Diagnostic Value of Parameters Related to White Blood Cell Counts for Troponin I Elevation in CO Poisoning.

Abstract

To assess myocardial injury related to acute carbon monoxide (CO) poisoning, serial troponin I is measured in patients not presenting with troponin I elevation. This retrospective study investigated whether parameters related to white blood cell (WBC) counts (total and differential WBC counts, neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio) improved predictive accuracy for troponin I elevation (> 0.04ng/ml) in patients not presenting with evidence of myocardial injury. Serial parameters, troponin I values, and clinical courses were collected in 241 patients. Troponin I was elevated in 33 (13.7%) patients after hospitalization. The median lag times to troponin I elevation in patients with undetectable and detectable troponin I (0.015ng/ml ≤ troponin I ≤ 0.04ng/ml) at presentation were 5.9h and 3.0h, respectively. Patients with troponin I elevation after presentation had higher total WBC and neutrophil counts and NLRs and a lower lymphocyte count during the first 4h after presentation than patients without troponin I elevation during hospitalization. Total WBC count, neutrophil count, and log NLR at presentation were selected as independent predictive factors for troponin I elevation after presentation. However, only the neutrophil count and log NLR at presentation improved the predictive accuracy in combination with clinical parameters compared with that achieved with a predictive model including only clinical parameters. The optimal cut-off neutrophil count and NLR were 5.21 × 103/uL and 4.02, respectively. The total neutrophil count and NLR, which are widely available and inexpensive parameters obtained in the emergency department (ED), are promising screening tools for predicting the risk of troponin I elevation in patients without evidence of myocardial injury-related acute CO poisoning at presentation.