How can we improve the diagnosis of deep vein thrombosis and pulmonary embolism at the primary level?

Abstract

c i n e ltd Venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), is not uncommon in primary care, with a reported incidence of almost one case per 1000 person-years [1]. VTE is highly feared for its associated mortality, since the risk of PE-related early mortality may be higher than 15%, as well as its associated morbidity, since patients with VTE may also develop long-term complications such as chronic thromboembolic pulmonary hypertension or postphlebitic syndrome [2]. On the other hand, anticoagulation is an effective treatment for VTE, but carries a non-negligible rate of major bleeding events, ranging from 2% in clinical trials to 7.4% in population cohort studies [3]. When VTE is suspected, it is therefore mandatory to quickly achieve the correct diagnosis. Nonetheless, this process is particularly difficult in primary care, because of the limited number of available diagnostic tools and the nonspecific clinical presentations of DVT and PE. Therefore, the ultimate aim is to determine in which patients DVT or PE can be safely ruled out and which patients should be referred for additional diagnostic workup in secondary care. The current diagnostic approach to VTE is based on the combination of three tools: diagnostic clinical prediction rules (CPRs) for assessing the pretest probability; d-dimer measurement; and imaging tests (i.e., venous ultrasono graphy for DVT and computed tomo graphy pulmonary angiography [CTPA] or ventilation-perfusion lung scan for PE). In patients who present with suspected DVT, the CPR developed by Wells and colleagues is currently the most used worldwide [4]. The Wells CPR includes information from patients’ medical history (active cancer, recent immobilization of the lower extremities, recent bedridden or major surgery), physical examination (tenderness along the deep venous system, entire leg swollen or calf difference greater than 3 cm, pitting edema and collateral superficial veins) and a subjective variable (i.e., an alternative diagnosis for patients’ symptoms) [4]. As the Wells CPR has been developed for secondary care outpatients, its applicability to primary care is still uncertain. A prospective validation study showed that the Wells CPR, in combination with a negative quantitative d-dimer test, did not adequately rule out DVT in the primary care setting [5]. This result could be partly due to different population characteristics, since secondary care patients are often selected by primary care physicians, and partly to the subjectively estimated probability of an alternative diagnosis, which might be assessed differently by general practitioners and by specialists. Oudega and colleagues have developed a specific primary care CPR for suspected DVT that included eight simple diagnostic indicators (male gender, oral contraceptive use, presence of malignancy, recent surgery, absence of leg trauma, vein Nicoletta Riva Research Center on Thromboembolic Disorders & Antithrombotic Therapies, Department of Clinical & Experimental Medicine, University of Insubria, Varese, Italy