Résumé
Venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), is a common clinical problem, associated with a significant mortality and morbidity. Hence, accurate diagnosis and appropriate treatment are essential for patients presenting with suspected VTE. Unfortunately, the diagnosis of VTE is challenging in routine practice because of the nonspecific signs and symptoms of this disease. A large number of epidemiologic studies have focused on VTE, contributing to better understanding of this disease and improving its management. Demonstrated risk factors for VTE have been included into clinical prediction rules derived to help physician identify patients that should be referred for objective diagnostic tests. Over the past decade, the extensive use of diagnostic tests combined with the recent advances in imaging technology have resulted in more frequent diagnosis and treatment of early presentation of VTE, including isolated distal DVT or isolated PE. However the clinical signification of various VTE presentations remains unclear, and knowledge on epidemiology of VTE needs to be improve. Therefore we prospectively investigated the relative frequency and risk factors of isolated distal DVT, proximal DVT, PE with DVT and without DVT. Between November 2004 and January 2006, all patients over 18 years old who were referred to 359 french board certified vascular physicians for a clinical suspicion of VTE were included. VTE presentations were categorized using validated clinical decision rules and objective tests including ultrasonography, lung scan and helical CT scan. Subjects without an objectively confirmed diagnosis of VTE were used as controls. We performed multivariate analysis of risk factors for each type of VTE. 8256 patients entered the study, among which 7532 were analysed. The median age for all patients was 65 years (49–77 years), 2923 (39%) were men, 2925 were inpatients (39%), and 1884 (25%) had a previous history of VTE. 933 had isolated distal DVT (12%), 710 proximal DVT (9.4%), 426 PE with DVT (5.7%), 148 PE without DVT (2.0%) and 5315 had no VTE (70.6%). Classically risk factors were comparable for all different types of DVT (distal, proximal, or associated with PE). Curiously, risk factors for isolated pulmonary embolism are opposite to those for DVT-associated PE. Specially isolated PE was not associated with age (> 75y, OR 1.2 [0.7–2.1, p 0.58), family history of VTE (OR 0.7 [0.4–1.3, p 0.26, bed confinement (OR 0.6[0.4–1.1, p 0.1),plaster (OR 0.3 [0.04–2.5, p 0.28), or acute respiratory or cardiac failure (OR 1.8 [0.9–3.3], p 0.07). Only personnal history of VTE (OR 1.7 [1.1–2.6], recent surgery (OR 1.7 [1.0–3.0], cancer [OR 1.7 [1.1–2.7, p 0.02) and contraceptive use (OR 6.3 [2.5–15.6] p< 0.01) were shwon as risk factors for isolated PE. So this multicenter prospective cohort study shows heterogeneity in the risk factor profile between different forms of VTE encountered in daily practice, providing new insight in the epidemiology of this disease. Specifically, our study underlines the specific risk factors profile of isolated PE comparing to DVT-associated PE.