|
Exclusion of deep vein thrombosis using the Wells rule in clinically important subgroups: individual patient data meta-analysis |
Geersing GJ, Zuithoff NP, Kearon C, Anderson DR, ten Cate-Hoek AJ, Elf JL, Bates SM, Hoes AW, Kraaijenhagen RA, Oudega R, Schutgens RE, Stevens SM, Woller SC, Wells PS, Moons KG |
|
|
CRD summary This review concluded that the Wells rule could be combined with D-dimer test results to rule out deep vein thrombosis in many patient subgroups. It could be adjusted for suspected recurrent thrombosis, but not cancer. The clear conclusions appear to be based on the evidence presented, and should be considered reliable despite potential flaws in the review process. Authors' objectives To assess the accuracy of the Wells rule for excluding deep vein thrombosis, and whether this was the same for different subgroups of patients. Searching A previous systematic review (see Other Publications of Related Interest) was used to identify studies published in English or French between January 1990 and July 2004. It appears that this search was updated in MEDLINE for studies from January 2006 to present (date not given). Experts and authors were contacted to identify other eligible studies; in the previous review, review articles and reference lists were checked. Study selection Diagnostic studies were eligible for inclusion if they recruited consecutive out-patients with suspected deep vein thrombosis. Studies had to include all the predictors that inform the Wells rule, and categorise patients using the Wells rule before venous imaging. They had to report the results of any D-dimer testing, and the presence or absence of proximal deep vein thrombosis according to a reference test. Acceptable reference tests were compression ultrasonography or venography at initial presentation, or uneventful follow-up for at least three months if venous imaging was not performed. Individual patient data were collated from studies conducted in the USA, Canada, the Netherlands or Sweden. The included studies either used the Wells rule to guide management or were validation studies. Hospital out-patients and primary care out-patients were assessed. Their median age was 59 years and 62% of them were female; 19% had proximal deep vein thrombosis. It was unclear how many reviewers screened the updated search results, and how many screened the results in the original review. Assessment of study quality Studies were not assessed for methodological quality (included studies, in the previous review, had to meet basic quality criteria). Data checking and cleaning procedures were not described. Data extraction Authors were contacted to provide original anonymous data. One eligible study, identified by the search update, could not be included as its authors were unwilling to provide their data. Missing data were imputed, using multivariable regression within each study. The percentage of imputed data ranged from one to five, depending on the variable. Where specific test results were not available for an entire study these were not imputed. It was unclear how many reviewers were involved in data extraction and construction of the individual patient database. Methods of synthesis Various logistic regression models were used to assess the accuracy of the Wells rule for the presence of deep vein thrombosis for each subgroup. Study-level data clustering was adjusted for, using multilevel logistic regression. The subgroups of interest were: care setting (primary, secondary or hospital), malignancy, gender, and history of deep vein thrombosis. Further analyses assessed the impact of adding a negative D-dimer test result to an unlikely Wells rule score. Heterogeneity was explored through the calculation of prediction intervals around the model estimates. Results of the review Thirteen studies contributed data from 10,002 patients. Increasing scores on the Wells rule were associated with an increasing probability of having deep vein thrombosis. Even with a Wells rule score of -2, the likelihood was around 5% (range 2.0 to 5.9), so deep vein thrombosis could not be excluded using the Wells rule alone. The estimated probabilities were almost twice as high in patients with cancer, in patients with suspected recurrent events, and (to a lesser extent) in males. An unlikely score on the Wells rule (≤1), combined with a negative D-dimer test result, was associated with an extremely low likelihood of deep vein thrombosis (1.2%, 95% CI 0.7 to 1.8). The findings were consistent in subgroups defined by type of D-dimer assay (quantitative or qualitative), gender, and care setting (primary or hospital care). For patients with cancer, the combination of an unlikely score on the Wells rule and a negative D-dimer test result occurred in only 9% of patients and was associated with a 2.2% likelihood of deep vein thrombosis. In patients with suspected recurrent events, only the modified Wells rule (adding one point for the previous event) was safe. Authors' conclusions Combined with a negative D-dimer test result (both quantitative and qualitative), deep vein thrombosis could be excluded for patients with an unlikely score on the Wells rule. This finding was true for both genders, and for primary and hospital care. This combination was not safe to use for patients with cancer, and a modified Wells rule (the addition of one point) was required for patients with suspected recurrent deep vein thrombosis. CRD commentary This review and individual patient data meta-analysis was based on a previous systematic review and an updated search. Only one database was searched for English or French papers, which may mean that eligible studies were omitted. The review processes were sparsely reported making it difficult to rule out reviewer error and bias. Studies were not assessed for quality in this review, and no details of any data checking or cleaning procedures were reported. The analyses were pre-specified and appear to have been appropriate. The clear conclusions appear to be based on the evidence presented, and should be considered to be reliable despite potential flaws in the review process. Implications of the review for practice and research Practice: The Wells rule could be used to assess the pretest probability of deep vein thrombosis and the results could drive further diagnostic examinations. The probability of deep vein thrombosis in patients with an unlikely Wells rule score (≤1), combined with a negative D-dimer test result, was low (<2%), enabling the exclusion of about one in three suspected patients. In patients with cancer, the combination was neither safe nor efficient, and in patients with suspected recurrent disease one extra point had to be added to the score to enable the safe exclusion of deep vein thrombosis. Research: Further prospective validation studies were recommended to explore the clinical value of using an age-adjusted cut-off in daily practice. Bibliographic details Geersing GJ, Zuithoff NP, Kearon C, Anderson DR, ten Cate-Hoek AJ, Elf JL, Bates SM, Hoes AW, Kraaijenhagen RA, Oudega R, Schutgens RE, Stevens SM, Woller SC, Wells PS, Moons KG. Exclusion of deep vein thrombosis using the Wells rule in clinically important subgroups: individual patient data meta-analysis. BMJ 2014; 348: g1340 Other publications of related interest Wells PS, Owen C, Doucette S, Fergusson D, Tran H. Does this patient have deep vein thrombosis? JAMA 2006; 295: 199-207. Indexing Status Subject indexing assigned by NLM MeSH Diagnosis, Differential; Fibrin Fibrinogen Degradation Products /metabolism; Humans; Medical History Taking; Predictive Value of Tests; Primary Health Care /methods; Probability; Risk Factors; Venous Thrombosis /blood /diagnosis AccessionNumber 12014018140 Date bibliographic record published 11/03/2014 Date abstract record published 25/03/2014 Record Status This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn. |
|
|
|