To examine the risk of non-steroidal anti-inflammatory drug-induced heart failure in patients with chronic kidney disease.
Embase, Medline, CENTRAL, Web of Science, and Google Scholar were searched for papers published in English between 1st January 1999 and 31st May 2020. Papers were included if some participants had chronic kidney disease, were exposed to non-steroidal anti-inflammatory drugs, and where heart failure was measured as an outcome. Papers were assessed for risk of bias using the Cochrane Risk of Bias 2 tool for randomised controlled trials, and ROBINS-I for observational studies.
A total of 2480 independent papers were retrieved. Following abstract screening, 165 full texts were reviewed to identify seven eligible papers: two randomised controlled trials, four cohort studies, and one case-control study. For chronic kidney disease (stage 3–5), relative risk for heart failure ranged from 0.3 to 1.9 with 95% confidence interval 0.04 to 15.1. Results were not pooled due to study heterogeneity. We attributed bias to heterogenous populations studied, probable confounding due to partially adjusted risk estimates, and heterogenous measurement of the heart failure outcome.
Overall, there are only a few studies to refute or support an increased risk of heart failure associated with taking non-steroidal anti-inflammatory drugs in patients with chronic kidney disease, and therefore no robust evidence was available.