Navigating the Efficacy and Safety Landscape of Duloxetine in Osteoarthritis Management: A Meta-Analysis of Randomized Controlled Trials.

Abstract

To evaluate the efficacy and safety of duloxetine in treatment of osteoarthritis compared with placebo. We conducted a comprehensive search for pertinent randomized controlled trials (RCTs) across multiple databases, including PubMed, Embase, Scopus, Cochrane Central Register of Controlled Trials, ClinicalTrials.gov and Google scholar. These selected RCTs aimed to assess the effectiveness and safety of duloxetine versus a placebo in the management of osteoarthritis (OA). We computed the mean difference (MD) for continuous outcome measures and calculated the risk ratio (RR) for dichotomous outcome measures. Data from 10 RCT including 2294 participants comparing duloxetine and placebo were pooled. Duloxetine showed improvement in BPI-S (Brief Pain Inventory-Severity)MD =  - 0.66; 95% CI, [- 0.75, - 0.57], BPI-I (Brief Pain Inventory-Interference)MD =  - 0.59; 95% CI [- 0.66, - 0.52], WOMAC (Western Ontario and McMaster Universities Arthritis Index) pain sub scale scores SMD =  - 2.56; 95% CI, [- 4.24, - 0.89] <i>P</i> = 0.003; physical function MD =  - 4.59; 95% CI, [- 5.83, - 3.34] <i>P</i> < 0.00001; SF 36 (Short Form) Physical function MD = 1.56; 95% CI, [0.14, 2.97] <i>P</i> = 0.03 and patients global impressions scores SMD =  - 0.47; 95% CI, [- 0.62, - 0.32]. Duloxetine resulted in higher number of treatment emergent adverse events RR = 1.34; 95% CI, [1.12,1.61] <i>P</i> = 0.002; and discontinuations RR = 2.54; 95%, CI, [1.89, 3.42] <i>P</i> < 0.00001. However duloxetine was no different from placebo in improving stiffness and in terms of incidence of serious adverse effects. Duloxetine demonstrates effectiveness in addressing chronic pain and mitigating the decline in physical function associated with knee osteoarthritis (OA), while maintaining a reasonable level of tolerable adverse events. However, it does not offer a distinct advantage in alleviating joint stiffness. The online version contains supplementary material available at 10.1007/s43465-025-01372-y.