Predicting the Prognosis of Diffuse Axonal Injury Using Automated Pupillometry

Abstract

Abstract Background Although previous studies have reported various predictive indicators of diffuse axonal injury (DAI), a consensus regarding the gold-standard predictive indicator has not yet been reached. The usefulness of automated pupillometry in patients with consciousness disorders has been widely reported; however, there are few reports of its use in patients with DAI. We thus investigated the usefulness of pupillary findings for predicting prognosis in DAI.Methods We included patients with a diagnosis of DAI admitted to our center from 1 June 2021 to 30 June 2022. Pupillary findings in both eyes were quantitatively measured by automated pupillometry every 2 hours after admission. We then examined the correlations between automated pupillometry values, patient characteristics, and outcomes such as the Glasgow Outcome Scale Extended (GOSE) score 6 months after injury and the time to follow commands.Results Of the 22 included patients, 5 presented oculomotor nerve palsy. Oculomotor nerve palsy was correlated with all outcomes. In contrast, Marshall computed tomography classification, injury severity score, and DAI grade were correlated with few outcomes. Some automated pupillometry values were significantly correlated with GOSE at 6 months after injury, and many values from the first 24 hours of measurement were correlated with the time to follow commands. In general, these results were not affected by adjusting for the sedation period, injury severity score, or Marshall computed tomography classification. A subgroup analysis of patients without oculomotor nerve palsy revealed that many automated pupillometry values from the first 24 hours of measurement were significantly correlated with most outcomes. The cutoff values that differentiated good prognosis (GOSE 5–8) from poor prognosis (GOSE 1–4) were a constriction velocity of 1.43 (area under the curve [AUC] = 0.81 [0.62–1], p = 0.037) and a maximum constriction velocity of 2.345 (AUC = 0.78 [0.58–0.98], p = 0.04). The cutoff values that differentiated the time to follow commands into within 7 days and 8 days or over were a percentage of constriction of 8 (AUC = 0.89 [0.68–1], p = 0.011), a constriction velocity of 0.63 (AUC = 0.92 [0.78–1], p = 0.013), a maximum constriction velocity of 0.855 (AUC = 0.9 [0.74–1], p = 0.017), and an average dilation velocity of 0.175 (AUC = 0.95 [0.86–1], p = 0.018).Conclusions Pupillary findings in DAI were a strong predictive indicator of prognosis. Quantitative measurements using automated pupillometry may facilitate the prediction of DAI prognosis.