Use of the King-Devick test as a concussion assessment tool in the pediatric emergency department: a pilot study
In the United States, an annual estimate of 1.36 million traumatic brain injuries present to the emergency department (ED), of which approximately 75% are concussions. Proper and timely treatment of concussion is especially important in pediatrics as children and adolescents under the age of 19 are at a higher risk for sustaining more severe and longer-lasting consequences. However, due to the wide range of symptoms at presentation, or to the potential lack of obvious symptoms, concussion can be especially difficult to diagnose in the ED setting. Neurocognitive tests provide a valuable supplement to the clinical diagnosis of concussion by objectively identifying aberrant brain activity. However, many of these tests are often too lengthy and impractical for use in the ED setting. The Immediate Post-Concussion Assessment and Cognitive Test (ImPACT) is a 20-minute computer test that is considered to be one of the gold-standard neurocognitive tests used to diagnose concussion and track recovery. The King-Devick test (KD) is a 1-2 minute test that uses saccadic eye movements to detect suboptimal brain impairment associated with concussion. To date, there have not been any studies that analyzed the relative usability of the KD and the ImPACT in the pediatric ED (PED). The present prospective pilot study investigates the use of the KD as a neurocognitive tool for concussion assessment in the PED and at a post-ED visit, relative to the ImPACT, the gold standard tool for concussion diagnosis. We hypothesize that the change in performance in the KD will correlate with the change in the ImPACT results. To date, 20 subjects between the ages of 11-18 years old presenting to the PED within 72 hours of sustaining a head injury have completed the study. The mean age of our study population was 13.6 years. The average change in test scores between PED and follow-up were: 7.2 seconds in the KD, 0.03 points in the ImPACT reaction time, 1.8 points in verbal memory, 8.3 points in visual memory, 0.8 points in visual motor speed, and 14.9 points in post-concussion symptom scale. Analysis of the correlation of the change in the KD scores to the change in the ImPACT measures revealed that the change in the KD was significantly correlated with the change in the ImPACT reaction time (p < 0.01), and with the change in the ImPACT verbal memory (p < 0.05) in the subjects that presented with LOC, 80% of whom were male. In conclusion, our findings report that the correlation between the results of the KD and the ImPACT is more pronounced in patients presenting with more severe head trauma, such as those leading to LOC. The usability of the KD as a reliable concussion assessment tool in the PED would require further investigation with a larger sample of participants.