Annals of African Medicine

: 2013  |  Volume : 12  |  Issue : 2  |  Page : 98--104

Risk of sensorineural hearing loss in infants with abnormal head size

Bolajoko O Olusanya 
 Department of Community Health and Primary Care, College of Medicine, University of Lagos, Surulere, Lagos, Nigeria

Correspondence Address:
Bolajoko O Olusanya
Director, Healthy Start Initiative, 286A Corporation Drive, Dolphin Estate, Ikoyi, Lagos


Objective: The purpose was to determine the risk of sensorineural hearing loss (SNHL) in young infants with abnormal head sizes in a developing country. Materials and Methods: A matched case-control study of two (hospital-based and community-based) cohorts of term infants who failed a two-stage hearing screening test with transient-evoked otoacoustic emissions and automated auditory brainstem response in Lagos, Nigeria. Abnormal head size (microcephaly or macrocephaly) was determined with World Health Organisation«SQ»s growth standards for head circumference. The adjusted odds ratios for the risk of SNHL in microcephalic and macrocephalic infants were established through unconditional and conditional logistic regression analyses. Results: Some 194 cases and 970 matched controls drawn from 8,872 term singletons 3 months or younger were studied. The median age of enrolment was 1 day in the hospital-based cohort and 17 days in the community-based cohort. Microcephalic infants in both cohorts were significantly at risk of SNHL while no significant risk was found among macrocephalic infants regardless of birth setting. Conclusions: Microcephalic infants should be routinely screened for potential hearing loss particularly where universal newborn hearing screening is not immediately practicable. Etiological investigation of abnormal head size in this and similar population is warranted. Routine screening and maternal immunization for congenital infections should also be considered.

How to cite this article:
Olusanya BO. Risk of sensorineural hearing loss in infants with abnormal head size.Ann Afr Med 2013;12:98-104

How to cite this URL:
Olusanya BO. Risk of sensorineural hearing loss in infants with abnormal head size. Ann Afr Med [serial online] 2013 [cited 2021 Jun 22 ];12:98-104
Available from:

Full Text


Abnormal head size in childhood, broadly grouped into microcephaly and macrocephaly, is associated with a range of neurodevelopmental disorders including abnormal brain growth, autism spectrum disorders, mental retardation, learning disability, epilepsy, and childhood brain cancer in the developed and developing countries. [1],[2],[3],[4],[5] The burden of abnormal head size also extends into adulthood with documented association with Alzheimer, schizophrenia, and coronary heart disease. [6],[7] Both microcephaly and macrocephaly are attributable to chromosomal abnormalities and a variety of environmental factors including congenital infections such as cytomegalovirus (CMV), rubella, and toxoplasmosis. [1],[2]

The association between sensorineural hearing loss and craniofacial anomalies, syndromes and congenital infections have been extensively documented. [8],[9] However, evidence on the direct association between sensorineural hearing loss and abnormal head size in a large cohort of infants is rare in the literature especially from the developing world. Available evidence is predominantly based on clinical case reports and studies with convenient samples. Moreover, the case definition for abnormal head size is highly variable across available studies with different local or national reference populations.

In 2007, the World Health Organization (WHO) published an International Child Growth Standard (WHO-CGS) for head circumference derived from a database of well-nourished children from developed and developing countries to facilitate comparability of studies. [10] Unlike the WHO's complementary and more widely utilized growth reference for length and weight in the first 5 years, the WHO-CGS is yet to be implemented or reported in many countries to facilitate early detection of infants with abnormal head size for appropriate intervention. This study therefore sets out to determine the risk of sensorineural hearing loss in two large cohorts of infants with abnormal head size in a low-income country based on the WHO-CGS. This is against the backdrop of recent WHO reports encouraging the routine hearing screening of high-risk infants in all countries where the more ideal universal newborn hearing screening currently mandatory in most of the developed world is not immediately practicable. [11]

 Materials and Methods

Study design and population

This case-control study was conducted in an inner-city area of Lagos, Nigeria with an estimated population of about 250,000. The participants were drawn from two previously reported cohorts under two universal newborn/infant hearing screening (UNHS) programs conducted between May 2005 and April 2008. [12],[13] The first project was a hospital-based UNHS program at a maternity hospital under which a total of 4718 infants were consecutively screened. [12] The second consisted of a community-based UNHS program for infants 3 months or younger attending their first immunization clinics shortly after birth at four primary healthcare or well-child centers under which 7179 infants were also tested. [12] Ethical approval was obtained from Lagos State Health Management Board, Nigeria and University College London, UK (which supervised the substantive studies) according to the guidelines laid down in the Declaration of Helsinki. Informed consent was obtained from all participating mothers in writing or by thumb printing before enrolment.

Cases and controls

All term (at least 37 completed weeks of gestation) singletons who failed the screening tests were considered as cases and matched with controls based on infant's age and sex at the ratio of 1:5 under the hospital and the community-based populations. Preterm (<37 weeks gestation) births and multiple gestations were excluded because of their confounding effects on growth consistent with the reference populations used for the WHO-MGS standards. [10] Although post hoc power analysis for retrospective studies is often debated, an indication of whether the study sample for each population was sufficiently powered to provide reliable estimates of statistically significant differences between cases and controls was attempted to complement the evidence from the confidence intervals of the estimates. [14] Thus, assuming an 11.4% exposure risk among infants with abnormal head size in the hospital-based program and 1.5% exposure risk among normal-sized infants with an odds ratio of 8.45, a sample size of 82 cases and 410 controls will have 85% power to detect approximately 10.0% difference in exposure risk between the two groups at 95% confidence interval. Similarly, assuming a 10.2% exposure risk among infants with abnormal head size in the community-based program and 2.0% exposure risk among normal-sized infants with an odds ratio of 5.6, a sample size of 112 cases and 560 controls will have 85% power to detect approximately 8.0% difference in exposure risk between the two groups at 95% confidence interval. If the effect of case-control matching were considered the required total sample size would be lower than 672 for the community-based population and 492 for the hospital-based population.

Hearing evaluation

In both programs, a two-stage screening protocol was implemented consisting of an initial screening with transient-evoked otoacoustic emissions (TEOAE) followed by a second-stage screening with automated auditory brainstem response (AABR) for all those referred from the first-stage screening. All children who failed AABR under both programs were referred to a diagnostic center for further evaluation and appropriate intervention. The TEOAE is a physiological test for the specific measure of the integrity of the outer hair cells in the cochlea while AABR is an electro-physiological measure of the function of the auditory pathway from the eighth cranial nerve through the brainstem. Due to poor follow-up compliance and resource constraints for effective patient tracking the present study has been limited to the outcomes of the two-stage screening tests. However, this two-stage screening protocol typically has a test sensitivity of 92%, specificity of 98%, and satisfactory positive likelihood ratio (>50). [15] Those who passed TEOAE or AABR were classified as "pass." Those who were referred with a AABR test were classified as "fail." Those who were referred after TEOAE failure but did not present for AABR were classified as "incomplete." Only infants who passed either TEOAE or AABR were eligible as controls.


Head circumference (occipito-frontal circumference) for each child was measured with a standard nonstretchable lasso tape (1 mm increments) (Child Growth Foundation, London, UK) at enrolment. One research assistant recruited and trained for each cohort by this author (a developmental pediatrician) carried out the measurement throughout the study period. Gender-specific z-scores for head circumference were obtained from the macro provided by WHO. [16] Each z-score represents the difference between the head circumference of a child and the median head circumference of a reference population (for the same age and sex) divided by the standard deviation of the reference population used by the WHO-MGS. Default settings in the software regarding cut-offs for out-of-range or biologically improbable values were used in the data analysis and all such values were recorded as missing data. Microcephaly was defined as z-score <-2SD while macrocephaly was defined as z-score >2 SD. [1],[2] Z-scores between -2SD and 2SD were considered to be within the "normal" range or "normocephalic" and used as the reference category for microcephaly and macrocephaly.

Statistical analysis

The characteristics of the study population examined included maternal age, marital status, educational attainment, social class, parity, antenatal care visits, mode of delivery (caesarean or vaginal), and the infant's sex. Social classes were determined based on mother's education and father's occupation as validated in our study population. Social class I was termed as "high," II or III as "middle" and IV or V as "low." The potential confounding variables of interest were guided by evidence of biological plausibility derived from this study population and the literature. [17],[18] These factors included social class, place of delivery (private/public hospital vs. outside hospital), type of attendant at delivery (skilled or unskilled), mode of delivery (elective or emergency caesarean and vaginal), history of hyperbilirubinemia requiring phototherapy/exchange blood transfusion, admission for special care at birth or for neonatal illness. "Skilled birth attendants" consisted of doctors, mid-wives, and trained nurses while unskilled birth attendants comprising traditional birth attendants, "untrained" auxiliary nurses, neighbors, and relations. Admission for special care at birth or during the neonatal period is a useful proxy for adverse perinatal conditions that cannot be readily ascertained in hospital settings with limited facilities for clinical/laboratory investigations or community-based settings. Other potential confounders such as consanguinity and family history of deafness were not considered in the analysis because they were sparsely documented in the study population due to widespread social stigma. The association between abnormal head sizes and hearing screening failure for each of the two cohorts was explored using both conditional (matched) and unconditional (unmatched) logistic regression analyses. Strength of the association was described with the odds ratios (OR) and the corresponding 95% confidence intervals (CI). Two-tailed P values at 5% significance level were reported. Model calibration was verified with the Hosmer-Lemeshow test. SPSS for Windows version 16.0 (SPSS Inc, Chicago, IL, USA) was used for all statistical analyses.


A total of 3196 term singletons in the hospital-based program and 5676 term singletons under the community-based cohort with complete anthropometric data and biologically plausible z-scores based on the WHO cutoff values were enrolled for this study. In the hospital-based cohort, 340 (10.6%) had microcephaly and 74 (2.3%) had macrocephaly while 597 (10.5%) and 107 (1.9%) of the community-based cohort were microcephalic and macrocephalic respectively. A total of 82 (2.6%) infants in the hospital-based cohort and 112 (2.0%) in the community-based cohort who failed the hearing screening tests were considered as cases and matched with 420 and 560 controls who passed the screening tests respectively.

The maternal and infant characteristics for both cohorts are presented in [Table 1]. Majority of the mothers of both cases and controls were married, had at least primary education/secondary education, attended antenatal clinics, and delivered vaginally. A statistically significant proportion of cases compared to controls belonged to the low social classes in the hospital (χ2 =4.59; P = 0.032) and community population (χ2 =15.12; P < 0.001). More than half of the infants in both cohorts were male. The median age at enrolment was 1 day in the hospital-based cohort and was significantly lower than the 17 days in the community-based cohort (P < 0.001). More than half (59.8%) of the cases under the community-based cohort were born outside hospital; majority (83.6%) of whom were delivered without skilled birth attendants. Only a small proportion (less than 4%) of cases in both cohorts had macrocephaly. Microcephaly was severe (z < -3SD) in 17.9% (7/39) of the hospital-based cohort compared to 24.5% (23/94) in the community-based cohort among all microcephalic infants.{Table 1}

In the hospital-based cohort, microcephalic infants had marginally significant risk of sensorineural hearing loss in the unadjusted analysis which was further strengthened after adjusting for confounders in both the conditional and unconditional logistic regression analyses [Table 2]. Macrocephalic infants consistently did not appear to be at risk of sensorineural hearing loss in both the unadjusted and adjusted analysis. The logistic models were satisfactorily calibrated (Hosmer-Lemeshow test: χ2 =2.45; P = 0.875). Similarly, microcephalic infants in the community-based cohort were significantly at risk of sensorineural hearing loss in both the unadjusted and adjusted analyses [Table 3]. While macrocephalic infants appeared not to be at risk of hearing loss, the consistent and strong association between infants with any abnormal head sizes and hearing screening failure would suggest the need for a more adequately sized study population to provide a more robust power to detect small differences in the comparison groups. The logistic regression models were nonetheless well calibrated (Hosmer-Lemeshow test: χ2 =1.28; P = 0.937). Only three infants in the hospital-based population were reported with craniofacial anomalies and they were all normocephalic. A total of six infants under the community-based population had craniofacial anomalies and two were microcephalic accounting for only 2.1% of all microcephalic infants.{Table 2}{Table 3}


This study is perhaps the first to explore the relationship between abnormal head size and the risk of hearing loss in early infancy from the developing world and one of the earliest studies in the literature to implement the latest WHO reference standards for head circumference. The principal finding is that microcephalic term infants among whom craniofacial anomalies were rarely reported are at risk of sensorineural hearing loss within the first 3 months of life. This would suggest that the underlying etiology is possibly congenital, acquired and/or early postnatal. This finding is strengthened by its consistency in two separate cohorts in hospital and community settings suggesting that the relationship is independent of whether the infants were born in or outside hospitals. This is significant considering that the vast majority of babies in low income countries are delivered outside hospitals with higher risk of diverse perinatal and childhood infections predisposing to a wide range of developmental disabilities. [19] Moreover, the onset of sensorineural hearing loss from infancy portends considerable life-long burden on the affected infants and their families which can be curtailed through timely detection and intervention. [11]

The main finding in this study would appear to be in agreement with the established association between microcephaly and developmental disabilities in both developed and developing countries. [1],[2],[3],[4],[19] In contrast, far fewer studies exist with respect to macrocephalic infants, which would not be unexpected especially in resource-constrained settings such as in the current study population due to the lower prevalence of survivors. Moreover, congenital anomalies of the central nervous systems are generally underreported in developing countries.

The causal pathways between microcephaly and sensorineural hearing loss in early infancy are probably characterized by the effects of various chromosomal abnormalities as well as congenital and postnatal infections on the auditory system. [8],[9],[20],[21],[22] Viral infections in pregnancy have devastating effects during the critical period of the embryonic development of the inner ear especially around the fifth to the eighth week of gestation. [22] However, the relative contributions of these factors particularly the role of infections in this retrospective study could not be ascertained. For example, congenital rubella is a potential contributor to hearing loss in microcephalic infants although it was less likely to play a prominent role in our study population of apparently healthy term infants without typical signs of multiorgan dysfunctions characteristic of this condition. Congenital/postnatal CMV has been extensively reported as a cause of microcephaly as well as unilateral, bilateral, fluctuating, progressive, and late-onset hearing loss in both developed and developing countries. [9],[20],[21] In fact, CMV is the most prominent infectious cause of hearing loss just as hearing loss is reputed as the most common sequelae in both symptomatically and asymptomatically infected infants. It is therefore not unlikely that the ubiquitous CMV virus albeit seldom diagnosed routinely in this study population would be a major contributor to the observed association as in comparable countries. [23],[24]

Since the heterogeneous causes of microcephaly cannot be completely prevented especially in resource-poor countries, early identification of the affected infants through routine growth monitoring should be considered as a priority in curtailing the burden of this abnormality. This should be within the context of overall developmental surveillance in early childhood which may be cost-effectively synchronized with the current schedules for routine immunization. Infants with abnormal head sizes should be screened promptly for potential hearing loss which is perhaps the most hidden neurodevelopmental problem that requires timely intervention for optimal outcomes within the first few months of life based on the robust evidence on the architecture of the developing brain. [25]

Notable features of the current study include the strength of the evidence derived from its case-control design, careful selection of controls, adequately-powered sample size, adjustment for important confounders, satisfactory model calibration, and narrow confidence intervals. However, a number of limitations of this retrospective study are worth noting to guide future studies. For example, comorbidity with other well-established sequelae of microcephaly such as cerebral palsy, mental retardation, and epilepsy could not be ascertained given the relatively young age of our study population. Detailed etiological investigations and neurologic examinations were not undertaken which would have provided a clearer picture of the clinical profile of the cases. Such diagnostic workups including magnetic resonance imaging and genetic testing are generally not available routinely in many clinical settings in resource-poor countries and are perhaps impracticable in community/nonhospital settings. As earlier noted case finding for hearing loss was limited to hearing screening outcomes due to the high rate of default for diagnostic hearing loss evaluation. However, testing with AABR as used in this study has been shown to facilitate the detection of a high proportion of infants with CMV infection who otherwise would not have been diagnosed. [26] Given the variable onset of CMV-related hearing loss, it was uncertain what proportion of cases and controls may subsequently develop hearing loss after the first 3 months of life. Lastly, while WHO-CGS may result in some false-positives and false-negatives for certain older children compared with various preexisting national monitoring charts there is little debate that the standard is reliable and accurate for evaluating growth in the first 6 months of life.

Overall, this study suggests that microcephalic infants are at risk of multiple disabilities including sensorineural hearing loss from early infancy. This may be indicative of a variety of "asymptomatic" genetic and environmental factors which cannot be detected readily in many clinical settings but the associated burden can be reduced by testing all microcephalic infants for potential hearing loss particularly in sub-Saharan Africa where universal newborn hearing screening is not immediately practicable. Infants who fail the screening tests should receive prompt diagnostic evaluation and enrolled for appropriate intervention if hearing loss is confirmed. Immunization against preventable causes such as rubella needs to be widely promoted especially in developing countries like Nigeria where it is currently not included in the national immunization programs. It is not unlikely that these conclusions will also be appropriate for preterm infants who were excluded from this study and would require modification to the WHO-CGS for a more accurate growth evaluation. Additionally, barring any residual confounding, the main finding from this study may be readily generalizable to other comparable settings given the use of the WHO-CGS as well as objective hearing screening tests.


1Williams CA, Dagli A, Battaglia A. Genetic disorders associated with macrocephaly. Am J Med Genet A 2008;146A: 2023-37.
2Abuelo D. Microcephaly syndromes. Semin Pediatr Neurol 2007;14:118-27.
3Jauhari P, Boggula R, Bhave A, Bhargava R, Singh C, Kohli N, et al. Aetiology of intellectual disability in paediatric outpatients in Northern India. Dev Med Child Neurol 2011;53:167-72.
4Abdel-Salam GM, Halász AA, Czeizel AE. Association of epilepsy with different groups of microcephaly. Dev Med Child Neurol 2000;42:760-7.
5Samuelsen SO, Bakketeig LS, Tretli S, Johannesen TB, Magnus P. Head circumference at birth and risk of brain cancer in childhood: a population-based study. Lancet Oncol 2006;7:39-42.
6Kunugi H, Takei N, Murray RM, Saito K, Nanko S. Small head circumference at birth in schizophrenia. Schizophr Res 1996;20:165-70.
7Fan Z, Zhang ZX, Li Y, Wang Z, Xu T, Gong X, et al. Relationship between birth size and coronary heart disease in China. Ann Med 2010;42:596-602.
8Nance WE. The genetics of deafness. Ment Retard Dev Disabil Res Rev 2003;9:109-19.
9Roizen NJ. Non-genetic causes of hearing loss. Ment Retard Dev Disabil Res Rev 2003;9:120-7.
10World Health Organization (WHO) Multicentre Growth Reference Study Group. WHO child growth standards: head circumference-for-age, arm circumference-for-age, triceps skinfold-for-age and subscapular skinfold-for-age: methods and development. Geneva; WHO, 2007. Available from: [Last accessed on 2011 Aug 10].
11World Health Organization. Neonatal and infant hearing screening. Current issues and guiding principles for action. Outcome of a WHO informal consultation held at WHO head-quarters, Geneva, Switzerland, 9-10 November, 2009. Geneva; WHO; 2010.
12Olusanya BO. Newborns at risk of sensorineural hearing loss in low-income countries. Arch Dis Child 2009;94:227-230.
13Olusanya BO, Ebuehi OM, Somefun AO. Universal infant hearing screening programme in a community with predominant non-hospital births: A three-year experience. J Epidemiol Comm Health 2009;63:481-7.
14Hoenig JM, Heisey DM. The abuse of power: the pervasive fallacy of power calculations for data analysis. Am Stat 2001;55:19-24.
15Kennedy C, McCann D, Campbell MJ, Kimm L, Thornton R. Universal newborn screening for permanent childhood hearing impairment: an 8-year follow-up of a controlled trial. Lancet 2005;366:660-2.
16World Health Organization. WHO Anthro 2005 software and macros. Geneva: WHO; 2006. Available from: [Last accessed on 2011 Aug 10].
17Morzaria S, Westerberg BD, Kozak K. Systematic review of the etiology of bilateral sensorineural hearing loss in children. Int J Pediatr Otorhinolaryngol 2004;67:1193-8.
18Olusanya BO. Making targeted screening for infant hearing loss an effective option in low-income countries. Int J Pediatr Otorhinolaryngol 2011;75:316-21.
19Committee on Nervous System Disorders in Developing Countries (CNSDDC), Board on Global Health. Neurological, psychiatric and developmental disorders: ZMEETING the challenge in the developing world. Washington DC: National Academy Press; 2001. p. 113-76.
20Dollard SC, Grosse SD, Ross DS. New estimates of the prevalence of neurological and sensory sequelae and mortality associated with congenital cytomegalovirus infection. Rev Med Virol 2007;17:355-63.
21Samileh N, Ahmad S, Mohammad F, Framarz M, Azardokht T, Jomeht E. Role of cytomegalovirus in sensorineural hearing loss of children: a case-control study Tehran, Iran. Int J Pediatr Otorhinolaryngol 2008;72:203-8.
22Newton VE, Vallely PJ editors. Infection and hearing impairment. West Sussex: Whurr Publishers Ltd; 2006.
23Cannon MJ, Schmid DS, Hyde TB. Review of cytomegalovirus seroprevalence and demographic characteristics associated with infection. Rev Med Virol 2010;20:202-13.
24Bello C, Whittle H. Cytomegalovirus infection in Gambian mothers and their babies. J Clin Pathol 1991;44:366-9.
25Shonkoff JP. Phillips DA. (Editors). From Neurons to Neighborhoods: The Science of Early Childhood Development. National Research Council and Institute of Medicine Committee on Integrating the Science of Early Childhood Development. Board on Children, Youth and Families, Commission on Behavioral and Social Sciences and Education: Washington DC: The National Academies Press; 2000.
26Stehel EK, Shoup AG, Owen KE, Jackson GL, Sendelbach DM, Boney LF, et al. Newborn hearing screening and detection of congenital cytomegalovirus infection. Pediatrics 2008;121:970-5.