Screening for intracranial hypertension using ambient-light-corrected computational pupillometry: a prospective pilot study

Le résumé fourni par la source

BackgroundIntracranial hypertension is a life-threatening condition. Invasive intracranial pressure (ICP) monitoring remains the reference standard; however, catheter-based monitoring is not available for all at-risk intensive care unit patients. We evaluated whether ambient-light-corrected computational pupillometry could provide a bedside rule-out signal using the Pupil Reactivity (PuRe) score.MethodsIn this prospective single-center pilot study (2025–2026), adults who underwent intraparenchymal ICP monitoring underwent serial computational pupillometry (PuRe Pupillometer, Solvemed Inc.). The PuRe score is a 0–5 measure of pupillary reactivity (0, non-reactive; 0.1–2.9, sluggish; and ≥3, brisk). The primary outcome was diagnostic performance for ICP ≥22 mm Hg at the clinical brisk-reactivity boundary (PuRe score <3), using analyses that accounted for repeated measurements within patients.ResultsTwenty-six patients were included. At PuRe score <3, negative predictive value was 97.1%, sensitivity 81.1%, specificity 64.5%, and area under the receiver operating characteristic curve 0.77 (95% CI, 0.55–0.92). Mean ICP fell from 30.7 mm Hg with unreactive pupils to 7.6 mm Hg with brisk reactivity, and the inverse PuRe–ICP association held in a mixed-effects model accounting for clustering (slope –1.41 mm Hg per unit, P<0.001). The PuRe score showed no detectable dependence on ambient lighting in the ICP- and patient-adjusted models (P=0.13), whereas the constriction-based metrics remained light-dependent (P<0.001).ConclusionIn this small single-center pilot cohort, preserved PuRe reactivity was infrequently accompanied by concurrent ICP ≥22 mm Hg. These preliminary findings support the further validation of ambient-light-corrected computational pupillometry as a potential adjunctive bedside screening, but not as a replacement for invasive monitoring.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.