Wells rule may not be accurate for excluding DVT in certain patients

The Wells rule may not be accurate for excluding deep venous thrombosis (DVT) in patients with cancer and those with previous DVT, according to a new study.

Researchers performed a meta-analysis of individual patient data to examine the accuracy of the Wells rule in excluding DVT in different subgroups of patients. Studies selected for the meta-analysis enrolled consecutive outpatients with suspected DVT, scored all of the variables of the Wells rule, and used an appropriate reference standard. Data from the 13 included studies were merged into a single dataset. The current study's main outcome measures were estimated differences in predicted probabilities of DVT according to the Wells rule, along with differing abilities to exclude DVT based on an unlikely Wells score and a negative D-dimer result. The study results were published online March 10 by BMJ.

Data from 10,002 outpatients were included in the study; no inpatients were included. A total of 1,864 patients (19%) had proximal DVT, 62% of patients were women, and the median patient age was 59 years. In the patient population overall, higher Wells rule scores were reliably associated with higher DVT risk. However, the actual probability of DVT was approximately 5% with a Wells rule score of −2, indicating that the Wells rule alone should not be used for DVT exclusion. Patients with cancer, those with suspected recurrent DVT, and, to a lesser extent, men had almost twofold higher predicted probabilities of DVT when the Wells rule score was low.

Patients with a Wells rule score of 1 or lower and a negative D-dimer results (29% of patients) were extremely unlikely to have DVT (probability, 1.2%; 95% CI, 0.7% to 1.8%). The authors noted that a failure rate of up to 2% is often considered the threshold of acceptability. Nine percent of patients with cancer had a Wells score of 1 or lower and negative D-dimer results, and this combination was associated with a 2.2% probability of DVT. The Wells rule combined with a negative D-dimer test result did not safely predict DVT in patients with suspected recurrent disease unless 1 point was added to the score.

The authors noted that the DVT assessor was not blinded in many of the included studies and that compression ultrasonography was often used as the reference standard although it is less reliable in assessing recurrent DVT events, among other limitations. They concluded that a low score on the Wells rule when combined with a negative D-dimer result can reliably exclude DVT in most patients. However, this combination was not found to be safe or efficient in patients with cancer. When the combination is used in patients with suspected recurrent DVT, the authors recommended that an extra point be added to the Wells score in this category, which preserves the ability to exclude DVT in this patient group.