Secondary Logo

Journal Logo

Maternal-Neonatal Reports

Long-term Outcomes after Postnatal Cytomegalovirus Infection in Low Birthweight Preterm Infants

A Systematic Review

Stark, Ashley MD, MS*; Cantrell, Sarah MLIS; Greenberg, Rachel G. MD, MB, MHS*,‡; Permar, Sallie R. MD, PhD*,§; Weimer, Kristin E. D. MD, PhD, MHS*

Author Information
The Pediatric Infectious Disease Journal: June 2021 - Volume 40 - Issue 6 - p 571-581
doi: 10.1097/INF.0000000000003072

Abstract

The clinical manifestations and long-term outcomes of postnatally acquired cytomegalovirus (pCMV) in preterm infants are not well described. Although most commonly acquired via breast milk, pCMV can be acquired through blood product transfusion (albeit rare with the use of CMV seronegative or leukoreduced blood products), genital secretions during childbirth, and horizontal transmission.1–33 Transmission of pCMV to term and preterm infants ranges from 10% to 60%.19,30 Full-term infants with pCMV are generally asymptomatic; however, 14%–50% of very low birthweight infants (VLBW) infants can present with fever, pneumonitis, thrombocytopenia, lymphocytosis, encephalitis, hepatosplenomegaly, hepatitis and opportunistic infections.3,7,10–13,15–17,21,24,25,27,28,31,34–37 An acute sepsis-like illness occurs in approximately 1.4% of infected VLBW infants.17 pCMV can have multisystem involvement; however, studies are conflicting and do not provide a cohesive narrative of its long-term outcomes.1–36,38–41 To address this gap, we conducted a systematic review to synthesize current information on the long-term morbidity of pCMV for neonatologists and general pediatricians. Understanding the full clinical impact of pCMV and the gaps in this knowledge will help guide future clinical trials.

MATERIALS AND METHODS

Search Criteria

A systematic review was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines (PRISMA).42 Searches were conducted in MEDLINE (PubMed), EMBASE (Elsevier), the Cochrane Central Register for Controlled Trials (Wiley), CINAHL Complete (EBSCO) and Web of Science Core Collection (Clarivate) from inception through December 2019. Editorials, comments and animal-only studies were excluded. No restrictions were placed by date, publication status or language. Analysis methods and inclusion/exclusion criteria were registered with PROSPERO (registration number: CRD42020145599). The search included keywords and database-specific subject headings for CMV and VLBW infants (Appendix/Supplementary Materials, Supplemental Digital Content 1, http://links.lww.com/INF/E342). Case reports, observational studies, experimental studies and randomized controlled trials on pCMV in preterm and LBW infants and long-term (>1 month) impact of pCMV were included.

Patient Selection

Preterm infants [gestational age (GA) ≤36 weeks and/or birthweight (BW, <2000 g)] diagnosed with pCMV were included. pCMV was defined as testing positive for CMV by polymerase chain reaction (PCR) and exclusion of congenital CMV (cCMV) either by urine PCR at birth, previous negative CMV testing, or retrospective PCR of dried blood spot collected at birth. The following were excluded: review articles, articles that did not adequately differentiate pCMV from cCMV or if pCMV and cCMV outcomes were combined in the data reporting.

Study Selection

Two blinded reviewers independently determined the eligibility of studies identified during the initial screening of abstracts using Covidence systematic review software (Melbourne, Australia).43 A third reviewer blind to the previous reviewers’ decisions resolved disagreements. The same reviewers subsequently performed full-text screening independently to further determine eligibility. Disagreements were resolved as above. Possible duplications were reviewed and if deemed duplicates, one was excluded. The following information was extracted: GA, BW, pCMV status, type of study, number of subjects and single versus multicenter. Although non-English studies were unable to be translated, information was extracted from the abstract if in English and included in the final analysis.

Of the 3113 studies screened, a total of 40 met inclusion criteria (Fig. 1). Twelve studies were case reports, 7 retrospective studies and 21 prospective studies. Outcomes of the included studies were pulmonary (21), gastrointestinal (GI, 12), neurodevelopment (20), auditory (17), length of stay (LOS, 12), ophthalmologic (13) and anthropomorphic (9) (Table 1).

TABLE 1. - Outcomes Based on Study Design and Association
Outcome (N Studies) Total No. Subjects Case Report Studies (Subjects, %) Retrospective (Identified Symptomatic) Studies (Subjects, %) Prospective (Screened Population) Studies (Subjects, %)
Pulmonary 3743 3 (3) 7 (1761) 11 (1602)
Increased respiratory support 2 (2, 66%) 3 (1074, 61%) 2 (170, 11%)
No change in respiratory support 0 (0, 0%) 1 (80, 5%) 2 (144, 9%)
Increased prevalence of BPD 1 (1, 33%) 5 (1483, 84%) 4 (246, 15%)
No difference in prevalence of BPD 0 (0, 0%) 1 (80, 5%) 5 (1086, 68%)
Gastrointestinal 2546 3 (4) 6 (1676) 3 (866)
Association with enterocolitis 3 (4, 100%) 0 (0, 0%) 0 (0, 0%)
Increased prevalence of NEC 0 (0, 0%) 0 (0, 0%) 1 (596, 69%)
No difference in prevalence of NEC 0 (0, 0%) 6 (1676, 100%) 1 (270, 31%)
Neurodevelopmental 1975 5 (5) 5 (1019) 10 (951)
No neurological impairment 2 (2, 40%) 2 (499, 49%) 5 (283, 30%)
Neurological impairment 3 (3, 60%) 2 (449, 49%) 2 (107, 11%)
No difference in imaging findings 0 (0, 0%) 1 (176, 17%) 0 (0, 0%)
Difference in imaging findings 0 (0, 0%) 2 (436, 43%) 3 (557, 59%)
Auditory 2268 4 (4) 5 (1282) 8 (709)
Increased prevalence of hearing loss at follow-up 1 (1, 25%) 0 (0, 0%) 0 (0, 0%)
No difference in the prevalence of hearing loss at follow-up 3 (3, 75%) 4 (1009, 79%) 8 (709, 100%)
Increased rate of hearing screen failure 0 (0, 0%) 1 (273, 21%) 0 (0, 0%)
LOS 2163 0 (0) 5 (1165) 7 (725)
No difference in LOS 0 (0, 0%) 2 (424, 36%) 3 (432, 60%)
Increased LOS 0 (0, 0%) 3 (741, 64%) 4 (293, 40%)
Eye 1899 1 (1) 4 (724) 8 (1174)
No difference in the prevalence of ROP 0 (0, 0%) 4 (724, 100%) 7 (986, 84%)
Increased severity of ROP 1 (1, 100%) 0 (0, 0%) 1 (188, 16%)
Anthropometry 1202 0 (0) 3 (949) 6 (253)
Increased weight 0 (0, 0%) 0 (0, 0%) 2 (95, 38%)
Decreased weight 0 (0, 0%) 1 (546, 58%) 1 (11, 4%)
No difference in weight 0 (0, 0%) 2 (403, 42%) 4 (202, 80%)
No difference in head circumference 0 (0, 0%) 2 (626, 66%) 5 (242, 96%)
Increased length 0 (0, 0%) 0 (0, 0%) 2 (95, 38%)
No difference in length 0 (0, 0%) 2 (869, 92%) 3 (140, 55%)
A total of 40 studies were included in our final systematic review. Outcomes are highlighted in gray if the majority of studies found positive association with the outcome. If there was no difference found in outcomes for the majority of studies, the row is white.

FIGURE 1.
FIGURE 1.:
PRISMA flow chart. A total of 3113 studies were produced. Of these, 1458 duplicates were removed, and the remaining 1655 studies were screened. 1570 studies were removed after abstract review, as it was clear that they did not meet eligibility criteria. The remaining 85 additional studies were assessed through full-text screening. Forty-five were excluded after it was determined they did not meet inclusion criteria. Of those, 9 studies were excluded due to inability for English translation; however, studies in which the abstract was available in English (1) were included.

Assessment of Bias

To assess the validity of articles included, 2 reviewers independently assessed risk of bias for each study using the Newcastle-Ottawa Scale (NOS). For case reports, a modified version of the NOS was used, per previous published guidelines.44 The following items were removed for case reports: selection of nonexposed cohort, comparability and outcome adequacy of follow-up for cohorts (Table 2).

TABLE 2. - Risk of Bias Assessment Using the NOS
Author Type of Study Publication Year Single or Multicenter NOS Score
DeCates et al7 Case report 1988 n/a 3
Griffin et al35 Case series 1990 Single center 4
Suzumura et al38 Case report 1996 n/a 2
Bradshaw et al3 Case report 2003 n/a 3
Gessler et al11 Case report 2004 n/a 1
Bergman et al1 Case report 2006 n/a 2
Takahashi et al39 Case report 2007 n/a 3
Fischer et al9 Case report 2010 n/a 3
Baerts et al41 Case report 2010 n/a 3
Torregrossa et al36 Case report 2015 n/a 3
Modrzejewska et al20 Case report 2018 n/a 2
Gaona-Álvarez et al10 Case report 2019 n/a 3
Chang40 Retrospective case control (abstract only) 2011 Multicenter 7
Goelz et al12 Retrospective cohort 2013 Single center 9
Turner et al29 Retrospective cohort 2014 Single center 9
Kelly et al17 Retrospective cohort 2015 Multicenter 9
Yoo et al33 Retrospective cohort 2015 Single center 8
Mukhopadhyay et al21 Retrospective cohort 2016 Single center 9
Weimer et al31 Retrospective cohort 2019 Multicenter 9
Yeager et al32 Prospective cohort 1983 Single center 8
Paryani et al37 Prospective cohort 1985 Single center 8
Sawyer et al28 Prospective cohort 1987 Single center 8
Jim et al15 Prospective cohort 2004 Single center 8
Vollmer et al30 Prospective cohort 2004 Single center 8
Meier et al19 Prospective cohort 2005 Single center 8
Neuberger et al22 Prospective cohort 2006 Single center 8
Buxmann et al5 Prospective cohort 2009 Single center 8
Capretti et al6 Prospective cohort 2009 Single center 8
Nijman et al25 Prospective cohort 2012 Single center 8
Nijman et al23 Prospective cohort 2012 Single center 5
Wakabayashi et al14 Prospective cohort 2012 Single center 8
Nijman et al24 Prospective cohort 2013 Single center 8
Bevot et al2 Prospective cohort 2012 Single center 8
Dorn et al8 Prospective cohort 2014 Single center 7
Brecht et al4 Prospective cohort 2015 Single center 7
Jim et al16 Prospective cohort 2015 Single center 8
Pilar et al27 Prospective cohort 2015 Single center 8
Martins-Celini et al18 Prospective cohort 2016 Multicenter 8
Gunkel et al13 Prospective cohort 2018 Single center 8
Patel et al26 Prospective cohort 2019 Multicenter 9
Possible scores ranged from 0 to 5 for case reports and case series, and 0 to 9 for cohort and case-control studies. Approximately, 66.6% (8/12) of case reports had a score of 3 or higher, while 96.4% (27/28) of cohort and case control studies had a score of 7 or higher.

RESULTS

Pulmonary

Case reports showed worsening respiratory outcomes after pCMV infection in preterm infants. Two case reports of 23-week and 26-week GA pCMV-infected infants described prolonged respiratory support and worsening of previously established bronchopulmonary dysplasia (BPD).7,39 A case report of a 28-week GA neonate with pCMV reported persistent emphysema and atelectasis requiring supplemental oxygen until 8 months corrected gestational age (CGA) (Table 3).38

TABLE 3. - Long-term outcomes after pCMV infection in preterm LBW infants
Study Authors Total No. Subjects Infected Uninfected Population Findings
Case reports/series
De Cates et al7 1 1 0 26-wk infant, BW 867 g - Increased respiratory support, worsening BPD on CXR. Died at 5 mo due to respiratory failure. Autopsy found CMV infection in liver, salivary gland, kidney, lungs and pituitary gland
Griffin et al35 385 235 (5 reported) 100 CMV seropositive infants, BW <1250 g - 1 infant died after multiple episodes of sepsis (including CMV), 1 infant with worsening BPD and increased respiratory support after CMV diagnosis
Suzumura et al38 1 1 0 28-wk GA infant, BW 690 g - Worsening respiratory status, persistent emphysema and atelectasis on CXR requiring supplemental O2 until 8 mo CGA
Bradshaw et al3 1 1 0 30-wk GA infant, 1590 g - Symptomatic pCMV infection. Normal growth and development at 16 mo
Gessler et al11 1 1 0 29-wk GA infant, BW 1490 g - NEC on DOL 6, CMV detected at 6 wks. Colonic stricture at 10 wks—CMV identified on immunohistochemical staining of endothelial cells. Continued with prolonged feeding intolerance, hepatosplenomegaly and ascites
Bergman et al1 1 1 0 25-wk GA infant, BW 875 g - Infant with reported pCMV. Course complicated by trachea-esophageal atresia and fistula, PDA, ROP, BPD and PVL. Mildly impaired psychomotor development at 2 yrs of age
Takahashi et al39 1 1 0 23-wk GA infant, BW 622 g - Worsening and prolonged respiratory support (intubated 3 mo), BPD
- Development “nearly age-appropriate” at 30 mo CGA
- ABR at 37 wks CGA showed hearing loss > 65-dB threshold bilaterally
- Normal ABR at 8 mo of age
Fischer et al9 1 1 0 24-wk GA infant, BW 780 g - Infant with sepsis- like-syndrome, pneumonitis and colitis found to be pCMV positive. Mild abnormalities in muscle tone at 6 and 12 mo of age. Global developmental delay on GSID at 18 mo of age, quotient = 74 (standardized quotient = 100)
- Bilateral hearing loss at 43 wks GA. Normal hearing on the right, unchanged on the left side at 4 mo of age. Normal hearing bilaterally at 12 mo of age
Baerts et al41 1 1 0 26-wk GA infant, BW 1050 g - Symptomatic pCMV infection. Failed hearing screen at TEA. Hearing loss due to auditory neuropathy at 6 wks CGA. Mild delayed language expression at 4 yrs of age
Torregrossa et al36 1 1 0 25-wk GA infant, BW 650 g - Sepsis-like- syndrome on DOL 45, CMV positive on DOL 55. Course complicated by malrotation and multiple abdominal surgeries for obstruction and perforation. Bowel histopathology from surgery on DOL 142 showed presence of CMV infection. Infant was treated with Ganciclovir and fully recovered, discharged on DOL 203
- Neurodevelopmental delay, predominantly in language and visuospatial fields, at 18 mo CGA
- Normal hearing examination at 3.5 mo CGA
Modrzejewska et al20 1 1 0 24 wks GA, BW 480 g - Bilateral stage 3, zone 2+ ROP successfully treated with diode laser therapy. Follow-up examination showing vitritis with vitreous haze, new hemorrhages and fibrous preretinal proliferations. Infant found to be pCMV positive with concern that CMV caused the vitroretinal lesions
Gaona-Alvarez et al10 1 1 0 31-wk GA infant, BW 1130 g - Hematemesis, melena and food intolerance 1 mo after diagnosis of pCMV. Proposed diagnosis of gastritis with peripheral eosinophilia secondary to CMV. Resolved 1 yr later without treatment
Retrospective studies
Chang40 136 Not reported Not reported <37 wks GA - CMV infection identified as important risk factor for moderate/severe BPD by multivariate logistic regression
Goelz et al12 84 42 42 <32 wks GA and BW <1500 g - Nonsignificant decrease in all subscales of K- ABC for pCMV infected infants at 6 yrs of age (P = not significant)
- In multivariate analysis, infants diagnosed with pCMV before discharge had significantly lower scores in simultaneous processing scale compare to infants infected after discharge (P = 0.029)*
Turner et al29 198 16 180 BW ≤ 1500 g - No difference in prevalence of BPD+death (P = 0.21)
- No difference in incidence of NEC (P = 1.00)
- Nonsignificant increase spontaneous GI perforation (13% vs. 2%, P = 0.07)
- No difference in neurological impairment by GMFCS (P = not reported)
- No difference in neuroimaging (cranial ultrasound) (P = 0.43)
- No difference in incidence of SNHL at 18–24 mo CGA (P = 0.47)
- No difference in incidence of ROP (P = 0.32)
Kelly et al17 (matched cohort) 606 303 303 <1500 g hospitalized on DOL 21 - Increase relative risk of death or BPD at 36 wks PMA (RR = 1.21, CI = 1.10–1.32, P < 0.001)
- Increase relative risk of BPD at 36 wks PMA (RR = 1.33, CI = 1.19–1.50, P < 0.001)*Note – same database as Weimer et al31, different inclusion criteria and years
Yoo et al33 323 27 296 BW < 1000 g - Increase in duration of mechanical ventilation days (P < 0.05)
- Increase in incidence of ≥ moderate BPD (P < 0.05)
- Increased odds of > moderate BPD with CMV infection (OR 6.36, 95% CI 2.18–18.54)
- No increased odds of NEC with CMV infection (OR 0.862, 95% CI 0.27–2.72)
- No difference in MDI on BSID-II at 24 mo CGA (P = not reported)
- No difference in PDI on BSID-II at 24 mo CGA (P = not reported)
- No difference in cerebral palsy at 24 mo CGA (P = not reported)
- No difference in hearing loss at 24 mo CGA (P = not reported)
- No difference in hospitalization days (P = not reported)
- No difference in incidence or odds of ROP (OR = 1.693, 95% CI = 0.69–4.11)
- No differences in weight at 36 wks CGA (P = not reported)
- No differences in weight, height or head circumference at 24 mo CGA (P = not reported)
Mukhopadhyay et al21 145 27 118 BW < 1500 g - Increase in duration of mechanical ventilation (P = 0.03)
- Increase in incidence of BPD (P < 0.001)
- Increase adjusted odds of BPD with CMV (OR = 4.0, 95% CI 1.3–12.4)
- No difference in incidence of NEC (P = 0.11)
- 2/27 pCMV infected infants failed hearing screen
- Increase LOS in days, 101 (92–119) vs. 78 (62–99) (P = 0.001)
- No difference in incidence of ROP (P = 0.35)
Weimer et al31 546 273 273 BW < 1500 g - Increased RR of BPD (RR = 1.30, 95% CI 1.17–1.44; P < 0.001)
- No difference in RR of NEC (RR = 2.00, 95% CI = 0.18–22.06, P = 0.57)
- Increased RR of failed hearing screen at discharge (RR = 1.8, 95% CI = 1.14–2.85, P = 0.01)
- Increased LOS, days (113 vs. 102, P < 0.001)
- Increased PMA at discharge (41.1 vs. 39.9, P < 0.001)
- Lower weight-for-age at discharge (P = 0.005)
- No difference in length-for-age at discharge (P = 0.70)
- No difference in head circumference at discharge (P = 0.17)*Note – same database used as Kelly et al17, different inclusion criteria and years
Yeager et al32 106 18 88 BW < 1500 g - No difference in O2 requirements between early and late excreters (P = 0.086)
Prospective studies
Paryani et al37 86 43 43 ≤30 wks GA - No difference in neuromuscular impairment at 3 yrs of age (P > 0.05)
- No difference in mild to moderate handicap at 3 yrs of age (P > 0.05)
- No difference in severe handicap at 3 yrs of age (P > 0.05)
- No difference in prevalence of sensorineural hearing loss at 3 yrs of age (P > 0.05)
- No difference in ROP (P = not reported)
Sawyer et al28 64 32 32 ≤33 wks - Increased number of inspired oxygen treatment days (P = 0.002)
- Increased number of intubation days (P = 0.056)
- Increase in PIP (P = 0.034)
- Increased incidence of BPD (P = 0.005)
- Increased LOS (d) (107 vs. 67.8, P = 0.001)
Jim et al15 42 6 34 <35 wks GA and BW < 1500 g - No difference in MDI or PDI on the BSID-II at 6 mo CGA (P = 0.85, P = 0.72, respectively)
- No difference in hearing impairment at 6 mo CGA (P = 0.93)
- No difference in ROP (P = 0.13)
- pCMV infants had higher weight and length at 6 mo CGA were of higher weight at 6 mo corrected age (P = 0.02, P = 0.05, respectively)
- No difference in head circumference at 6 mo corrected age (P = 0.17)§§The same initial cohort as Goelz et al,12 Bevot et al2 and Brecht et al.4
Vollmer et al30 44 22 22 <32 wks or BW < 1500 g - No differences in neurological impairment, motor development or speech and language development at 2 and 4.5 yrs, no specific test documented (P = not reported)
- No difference in incidence of SNHL at 2–4.5 yrs (P = not reported)
- No difference in weight, length or head circumference at 2- and 4.5-yr follow-up (P > 0.1 for all comparisons)
Meier et al19 89 22 67 ≤33 wks - No difference in BPD defined as O2 requirement ≥25% on DOL 28 or O2 requirement at 36 wks PCA (P = 0.5, P = 0.17, respectively)
- No difference in mechanical ventilation days (P = 0.39)
- No difference in ROP (P = 0.19)
Neuberger et al22 80 40 40 GA ≤ 31 wks or BW < 1500 g - No difference in duration of mechanical ventilation or duration of oxygen therapy (P = not significant)
- No difference in prevalence of BPD (P = not significant)
- No difference in incidence of NEC (P = not significant)
- No difference in LOS or CGA at discharge (P = not significant)
- No difference in incidence of ROP > stage 2 (P = not significant)
- No difference in weight or head circumference at discharge (P = not significant)
Buxmann et al5 58 5 53 <31 wks GA infants
3 Groups—infants of seronegative women (study group), CMV + infants of seropositive women (infected), CMV negative infants of seropositive women (uninfected)
- No difference in BPD between groups (P = not significant)
- Increased median LOS for infected infants (96 d) vs. study group (77 d) and uninfected (73 d) (P = 0.025)
- No difference in ROP stage 3/4 between groups (P = not significant)
- No difference between groups in weight or head circumference at discharge (P = not significant)
Capretti et al6 62 9 53 <32 wks GA and BW < 1500 g born to seropositive mothers - No difference between groups for O2 at 30 DOL (P = 1.000), 36 wks PMA (P = 0.695), length of oxygen supplementation (P = 0.795)
- No abnormalities in GMDS scores at 2 yrs of age for any pCMV positive infants
- No hearing loss at 2 yrs for any pCMV positive infants
- No difference in LOS or CGA at discharge between infected and uninfected infants (P = 0.65 and P = 1.00, respectively)
- No difference in ROP grade 1–2 between infected and uninfected infants (P = 1.00)
- No difference in weight at discharge (P = 0.572)
Nijman et al25 315 39 276 <32 wks GA - No difference in needing mechanical ventilation (P = 0.948) or incidence of BPD (P = 0.352)
- Increased incidence of LSV at term age equivalent (P < 0.001)
- No difference in incidence of germinolytic cysts (P = 0.792)
- No difference in abnormal newborn hearing screen (P = 0.453)
- No difference in LOS (d) (P = 0.09)
Nijman et al23 64 64 (1 yr)
18 (2 ys)
0 <34 wks GA - No infants with pCMV developed sensorineural hearing loss at 1- and 2-yr follow-up (no uninfected comparison group)
Wakabayashi et al14 11 4 7 BW < 1500 g - Increased LOS for symptomatic pCMV positive infants vs. asymptomatic pCMV positive and pCMV negative infants (median 193.5 vs. 65 d, P < 0.01)
- Decreased weight gain in symptomatic pCMV infected infants (16.5 g/d) compared with asymptomatic pCMV positive infants and CMV negative infants (26 g/d) (P < 0.05)
Nijman et al24 82 21 61 <32 wks GA - No difference in mechanical ventilation > 7 d (P = 0.943) or chronic lung disease (P = 0.473)
- No difference in incidence of NEC (P = 0.555)
- Increased incidence of LSV at term age equivalent (P = 0.007)
- No difference in incidence of germinolytic cysts at term age equivalent (P = 0.851)
- Reduced fractional anisotropy on cerebral diffuse tensor imaging in occipital regions (P = 0.002)
- No difference in neurological development by GMDS at 16 mos corrected age (P = 0.918)
Bevot et al2 41 20 21 <32 wks GA or BW < 1500 g - pCMV positive infants (N = 18) showed poorer performance on M-ABC at 8 yrs of age [P = 0.053 (total impairment), P = 0.053 (manual dexterity), P = 0.037 (ball skills)]
- Mean scores for both groups in normal range for K-ABC at 8 yrs of age; pCMV negative infants had higher mean scores [P = 0.046 (intellectual abilities), P = 0.020 (simultaneous scale), P = 0.646 (sequential processing), P = 0.349 (acquired skills)]
- All pCMV positive infants had normal hearing at school age
- No differences in any anthropomorphic measurement at 8 yrs [P = 0.43 (weight), P = 0.29 (length), P = 0.53 (head circumference)]*Note: same initial cohort as Goelz et al12, Vollmer et al30 and Brecht et al4
Dorn et al8 71 15 19 (+37 term controls) ≤32 wks GA or BW < 1500 g
CMV positive preterm infants, CMV negative preterm infants and term controls
- Stronger activation in left hippocampus in preterm CMV positive infants compared with term uninfected controls (P < 0.001)
- Stronger activation in right anterior cingulate cortex in preterm CMV positive infants when performing language tasks compared with pCMV negative preterm infants (P < 0.001)
Brecht et al4 66 19 23 (+24 term controls) ≤32 wks GA or BW < 1500 g
CMV positive preterm infants, CMV negative preterm infants and term controls
- Lower mean IQ scores in infants with pCMV compared with preterm pCMV negative and term controls at 11–17 yrs (P = 0.030)
Jim et al16 55 14 41 <35 wks GA and BW < 1500 g - No difference in incidence of BPD (P = 0.21)
- No difference in MDI or PDI on the BSID-II at 12 mo CGA (P > 0.05 for both)
- No difference in MDI or PDI on the BSID-II at 24 mo CGA (P > 0.05 for both)
- No differences of mild hearing impairment at 12 or 24 mo CGA (P = not reported)
- No difference in LOS, (68.3 ±23.8 d infected vs. 66.2 ±25 d uninfected, P = 0.78)
- No difference in ROP (any stage) (P = 0.48)
- No differences in weight, length or head circumference at 12 and 24 mo CGA (P > 0.05)
Pilar et al27 160 Not reported Not reported <32 wks GA - Increased incidence of LSV (P < 0.001)
- Increased mean LOS for symptomatic infants with pCMV compared with asymptomatic pCMV infected infants (P < 0.05)
Martins-Celini et al18 188 24 164 (31 CMV uninfected mothers, 133 CMV infected mothers) ≤30 wks, 2 centers. CMV negative infants of seronegative mothers, CMV negative infants of seropositive mothers and CMV positive infants of CMV seropositive mothers - Increase in incidence of BPD in CMV positive (P = 0.05)
- Increase RR of BPD with CMV infection in unadjusted (RR = 1.88, 95% CI 1.08–-3.27), but not adjusted outcomes (RR = 1.46, 95% CI 0.82–2.55)
- No difference in incidence of NEC (P = not reported)
- Increased incidence of stage 2 or 3 ROP in CMV infected vs. other 2 groups (P < 0.01)
- CMV infected infants had a higher unadjusted and adjusted relative risk of stage 2 or 3 ROP (RR = 1.88, 95% CI 1.08–3.27; aRR = 2.51, 95% CI 1.07–5.91)
Gunkel et al13 344 (24 mo) 213 (6 yrs) 55 (24 mo)
41 (6 yrs)
289 <32 wks GA Infants assessed at 24 mo CGA and 6 yrs - No difference in incidence of NEC (P = 0.890 in 24 mo cohort, P = 0.435 in 6 yrs cohort)
16 mo:
- No difference in mean developmental quotient of GMDS (P = 0.320)
- Higher mean locomotor subscale quotient in pCMV infected infants (P = 0.025)
24 mo:
- Mean general developmental quotient of GMDS within normal range in both infected and uninfected infants (P = 0.634)
- No significant differences in BSID-III scores (P = 0.831)
- Increased incidence of LSV at TEA (P = 0.001)
- No difference in incidence of germinolytic cysts at TEA (P = 0.729)
6 yrs:
- Mean WPPSI-III scores in the normal range for infected and uninfected infants, but verbal IQ significantly lower for CMV infected infants (P = 0.046)
- No difference in mean M-ABC-II score and median in the normal range for infected and uninfected infants.
- Increased incidence of LSV at TEA (P = 0.001)
- No difference in incidence of germinolytic cysts at TEA (P = 0.729)
- No children at 6 yrs of age with pCMV infection had sensorineural hearing loss
- Slight impairment in (9/56) 16% of children, 7 unilateral conductive hearing loss and 2 bilateral
- Moderate impairment in 5% (3/56), 1 unilateral conductive hearing loss and 2 bilateral
- No difference in LOS for groups of infants evaluated at 24 mo CGA (P = 0.890) and 6 yrs (P = 0.435)
Patel et al26 596 139 457 BW ≤ 1500 g - No association between pCMV and BPD (any type) (RR = 1.00, 95% CI = 0.69–1.46, P = 0.98) or moderate/severe BPD (RR = 1.02, CI = 0.57–1.80, P = 0.95)
- Increased incidence of NEC in infants with pCMV (6/33 vs. 37/563).
- Increased hazard of NEC with pCMV infection in multiple models that vary with timing of exposure and covariates (HR 2.23–8.45)
- Exposure to higher breast milk CMV load was associated with higher risk of NEC (aHR = 1.28, 95% CI = 1.06–1.54)
- No difference in incidence of ROP stage 2+ (P = 0.72) or ROP stage 3+ (P = 0.75)
Outcomes are divided by study design.
*The same initial cohort as Vollmer et al,30 Bevot et al2 and Brecht et al.4
The same database as Weimer et al,31 different inclusion criteria and years.
The same database used as Kelly et al,17 different inclusion criteria and years.
§The same initial cohort as Goelz et al,12 Bevot et al2 and Brecht et al.4
The same initial cohort as Goelz et al,12 Vollmer et al30 and Brecht et al.4
ABR indicates auditory brainstem response;aHR, adjusted hazard ratio; BSID, Bayley Scales of Infant Development; CXR, chest radiograph; DOL, day of life;GMDS, Griffiths Mental Development Scale; GMFCS, Gross Motor Function Classification System; GSID, Griffiths Scale of Infant Development; IQ, intelligence quotient; K-ABC, Kaufman Assessment Battery for Children; MDI, Mental Development Index; PDA, patent ductus arteriosus; PDI, Psychomotor Development Index; PIP, peak inspiratory pressure; PMA, postmenstrual age; PVL, periventricular leukomalacia; RR, relative risk; TEA, term age equivalent; WPPSI, Wechsler Preschool and Primary Scale of Intelligence.

Despite case reports of worsening respiratory status in preterm infants with and without established lung disease, prospective and retrospective studies are conflicting. One retrospective and 3 prospective studies found no difference in the duration of mechanical ventilation between preterm infants with and without pCMV.6,22,24,25 In contrast, 2 retrospective studies found an increased duration of mechanical ventilation days for pCMV infants.13,21,33 A prospective study of 64 infants showed an additional 10 intubation days in pCMV infants compared with those without, although this barely reached significance.28 This same study showed a significant increase in oxygen and peak inspiratory pressure requirements in pCMV infants.28 A prospective cohort of pCMV-infected infants considered early viral shedders (excretion of virus in urine ≤6 weeks of age) showed a nonsignificant trend towards longer oxygen requirements than late viral shedders (first excretion at >6 weeks of age).32

Multiple retrospective studies found an increased incidence of BPD in preterm-pCMV infants compared with those without.21–23 Additionally, 2 propensity score-matched retrospective studies using cohorts from the same database of 606 and 546 infants, respectively, found an increase in the relative risk of BPD.17,31 Another study of 323 infants found an increased incidence of moderate BPD and increased odds of greater than moderate BPD with pCMV.33 Some prospective studies also demonstrated increased incidence or worsening BPD, while other prospective studies found no difference in the rates of BPD in pCMV-infected versus uninfected infants.5,6,16,18,19,25,26,28,29,35 Notably, one study of 136 infants found that preterm infants with BPD were more likely to have pCMV via multivariate logistic regression.40

Several hypotheses have been proposed to explain the association between pCMV and worsening respiratory status and BPD in preterm infants. Worsening respiratory status may be secondary to mononuclear inflammatory processes that occur with pCMV infection.17,28 Other hypotheses highlight the direct effects of CMV infection, such as necrotizing pneumonia and fibrosis, which may lead to deterioration in respiratory status and thus prolonged mechanical ventilation and increased oxygen requirements, which are known risk factors for BPD.17,28 Differing results from these studies may reflect differences in design, as prospective studies screened all preterm infants for pCMV, while retrospective studies included only symptomatic infants later found to have pCMV. While these studies differed, the collective evidence suggests that pCMV infected preterm infants, especially those that are symptomatic, are at increased risk for pulmonary complications, including BPD.

Gastrointestinal

Case reports of extremely LBW (ELBW) infants with pCMV support concern for worsening enterocolitis, resulting in small bowel obstruction and rectal bleeding.11,36 In one case report, a pCMV infant presented with hematemesis, melena and feeding intolerance 1 month after diagnosis.10 All case reports demonstrated CMV inclusion bodies in the GI tract.10,11,36 Although 1 prospective study of 596 infants found that preterm infants with high maternal breast milk CMV loads in the first 14 days of life were 3.3 times more likely to have necrotizing enterocolitis (NEC) than those with lower levels, most studies found no association between pCMV and NEC.13,18,21,22,24,26,29,31,33

One hypothesis to explain the GI complications of pCMV is that CMV may increase intestinal permeability allowing bacterial translocation and further exaggeration of the inflammatory response.26 Long-term GI complications may be secondary to viral persistence in the intestinal mucosa. Taken together, previous studies suggest that while pCMV might be associated with feeding intolerance and abdominal complications, it is unlikely a significant cause of NEC in preterm infants.

Neurodevelopmental

Case reports following long-term neurodevelopmental impairment (NDI) of infants with pCMV found differing results. Some case reports found pCMV-infected infants were meeting developmental milestones (ages 16–30 months); however, these infants tended to be of older GA at birth (range: 23–30 weeks) compared with those showing NDI (ages 24–25 weeks), highlighting extreme prematurity as a confounding factor.1,3,9,36,39 Reported NDI included global developmental delays, hypotonia, psychomotor, language and visuospatial delays.1,9,36

Prospective and retrospective studies evaluating NDI after pCMV infection were split.2,4,6,12,13,15,16,24,29,30,33,37 Importantly, studies that found no differences in NDI had shorter follow-up than those that did (6 months–4.5 years vs. 3–17 years, respectively), highlighting the importance of long-term follow-up.2,4,6,12,13,15,16,24,29,30,33,37 One prospective cohort of 20 preterm infants diagnosed with pCMV underwent neurodevelopmental testing from age 2 to 17 years. Initially, no differences were seen in NDI compared with uninfected preterm infants.2,4,12,30 However, beginning at age 6, pCMV-infected infants showed subtle differences on the Kaufman Assessment Battery for Children (K-ABC).12 At 12 years, this continued cohort showed significantly worse performance compared with their pCMV-uninfected peers.2 Differences were also noted in the Movement Assessment Battery for Children (M-ABC) in the total impairment score.2 For cognitive skills, evaluated using the K-ABC, mean scores were all within normal range for both groups, yet the uninfected group had higher scores in all subdomains.2 For this same cohort at 11 and 17 years, the pCMV-infected preterm infants scored significantly lower on IQ.4

Differences in neuroimaging in pCMV infants were identified as early as age 3 months, including lenticulostriate vasculopathy (LSV), germinolytic cysts and differences in the occipital lobe, hippocampus and cingulate cortex.8,24,25,27,29 Three prospective studies found an increased incidence of LSV between the ages of 3 and 16 months via ultrasound in infants <34 weeks GA in pCMV-infected infants compared with pCMV-uninfected infants.24,25,27 Another study found increased LSV in infants <32 weeks GA at ages 24 months and 6 years.13 Two studies found no difference in germinolytic cysts at term age equivalent (TEA), 24 months and 6 years in preterm infants with and without pCMV.13,25 One study of 82 infants found significantly reduced fractional anisotropy on cerebral diffuse tensor imaging in the occipital regions of preterm pCMV infants compared with uninfected infants; however, these same infants demonstrated no NDI.24 Interestingly, a functional magnetic resonance imaging study assessing adolescents diagnosed with pCMV as preterm infants demonstrated significantly stronger activation of the left hippocampus compared with term uninfected controls that was not seen when pCMV-uninfected preterm infants were compared with term controls.8 The authors speculated this stronger activation was indicative of higher cognitive demand or impairment in working memory.8 This same study found stronger activation in the right anterior cingulate cortex when performing language tasks in pCMV-infected preterm infants, again suggestive of higher cognitive demands.8

Studies suggest that pCMV-infected infants are at increased risk for NDI and the impact may not be fully realized until school age. However, many potential confounders affect long-term NDI, including but not limited to, degree of prematurity, additional comorbidities, nutritional status, socioeconomic status and early interventional access. While speculation has been given to the imaging differences seen in pCMV-infected preterm infants, abnormalities seen on neuroimaging do not always translate to NDI. Necrotizing inflammation secondary to CMV infection may be responsible for the increased prevalence of LSV seen in pCMV-infected infants.25 Preterm infants miss crucial time for brain maturation in utero, resulting in highly vulnerable oligodendrocytes, synapse formations and glial proliferation and differentiation which may be more vulnerable to the effects of CMV.2,24

Auditory

Although cCMV is the most common cause of nongenetic hearing loss in infants, most studies of pCMV in preterm infants (follow-up: 4 months–12 years), found no increased risk of hearing loss.2,6,9,13,15,16,21,23,25,29,30,33,36,37,39,41 One case report of a 23-week GA infant found bilateral hearing loss at 37-weeks CGA.39 In another case report of a 26-week GA infant, retrocochlear auditory neuropathy requiring cochlear implants was noted at TEA.41 In a prospective study, preterm pCMV-infected infants had 16% slight and 5% had moderate hearing impairment at age 6, and no infants had sensorineural hearing loss.13 Last, a retrospective cohort study found an increased rate of hearing screen failure at discharge in preterm pCMV-infected infants.31 However, a failed hearing screen does not directly imply permanent hearing loss; thus, it is possible that with long-term follow-up, these infants would have normal hearing.31

The direct viral cytopathic effects and the resultant inflammatory response thought to be responsible for hearing loss in cCMV may explain hearing loss associated with pCMV.31 However, the majority of studies suggest that pCMV infection does not increase rates of long-term hearing loss in preterm infants.

Length of Stay

Although most studies compared pCMV infants with those of similar GA and comorbidities, we found a wide range in LOS outcomes.5,6,14,16,21,22,25,27,28,31,33 Prospective studies found an increase in LOS from 23 to 128 days for pCMV infected infants.5,14,27,28 Two retrospective studies found an increase in LOS for pCMV infants, with one also finding an increased postnatal age at discharge.21,31 Six studies did not find a difference.6,13,16,22,25,33 The conflicting results seen for LOS may be secondary to GA at birth, the wide range in practice variation amongst neonatal intensive care units, and lack of guidelines for the treatment of pCMV. Overall, studies suggest pCMV infection may increase LOS in preterm infants.

Ophthalmology

Although cCMV is known to cause chorioretinitis,20 almost all retrospective and prospective studies found no differences in the prevalence of eye complications and retinopathy of prematurity (ROP).5,6,15,16,18,19,21,22,26,29,33,37 One case report found vitritis and new hemorrhages in a pCMV infant after ROP treatment.20 Notably, one prospective study of 188 infants found an increased incidence of stage 2 or 3 ROP and 2.5 higher risk of severe stages of ROP in pCMV-infected preterm infants.18 The authors speculated that high viral replication contributed to increasing inflammation leading to more severe stages; however, they acknowledged that the inflammation could not directly or solely be explained by the presence of CMV.18 Together these studies suggest pCMV in preterm infants does not cause significant ophthalmologic morbidity.

Anthropometry

For anthropomorphic measures, one study of 546 infants found those with pCMV had lower weight-for-age at discharge compared with uninfected infants,31 while another study of 11 subjects suggested slower weight gain in pCMV infants that were symptomatic compared with asymptomatic pCMV-positive and pCMV-negative infants.14 In contrast, one study of 40 infants found that pCMV-infected preterm infants had significantly higher body weight than their uninfected counterparts at 6 months CGA15; however, an additional study of 55 infants (including the same initial cohort) found no difference in weight at 12- and 24-month follow-up.16 Six remaining studies did not find a difference in weight at follow-up (range: 2 months–12 years) between pCMV-infected and uninfected preterm infants.2,5,6,22,30,33 Overall, these findings suggest that symptomatic pCMV-infected infants may have lower weight gain while hospitalized, but that pCMV does not have a long-term impact on weight gain trajectory.

No study found a difference in head circumference between pCMV-infected and uninfected infants.2,5,15,16,22,30,31,33 One study of 40 infants found an increase in length in pCMV-infected infants at 6 months CGA, while another study of 55 infants (including the same initial cohort) found no difference in length at 12- and 24-month follow-up.15,16 No other studies found a difference in length (follow up: 2 months–12 years).2,30,31,33 Taken together, there does not seem to be a difference in head circumference nor length in preterm pCMV-infected infants compared with their uninfected counterparts.

DISCUSSION

In this systematic review, we synthesized the current state of research regarding the long-term impact of pCMV infection on preterm infants. Preterm infants with pCMV infection likely have an increased risk for pulmonary and neurologic complications, but no association with NEC, ophthalmologic, audiologic and anthropomorphic complications. Studies were evenly split on the impact of pCMV infection on LOS, but suggest it is likely increased for symptomatic pCMV preterm infants.

Our systematic review is the first to our knowledge to compile the long-term impact of pCMV on preterm infants. However, our review has several limitations. We were unable to conduct a meta-analysis due to the heterogeneity of studies and large number of case reports; however, it is our hope that this compilation will inspire further research that includes well-matched pCMV-uninfected controls and adequate follow-up times. Differences in outcomes between studies may be due to differences in study population and study design. For example, as mentioned earlier, GA at birth may influence the timing and severity of pCMV infection and NDI outcomes. Prospective studies were more likely than retrospective studies to report null findings, demonstrating the degree of symptomatology is likely an important risk factor for long-term sequelae, as exemplified by BPD.5,6,16–19,21,22,24–26,28,29,31–33,37,39 The reported differences may also be due to variations in neonatal intensive care unit practices. In addition, most of the studies are of single-center design, limiting their widespread applicability. Last, it is difficult to determine whether differences were due to pCMV infection, extreme prematurity or a combination.

There are no guidelines or consensus recommendations by any major academic society for the prevention of pCMV infection or direct treatment (or not) of pCMV infection in preterm infants. Outside case reports, there is little data on treatment of pCMV in preterm infants, making it difficult to determine which infants would benefit from treatment and what interventions are needed as these infants age. In our current review, there was unfortunately not enough information regarding treatment gathered from the included studies to make any conclusions, which speaks to the pressing need for further research. A preterm infant identified with pCMV infection could have a very different course depending on their location and provider. Although studies varied in design and cohorts, the data presented in this systematic review support an increased risk for respiratory complications, NDI and LOS in preterm infants with pCMV infection. These findings support the need for larger prospective studies to confirm these findings and determine if treatment improves outcomes. More importantly, strategies are needed to dampen or eliminate this infection that is a major contributor to morbidity in the preterm population. Neonatologists and primary care pediatricians who care for this population should be aware of the possible long-term implications of pCMV infection and attention should be paid to increased neurodevelopmental surveillance and services.

REFERENCES

1. Bergman KA, Benne CA, Doedens RA, et al. Primary cytomegalovirus infection in the postnatal period. Ned Tijdschr Geneeskd. 2006;150:909–912.
2. 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.
3. Bradshaw JH, Moore PP. Perinatal cytomegalovirus infection associated with lung cysts. J Paediatr Child Health. 2003;39:563–566.
4. Brecht KF, Goelz R, Bevot A, et al. Postnatal human cytomegalovirus infection in preterm infants has long-term neuropsychological sequelae. J Pediatr. 2015;166:834–839.e1.
5. Buxmann H, Miljak A, Fischer D, et al. Incidence and clinical outcome of cytomegalovirus transmission via breast milk in preterm infants </=31 weeks. Acta Paediatr. 2009;98:270–276.
6. Capretti MG, Lanari M, Lazzarotto T, et al. Very low birth weight infants born to cytomegalovirus-seropositive mothers fed with their mother’s milk: a prospective study. J Pediatr. 2009;154:842–848.
7. de Cates CR, Roberton NR, Walker JR. Fatal acquired cytomegalovirus infection in a neonate with maternal antibody. J Infect. 1988;17:235–239.
8. 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.
9. 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.
10. Gaona-Álvarez C, González-Velasco C, Mesa-Vázquez JF, et al. Postnatal cytomegalovirus infection with gastritis in a preterm newborn: a case report and literature review. Clin Microbiol Newslett. 2019;41:90–92.
11. Gessler P, Bischoff GA, Wiegand D, et al. Cytomegalovirus-associated necrotizing enterocolitis in a preterm twin after breastfeeding. J Perinatol. 2004;24:124–126.
12. 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.
13. Gunkel J, de Vries LS, Jongmans M, et al. Outcome of preterm infants with postnatal cytomegalovirus infection. Pediatrics. 2018;141:e20170635.
14. Wakabayashi H, Mizuno K, Kohda C, et al. Low HCMV DNA copies can establish infection and result in significant symptoms in extremely preterm infants: a prospective study. Am J Perinatol. 2012;29:377–382.
15. Jim WT, Shu CH, Chiu NC, et al. Transmission of cytomegalovirus from mothers to preterm infants by breast milk. Pediatr Infect Dis J. 2004;23:848–851.
16. 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.
17. Kelly MS, Benjamin DK, Puopolo KM, et al. Postnatal cytomegalovirus infection and the risk for bronchopulmonary dysplasia. JAMA Pediatr. 2015;169:e153785.
18. Martins-Celini FP, Yamamoto AY, Passos DM, et al. Incidence, risk factors, and morbidity of acquired postnatal cytomegalovirus infection among preterm infants fed maternal milk in a highly seropositive population. Clin Infect Dis. 2016;63:929–936.
19. 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.
20. Modrzejewska M, Kulik U, Modrzejewska A. Nosocomial cytomegalovirus infection resulting in worsening of retinopathy of prematurity. Indian J Pediatr. 2018;85:401–402.
21. 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.
22. 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.
23. 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.
24. Nijman J, Gunkel J, de Vries LS, et al. Reduced occipital fractional anisotropy on cerebral diffusion tensor imaging in preterm infants with postnatally acquired cytomegalovirus infection. Neonatology. 2013;104:143–150.
25. Nijman J, de Vries LS, Koopman-Esseboom C, et al. Postnatally acquired cytomegalovirus infection in preterm infants: a prospective study on risk factors and cranial ultrasound findings. Arch Dis Child Fetal Neonatal Ed. 2012;97:F259–F263.
26. Patel RM, Shenvi N, Knezevic A, et al. Observational study of cytomegalovirus from breast milk and necrotising enterocolitis. Arch Dis Child Fetal Neonatal Ed. 2019;105:259–265.
27. Pilar RGM, Marta C, María Teresa MB, et al. Evaluation of cytomegalovirus infection in low-birth weight children by breast milk using a real-time polymerase chain reaction assay. J Med Virol. 2015;87:845–850.
28. Sawyer MH, Edwards DK, Spector SA. Cytomegalovirus infection and bronchopulmonary dysplasia in premature infants. Am J Dis Child. 1987;141:303–305.
29. Turner KM, Lee HC, Boppana SB, et al. Incidence and impact of CMV infection in very low birth weight infants. Pediatrics. 2014;133:e609–e615.
30. 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.
31. 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.
32. Yeager AS, Palumbo PE, Malachowski N, et al. Sequelae of maternally derived cytomegalovirus infections in premature infants. J Pediatr. 1983;102:918–922.
33. Yoo HS, Sung SI, Jung YJ, et al. Prevention of cytomegalovirus transmission via breast milk in extremely low birth weight infants. Yonsei Med J. 2015;56:998–1006.
34. Osterholm EA, Schleiss MR. Impact of breast milk-acquired cytomegalovirus infection in premature infants: pathogenesis, prevention, and clinical consequences? Rev Med Virol. 2020;30:1–11.
35. Griffin MP, O’Shea M, Brazy JE, et al. Cytomegalovirus infection in a neonatal intensive care unit. Subsequent morbidity and mortality of seropositive infants. J Perinatol. 1990;10:43–45.
36. Torregrossa A, Reinberg O, Alamo L, et al. Surgical complications due to postnatal cytomegalovirus infection in a preterm infant with malrotation. Pediatr Int. 2015;57:993–995.
37. Paryani SG, Yeager AS, Hosford-Dunn H, et al. Sequelae of acquired cytomegalovirus infection in premature and sick term infants. J Pediatr. 1985;107:451–456.
38. Suzumura H, Sakurai K, Kano K, et al. Chronic respiratory failure after acquired cytomegalovirus infection in a very low birthweight infant. Acta Paediatr Jpn. 1996;38:677–680.
39. Takahashi R, Tagawa M, Sanjo M, et al. Severe postnatal cytomegalovirus infection in a very premature infant. Neonatology. 2007;92:236–239.
40. Chang L. Incidence and risk factors of bronchopulmonary dysplasia in premature infants in 10 hospitals in China. Zhonghua Er Ke Za Zhi. 2011;49:655–662.
41. Baerts W, van Straaten HL. Auditory neuropathy associated with postnatally acquired cytomegalovirus infection in a very preterm infant. BMJ Case Rep. 2010;2010:bcr0120102689.
42. Liberati A, Altman D, Tetzlaff J, Mulrow C. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ. 2009;339:b2700.
43. Covidence Systematic Review Software. Veritas Health Innovation.
44. Murad MH, Sultan S, Haffar S, et al. Methodological quality and synthesis of case series and case reports. BMJ Evid Based Med. 2018;23:60–63..
Keywords:

postnatal cytomegalovirus; neonatal intensive care unit; low birthweight; long-term outcomes

Supplemental Digital Content

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.