Derivation and Validation of a Prediction Model for Risk Stratification of Post-Thrombotic Syndrome in Elderly Patients with a First Deep Vein Thrombosis

Abstract

Background Not all patients carry the same risk of developing a post-thrombotic syndrome (PTS), we therefore aimed to derive a prediction rule for risk stratification of PTS in patients with deep vein thrombosis (DVT). Methods Our derivation sample included 276 patients with a first acute symptomatic lower limb DVT enrolled in a prospective cohort. We derived our prediction rule using regression analysis, with the occurrence of PTS within 24 months of a DVT based on the Villalta score as outcome, and 11 candidate variables as predictors. We used bootstrapping methods for internal validation. Results Overall, 161 patients (58.3%) developed a PTS within 24 months of a DVT. Our prediction rule was based on five predictors (age ≥ 75 years, prior varicose vein surgery, multi-level thrombosis, concomitant antiplatelet/non-steroidal anti-inflammatory drug therapy and the number of leg symptoms and signs). Overall, 16.3, 31.2 and 52.5% of patients were classified as low- (score, 0–3), moderate (score, 4–5) and high-risk (score, ≥ 6), for developing a PTS. Within 24 months of the index DVT, 24.4% of the patients in the low-risk category developed a PTS, 38.4% in the moderate and 80.7% in the high-risk category. The prediction model showed good predictive accuracy (area under the curve, 0.77; 95% confidence interval, 0.71–0.82, calibration slope, 0.90 and Brier score, 0.20). Conclusion This easy-to-use clinical prediction rule accurately identifies patients with DVT who are at high risk of developing PTS within 24 months who could potentially benefit from special educational or therapeutic measures to limit the risk of PTS.