NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Dolor RJ, Melloni C, Chatterjee R, et al. Treatment Strategies for Women With Coronary Artery Disease [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2012 Aug. (Comparative Effectiveness Reviews, No. 66.)

Cover of Treatment Strategies for Women With Coronary Artery Disease

Treatment Strategies for Women With Coronary Artery Disease [Internet].

Show details

Future Research

This comprehensive review of the comparative effectiveness of treatment modalities for women with CAD identified numerous gaps in evidence that would be suitable for future research and for improving the reporting of women findings of cardiovascular therapies in the published literature.

Studies With Sufficient Representation of Women

Sex subgroup analyses are often limited by the number of men or women in each treatment group to allow for adequate power to detect a statistically significant difference in outcome. While we were able to find RCTs that reported risk ratios in women, the enrollment numbers were insufficient to have adequate power to detect a difference, thus resulting in large confidence intervals that often crossed the null effect, with a potential type II error. To better understand the clinical outcomes of women treated by medical therapy or revascularization, trials should be either (1) women-only enrollment or (2) of large enough sample size with stratification of randomization by sex to allow for meaningful sex-based analyses. In order to assess sex differences in treatment modalities and their impact on clinical outcomes, a sufficient sample size is required in order to have adequate statistical power for subgroup analyses.

Patient-Level Meta-Analysis

Given the small representation of women in these RCTs, the heterogeneity of clinical outcomes (e.g., definition of composite outcome) and different measurement time points (e.g., 30 days, 6 weeks for short-term outcomes), we are aware that our group-level meta-analysis may be inadequate (when too few studies are available) to address the comparative effectiveness of medical therapy and revascularization. Therefore, patient-level analysis of trials comparing similar interventions for the same CAD presentation may be more appropriate for assessing the sex differences as well as for conducting subgroup analyses on demographic and clinical factors that influence treatment outcomes, or for evaluating safety concerns/harms of these treatment strategies. Subgroup analyses across trials can be done similarly to a previous AHRQ report on the comparative effectiveness of PCI and CABG, which included an addendum study that pooled individual patient data from 10 randomized trials to compare the effectiveness of CABG with PCI according to patients’ baseline clinical characteristics (e.g., age, diabetes, sex, individual cardiac risk factors, angioplasty versus bare metal stents).34,109,110

Reporting Sex by Treatment Results Separately

Our review excluded trials that looked for a sex effect yet failed to provide results of women and men by treatment arm. An example is a trial that did a multivariate analysis to assess factors that influenced clinical outcomes and included male (or female) sex in the model, with a finding that it was nonsignificant or significant. We did not contact the corresponding authors of the articles that did not report sex results separately. It would aid future comparisons of treatment modalities if study authors were to report the primary data for women and men separately either within the article itself or in an online supplementary appendix. The 2010 report by the Institute of Medicine on Women’s Health Research recommended that funding agencies ensure adequate participation of women and reporting of sex-specific analyses in health research.111

Reporting of Demographic and Clinical Factors That Influence Cardiovascular Outcomes

We found a few studies that conducted subgroup analyses of age, diabetes, and risk stratification in women populations. We did not find any data specific to women on race, socioeconomic, chronic kidney disease, angiographic-specific factors, or CABG-specific factors that were listed in KQ 2. Knowing the influence of these factors on cardiovascular outcomes is important for determining the proper treatment strategy and prognosis of women patients who present with various risk factors and comorbidities.

Reporting of Safety Concerns/Risks by Sex

Medical therapy can result in adverse drug reactions, and use of fibrinolytics can result in bleeding or intracranial hemorrhage. PCI can cause access site complications, radiation exposure, contrast-related anaphylaxis, bleeding, and stent thrombosis. CABG can result in wound infections, renal dysfunction, and bleeding. Most studies reported the bleeding risk of revascularization strategies but not the other safety concerns. Systematic reporting of adverse events in publications—in total and by sex—should continue to clarify which treatment modalities are safe for use in clinical practice.

To summarize, these evidence gaps could be addressed in various ways. First, more primary research with adequate representation of women for any of the three CAD clinical presentations could be conducted to achieve adequate statistical power for a sex-based analysis. Second, authors of the comparative trials that were excluded for not reporting sex-based results could be contacted to provide results of women and men by treatment arm, and the group-level meta-analysis could be repeated with a larger number of trials. Alternatively, these authors could be contacted to provide compatible (deidentified) datasets that could be combined for a patient-level analysis to assess the comparative effectiveness, modifiers of effectiveness, and risks of the various treatment strategies available. Finally, the use of observational cohorts from electronic health records could inform the real-world effectiveness of the treatment strategies chosen by clinicians and patients in a nonrandom fashion.

PubReader format: click here to try


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.7M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...