Mortality at 30 and 90 days in elderly patients with pulmonary embolism: a retrospective cohort study

Abstract

Pulmonary Embolism (PE) incidence increases with age. Data on mortality and prognosis in elderly patients with suspected PE are lacking. (1) To assess 30- and 90-day mortality in subjects with PE from an elderly population seen in the emergency department (ED); (2) to test the prognostic accuracy of a simplified Pulmonary Embolism Severity Index (sPESI) coupled to a highly sensitive cardiac Troponin T (hs-cTnT) level. A retrospective cohort study was performed, including patients evaluated in the ED of Vimercate Hospital for clinically suspected PE from 2010 to 2012. Study population: n = 470, 63.4 % women, mean age ± SD 73.06 ± 16.0 years, 40 % aged ≥80 and 77.7 % ≥65 years old, confirmed PE: 22.6 % (106 cases). Within 30 and 90 days, mortality among patients with confirmed PE was 14.2 % (8.8–22.0) and 20.8 % (16.5–41.7). In subjects aged ≥80 years, 30-day mortality was 18.9 % among patients with confirmed PE, and 12.6 % among those with PE excluded (p = 0.317). Ninety-day mortality rates were 29.7 and 19.9 %, respectively (p = 0.193). In patients with confirmed PE, Negative Predictive Value of sPESI was 94.1 % (80.3–99.3) for 30 days and 88.2 % (72.3–96.7) for 90-day mortality. Adding the hs-cTnT level to sPESI did not improve its performance. (1) In an elderly population referring to the ED with clinically suspected PE, mortality was high both in subjects with and without confirmed PE; (2) the ability of sPESI and hs-cTnT to predict PE mortality seems to be lower than reported in studies based on data from younger populations. Better risk stratification tools will be necessary to improve clinical management in this setting.