OBJECTIVES: To assess the quality of recent systematic reviews (SRs) of randomised controlled trials (RCTs) of blood pressure lowering drugs (BPLDs) published in high-impact journals, and to assess what proportion were used in hypertension guideline development.

METHODS: Medline and Epistemonikos databases were searched for SRs of RCTs of BPLDs. Studies were included if they were published in high impact journals (at least 20-impact factor) in the last 10 years. High-impact journals were identified from Thomson Reuters InCites Journal Citation Report-2017. Two authors independently performed critical appraisals using ‘risk of bias in systematic reviews (ROBIS) tool’, which is based on four domains (eligibility criteria, identification and selection of studies, data collection and study appraisal, synthesis and findings), and 21 signalling questions. Each domain and overall SR/MAs were rated “high”, “unclear”, and “low” risk of bias based on the rating given to each signalling question.

RESULTS: Among journals with more than 20 impact factor, five (Lancet, JAMA, Lancet oncology, Lancet neurology, and BMJ) published SRs of RCTs of BPLDs. Of the 35 citations identified from the above journals, 27 articles were included. Only six (22%) SRs had low risk of bias, five (19%) were unclear and the rest were found to have ‘high’ (n=16, [59%]) risk of bias. Most of the concerns were observed in the ‘identification and selection of studies’ (n=16, [59%]), followed by ‘data collection and study appraisal’ (n=15, [55%]), ‘synthesis and findings’ (n=7, [26%]), and ‘eligibility criteria’ (n=4, [14%]). Overall, nine of 27 SRs were considered for guideline development (six in ACC/AHA-2017, two in ESH-ESC-2013, five in Australian-2016, and one in Canadian-2018 guidelines).

CONCLUSIONS: More than half of included SRs are rated as having high risk of bias on the ROBIS toolFew SRs were used in clinical practice guideline development.