PC1013: Diagnostic and Prognostic Utility of Objective Physiologic Measures for Acute Pediatric Concussion

Background: Diagnosis and prognosis of acute pediatric concussion in the emergency department (ED) relies primarily on subjective symptom reporting. Few objective physiologic measures that have shown utility in specialty settings, specifically pupillary light reflex (PLR) and visio-vestibular examination (VVE) deficits, have been comprehensively evaluated in the ED. Our objective was to assess the diagnostic and prognostic utility of PLR and VVE among a group of acutely concussed ED patients.

Methods: We enrolled participants age 8-18 years with concussion in the ED of a tertiary care children’s hospital within 72 hours of injury. Measures assessed included: (1) PLR, assessing pupillary dynamics in response to a brief white light stimulus (using a Neuroptics PLR-3000 pupillometer) with the following 9 measures: maximum (pre-stimulus) and minimum (post-stimulus) pupil size, percent constriction, maximum/minimum constriction and dilation velocities, latency (time to maximum constriction), and T75 (time to reach 75% of maximum size on re-dilation); (2) A visio-vestibular exam (VVE), measuring smooth pursuits, horizontal/vertical saccades and gaze stability, near point of convergence, monocular accommodation, and complex tandem gait; and (3) Symptom burden, using the Postconcussion Symptom Inventory (PCSI), both at presentation and 28 days post-injury. PLR and VVE metrics were compared to established measures in healthy controls via Student’s t-tests. Within the injured sample, we compared patients seen <=12 hours of injury to those seen >12 hours. Finally, to evaluate prognosis, we compared enrollment PLR metrics, composite VVE score, and PCSI among clinically recovered participants at 28 days (symptoms returning to pre-injury baseline) and those not recovered at 28 days, using both Student’s t-tests and areas under the receiver operating characteristic curves (AUC).

Results: We enrolled 33 participants, mean age 13.0 years, 51.5% female. Three injured participants could not complete PLR due to blinking, and 8 were lost to follow-up in determining recovery. Seven PLR metrics were significantly different between concussed participants and normative data, as was the composite VVE score (Table 1). Though not statistically significant, there was a trend toward control-level values for many PLR metrics and the composite VVE deficits for those evaluated less acutely (>12-72 hours) post-injury. VVE deficits (AUC = 0.750) outperformed initial symptom score (AUC = 0.504) and nearly all PLR measures (latency AUC = 0.771) in predicting recovery (Table 2).

Conclusion: Both PLR and VVE have diagnostic utility in the ED, with PLR measures showing enhancement (larger pupil sizes, faster constriction/dilation velocities) in acutely concussed youth. The time trajectory of both PLR and VVE measures suggest these begin to normalize even within 72 hours post-injury. VVE deficits show significant promise in identification of those at risk for prolonged recovery. Further identification of the ideal combination of symptom scores and objective physiologic measures is needed to determine the optimal concussion assessment battery for ED providers.