Abstract
Pulmonary embolism (PE) is a fatal condition particularly in critically ill patients. We determined the association among the cardiac troponin T (cTnT) level, acute physiologic and chronic health evaluation (APACHE III) scoring system, and in-hospital mortality in critically ill patients with acute PE. A total of 501 patients with PE were initially enrolled. According to the multivariable logistic regression model for in-hospital mortality, the odds ratio of the cTnT level and APACHE III score was 1.96 (95% confidence interval [CI] = 1.18-3.24, P = 0.008) and 1.03 (95% CI = 1.02-1.05, P < 0.001), respectively. The area under the curve (AUC) of cTnT and APACHE III score for in-hospital mortality was 0.630 (95% CI = 0.586-0.672, P = 0.03) and 0.740 (95% CI = 0.699-0.778, P = 0.02), respectively. The discriminatory cTnT and APACHE III threshold values for in-hospital mortality were 0.08 ng/L and 38 score, respectively; the sensitivities and specificities of cTnT were 46.48 and 83.10%, respectively, whereas those of the APACHE III score were 74.88 and 54.19%, respectively. The cTnT and APACHE III scores were combined in the logistic analysis model, and a regression equation was derived to calculate the in-hospital mortality. The AUC was found to increase to 0.788 (95% CI = 0.734-0.840, P = 0.025). The sensitivity and specificity increased to 84.5 and 71.4%, respectively. The cTnT and APACHE III scores exhibited a significant association with in-hospital mortality of critically ill patients with PE. In conclusion, these parameters in combination can significantly improve the in-hospital mortality prediction.