I Efficacy And Safety Of Over-the-counter NSAIDs For Primary Dysmenorrhoea: A Network Meta-analysis

Objective The purpose of this study was to conduct a network Meta-analysis to assess theefficacy and safety of the six over-the-counter non-steroidal anti-inflammatory drugs(NSAIDs) in patients with primary dysmenorrhea.Methods According to the literature search strategy, the relative randomized controlledtrials(RCTs) of the six NSAIDs were searched in PubMed、Embase、Cochrane Library、Webof Science、CINAHL、CNKI、VIP、CBM、CSCD and Wanfang Database from the date ofthe establishment of the databases to December2013, the related references and website werealso traced. The RCTs were screened according to the predefined inclusion and exclusioncriteria and the related items were used to assess the quality of RCTs. The data were analyzedby Revman，WinBugs and R software.Results①A total of1116related articles were found and22were finally included. In thesestudies,36.4%RCTs reported the details of their methods of randomization,9.1%describedallocation concealment,95.5%reported blinding of participants and personnel,54.5%tookblinding of outcome assessment,40.9%reported in complete outcome data,72.7%had noselective reporting and22.7%had no bias from other sources of bias.②The result ofMeta-analysis showed that all the six included NSAIDs were significantly more effectivecompared with the placebo [OR=3.70，95%C（I2.59，5.27），P<0.00001]. Naproxen[OR=3.75，95%CI（2.49，5.65），P<0.00001], ibuprofen[OR=3.67，95%CI（1.13，11.99），P=0.03],diclofenac potassium [OR=15.24，95%C（I1.23，189.53），P=0.03], aspirin[OR=1.80，95%CI（1.11，2.91），P=0.02]and ketoprofen[OR=8.21，95%C（I3.78,17.83），P<0.00001]were moreeffective than placebo. The Efficacy of Acetaminophen [OR=1.94，95%CI（0.73,5.18），P=0.19] in relieving pain was still unclear. Except Naproxen was more effective than aspirin,there was no statistically difference between ibuprofen and acetaminophen, acetaminophenand aspirin, ketoprofen and naproxen. In the aspect of safety, no significant difference wasobserved in all the six included NSAIDs compared with the placebo [OR=0.83，95%C（I0.67， 1.04，P=0.11]. No statistical difference was showed in naproxen[OR=0.88，95%CI（0.66，1.88），P=0.40], ibuprofen[OR=0.71，95%C（I0.33，1.50），P=0.37], acetaminophen[OR=0.77，95%C（I0.42，1.43），P=0.41], diclofenac potassium[OR=2.17，95%C（I0.61，7.66），P=0.23],aspirin[OR=0.81，95%CI（0.43，1.53），P=0.52]and ketoprofen[OR=0.33，95%CI（0.10，1.02），P=0.05] compared to the placebo. There was no statistically difference betweenacetaminophen and aspirin, acetaminophen and ibuprofen, naproxen and ibuprofen, naproxenand acetaminophen, diclofenac potassium and ibuprofen, ketoprofen and naproxen in theaspect of safety.③The result of Network Meta-analysis showed that naproxen[OR=0.2196，95%CI（0.0976，0.4106）], ibuprofen[OR=0.2272，95%CI（0.0633，0.5777）], diclofenacpotassium [OR=0.0674，95%C（I0.0129，0.1901）]and ketoprofen[OR=0.1989，95%C（I0451，0.5670）]were more effective than the placebo. Compared with aspirin, the efficacy ofnaproxen [OR=0.3714，95%CI（0.0838，0.9561）]and diclofenac potassium[OR=0.1161，95%CI（0.0151，0.4074）] was statistically more obvious, and the efficacy among remainingNASIDs was showed no significant difference. No statistical difference was found innaproxen[OR=1.1860，95%CI（0.8342，1.5619）], ibuprofen[OR=0.71，95%CI（0.33，1.50），P=0.37[OR=1.1860，95%CI（0.8342，1.5619）], acetaminophen[OR=1.2518，95%CI（0.7228，2.0249）], diclofenac potassium[OR=0.5007，95%CI（0.1061，1.4018）],aspirin[OR=1.3998，95%CI（0.6489，2.7368）]and ketoprofen[OR=2.4079，95%CI（0.6726，6.0926）] compared to the placebo for safety. Compared to diclofenac potassium, it is acceptedthat ibuprofen[OR=4.7878，95%CI（1.1214，13.7577）] had low probability of causing sideeffects, and the incidence of adverse reaction among the remaining NASIDs was showed nosignificant difference.Conclusion①Combined the result of conventional and network Meta-analysis, naproxen,ibuprofen, diclofenac potassium and ketoprofen were more effective in relieving pain inprimary dysmenorrhea, and the efficacy of aspirin and acetaminophen was still unclearcompared to the placebo.②All the six included NSAIDs were safety in relieving pain inprimary dysmenorrhea. Except that diclofenac potassium had more probability of causing sideeffects than ibuprofen to some extent, the incidence of adverse reaction among the remainingNASIDs was showed no significant difference.