The predictive value of scores based on peripheral complete blood cell count for long-term neurological outcome in acute carbon monoxide intoxication.

Abstract

Growing evidence indicates that inflammation is associated with neurological sequelae after CO poisoning. Several scores incorporating the peripheral complete blood cell (CBC) count have been introduced as indicators of systemic inflammation. This study investigated whether these scores can improve the predictive accuracy for long-term neurological outcome of acute carbon monoxide (CO) poisoning. Two hundred seventy-nine patients who suffered from acute CO poisoning were included. Data on the demographic details, serial peripheral CBC counts and scores based on peripheral CBC counts (neutrophil-lymphocyte ratio (NLR), monocyte-lymphocyte ratio (MLR), platelet-lymphocyte ratio (PLR), systemic immune inflammation index [SII]) over the first 12hours after presentation together with the clinical course during hospitalization and long-term neurological outcome were collected. Both a multivariate logistic regression model with only significant univariate predictors and a model with univariate predictors plus each score for long-term neurological outcome were constructed. Patients with a poor long-term neurological outcome had higher neutrophil and monocyte counts and lower lymphocyte counts over the first 12hours after admission than patients with a good outcome. The diagnostic performance of the NLR, MLR and SII over the first 12hours for predicting long-term neurological outcome was acceptable. These scores at presentation were independently associated with the long-term neurological outcome. Among these scores, only the SII at presentation significantly improved the predictive accuracy of the model when combined with clinical parameters (AUC 0.949, 95% CI 0.916-0.972 vs AUC 0.923, 95% CI 0.884-0.952 for the model with clinical parameters only, P=0.0476). The optimal cut-off value for SII at presentation was 1012.2, resulting in a sensitivity of 97% (84.2%-99.9%) and specificity of 64.6% (58.2%-70.6%). The SII at presentation could significantly improve the prognostic accuracy for predicting the long-term neurological outcome in patients with acute CO poisoning. Because the SII is an inexpensive and easily measurable parameter, it might be used as a prognostic tool in clinical fields.