Risk factors and timing of venous thromboembolism after radical cystectomy in routine clinical practice: a population-based study.

OBJECTIVE

To describe the risk factors and timing of perioperative venous thromboembolism (VTE) and its association with survival for patients undergoing radical cystectomy (RC) in routine clinical practice.

PATIENTS AND METHODS

The population-based Ontario Cancer Registry was linked to electronic records of treatment to identify all patients who underwent RC between 1994 and 2008; VTE events were identified from hospital diagnostic codes. Multivariate logistic regression analysis was used to determine the factors associated with perioperative VTE. A Cox proportional hazards regression model explored the associations between VTE and survival.

RESULTS

Of the 3 879 patients included in the study, 3.6% (141 patients) were diagnosed with VTE at ≤1 month of their surgical admission date. This increased to 4.7% (181) at ≤2 months and 5.4% (211) at ≤3 months. In all, 55% of VTE events presented after hospital discharge. In multivariate analysis, factors associated with VTE included higher surgeon volume (P = 0.004) and increased length of hospital stay (LOS; P < 0.001). Lymph node yield and adjuvant chemotherapy were not associated with VTE. VTE was associated with an inferior cancer-specific survival [hazard ratio (HR) 1.35, 95% confidence interval (CI) 1.13-1.62] and overall survival (HR 1.27, 95% CI 1.08-1.49).

CONCLUSIONS

Over half of VTE events in RC patients occur after hospital discharge, with a substantial incidence up to 3 months after surgery. Limited actionable risk factors for VTE were identified other than LOS. In this population-based cohort, VTE was associated with inferior long-term survival.