Long-term mortality from imaging-diagnosed pulmonary embolism in women and men

Pulmonary embolism (PE) is a heterogeneous disease with variable severity. Risk factors and occurrence of PE differ in patients by gender, age, and comorbidity. Our goal was to determine long-term mortality in imaging diagnosed PE in men and women while accounting for differences in age and health status. This was a retrospective study of adults imaged for PE with ventilation/perfusion scintigraphy or CT pulmonary angiography over a 7-year period at our multisite urban academic medical center. Electronic health records (EHR) were reviewed for patient data, including Charlson comorbidity index (CCI). Mortality at one-year was determined from EHR and National Death Index. Association between PE and survival was assessed using Cox proportional hazard model. 33,628 patients (55yo +/- 19, 69 % women) were imaged for PE, 9.5 % of exams were positive. One-year mortality for the cohort was 9.9 % for women vs. 16.8 % for men, p < 0.001. Women with PE had higher increased one-year mortality (HR 2.75[2.48-3.06]) than men with PE (HR:1.38[1.21-1.58]). Women with PE also had higher hazard ratio after inclusion of age and CCI as covariates in the Cox proportional hazard model (women HR 1.68 [1.51-1.86] versus men HR 1.20 [1.05-1.36]). The higher impact of PE on the mortality of women yielded a similar one-year mortality for women and men with PE, 21.9 % vs. 21.5 %, respectively (P = 0.83). Positive PE study was associated with significantly increased one-year mortality, however this effect was much stronger in women. This gender difference persists after accounting for age and comorbidities.