Appropriate hearing screening in the pediatric patient with head trauma

Results Seventeen (34%) of the 50 children had abnormal hearing test results. Fischer's exact test showed abnormal test results were most strongly related to temporal bone fracture ( p = 0.0041), non-temporal bone skull fracture ( p = 0.0211) and younger age ( p = 0.0638). Conclusions Any child with head trauma and clinical or radiologic evidence of temporal bone fracture should have early hearing evaluation. Using the proposed algorithm to test children with head trauma and GCS ≤13 and/or LOC and age <3 years or any type of skull fracture may help identify children with hearing loss in a more cost effective manner. Keywords Hearing loss Head trauma Hearing screening Temporal bone fracture Algorithm 1 Introduction Each year in the United States, an estimated 200 in every 100,000 children suffer head injury [1] , and hearing loss has been reported in about 23–64% of cases [2–8] . Most authors agree that patients with temporal bone fractures should have audiometric and otologic evaluation. On the other hand, resolution of hearing loss without treatment has been reported in patients with minor head injury [6,9] . Furthermore, researchers have been unable to correlate the presence of skull fracture not involving the temporal bone, mechanism of injury, or severity of injury with predictable hearing loss patterns [2,4,9] . Thus, there is no consensus regarding which patients with head injury, other than those with a fracture involving the temporal bone, should undergo audiometric evaluation [2–11] . Because early recognition of loss in children can greatly decrease its morbidity, we conducted a study to identify which children with head trauma are most likely to suffer hearing loss and therefore should undergo early hearing testing. We propose an algorithm for evaluating hearing in children who have suffered head trauma may help diagnose and treat post-traumatic hearing loss in a cost efficient manner. 2 Methods This prospective cohort study was approved by the Children's Hospital of Alabama Institutional Review Board and all parents/guardians of patients gave written consent to participation in the study and authorization for release of health information pursuant to HIPAA. The study population was the first 50 consecutive patients admitted to the Pediatric General Surgery service at the Children's Hospital of Alabama after October 1, 2005, who had a head injury with Glasgow Coma Scale (GCS) score of 13 or less or loss of consciousness (LOC), excluding those who were not expected to survive (i.e., those being evaluated for organ donation) and excluding those with a history of hearing loss (by parental report or documented hearing test results) and those whose parent/guardian refused consent. The algorithm used to screen patients for study inclusion and hearing testing is presented in Fig. 1 . A total of 382 patients were admitted to the pediatric trauma service during the study period. Fifty (50) of the patients (13%) met the criteria for hearing evaluation in the study protocol. Each patient was treated by the pediatric general surgery team according to the Advanced Trauma Life Support (ATLS) protocol. Radiologic examinations included chest X-ray, head computed tomography (CT), and cervical spine (C-spine) films in every case, plus abdominal and chest CT when injuries were suspected in these regions. Each patient's chart was reviewed and the following information recorded: mechanism of current injury, past medical and surgical history, other traumatic injuries, and results of the physical exam, including condition of the external ear canal, tympanic membrane, and facial nerve function, when documented. The Pediatric Trauma Database (Trauma Registry of the American College of Surgeons, TRACS Version 3.4; Chicago, IL), which is maintained by the Children's Hospital of Alabama Department of Pediatric General Surgery, was queried to obtain the ICD-9 codes and was used to calculate the Injury Severity Score (ISS), length of hospital stay (LOS), and Pediatric Trauma Score (PTS) for each patient. 2.1 Hearing testing All hearing tests were performed by an audiologist using Grason-Stadler (Madison, WI) equipment, either in the audiology clinic (located in the hospital) or at the bedside if the patient was unable to travel to the clinic. Otoacoustic emissions (OAEs) were recorded using the GSI 60 DPOAE System with software version 5.0 and tympanograms were recorded with the GSI TympStar Middle Ear Analyzer. If/when the patient's age and mental status permitted, an age-appropriate audiologic exam was also performed using the GSI 61 Clinical Audiometer. 2.2 Otolaryngology evaluations Patients with temporal bone fracture and those without who had abnormal hearing test results were referred for follow-up evaluation by an otolaryngologist. 2.3 Data analysis Data analysis was performed using SAS software (version 7.1). Univariate analyses using Fisher's exact test (two-sided p value) were conducted to identify the association between hearing loss and each of these 10 variables: Age, sex, LOS, PTS, ISS, presence of temporal bone fracture, skull fracture (excluding temporal and midface fractures), midface and mandible fractures, head and neck soft tissue injury, and intracranial pathology identified by CT (e.g., pneumocephalus, subdural hematoma, epidural hematoma). Odds ratios and 95% confidence intervals were calculated where applicable. 3 Results The final patient in the series was enrolled in January 2006. During the study period, a total of 318 children were admitted to the Pediatric General Surgery trauma service. Only 1 patient who was eligible declined enrollment in the study. The study group consisted of 32 (64%) boys with a mean (SD) age of 12.6 (±4.4) years and 18 (36%) girls with a mean (SD) age of 10.2 (±5.4) years. Mechanisms of injury in order of frequency included accidents involving motor vehicles, all-terrain vehicles (ATVs), motorcycles, bicycles or other injuries (including falls, sports injuries and golf carts). Of the 50 patients, 17 (34%) had abnormal hearing test results. Eight of the 50 patients were lost to follow-up before any hearing testing could be performed. Of 36 children with abnormal hearing evaluations, a temporal bone fracture was found in 6. All 6 had their hearing tested and were found to have hearing loss and were evaluated by an otolaryngologist. Another 11 patients had indeterminate or abnormal results on one or more of their hearing tests and were scheduled for otolaryngology follow up. The types of hearing loss in the 17 patients are shown in Table 1 . The degree of hearing loss in patients with non-temporal bone fractures was mild to moderate and usually confined to high frequencies. Table 2 shows the association of hearing loss with each of the 10 variables. Significant associations were found for abnormal audiogram results and temporal bone fracture ( p = 0.0041) or non-temporal bone skull fracture ( p = 0.0211). In addition, age less than 3 years was associated with increased risk of abnormal hearing test results ( p = 0.0638). 4 Discussion Hearing loss secondary to temporal bone fractures is well documented [1,2] , and hearing loss is known to be a possible consequence of head trauma without temporal bone fracture. However, the reported incidence of hearing loss due to head trauma varies greatly, and there is no consensus on which patients other than those with temporal bone fracture are at risk and therefore should be evaluated for hearing loss [3–11] . There is consensus on the importance of early recognition and intervention for children with hearing loss. Brookhouser et al. reported that 31% of children with unilateral sensorineural hearing loss experienced “scholastic or behavior problems in school” [12] . Thus, it is essential to test hearing and provide appropriate follow-up care to children with significant hearing loss. Previous studies have examined injury characteristics as possible risk factors for traumatic hearing loss. Neither the mechanism of injury nor the severity of injury was predictive of hearing loss [2,4–8] , although one study found a trend for higher incidence of hearing loss in patients who suffered head injury from falling [6] . Two studies found a trend for higher incidence of hearing loss following injuries involving the parieto-occipital or temporal region [4,6] but no analysis was performed to determine if these findings were statistically significant. From our study results, we developed a new algorithm that may help identify children at significantly higher risk for hearing loss after head trauma ( Fig. 2 ). According to this algorithm, in addition to children with evidence of temporal bone fracture, those with GCS ≤13 and/or LOC on admission or who are younger than 3 years old or who have a non-temporal bone skull fracture are at higher risk of hearing loss. All of these children should have full audiologic testing and otolaryngologic evaluation. Our study does have certain limitations. Because our goal was to identify children at higher risk for hearing loss, to develop an effective algorithm for early hearing testing, we excluded children we considered to be at lower risk, i.e., those without LOC and a GCS 14 or greater, and those who would not benefit from audiologic testing, i.e., those unlikely to survive head trauma. Although children may suffer hearing loss after less severe head trauma, in our algorithm we recommend hearing testing only when there is clinical evidence of hearing loss, such as on physical examination or a parent complaint. We also excluded children with a history of hearing loss, who should have their hearing tested periodically in any case and certainly after head trauma. Another limitation of our study is that children were considered to have “normal” hearing if they had no complaints of hearing loss and normal tympanograms and OAE recordings. However, OAE recordings might be in the range of normal in the presence of a small degree of hearing loss. In addition, type B tympanograms and abnormal OAEs could have been present due to the presences of a middle ear effusion, which has a higher incidence in younger children. Some studies have demonstrated transient sensorineural hearing loss after head trauma. We did not identify any such cases in our study, but our follow-up was less than optimal, due to the wide geographic area our hospital serves and the typically transient nature of the trauma population. For this reason, we also could not determine whether medical therapy for any middle ear disease that might have developed after discharge from the hospital would have altered the hearing loss caused by head trauma. Finally, our study population is relatively small. In summary, we believe our study to be the first to present an algorithm for early hearing testing in children who have suffered head trauma. To address the limitations of this study, we hope that our algorithm will be tested in larger populations with longer follow-up and assessment of the effects of post-acute treatments for middle ear disease. Acknowledgements Special thanks to Dr. Eben Rosenthal, Heather Baty, Caroline King, and Geni Smith. References [1] J. Singh A. Stock Head trauma eMed. J. 2001 1 11 Retrieved from http://emedicine.medscape.com/article/907273-overview (last updated 2006). [2] L. Podoshin M. Fradis Hearing loss after head injury Arch. Otolaryngol. 101 1 1975 15 18 [3] L.K. Kochhar R.C. Deka S.K. Kacker E.V. Raman Hearing loss after head injury Ear Nose Throat J. 69 8 1990 537 542 [4] M.V. Griffiths The incidence of auditory and vestibular concussion following minor head injury J. Laryngol. Otol. 93 3 1979 253 265 [5] H.O. Barber Head injury audiological and vestibular findings Ann. Otol. Rhinol. Laryngol. 78 2 1969 239 252 [6] E.B. Dorman R.P. Morton Hearing loss in minor head injury N. Z. Med. J. 95 711 1982 454 455 [7] E. Vartiainen S. Karjalainen J. Karja Auditory disorders following head injury in children Acta Otolaryngol. 99 5–6 1985 529 536 [8] W.D. Zimmerman T.M. Ganzel I.M. Windmill G.B. Nazar M. Phillips Peripheral hearing loss following head trauma in children Laryngoscope 103 1 Pt 1 1993 87 91 [9] G.G. Browning I.R. Swan S. Gatehouse Hearing loss in minor head injury Arch. Otolaryngol. 108 8 1982 474 477 [10] J.L. Lancaster D.J. Alderson J.W. Curley Otological complications following basal skull fractures J. R. Coll. Surg. Edinb. 44 2 1999 87 90 [11] O. Schubiger A. Valavanis G. Stuckmann F. Antonucci Temporal bone fractures and their complications. Examination with high resolution CT Neuroradiology 28 2 1986 93 99 [12] P.E. Brookhouser D.W. Worthington W.J. Kelly Unilateral hearing loss in children Laryngoscope 101 12 Pt 1 1991 1264 1272