PO-09 - Incidence of pre-operative and post-operative deep vein thrombosis in colorectal cancer surgery. Interim results of a prospective clinical study

Abstract

Venous thromboembolism (VTE) is a major cause of death in patients undergoing colorectal cancer surgery and usually arise from Deep Vein Thrombosis's (DVTs). In patients with cancer below knee and asymptomatic DVTs are at risk of propagating and result in a VTE. Retrospective, population based studies predating extended course venous thromboprophylaxis report an incidence of symptomatic VTE in colorectal cancer patients of approximately 5.5%. Clinical studies have suggested that pre-operative d-dimer may predict the development of post-operative DVTs. We undertook a multicentre prospective study to screen colorectal cancer patients pre- and post-operatively for the presence of DVTs. The aim of the study was to determine the incidence of DVTs in patients undergoing elective, curative surgery for colorectal cancer. The study also aimed to identify histological and clinical factors that pre-dispose to development of a DVT. The role of pre-operative d-dimer in predicting the development of post-operative DVTs was also analysed. Patients at four hospitals undergoing elective curative surgery for colorectal cancer were recruited pre-operatively. Exclusion criteria were: previous VTE, previous malignancy, anti-coagulants. Bilateral full leg venous duplex was performed pre-operatively and at six weeks post-surgery. Plasma D-dimer was measured pre-operatively. Of 60 patients undergoing pre-operative duplex, five (8.3%) had below knee, asymptomatic DVTs. These patients were then excluded from further analysis. Of the remaining 55, 48 had post-operative duplex examination. Three (6.3%) developed post-operative DVTs all of which were asymptomatic and below knee. There were no symptomatic DVTs. Two of these patients had received extended course venous thromboprophylaxis and the other received inpatient thromboprophylaxis. Development of post op DVT was associated with lymph node positivity (p=0.02). There were no other histological, surgical or demographical factors that predicted the development of post-operative DVTs. Pre-operative D-dimer was higher in patients who developed a post-operative DVT compared to those who did not [1,275 ng/L (95%CI: 780 - 1770 ng/L) vs 805 ng/L (95% CI: 632 - 980 ng/L) p=0.03]. Of patients diagnosed with operable colorectal cancer, 8% have pre-operative asymptomatic DVT, and a further 6% develop DVT despite thromboprophylaxis. The subgroup of patients with lymph node involvement may benefit from more aggressive anticoagulation as they are at increased risk of DVT despite extended course anticoagulation. Pre-operative D-dimer may offer a predictive method to identify patients at risk of post-operative DVT.