The gradient between arterial and end-tidal carbon dioxide predicts in-hospital mortality in post-cardiac arrest patient

Abstract

PurposeWe investigated the predictive value of the gradient between arterial carbon dioxide (PaCO2) and end-tidal carbon dioxide (ETCO2) (Pa-ETCO2) in post-cardiac arrest patients for in-hospital mortality. MethodsThis retrospective observational study evaluated cardiac arrest patients admitted to the emergency department of a tertiary university hospital. The PaCO2 and ETCO2 values at 6, 12, and 24 h after return of spontaneous circulation (ROSC) were obtained from medical records and Pa-ETCO2 gap was calculated as the difference between PaCO2 and ETCO2 at each time point. Multivariate logistic regression analysis was performed to verify the relationship between Pa-ETCO2 gap and clinical variables. Receiver operating characteristic (ROC) curve analysis was performed to determine the cutoff value of Pa-ETCO2 for predicting in-hospital mortality. ResultsThe final analysis included 58 patients. In univariate analysis, Pa-ETCO2 gaps were significantly lower in survivors than in non-survivors at 12 h [12.2 (6.5–14.8) vs. 13.9 (12.1–19.6) mmHg, p = 0.040] and 24 h [9.1 (6.3–10.5) vs. 17.1 (13.1–23.2) mmHg, p < 0.001)] after ROSC. In multivariate analysis, Pa-ETCO2 gap at 24 h after ROSC was related to in-hospital mortality [odds ratio (95% confidence interval): 1.30 (1.07–1.59), p = 0.0101]. In ROC curve analysis, the optimal cut-off value of Pa-ETCO2 gap at 24 h after ROSC was 10.6 mmHg (area under the curve, 0.843), with 77.8% sensitivity and 85.7% specificity. ConclusionThe Pa-ETCO2 gap at 24 h after ROSC was associated with in-hospital mortality in post-cardiac arrest patients.