Association of Cerebral Oxygenation During Prehospital Anaesthesia and Functional Outcome: A Prospective, Observational Multi‐Centre Cohort Study of 1014 Patients

Abstract

<h3>Background</h3><p>Many patients undergoing prehospital anaesthesia may be at risk of inadequate cerebral oxygenation due to underlying conditions or adverse events like hypotension or hypoxia. This study examined whether a decrease in regional cerebral oxygen saturation (rSO₂) measured with near-infrared spectroscopy (NIRS) during prehospital anaesthesia associates with worse outcomes.</p><h3>Methods</h3><p>We conducted a prospective, observational study including adult patients anaesthetised by six prehospital critical care teams. A relative cerebral desaturation event (rCDE) was defined as a ≥ 10% decrease in rSO₂ for ≥ 5 min from baseline. An absolute cerebral desaturation event (aCDE) was defined as rSO₂ < 60% during anaesthesia or lower than baseline if already < 60%. The primary outcome was favourable functional outcome (modified Rankin Scale ≤ 2) at 30 days and secondary outcomes included 30-day survival, 1-year functional outcome, and 1-year survival.</p><h3>Results</h3><p>Among 1014 patients, 199 experienced an rCDE, with 125 (63%) having supraphysiological baseline. rCDE was not associated with outcomes. Of 182 patients with aCDE, 30-day favourable outcomes were not significantly different (30% vs. 36%, <em>p</em> = 0.14, adjusted OR 0.92, 95% confidence interval 0.62–1.34). However, aCDE was associated with lower 30-day survival (46% vs. 58%, <em>p</em> = 0.006) and less favourable 1-year outcomes (31% vs. 41%, <em>p</em> = 0.043). Adjusted analyses showed no significant associations.</p><h3>Conclusion</h3><p>An rCDE was not associated with worse functional outcomes. While aCDEs were linked to unfavourable outcomes in unadjusted analyses, these associations were not significant after adjustment, highlighting the complexity of interpreting NIRS in heterogeneous populations. Condition-specific studies are needed to clarify its role.</p>