infants in neonatal intensive care units, California, 2005 to 2016. Am J Perinatol. 2020;37:146–150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Neuberger P, Hamprecht K, Vochem M, et al. Case-control study of symptoms and neonatal outcome of human milk-transmitted cytomegalovirus infection in premature infants. J Pediatr. 2006;148:326–331. [DOI] [PubMed] [Google Scholar]
- 33.Kelly MS, Benjamin DK, Puopolo KM, et al. Postnatal cytomegalovirus infection and the risk for bronchopulmonary dysplasia. JAMA Pediatr. 2015;169:e153785. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Mukhopadhyay S, Meyer SA, Permar SR, et al. Symptomatic postnatal cytomegalovirus testing among very low-birth-weight infants: indications and outcomes. Am J Perinatol. 2016;33:894–902. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Nijman J, van Zanten BG, de Waard AK, et al. Hearing in preterm infants with postnatally acquired cytomegalovirus infection. Pediatr Infect Dis J. 2012;31:1082–1084. [DOI] [PubMed] [Google Scholar]
- 36.Paradiž KR, Seme K, Puklavec E, et al. Prevalence of congenital cytomegalovirus infection in Slovenia: a study on 2,841 newborns. J Med Virol. 2012;84:109–115. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.