Systematic review and meta-analysis of randomized trials versus placebo

Context. Paracetamol is the largest-selling drug in France, while its mechanism of action as an analgesic is still imperfectly known, and its analgesic effect has been poorly studied in primary care. Objective. To quantify, the effect of paracetamol on acute pain in adult primary care patients, and to assess the influence of the administered dose and the clinical context on this effect. Methods. For this systematic review including a meta-analysis, we searched 3 databases (Central, EMbase and Medline) in November 2016. All randomized clinical trials (RCTs), parallel and double-blind, assessing the effect of a single dose oral paracetamol versus placebo on acute pain, within the 6 h after drug intake, were included. Considering the diversity of outcomes reported in RCTs, 12 criteria were assessed in this meta-analysis. Results. Twenty-one RCTs were included in the review, including 19 in the meta-analysis ( 4359 patients). Their methodological quality was frequently assessed as "uncertain", due to the lack of information available in study reports. Two primary care situations were considered: headache (migraine and tension headaches) and upper respiratory tract infections. Paracetamol was significantly more effective than placebo for all endpoints, but heterogeneity between studies was high (I2 > 50%). Subgroup analyses did not identify a dose effect or an influence of the clinical context on paracetamol efficacy. Factors influencing the placebo effect could be responsible for the heterogeneity observed. Conclusion. While there is evidence of paracetamol analgesic efficacy in primary care, the poor methodological quality and the lack of standardization of outcome measures in RCTs limit interpretation of results. The standardization of RCTs on pain and a better understanding of the factors influencing the placebo effect seem to be necessary.