BACKGROUND: Neonatal hearing screening occasionally fails. We noticed that infants born by cesarean delivery (CD) appeared to fail the first otoacoustic emissions (OAE) test more frequently than infants delivered vaginally (VD). This might increase maternal anxiety. We aimed to evaluate the influence of mode of delivery on failure to pass the first OAE.
METHODS: Overall, 1653 infants of >35 weeks’ gestation were evaluated. OAE was performed before discharge. Perinatal-neonatal variables and results of OAE were recorded retrospectively.
RESULTS: Compared with VD infants, CD infants had lower gestational age and 1-minute Apgar, more small for gestational age (SGA), and earlier age at first OAE. Univariate analysis: CD, male gender, gestational age 35 to 37 weeks, birth weight ≤2500 g, SGA, and younger age at first OAE were significantly associated with failed first OAE. In infants weighing 2501 to 4000 g, CD infants had 3-fold higher rates of failed first OAE in comparison with VD infants (20.7% vs 7.1%) before 48 hours of age. In the multivariate analysis, variables independently significantly associated with failed first OAE were male gender (odds ratio [OR] 1.42 [1.02–1.98]), CD (emergency CD: OR 3.18 [2.21–4.57], elective CD: OR 3.32 [2.04–5.42]), age 12 to 23 hours at first OAE (OR 3.1 [2.1–4.58]) and SGA (OR 2.2 [1.15–4.28]).
CONCLUSIONS: CD infants had significantly higher failure rates on first OAE. We speculate that CD is accompanied by retained fluid in middle ear which may impair neonatal hearing. The timing of first OAE after CD should preferably be postponed beyond 48 hours of age to improve OAE passage and minimize maternal anxiety and costs.
- universal newborn hearing screening
- otoacoustic emissions
- automated auditory brainstem responses
- mode of delivery
- vaginal delivery
- cesarean delivery
- AABR —
- automated auditory brainstem responses
- AAP —
- American Academy of Pediatrics
- CD —
- cesarean delivery
- OAE —
- otoacoustic emissions
- OR —
- odds ratio
- SGA —
- small for gestational age
- TM —
- tympanic membrane
- TTN —
- transient tachypnea of newborn
- UNHS —
- universal newborn hearing screening
- VD —
- vaginal delivery
What's Known on This Subject:
Neonatal hearing screening occasionally fails because of several perinatal and neonatal factors. However, the effect of mode of delivery on hearing screening has not yet been established.
What This Study Adds:
We show significantly more failures on hearing screening in cesarean delivery infants. Hence, the timing of screening after cesarean delivery should preferably be postponed beyond 48 hours to improve success rate, minimize maternal anxiety, and decrease costs.
Starting January 1, 2010, universal newborn hearing screening (UNHS)1–5 was implemented in our well-baby nursery, wherein all newborn infants undergo an otoacoustic emissions (OAE) test before discharge from the hospital. Infants who fail OAE or have risk factors for hearing loss undergo an automated auditory brainstem responses (AABR) test.
The audiological technicians who perform OAE have observed that infants born by cesarean delivery (CD) appeared to fail first OAE test more frequently than infants born by vaginal delivery (VD). A “fail” result on first OAE mandates repeated hearing screening tests, thus increasing anxiety and stress among mothers of these infants.
Previous reports have referred to factors that are associated with failed hearing screening in newborn infants. These factors include familial deafness, malformations of face/auricle, secretions in external ear canal,6 middle ear effusion,6 epidural anesthesia in CD births,7 VD,8,9 emergency CD,9 5-minute Apgar score <5,9 need for intensive care, significant hyperbilirubinemia,8,9 and OAE before 24 hours of age.10 In this regard, referral rates for audiological follow-up were 5% to 20% when OAE was conducted <24 hours and <3% when OAE was conducted 24 to 48 hours after birth.1 Hence, it is vital to isolate and categorize factors that may influence the success or failure of newborn hearing screening programs.7
The aim of this study was to identify perinatal and neonatal factors that are associated with failure on first OAE in neonatal hearing screening, while focusing on the role of mode of delivery.
This retrospective observational study was performed in the well-baby nursery at Rambam Health Care Campus, Haifa, Israel. The study was approved by the Institutional Helsinki Committee who gave waiver from obtaining written consent from parents of studied newborn infants.
All 1790 inborn newborn infants born beyond 35 weeks’ gestation between January 1, 2011, and May 31, 2011 were considered for inclusion in the study. In our institution, infants born vaginally (spontaneous or vacuum delivery) are generally discharged 48 to 72 hours after birth, whereas those born by CD are discharged on the fifth day of age.
Overall, 137 neonates were excluded from study as follows: hospitalization in the NICU (n = 89); significant congenital malformation/chromosomal aberration (n = 8); congenital cytomegalovirus infection (urine culture positive for congenital cytomegalovirus) (n = 2); apparent congenital hearing loss: abnormal OAE with abnormal AABR in 1 or both ears (n = 11); OAE before 12 hours of age (n = 3); auricular skin tag (n = 12); and history of parental hearing loss (n = 12).
Universal Newborn Hearing Screening
OAE tests (Otoport Lite, Otodynamics Ltd, UK) were performed on all newborn infants before discharge from nursery according to the guidelines of the Israeli Ministry of Health. OAE tests were performed by 2 audiological technicians daily, excluding Saturday. Infants who failed first OAE (at least in 1 ear) had repeated OAE within 6 to 16 hours (up to 4 OAE tests). If failure persisted, AABR (GSI Audio Screener, Grason-Stadler, Eden Prairie, MN) was then performed at a mean age of 47.1 ± 20 hours (range, 24–99 hours). If AABR was abnormal (at least in 1 ear), the infant was referred to our audiological institute for further hearing examinations and close follow-up. On Fridays, some OAE tests were performed earlier, at 12 to 24 hours of age, to ensure screening of all newborn infants discharged on the weekends. If an infant failed OAE on Friday, the parents were advised to return to the nursery the following Sunday for a repeated OAE.
For each infant included in the study, perinatal and neonatal variables and the timing and results of OAE and AABR tests were recorded. Transient tachypnea of newborn (TTN) was defined as tachypnea after birth with mild to moderate respiratory distress, without risk factors for early-onset sepsis, often necessitating oxygen supplementation and resolving within 48 hours. Phototherapy was initiated according to American Academy of Pediatrics (AAP) guidelines in 200411 and indicated significant hyperbilirubinemia. Gentamicin was administered empirically to infants with risk factors for early-onset sepsis for 48 hours pending blood culture result.
Statistical analysis was performed by using SPSS (Statistics Products Solutions Services) 18.0 software for Windows. The Student t test, the Mann-Whitney test, and χ2 test were used for comparison of 2 modes of delivery (VD and CD) as to perinatal and neonatal variables. A P value of <.05 was considered statistically significant. Binary logistic regression was used for calculation of the odds ratios (ORs) of failure on first OAE with 95% confidence intervals and P values in univariate analyses. All variables with a P value of ≤.1 in the univariate analysis were selected as candidates for the multivariate analysis model. Multivariate forward logistic regression analysis was performed to identify those variables that were independently significantly associated with failure on first OAE. The area under curve-receiver operating characteristic was used as a measure of model discrimination. The Hosmer-Lemeshow goodness-of-fit statistic was calculated. A P > .05 means that a good correlation exists between the expected and the observed findings.
Overall, 1653 infants were studied. Comparison of perinatal and neonatal variables between infants born by VD versus those born by CD is shown in Table 1. Compared with VD infants, those born by CD had significantly lower gestational age, lower 1-minute Apgar score, more small for gestational age (SGA) status and TTN, and earlier age of first OAE performance. Birth weight, 5-minute Apgar score, phototherapy, and gentamicin treatment were not different between the 2 groups.
Table 2 shows that CD delivery (both emergency and elective), male gender, gestational age of 35 to 37 weeks, birth weight of ≤2500 g, SGA status, and decreasing age at first OAE were significantly associated with failure on first OAE (univariate analyses). Conversely, Apgar scores at 1 and 5 minutes, TTN, phototherapy and gentamicin treatment were not significantly associated with failure on first OAE. Table 3 shows the adjusted ORs and 95% confidence intervals for failure on first OAE by perinatal variables. Variables that were independently significantly associated with failure on first OAE included male gender (OR 1.42 [1.02–1.98]), birth by CD (emergency CD: OR 3.18 [2.21–4.57], elective CD: OR 3.32 [2.04–5.42]), early age (12–23 hours) of first OAE (OR 3.1 [2.1–4.58]), and SGA status (OR 2.2 [1.15–4.28]).
Failure on first OAE was 3-fold higher in CD infants (20.7%) than in VD infants (7.1%) (Table 2). The mean ± SD age at “pass” in OAE after up to 4 tests was 34.1 ± 1.1 hours in VD infants and 36.7 ± 16.6 hours in CD infants (P < .001). Figure 1 compares failure rates on first OAE test by postnatal age of VD- versus CD-delivered infants. Infants born by CD had significantly higher rates of OAE failure on the first test up to 47 hours of age. Of note is the markedly high failure rate (41.6%) in CD infants when OAE was performed before 24 hours of age.
Repeated OAE tests were performed for those infants who failed the first OAE test (Table 4). The failure rates on the second and third OAE tests were still higher in CD infants even though performed at older ages than in VD infants. Of the 1653 infants in the study, 10 were referred for AABR, 5 of VD infants and 5 of CD infants. All of these 10 infants passed the AABR test. Eleven additional infants who failed both the OAE and AABR tests were a priori excluded from study. Of these 11 infants, 6 eventually had normal hearing, 3 did not attend follow-up, and 2 infants had confirmed hearing loss.
The incidence of congenital hearing loss is 1 to 6 per 1000 live births5 and might lead to significant developmental delay in infancy and childhood, including verbal, mental, emotional, and social impairments.3–5 UNHS was firstly recommended by the AAP in 19991 and has proved to be safe and efficient in the early detection of hearing loss after birth.1–5 Such detection allows early intervention and has proved to be advantageous as to the verbal development of infants with hearing loss.1–3,5
Our results show that, in comparison with VD, birth by CD increased the risk of failure on first OAE by ∼3.2-fold, regardless of type of CD (emergency or elective). This was true up to 47 hours of age in infants weighing 2501 to 4000 g and born before 41 weeks’ gestation. The higher failure rate in CD infants persisted beyond 47 hours of age (VD 1.0% and CD 8.3%), although with marginal significance (P = .06). This finding might be due to the small number of infants with first OAE performed at this age (1/101 vs 3/36). In addition, higher failure rates were recorded among CD-delivered infants on repeat OAE tests. Our findings were not in complete agreement with those of Olusanya et al9 who found that, in addition to nonelective CD, VDs were also associated with a more than 2-fold excess risk of hearing loss (OAE and AABR).The authors suggest that “the affected babies were unduly subjected to the intrapartum stress of the trial of labor.” The low Apgar scores reported (27.5% had <7 at 5 minutes, compared with only 3 of 1653 infants in our study) certainly may support this contention.
Failure on OAE in neonates can also be caused by external ear canal secretions, and removal of external ear debris increased the OAE pass rate from 76% to 91%.6 As to the middle ear, reports regarding the status of the tympanic membrane (TM) (appearance and mobility) in term infants are rare. Balkany et al12 examined 50 term infants before 24 hours of age to establish the normal micro-otoscopic appearance of the term newborn TM. These authors found that, compared with healthy term infants, 30% of NICU infants had middle ear effusion. Cavanugh et al13 found most term neonates to have limited mobility of TM and 93% to have normal TM appearance (pink or gray color). In another study, Chang et al6 found only 77% of 45 term neonates to have normal TM. In this series, the 1 newborn infant who showed absent TM movement and middle ear effusion failed OAE. Nonetheless, the authors did not evaluate the effect of mode of delivery on hearing screening.
The results of our study confirm the observation of our audiological technicians regarding higher failure rates on first OAE in CD infants even after adjusting for potential confounders including gender, SGA, and age at first OAE. This raises the question of whether infants born by CD have delayed fluid resorption from their middle ear similar to that occurring in their lungs, a condition known as TTN. If our assumption is correct, this may cause a conductive interference of sound transmission through the middle ear and failure on OAE test.
Failure on first OAE mandates a repeated OAE with an estimated cost of $8.00 for each test. Some proceed to AABR test if they still fail OAE (whenever the infant fails 2 consecutive OAE tests). It is well known that failure on first OAE increases maternal anxiety and worry. The degree of maternal worry was reportedly greater at the rescreen as compared with first hearing screen, although maternal knowledge about hearing screening increased between the 2 times.14 Efforts to minimize the neonatal false-positive hearing screen rates and to educate mothers about hearing screening are indicated to minimize unnecessary parental worry. This can be achieved, in part, if parents know that first OAE screening failures are not definitive until confirmed by appropriately timed second-stage screening preferably with AABR. In this regard, Khairi et al15 conducted face-to-face interviews with mothers whose infants failed on first OAE. The authors reported a significantly lower anxiety level before hearing rescreen in comparison with that after first screen.
Other known factors that might negatively affect the result of first OAE test, such as significant hyperbilirubinemia and gentamicin treatment, were not associated with failure on first OAE in our study. Nonethelesss, the majority of first OAE tests (92%) were performed up to 48 hours of age, an age too early for evaluation of the effects of these 2 above-mentioned factors on hearing screening. Apgar scores were not found to be associated with failure on first OAE. Nevertheless, most sick infants, some of whom had low Apgar scores, were transferred to NICU and were not included in analysis. Male gender was associated with a 1.4-fold increased risk for failure on first OAE. This finding is in agreement with that of Saitoh et al16 who found that, in OAE tests, female neonates displayed higher whole-wave reproducibility, response level, band reproducibility, and signal-to-noise ratio values in comparison with male mates. SGA status was significantly associated with failure on first OAE (2.2-fold increase). One previous report17 showed that compared with appropriate for gestational age neonates, SGA mates tended to have shorter V wave and interwave latencies during automated brainstem response tests.
There is no consensus worldwide regarding the optimal timing for performance of first OAE test, and the recent policy recommendation statement of the AAP in 20083 mandates hearing screening before 1 month of age. Shulman et al18 recently evaluated the 55 state and territorial UNHS programs and reported that most infants (92%) were screened for hearing loss before discharge from the hospital, and, of these, 2% were referred for diagnostic evaluation. Approximately 62% of infants who needed a diagnostic evaluation received one, suggesting that there is still a substantial loss to follow-up for infants who did not pass the initial hearing screen.
The Israel Ministry of Health directive (2009) requires that every neonate be screened before discharge from the hospital, with no reference to the optimal timing for screening. To achieve this goal as well as to complete the screening protocol and minimize the necessity for referral for further evaluation, 80% of the infants in our department were tested at ≤48 hours of age. Paul et al19 reported a newborn nursery length of stay of <48 hours in >50% of VD infants and <72 hours in ∼40% of CDs. This relatively short time period may limit the possibility of completing the screening process during the infants' birth hospitalization. In the United States, only ∼40% of programs did all of the screening before hospital discharge and 60% completed the 2-stage protocol after discharge.20
The results of the current study suggest that, although VD infants can be tested at 36 to 42 hours of age, OAE screening of CD-delivered infants should be performed no earlier than 48 hours of age. A further study of all births by using these new timelines for OAE studies might confirm a lower referral rate for CD-delivered infants, hence, lowering costs of the screening program and decreasing anxiety and stress to the parents. Alternatively, in resource-limited settings, a 2-stage OAE with the repeat test(s) delayed as far as practicable might be a reasonable primary outcome for future exploration of the role of mode of delivery or other pregnancy-related factors to screening failure.
Limitations of our study include, first, the retrospective design of the study; second, the lack of otoscopic examination of the infants for removal of any external ear secretions or debris, or to exclude middle ear effusion; and third, we did not take into consideration the possible frequency-dependent increase in transient evoked OAEs noted during the first month of life.21 In addition, 89 infants transferred to the NICU and were not included in the study. This might explain the low number of TTN cases in the study population (1%) and the lack of association between TTN and failed first OAE.
We conclude that, in comparison with VD, birth by CD was associated with significantly higher rates of failure on first OAE in neonates up to 47 hours of age. We speculate that delivery by CD may be accompanied by retained fluids in the middle ear and hence impaired neonatal hearing screening. The timing of first OAE after CD should be reconsidered and preferably postponed beyond 48 hours to improve success rate, minimize maternal anxiety, and decrease costs.
- Accepted March 20, 2012.
- Address correspondence to Tatiana Smolkin, MD, Department of Neonatology, Rambam Medical Center, Bat-Galim, Haifa 31096, Israel. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
- Erenberg A,
- Lemons J,
- Sia C,
- Trunkel D,
- Ziring P
- Wolff R,
- Hommerich J,
- Riemsma R,
- Antes G,
- Lange S,
- Kleijnen J
- US Preventive Services Task Force
- Cunningham M,
- Cox EO,
- Committee on Practice and Ambulatory Medicine and the Section on Otolaryngology and Bronchoesophagology
- Olusanya BO
- Owen M,
- Webb M,
- Evans K
- American Academy of Pediatrics Subcommittee on Hyperbilirubinemia
- Cavanaugh RM Jr
- Shulman S,
- Besculides M,
- Saltzman A,
- Ireys H,
- White KR,
- Forsman I
- Paul IM,
- Lehman EB,
- Hollenbeak CS,
- Maisels MJ
- Copyright © 2012 by the American Academy of Pediatrics