The utility of pulmonary computed tomography angiography (CTA) in the prediction of short- and long-term outcomes after pulmonary embolism (PE) is controversial. Between November 2011 and September 2014, 190 normotensive patients (age, 61 ± 16.90 years, 53.7 % female) were diagnosed with acute PE using a 128-slice dual-source pulmonary CTA scanner. All the related clinical and cardiovascular measurements were recorded. Primary endpoints were 30-day PE-related death, 30-day composite complications (death, hemodynamic instability, thrombolysis and thrombectomy, inotrope, and mechanical ventilation use), and long-term all-cause mortality during a median follow-up of 14.78 months. Overall 1-month mortality is 5.8 %, and death is PE-related in 4.7 % of total patients. Although non-significant, O2 saturation <90 % and the right ventricular short-axis to left ventricular short-axis diameters (RV/LV) ratio increase the risk of PE-related death by 3.5 and 2 times, respectively. The independent predictors of 30-day complications (15.8 %) are O2 saturation <90 % (OR: 3.924, 95 % CI 1.505-10.229), RV/LV ratio (OR: 3.018, 95 % CI 1.455-6.263), and heart rate ≥ 110 beats/min (OR: 2.607, 95 % CI 1.063-6.391). For long-term mortality (13.7 %), O2 saturation <90 % is an independent predictor (HR: 4.454, 95 % CI 2.016-8.862). The independent impact of the RV/LV ratio on the long-term mortality has a trend towards statistical significance (HR: 1.762, 95 % CI 0.968-4.218; p value = 0.064). The PE-related death is 4.7 % within 30 days after admisson and 13.7 % after a median follow-up of 14 months. Among the pulmonary CTA parameters, only the RV/LV ratio and among the clinical and paraclinical measures, O2 saturation <90 % remain independent predictors of short- and long-term mortality and complications after the diagnosis of PE.
Keywords: Computed tomographic pulmonary angiography; Mortality; Pulmonary embolism; Right ventricular dysfunction.