ABSTRACT
Objective
To assess the association between the oxygenation index (OI) and survival in mechanically ventilated dogs with primary pulmonary parenchymal disease.
Design
Retrospective, multicenter study.
Setting
Three veterinary medical teaching institutions and one private veterinary referral center.
Animals
Seventy-nine client-owned dogs.
Interventions
None.
Methods
OI, calculated as (mean airway pressure × FiO2 × 100) / PaO2, and the PaO2/FiO2 (PF) ratio were calculated for dogs undergoing mechanical ventilation for primary pulmonary disease. Median OI was lower in survivors (2.6) than nonsurvivors (6.6; p < 0.001), while PF was higher in survivors (317.9 vs. 177.9; p < 0.001). OI predicted mortality with an area under the receiver operating characteristic curve of 0.73, sensitivity of 65%, and specificity of 80% at an optimal cutoff of 4.3. Median PF had an area under the receiver operating characteristic curve of 0.72, sensitivity of 70%, and specificity of 73% at an optimal cutoff of 237.8. Each 1-unit increase in OI was associated with a 35% higher mortality risk (odds ratio: 1.35; 95% confidence interval: 1.14–1.61). Survivors showed greater improvement in OI during ventilation (p = 0.004). Using Pediatric Acute Lung Injury Consensus Conference and pediatric acute respiratory distress syndrome thresholds, survival likelihood declined with increasing severity, with no survivors in the severe category (OI >16). Similar trends were observed using updated Pediatric Acute Lung Injury Consensus Conference-2 criteria and acute respiratory distress syndrome severity classifications.
Conclusions
Higher OI values and lower PF ratios were associated with mortality in this group of mechanically ventilated dogs, with both metrics demonstrating similar predictive accuracy. Results suggest species-specific OI and PF thresholds are needed.
Journal of Veterinary Emergency and Critical Care, EarlyView.Wiley: Journal of Veterinary Emergency and Critical Care: Table of Contents