P097 The risks of adverse events with analgesics in patients with inflammatory arthritis: a systematic review of observational studies

Abstract

Abstract Background/Aims Analgesics are widely prescribed in patients with inflammatory arthritis (IA) despite limited evidence of efficacy and uncertainties about potential harms. The assessment of harms is complex as some drug-related adverse events (AE), including myocardial infarctions, are part of the IA comorbidity spectrum. As trials have limited follow-up periods and often exclude patients with comorbidities, safe prescribing requires an understanding of analgesic harms in “real-world” settings. We consequently undertook the first systematic review of observational studies reporting risks of adverse events with analgesics in adults with rheumatoid arthritis (RA) and spondyloarthritis (SpA). Methods A pre-specified protocol was PROSPERO-registered (CRD42019154277). Systematic searches were conducted in Web of Science/Embase/Medline without language/date restriction on the 18/1/22. Due to marked methodological and statistical heterogeneity, studies were combined in a structured narrative synthesis. Results Studies: Of 6,444 citations screened, thirty-five studies met eligibility criteria, comprising 21 cohort (200 to 143,433 patients); 12 case-control (41 to 7,102 cases); 1 self-controlled case series; 1 case-crossover. Two evaluated opioids in RA, 23 oral NSAIDs in RA, 9 oral NSAIDs in SpA, and 1 NSAIDs and opioids in RA. None evaluated paracetamol/gabapentinoids. Fourteen were rated “good”, 19 “fair” and 1 “poor” quality. Adverse Event Risks: All studies of opioids reported significantly increased risks of fractures/serious infection. For NSAIDs renal, upper gastrointestinal (UGI), and cardiovascular (CV) events were most often reported. There was consistent evidence that NSAIDs significantly increased renal (3/3 studies) and UGI (3/4 studies) AEs. Findings with CV AEs varied: 7/18 studies showed significantly increased risks; 3/18 showed significantly reduced risks, and 3/18 showed significantly increased risks with some NSAIDs and reduced risks with others. There was no evidence NSAIDs increased risks of lower/functional gastrointestinal AEs, or fractures. With regards to mortality, one study showed a significantly reduced risk of death with NSAID use in SpA; the other showed a significantly reduced risk of death with NSAID use in inflammatory polyarthritis (but not RA). Differences Between RA And SpA: More studies reported a significantly reduced risk of CV events/fractures/deaths with NSAIDs in SpA (6/10) than RA (4/24). Differences Between IA and Non-IA Populations: Two of four studies found significant differences in AE risks in RA/SpA compared to OA/controls. One reported a significantly higher risk of myocardial infarction in SpA than OA and another a significantly lower risk of composite CV events in RA than controls. Conclusion Observational study data on risks of analgesic harms in IA are limited. All studies of opioids showed increased risks of harms. Studies of NSAIDs gave variable results, reflecting heterogeneity and potential confounding. Overall, oral NSAIDs increased renal and UGI AEs, with some evidence they reduced fractures/mortality in SpA. Further high-quality observational studies are needed to better characterise analgesic harms in IA. Disclosure C.R. White: None. O. Babatunde: None. J. Higgs: None. S. Muller: None. C.D. Mallen: None. S.L. Hider: None. I.C. Scott: None.