Kernicterus Spectrum Disorders Diagnostic Toolkit: validation using retrospective chart review

Background Kernicterus Spectrum Disorders (KSDs) result from hyperbilirubinemia-induced brain injury. We developed a Toolkit (KSD-TK) to predict the likelihood of KSDs. This study aims to validate the KSD-TK by comparing it to clinical diagnoses made by the Kernicterus Clinic in the Division of Neurology. Methods Through retrospective chart review, we completed a KSD-TK for 37 patients evaluated between 2011 and 2019 using highest bilirubin, newborn risk factors, neonatal exam, follow-up exam, auditory testing, tooth enamel, and MRI brain results. KSD-TK results were compared to the clinical diagnoses given by a kernicterus expert (SS). Results Of 37 patients, 29 were clinically diagnosed with kernicterus, including 14/14 with KSD-TK scored as “definite”, 14/15 “probable”, and 1/2 with “possible” kernicterus. None of 6 patients with KSD-TK “not kernicterus” were clinically diagnosed with kernicterus. Combining KSD-TK “definite” and “probable”, the KSD-TK has 96.6% sensitivity and 87.5% specificity. Each KSD-TK component had high sensitivity, but only three had specificity ≥0.75: auditory neuropathy spectrum disorder, abnormal movements and/or tone on follow-up exam, and abnormal globus pallidus and/or subthalamic nucleus on MRI. Conclusion The KSD-TK is a promising screening tool for patients at risk for kernicterus. Impact This study provides validation of a Kernicterus Spectrum Disorders (KSDs) Toolkit. The toolkit provides screening criteria for predicting KSD diagnosis. Scores of definite or probable have high sensitivity and specificity for KSDs. Abnormal auditory processing, exam, and MRI were most specific for KSDs.