|Risk of deep vein thrombosis following a single negative whole-leg compression ultrasound: a systematic review and meta-analysis
|Johnson SA, Stevens SM, Woller SC, Lake E, Donadini M, Cheng J, Labarere J, Douketis JD
This review concluded that withholding anticoagulant therapy in patients with suspected deep vein thrombosis following a single negative whole-let compression ultrasound result was associated with a low risk of venous thromboembolism during three-month follow-up. Although the authors acknowledged a number of limitations with the included studies, this was generally a well-conducted review and their conclusions appear to be reliable.
To determine the risk of venous thromboembolism after withholding anti-coagulant therapy in patients with suspected lower extremity deep vein thrombosis (DVT) following a single negative whole-leg compression ultrasound result.
MEDLINE, EMBASE, CINAHL, LILACS, the Cochrane Library, and Health Technology Assessments databases were searched for articles published between January 1970 and November 2009, without language restrictions. Search terms were reported. In addition, Internet searches were conducted, including Google, Google Scholar, ClinicalTrials.gov, meeting abstracts and conference proceedings. Reference lists of included studies were manually searched. Experts in the field were contacted for any additional studies.
Randomised controlled trials (RCTs) or prospective cohort studies of patients with suspected lower extremity deep vein thrombosis (DVT), evaluated with a whole-leg compression ultrasound result, who were followed up for at least 90 days and did not receive anticoagulant therapy, were eligible for inclusion. Eligible studies were also required to report objective confirmation of venous thromboembolism during follow-up. Studies of asymptomatic patients or trials using only limited (above the knee) compression ultrasound evaluation of the lower limbs, were excluded.
The primary outcome was venous thromboembolism events and venous thromboembolism-attributable death. Methods for objective confirmation of suspected venous thromboembolism included repeat compression ultrasound or venography for suspected DVT and computer tomography pulmonary angiogram, ventilation perfusion scan, or conventional pulmonary angiogram for suspected pulmonary embolism.
Included studies were of ambulatory patients and inpatients, with mean ages ranging between 53 and 67 years. Included studies had a higher proportion of females (ranging between 59.2% and 69.9%). Where reported, between 6.5 and 12% of patients had experienced prior venous thromboembolism; between 4.4 and 29.9% of patients reported active cancer. The majority of included studies did not report baseline risk, but (where reported) this ranged from 30.4 to 42.0% (low risk), between 48.0 and 56.4% (moderate risk), and between 10.1 and 21.2% (high risk). Other baseline characteristics were reported in the review. The majority of studies followed up patients using telephone interviews or by reviewing medical records.
Two reviewers independently screened publications for inclusion and disagreements were resolved by a third reviewer.
Assessment of study quality
Two reviewers independently assessed the quality of the studies using the Newcastle-Ottawa Quality Assessment Scale for Cohort Studies, including assessment on the representativeness of cohorts, comparability of cohorts, outcome assessment method, duration and adequacy of follow-up. Study design, method of enrolment, diagnostic criteria for presence of deep vein thrombosis, and funding source were also assessed. An overall quality score was not calculated for the individual studies. The authors did not state how disagreements were resolved.
Two reviewers independently extracted the number of venous thromboembolism events during follow-up and possible venous thromboembolism-related deaths to calculate percentage incidence rates and their 95% confidence intervals (CIs). Authors were contacted for individual patient data and, where necessary, any additional data. The authors did not state how disagreements were resolved.
Methods of synthesis
A random-effects model was used to pool incidence rates and their 95% confidence intervals, weighted by the inverse of the variance. Subgroup analyses were conducted on studies providing individual patient data, grouped by pre-test probability assessment (high risk, moderate risk, low risk). Odds ratios (ORs) with their 95% confidence intervals were estimated using the probit regression model, with adjustments for clustering effects, to compare differences between risk groups (moderate risk versus low risk, and high risk versus low risk).
Statistical heterogeneity was assessed using the X2 and I2. Publication bias was assessed using a funnel plot and the Egger test.
Results of the review
Seven studies, six prospective cohort studies and one RCT (n=4,731 patients; range 375 to 1,243) were included in the analysis. The authors reported the quality of included studies to be generally high: all studies reported on quality assessment criteria, with the exception of funding source, which was not reported in three cohort studies. Between 0 and 1.8% of patients were lost to follow-up.
Objectively confirmed venous thromboembolism and venous thromboembolism-attributable death occurred in 34 (0.7%) patients, indicating low risk of venous thromboembolism during three-month follow-up, with a pooled incidence rate of 0.57% (95% CI 0.25 to 0.89, seven studies). There was no evidence of statistical heterogeneity.
The risk of venous thromboembolism remained low for patients based on pre-test probability (two studies) at: 0.29% (95% CI 0 to 0.70) for the low risk group; 0.82% (95% CI 0 to 1.83) for the moderate group; and 2.49% (95% CI 0 to 7.11) for the high risk group. There were statistically significant differences between risk groups, with greater risk among patients with moderate pre-test risk (OR 3.80, 95% CI 1.10 to 8.61) or high pre-test risk (OR 9.13, 95% CI 4.64 to 17.96).
There was evidence of publication bias using the funnel plot and Egger test (p=0.05).
Withholding anticoagulant therapy in patients with suspected deep vein thrombosis following a single negative whole-let compression ultrasound result was associated with a low risk of venous thromboembolism during three-month follow-up.
The review question was clear and was supported by appropriate inclusion criteria. A thorough search of the literature was undertaken using various appropriate sources. Searches were not restricted by language and unpublished data were sought, which reduced the possibility of missing potentially relevant data. However, there was evidence of publication bias, but the authors suggested that this may have been a result of small sample size. The authors used appropriate methods to assess the quality of cohort studies and these were deemed to be of generally high quality. The authors undertook each stage of the review in duplicate, reducing the potential for reviewer error and bias.
Appropriate methods appear to have been used to synthesise the data and assess statistical heterogeneity. The authors acknowledged a number of limitations with the included studies, such as small sample size, wide 95% confidence intervals, no long-term outcomes, and the variability among whole-leg compression ultrasound techniques, which may limit the validity and generalisability of their findings. This was generally a well-conducted review and the authors' conclusions appear to be reliable.
One author disclosed consultation or speaker's fees from AGEN Biomedical, Janssen-Ortho, Boehringer-Ingelheim, Sanofit-Aventis, AstraZeneca, Pfizer, and Leo Pharma.
Implications of the review for practice and research
Practice: The authors did not state any implications for practice.
Research: The authors stated that further research is needed in patients with moderate or high pre-test risk for venous thromboembolism, and that studies are needed that prospectively combine whole-leg compression ultrasound with D-dimer testing and formal pre-test risk assessment. They also recommended evaluating the cost-effectiveness of whole-leg compression ultrasound.
Johnson SA, Stevens SM, Woller SC, Lake E, Donadini M, Cheng J, Labarere J, Douketis JD. Risk of deep vein thrombosis following a single negative whole-leg compression ultrasound: a systematic review and meta-analysis. JAMA 2010; 303(5): 438-445
Subject indexing assigned by NLM
Aged; Anticoagulants /therapeutic use; Cohort Studies; Female; Humans; Male; Middle Aged; Pressure; Randomized Controlled Trials as Topic; Risk; Ultrasonography /methods; Venous Thrombosis /drug therapy /epidemiology /ultrasonography
Database entry date
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.