Raw data extraction form.

Introduction
Osteoarthritis (OA) as a degenerative disease, has seen a continuous rise in incidence and prevalence globally since 1990, imposing a significant disease burden. NSAIDs (Nonsteroidal anti-inflammatory drugs) as symptomatic medications for OA treatment, hold an indispensable position in clinical practice.
Objective
To evaluate the efficacy and safety of different NSAIDs in the treatment of OA through Bayesian Network Meta-Analysis (NMA).
Methods
Randomized controlled trials (RCTs) on NSAIDs for OA treatment were retrieved from PubMed, Web of Science, Embase, and the Cochrane Library databases. The search timeframe was from the inception of each database up to June 1, 2024. Outcome indicators for NMA were all conducted using a random-effects model. MetaInsight and Stata 14.0 software were used in R for calculations and plotting of NMA. Measurement data were represented by mean difference (MD), and count data by odds ratio (OR); a 95% confidence interval (CI) was also calculated for each effect size.
Results
This study included 31 studies, involving 68,539 patients with knee osteoarthritis (KOA) and 16 interventions. NMA results showed that compared to the placebo, Tiaprofenic reduced the VAS score (MD =  -0.16, 95% CI: (-0.46 to 0.14), P >  0.05), albeit without significant difference; meanwhile, Diclofenac reduced the total WOMAC score in KOA patients (MD =  -0.41, 95% CI: -1.05 to 0.24, P >  0.05). Compared to the placebo, Etoricoxib was the best medication for improving the WOMAC pain subscale score (MD =  -0.44; 95% CI: -0.61 to -0.26); Naproxen significantly improved the WOMAC Function score in KOA patients after administration (MD = -0.43; 95% CI: -0.82 to -0.04); Diclofenac intervention significantly reduced the WOMAC Stiffness score in KOA patients (MD =  -0.40; 95% CI: -0.67 to -0.13). In terms of adverse event rates, compared to the placebo, the use of Etoricoxib significantly increased the incidence of cardiovascular adverse events (OR =  0.56, 95% CI: 0.32–0.99); Ketoprofen had fewer gastrointestinal adverse events during the medication process (OR =  0.09, 95% CI: 0.04–0.20); Licofelone had a lower rate of other adverse events during the medication process (OR =  0.80, 95% CI: 0.45–1.40, P >  0.05). Therefore, the results indicate that Etoricoxib, Tiaprofenic, Naproxen, Diclofenac, and Ketoprofen have better clinical efficacy and safety.
Conclusion
Compared to other NSAIDs, Etoricoxib, Tiaprofenic, Naproxen, and Diclofenac play a more effective role in improving clinical symptoms of OA; in terms of reducing the incidence of adverse events, Ketoprofen has a lower chance of adverse events. However, the possibility of these results still needs further clinical and basic research for verification.

Introduction
骨关节炎（Osteoarthritis, OA）作为一种退行性疾病，自1990年以来全球发病率与患病率持续攀升，给社会带来了沉重的疾病负担。非甾体抗炎药（Nonsteroidal anti-inflammatory drugs, NSAIDs）作为OA治疗的对症用药，在临床实践中占据不可替代的地位。

Objective
本研究旨在通过贝叶斯网络荟萃分析（Bayesian Network Meta-Analysis, NMA），评估不同非甾体抗炎药治疗骨关节炎的有效性与安全性。

Methods
研究从PubMed、Web of Science、Embase及Cochrane Library数据库中检索用于OA治疗的非甾体抗炎药相关随机对照试验（Randomized controlled trials, RCTs），检索时限设置为各数据库建库起至2024年6月1日。本研究的网络荟萃分析均采用随机效应模型进行分析。使用MetaInsight及Stata 14.0软件在R环境中完成网络荟萃分析的计算与绘图。计量资料以均数差（mean difference, MD）表示，计数资料以比值比（odds ratio, OR）表示，并对每一项效应量计算95%置信区间（confidence interval, CI）。

Results
本研究共纳入31项研究，涉及68539例膝骨关节炎（knee osteoarthritis, KOA）患者，涵盖16种干预方案。网络荟萃分析结果显示：与安慰剂组相比，替诺昔康（Tiaprofenic）可降低视觉模拟评分（Visual Analogue Scale, VAS）（MD = -0.16，95%CI：-0.46~0.14，P>0.05），但差异无统计学意义；双氯芬酸（Diclofenac）可降低膝骨关节炎患者的总WOMAC评分（MD = -0.41，95%CI：-1.05~0.24，P>0.05）。与安慰剂组相比，依托考昔（Etoricoxib）是改善WOMAC疼痛维度评分的最优药物（MD = -0.44；95%CI：-0.61~-0.26）；萘普生（Naproxen）给药后可显著改善膝骨关节炎患者的WOMAC功能维度评分（MD = -0.43；95%CI：-0.82~-0.04）；双氯芬酸干预可显著降低膝骨关节炎患者的WOMAC僵硬维度评分（MD = -0.40；95%CI：-0.67~-0.13）。在不良事件发生率方面，与安慰剂组相比，依托考昔可显著增加心血管不良事件的发生风险（OR = 0.56，95%CI：0.32~0.99）；酮洛芬（Ketoprofen）用药过程中的胃肠道不良事件发生率更低（OR = 0.09，95%CI：0.04~0.20）；利可洛芬（Licofelone）用药过程中其他不良事件的发生率更低（OR = 0.80，95%CI：0.45~1.40，P>0.05）。因此，研究结果表明依托考昔、替诺昔康、萘普生、双氯芬酸及酮洛芬具有更优的临床有效性与安全性。

Conclusion
与其他非甾体抗炎药相比，依托考昔、替诺昔康、萘普生及双氯芬酸在改善OA临床症状方面表现更为突出；在降低不良事件发生率方面，酮洛芬的不良事件发生风险更低。但上述研究结论仍需进一步的临床与基础研究加以验证。