How well the brachycephalic risk score predicts surgery complications

Abstract

OBJECTIVE: To assess the validity of the brachycephalic risk (BRisk) score to correctly predict the risk of complications in dogs undergoing surgery for brachycephalic obstructive airway syndrome (BOAS). METHODS: Medical records from January 1, 2020, to December 31, 2024, were retrospectively reviewed. Variables for the BRisk score were collected, and scores were assigned. Area under the curve (AUC) was calculated for the ability of the BRisk score to correctly predict major postoperative complications (oxygen required for > 48 hours, temporary tracheostomy, permanent tracheostomy, and/or death). RESULTS: 228 brachycephalic dogs undergoing BOAS surgery were included, of which 29 dogs (12.7%) had major postoperative complications. The median BRisk score for all dogs was 3.5 (range, 0.5 to 9.5). Dogs with a score > 3 were 3.0 times more likely to have major postoperative complications (OR, 3.0; 95% CI, 1.2 to 7.3) compared to dogs with scores ≤ 3. The AUC of the BRisk score for predicting major postoperative complications was 0.719 (95% CI, 0.620 to 0.819). The AUC for French and English Bulldogs was 0.667 (95% CI, 0.527 to 0.806) and 0.754 (95% CI, 0.556 to 0.953), respectively. For breeds other than French or English Bulldog, the AUC was 0.882 (95% CI, 0.757 to 1.000). CONCLUSIONS: The BRisk score had acceptable performance in this population. Performance varied by breed and was worst for French Bulldogs. CLINICAL RELEVANCE: The BRisk score had acceptable performance in this external validation group. A breed-specific BRisk score with excellent to outstanding discrimination may better identify dogs at risk for major postoperative complications following BOAS surgery.