Contribution of Congenital Cytomegalovirus Infection to Permanent Hearing Loss in a Highly Seropositive Population: The Brazilian Cytomegalovirus Hearing and Maternal Secondary Infection Study

Abstract

Background The exact contribution of congenital cytomegalovirus infection (cCMVI) to permanent hearing loss (HL) in highly seropositive populations is unknown. We determined the contribution of cCMVI to HL and estimated the effectiveness of newborn hearing screening (HS) in identifying neonates with CMV-related HL. Methods A total of 11 900 neonates born from a population with ≥97% maternal seroprevalence were screened for cCMVI and HL. cCMVI was confirmed by detection of CMV-DNA in saliva and urine at age <3 weeks. Results Overall, 68 (0.6%; 95% confidence interval [CI], 0.4–0.7) neonates were identified with cCMVI. Of the 91 (0.8%) newborns who failed the HS, 24 (26.4%) were confirmed with HL, including 7 (29.2%; 95% CI, 17.2–59.3) with cCMVI. Another newborn with cCMVI passed the HS but was confirmed with HL at age 21 days. Of the 62 neonates with cCMVI who underwent a complete hearing evaluation, 8 (12.9%; 95% CI, 6.7–23.4) had HL and most (7/8; 87.5%; 95% CI, 46.6–99.7) were identified by HS. The rate of CMV-related HL was 8 per 11 887 neonates (0.7 per 1000 live births). The prevalence ratio of HL among neonates with cCMVI compared to CMV-uninfected neonates was 89.5 (95% CI, 39.7–202.0). No late-onset cCMVI-related HL was detected during a median follow-up of 36 months. Conclusions cCMVI is an important cause of HL in childhood in all settings. Integrating targeted cCMVI screening among neonates who fail a HS could be a reasonable, cost-effective strategy to identify newborns with early-onset cCMVI-related HL.

Natural history studies bring universal screening for congenital CMV infection closer

“…Overall, 7/24 (29%) of cases of SNHL in Brazil were attributable to CMV, a figure similar to that reported from other countries . What was different from the published literature was that none of 49 babies in the study with congenital CMV who had normal hearing at birth developed SNHL when followed prospectively with serial hearing tests for a median of 36 months . This means that, once universal screening is introduced, the parents of children born to mothers with recurrent infection could be reassured that close monitoring for hearing loss is not required if the protocol followed in Brazil had been followed.…”

“…Most babies born with congenital CMV infection cannot be scheduled to receive both tests because they do not have symptoms. However, by testing all neonates for congenital CMV, the Brazilian investigators were able to categorize them as high risk and so get both hearing tests done . This is one potential difference from other published studies …”

“…Now, a paper from Brazil, a country with virtually 100% seroprevalence, reports a difference in the development of progressive SNHL . A total of 11,900 babies were screened at birth for both SNHL and congenital CMV infections.…”

“…The results from these two new important studies suggest that universal screening for congenital CMV could be implemented by testing the infected cases using the high‐risk protocol of both otoacoustic emissions plus automated auditory brainstem responses . Thereafter, close monitoring could be offered only to those whose mothers had primary infection in the first trimester.…”

Estimated Cost-effectiveness of Newborn Screening for Congenital Cytomegalovirus Infection in China Using a Markov Model

“…11 Meanwhile, the targeted screening has been used by some health systems as an alternative to the universal screening. [12][13][14][15] It was hailed as an important intervention milestone in the US, 16 but the results from 7 US medical centers suggest that targeted screening could detect only around 40% of infants with CMV-related SNHL. 17 It may also not be able to identify infants with cCMVi who are at risk of late-onset SNHL.…”