Complete blood count-derived inflammatory markers in canine cerebrovascular accidents: diagnostic utility and prognostic implications.

Background: Cerebrovascular accident (CVA) is an under-recognized cause of acute neurological dysfunction in dogs. Although magnetic resonance imaging (MRI) supports diagnosis, the limited access and overlapping imaging features complicate its interpretation. Blood-based inflammatory indices such as the neutrophil-to-lymphocyte ratio (NLR) and systemic immune-inflammation index (SII) have shown diagnostic and prognostic value in human stroke; however, their clinical relevance in canine CVA has not been validated. Objectives: To evaluate the diagnostic and prognostic utility of the NLR and SII in dogs with MRI-confirmed CVA compared to healthy controls. Methods: Seventeen dogs with CVA and 20 healthy controls were analyzed retrospectively. The NLR and SII were calculated from complete blood counts at presentation. Functional outcomes were assessed using the Canine Functional Stroke Score (CFSS₀, CFSS₃₀). Associations with 30-day survival were analyzed between groups. Results: Dogs with CVA had significantly higher NLR and SII values than healthy controls. The receiver operating characteristic curves yielded areas under the curves of 0.82 (95% CI: 0.68-0.96) for NLR and 0.85 (95% CI: 0.71-0.99) for SII. The optimal cutoffs determined by the Youden index were 3.92 for NLR (sensitivity 70.6%, specificity 90.0%) and 1,352 for SII (sensitivity 76.5%, specificity 90.0%). Both markers were moderately correlated with poorer 30-day recovery. The SII was significantly higher in non-survivors than in survivors. Baseline CFSS was not significantly correlated with the degree of functional improvement but was significantly higher in non-survivors. Multivariate Cox proportional hazards regression model, neither the NLR nor the SII were significantly associated with survival. Although individual predictors did not reach statistical significance, the overall model demonstrated a significant association with 30-day survival. In the dichotomized analysis, the SII model demonstrated an improved model fit for predicting mortality, whereas the NLR model showed a poorer fit. Kaplan-Meier analysis showed that the high-SII group had lower survival rates than the low-SII group, while no significant difference in survival was observed between the NLR groups. Conclusion: NLR and SII may help differentiate dogs with CVA from healthy dogs and reflect clinical outcomes. The SII also shows potential prognostic value for short-term survival.