Background: Using data from secondary care outpatients, Wells and colleagues developed a diagnostic rule to estimate the probability of the presence of deep venous thrombosis (DVT). The accuracy of the Wells rule has not been properly validated for use in primary care patients in whom DVT is suspected.
Objective: To validate the diagnostic accuracy of the Wells rule, with and without D-dimer testing, in a primary care setting.
Design: Cross-sectional study with prospective data collection from 1 January 2002 to 1 March 2003.
Setting: 110 primary care practices in a circumscribed geographic region in The Netherlands.
Participants: 1295 consecutive patients who consulted their primary care physician about symptoms suggestive of DVT.
Measurements: All patients underwent history-taking and physical examination to calculate the Wells rule score, and D-dimer testing. Repeated leg ultrasonography was the reference standard to determine the true presence or absence of DVT.
Results: In the primary care setting, 12.0% of patients in the low-risk group had DVT; the original study by Wells and colleagues reported a rate of 3% among such patients. When combined with negative results on a D-dimer test, the Wells rule yielded a prevalence of DVT of 2.9% in the lowest-risk group, whereas the prevalence was 0.9% in the original study.
Limitations: Patients with previous DVT were included, and the diagnostic reference standard was different from that used in Wells and colleagues' original study.
Conclusion: The Wells rule, alone or in combination with D-dimer testing, does not guarantee accurate estimation of risk in primary care patients in whom DVT is suspected.