PURPOSE: To 1) conduct a thorough search of the literature for randomized controlled trials (RCTs) addressing the efficacy of garlic as an antihyperlipidemic agent, 2) critically appraise those studies, and 3) make a recommendation for practicing health care professionals. DATA SOURCES: Two independent reviewers extracted data from the articles identified from several data bases, using the previously tested Boyack and Lookinland Methodological Quality Index (MQI) as the standard. RESULTS: Six of ten studies found garlic to be effective. The average drop in total cholesterol was 24.8 mg/dL (9.9%), LDL 15.3 mg/dL (11.4%), and triglycerides 38 mg/dL (9.9%). The overall average MQI score was 39.6% (18%-70%). Major shortcomings of many of the RCTs included short duration, lack of power analysis and intention to treat analysis, as well as lack of control of diet as a confounding variable. CONCLUSION/IMPLICATIONS: The low methodological quality of the studies make it difficult to recommend garlic as an antihyperlipidemic agent. Until larger RCTs of longer duration, which correct the existing methodological flaws, are designed and carried out, it is best not to recommend garlic be used to treat mild to moderate hyperlipidemia.
PURPOSE: To 1) conduct a thorough search of the literature for randomized controlled trials (RCTs) addressing the efficacy of garlic as an antihyperlipidemic agent, 2) critically appraise those studies, and 3) make a recommendation for practicing health care professionals. DATA SOURCES: Two independent reviewers extracted data from the articles identified from several data bases, using the previously tested Boyack and Lookinland Methodological Quality Index (MQI) as the standard. RESULTS: Six of ten studies found garlic to be effective. The average drop in total cholesterol was 24.8 mg/dL (9.9%), LDL 15.3 mg/dL (11.4%), and triglycerides 38 mg/dL (9.9%). The overall average MQI score was 39.6% (18%-70%). Major shortcomings of many of the RCTs included short duration, lack of power analysis and intention to treat analysis, as well as lack of control of diet as a confounding variable. CONCLUSION/IMPLICATIONS: The low methodological quality of the studies make it difficult to recommend garlic as an antihyperlipidemic agent. Until larger RCTs of longer duration, which correct the existing methodological flaws, are designed and carried out, it is best not to recommend garlic be used to treat mild to moderate hyperlipidemia.