Association between Pulmonary Vascular Obstruction Phenotype and Pulmonary Embolism Recurrence: Results from the PADIS-PE Study

Pulmonary vascular obstruction index (PVOI), assessed using the Qanadli index on computed tomography pulmonary angiography (CTPA) at pulmonary embolism (PE) diagnostic, has been associated with recurrence after unprovoked PE, but its complexity limits routine use. We aimed to evaluate simplified semiquantitative (Sq) measures of initial PVOI for predicting recurrent PE after anticoagulation discontinuation. This post hoc analysis, based on the double-blind, randomized PADIS-PE trial where patients with a first unprovoked PE initially treated during 6 months were allocated to receive either an additional 18-month warfarin or placebo, was restricted on the 180 patients who had PE diagnosed by CTPA. Initial PVOI was assessed using four methods: Qanadli score, modified Qanadli score, anatomical-based SqPVOI1, and lobe-based SqPVOI2. Accuracy and associations with recurrent PE were evaluated using area under the curve (AUC) analysis and Cox regression models. Among the 180 included patients (mean age 66.2 ± 16.5 years; 45.2% female), recurrent PE occurred in 42 patients (36-month median follow-up). Accuracy was comparable across the four PVOI scores, with AUCs ranging from 0.70 to 0.74. All scores showed comparable ability to predict recurrent PE. When using the anatomical-based SqPVOI1, recurrence risk increased progressively with more proximal thrombus location: compared with segmental artery involvement (36-month cumulative incidence, 10.5%; n = 76), risk was 2- to 3-fold with lobar artery involvement (HR 2.90, 95%CI, 1.20–7.05, p = 0.005; 36-month cumulative incidence, 25.6%; n = 63) and 4- to 5-fold increased risk (HR 4.59, 95%CI, 1.85–11.42, p = 0.001; 36-month cumulative incidence, 46.8%; n = 41) with main pulmonary artery involvement. Simplified anatomical assessment of PVOI showed prognostic information comparable to quantitative indices. Further prospective validation is needed.