|
Effect of statins on venous thromboembolic events: a meta-analysis of published and unpublished evidence from randomised controlled trials |
Rahimi K, Bhala N, Kamphuisen P, Emberson J, Biere-Rafi S, Krane V, Robertson M, Wikstrand J, McMurray J |
|
|
CRD summary The evidence did not support the suggestion that statins (or high doses of statins) substantially reduced the risk of venous thromboembolism, but a modest, clinically relevant, reduction in risk, could not be ruled out. There was some uncertainty around the methods, but those that were reported were good. The conclusions reflect the evidence and are likely to be reliable. Authors' objectives To assess the effects of statins on venous thromboembolic events. Searching MEDLINE (from 1966), EMBASE (from 1985), and Cochrane Central Register of Controlled Trials (CENTRAL) were searched up to March 2012, without language restrictions. Search terms were reported. Reference lists of relevant articles were manually screened. Study selection Eligible for inclusion were randomised controlled trials (RCTs) comparing the effects of different doses of statins (high versus low intensity) or the effects of statins versus a control (placebo or usual care) on venous thromboembolic events, including deep vein thrombosis, and fatal or non-fatal pulmonary embolus. Trials had to include at least 100 participants and have at least six months of follow-up. There were no restrictions on the participants and the study outcomes. Trials were excluded if they reported no events for both treatment arms or where the authors did not respond to requests for data. A third of included trials were multinational, and most of the others were conducted in Europe or North America. The included trials were published between 1998 and 2010. The mean age of patients ranged from 48 to 75 years, and most of them were male. Some trials were in primary prevention settings, and patients had conditions such as diabetes, hypertension, ischaemic heart failure, Alzheimer's disease, or acute coronary syndrome. The difference in low-density lipoprotein cholesterol, between the two treatment groups, at baseline, ranged from 0.30 to 1.79 millimoles per litre. The trials compared atorvastatin (10 to 80mg), lovastatin (20 to 40mg), pravastatin (10 to 40mg), rosuvastatin (10 to 40mg), or simvastatin (40 to 80mg) versus placebo, usual care, no treatment, different statins or different doses of the same statin. Most of the included trials reported venous thromboembolic events as adverse outcomes. Two reviewers independently screened studies for inclusion. Disagreements were referred to a third reviewer. Assessment of study quality The risk of bias in the trials was assessed using the Cochrane risk of bias tool, which included criteria on sequence generation and allocation concealment, blinding, attrition, and selective outcome reporting. Each criterion was rated as low, unclear or high risk of bias, and the overall risk of bias was rated for each trial. The authors stated that they followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines, but they did not state how many reviewers performed the quality assessment. Data extraction The data on deep vein thromboses and pulmonary embolisms were extracted, on an intention-to-treat basis, to calculate odds ratios and their 99% confidence intervals. Authors were contacted for missing data, where necessary. The authors followed PRISMA guidelines, but did not state how many reviewers extracted the data. Methods of synthesis A random-effects model (Peto's method) was used to pool the odds ratios and calculate 95% confidence intervals. Statistical heterogeneity was assessed using Χ² and Ι². Subgroup analyses were undertaken by type of comparator and outcome (pulmonary embolism or deep vein thrombosis), for trials that specifically excluded patients with a known history of vascular disease (primary prevention trials) compared with other trials, and by type of statin. Additional analyses were undertaken by removing the trial that motivated the review (considered a hypothesis generating trial), and by weighting trials based on one-year differences in low-density lipoprotein cholesterol, between the two treatment groups. Results of the review Twenty-two RCTs, with 105,759 participants, were found that compared statin with control, and seven RCTs, with 40,594 participants (calculated as 40,599), were found that compared an intensive versus a standard dose of statin. Sample sizes ranged from 108 to 20,536 patients. All RCTs were deemed to be at a low risk of bias. There were no statistically significant differences in venous thromboembolic event rates, when comparing statins with control (22 RCTs), but inconsistency between trials could not be ruled out (Ι²=0, 95% CI 0 to 43%). There were no statistically significant differences in venous thromboembolic events, when comparing higher intensity versus standard intensity statin therapy (seven RCTs), but inconsistency between trials could not be ruled out (Ι²=0, 95% CI 0 to 61%). Sensitivity analyses did not significantly alter the findings. Subgroup analyses assessing trials of patients with no previous history of cardiovascular disease, indicated that statin therapy significantly reduced the risk of venous thromboembolism by 38% (OR 0.62, 95% CI 0.41 to 0.94; seven RCTs). No other subgroup analyses significantly altered the main findings. Authors' conclusions The evidence did not support the suggestion that statins (or high doses of statins) substantially reduced the risk of venous thromboembolic events, but a modest, but clinically relevant, reduction in risk, for some or all patients, could not be ruled out. CRD commentary The review question was clear and was supported by appropriate inclusion criteria. A satisfactory search of the literature was undertaken, without language and publication restrictions, reducing the potential for missed data. A number of trials that had no events for both treatment arms were not included in the meta-analyses, but it is unlikely that their inclusion would have altered the non-significant findings. The authors suggested that the exclusion of 16 trials, for which missing data could not be retrieved, should not have biased the results, but this cannot be confirmed. Study selection was performed in duplicate, but it was not clear whether this was true for data extraction and quality assessment, so the possibility of reviewer error and bias cannot be ruled out. The risk of bias in the trials was assessed using appropriate criteria and was found to be low. There were considerable differences in the participant characteristics, but appropriate methods seem to have been used to pool the trial data. Appropriate methods were used to investigate statistical heterogeneity, and the authors appropriately recommended caution when interpreting the findings from the subgroup analyses. There was some uncertainty around the review methods, but those that were reported were good and there was a large amount of evidence. The authors' conclusions reflect the findings and are likely to be reliable. Implications of the review for practice and research Practice: The authors did not state any implications for practice. Research: The authors stated that further RCTs were needed to assess the effects of statins in subpopulations. Funding Funded by Pfizer, the UK Department of Health, and Diabetes UK. Bibliographic details Rahimi K, Bhala N, Kamphuisen P, Emberson J, Biere-Rafi S, Krane V, Robertson M, Wikstrand J, McMurray J. Effect of statins on venous thromboembolic events: a meta-analysis of published and unpublished evidence from randomised controlled trials. PLOS Medicine 2012; 9(9):e1001310 Indexing Status Subject indexing assigned by NLM MeSH Humans; Hydroxymethylglutaryl-CoA Reductase Inhibitors /therapeutic use; Randomized Controlled Trials as Topic; Thromboembolism /drug therapy AccessionNumber 12012046913 Date bibliographic record published 31/10/2012 Date abstract record published 15/01/2013 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. |
|
|
|