Agreement Between Caloric and Horizontal Video Head Impulse Testing in School-Aged Children Presenting With Dizziness

Abstract

Objective: The purpose of this investigation was to determine the relationship between the horizontal semicircular canal video head impulse test (hvHIT) and the caloric test in a sample of school-age children evaluated due to complaints of dizziness or imbalance. In adults, these two tests provide different but complementary information regarding vestibular function and/or pathology. However, the most common causes of vestibular loss in children are different than those in adults, and it is not clear how these two tests of horizontal semicircular canal function are related in a pediatric population. Method: This is a retrospective study of electronic records of pediatric patients (< 18 years old) who were referred for vestibular function testing and completed both hvHIT and caloric testing. Agreement between both test results was evaluated by examining measures of sensitivity, specificity, and calculating Cohen's kappa. Results: Results from 117 patients (mean age = 13.7 years, age range: 8–17 years, 37 males) were analyzed. Most patients (97/115 [84.3%]) demonstrated both normal hvHIT and caloric test results. Compared to caloric testing, the hvHIT had poor sensitivity (38.9%) and excellent specificity (98.0%). Cohen's kappa analysis revealed moderate agreement between the methods (coefficient = 0.49 [0.24, 0.73]). Where disagreement between the two tests occurred, the final diagnosis was not contingent on either test result. Conclusions: There is a moderate agreement between the hvHIT and the caloric test in pediatric patients presenting to a tertiary care medical center for a vestibular evaluation. Most patients in this cohort presented with normal findings. The specificity of hvHIT was excellent, although the sensitivity was low. This may be related to the low prevalence of peripheral vestibular system disease in this study cohort, and these results do not generalize to centers that serve populations of children where vestibular disorders are more prevalent (e.g., children with sensorineural hearing loss).