Intraoperative monitoring and anaesthetic complications in horses: A secondary analysis of the CEPEF4 multicentre cohort

Abstract

Abstract Background CEPEF4 provides the most comprehensive contemporary dataset on perianaesthetic risk in horses. Intraoperative complications, monitoring and American Society of Anesthesiologists (ASA)‐stratified associations with death remain incompletely described. Objectives To describe monitoring and complication incidence, assess associations with ASA physical status, and compare ASA‐stratified mortality by complication status. Study Design Secondary analysis of a prospective, multicentre cohort study. Methods Data from 47,371 equine general anaesthetics were analysed. Monitoring was described overall and by ASA grade. Complication incidence was calculated within monitored subsets for hypotension, hypoxaemia, hypercapnia, arrhythmias and hypothermia. Intraoperative awakening and severe haemorrhage were analysed in the full cohort. Logistic regression estimated ASA‐graded odds ratios for complications. The primary outcome analysis used the predefined CEPEF4‐DEAD endpoint and compared complication‐positive and complication‐negative horses within ASA I–II and ASA III–V strata. Sensitivity analyses used DEAD‐or‐EUTHANASIA, centre‐cluster and monitor‐restricted approaches. Results Monitoring coverage was high for electrocardiography (92.2%), pulse oximetry (89.2%), capnography (86.8%) and invasive arterial pressure (82.0%). Among monitored cases, hypotension occurred in 16,926/41,939 (40.4%), hypoxaemia in 2346/44,279 (5.3%), hypercapnia in 2512/42,941 (5.8%), arrhythmias in 720/43,666 (1.6%) and hypothermia in 851/18,358 (4.6%). Intraoperative awakening and severe haemorrhage occurred in 5499/47,371 (11.6%) and 378/47,371 (0.8%) cases, respectively. The odds of most complications increased with ASA grade, whereas intraoperative awakening showed no clear association. Within ASA strata, the main DEAD‐only associations were observed for hypotension, hypoxaemia, arrhythmias and severe haemorrhage; hypercapnia was associated with DEAD‐only outcome in ASA III–V horses. Sensitivity analyses generally supported these findings, although marginal associations and estimates based on the broader DEAD‐or‐EUTHANASIA endpoint were interpreted cautiously. Main Limitations Event detection depended on monitoring availability and clinician recording; causal inference was precluded. Conclusions Intraoperative complications increased with ASA grade, with hypotension most frequent. Findings provide contemporary reference values and identify clinically relevant complication patterns for future outcome‐focused analyses.