Pulmonary Embolism Severity Index and troponin testing for the selection of low-risk patients with acute symptomatic pulmonary embolism.

BACKGROUND: The combination of the Pulmonary Embolism Severity Index (PESI) and troponin testing could help physicians identify appropriate patients with acute pulmonary embolism (PE) for early hospital discharge. METHODS: This prospective cohort study included a total of 567 patients from a single center registry with objectively confirmed acute symptomatic PE. On the basis of the PESI, each patient was classified into one of five classes (I-V). At the time of hospital admission, patients had troponin I (cTnI) levels measured. The endpoint of the study was all-cause mortality within 30 days after diagnosis. We calculated the mortality rates in four patient groups: group 1, PESI class I-II plus cTnI < 0.1 ng mL(-1); group 2, PESI classes III-V plus cTnI < 0.1 ng mL(-1); group 3, PESI classes I-II plus cTnI > or = 0.1 ng mL(-1); and group 4, PESI classes III-V plus cTnI > or = 0.1 ng mL(-1). RESULTS: The study cohort had a 30-day mortality of 10% [95% confidence interval (CI), 7.6-12.5%]. Mortality rates in the four groups were 1.3%, 14.2%, 0% and 15.4%, respectively. Compared with non-elevated cTnl, the low-risk PESI had a higher negative predictive value (NPV) (98.9% vs. 90.8%) and negative likelihood ratio (NLR) (0.1 vs. 0.9) for predicting mortality. The addition of non-elevated cTnI to low-risk PESI did not improve the NPV or the NLR compared with either test alone. CONCLUSIONS: Compared with cTnl testing, PESI classification more accurately identified patients with PE who are at low risk of all-cause death within 30 days of presentation.