Venous Thromboembolism in Northeast Melbourne, Australia: Evaluation of Epidemiology, Risk Factors and Treatment Strategies in the Warfarin Era

Aim

Venous thromboembolism (VTE) is a major cause of morbidity and mortality and is estimated to cost approximately AUD$117 000 per person (Access Economics 2008). Despite advances in thrombosis management, recurrence rates remain high and management strategies are often heterogenous even within a single institution. While most studies have analysed specific aspects of VTE management, we aim to provide a holistic evaluation of real-world VTE management in the warfarin era including identifiying potential causal effects and complications.

Method

Retrospective evaluation of VTE over an 18-month period, from July 2011 to December 2012, at two major hospitals in Northeast Melbourne, Australia, including demographics, provoking factors, management, complications and mortality. Comparisons were made between cancer and non-cancer populations.

Result

1003 patients, with median age of 63 (range 19-97) years, were identified including 26 recurrences (total 1029 episodes) - 577 (56%) pulmonary embolism (PE), 428 (42%) deep venous thrombosis (DVT). There was an overall male predominance (52%), particularly affecting the DVT subgroup (57% vs 48%, p=0.003) with no gender differences detected in the PE subgroup. 20% reported prior VTE and left limb DVT was more common (49% vs 43%, p=0.0008). The median follow up was 20 (range 10-32) months.

NON-CANCER PATIENTS

In this cohort, 63% had provoked VTE and thrombophilia screen was performed in 41%. The median duration of anticoagulation was 6 and 7 months for DVT and PE respectively. The majority (90%) was on warfarin for long-term anticoagulation. 5% required further interventions – IVC filter (n=28) and thrombolysis (n=15).

38% had end-of-treatment repeat imaging and residual clot was observed in 40%. Clot persistence was associated with increased recurrence risk, with an odds ratio of 2.64 (1.15 – 6.04, p=0.02). 8% had recurrent thrombosis with no difference between provoked versus unprovoked VTE (7.5% vs 9.0%, p=0.45). 5% reported grade III/IV bleeding, independent of duration of anticoagulation. Patients on enoxaparin had higher risk of bleeding (28% vs 10%, p<0.001). The all-cause mortality rate was 11%.

In the sub-analysis of below knee DVT (BKDVT), 3 patients (1.5%) were subsequently diagnosed with cancer, similar prevalence to those with major VTE (1.7%), defined as proximal DVT and/or PE. BKDVT were more likely to be provoked (72% vs 55%, p<0.001) and recurrence was similar to major VTE (6.8% vs 8.7%, p=0.42) although patients with major VTE were more likely to experience grade III/IV bleeding complications (6.3% vs 1.0%, p=0.003) despite similar duration of therapy. Mortality rate in this cohort was 5.5% with no thrombosis-related deaths.

CANCER PATIENTS

233 (23%) patients had active malignancy at time of VTE with 14 patients (1.4%) subsequently diagnosed. Cancer patients were older (67 vs 61 years, p<0.001) and had higher clot burden with more PE (64% vs 53%, p=0.004), proximal DVT (63% vs 46%, p=0.0008) and bilateral DVT (16% vs 5%, p<0.001) compared to non-cancer patients. Patients with metastatic cancer were also more likely to have unprovoked events (p=0.015). Cancer patients were more likely to require IVC filters (9% vs 3.6%, p<0.001) and lifelong anticoagulation (35% vs 18%, p<0.001).

Recurrent thrombosis (16%, p<0.001) and Grade III/IV bleeding (9%, p=0.025) were more common. However, bleeding rates in cancer patients on long-term enoxaparin were similar to warfarin. Mortality rate in the cancer patients was 63% and were related to with higher incidence of complications-related deaths (p<0.001).

Conclusion

VTE is associated with significant mortality, even in non-cancer patients (11%). The annual recurrence rate is 6%, with malignancy and residual clot on repeat imaging being major risk factors. Cancer patients experience more Grade III/IV bleeding complications (9% vs 5%) and had higher clot burden and all-cause mortality compared to non-cancer patients.

Interestingly, BKDVT, often considered a minor VTE, had comparable rates of recurrence and subsequent cancer detection to major VTE suggesting that the investigation and treatment of BKDVT should not differ from major VTE.

Further evaluation of new treatment strategies as well as clinical and laboratory risk assessments are required to improve the management for VTE. This data will serve as an important baseline for future comparison in the new era of novel oral anticoagulants.