Objective: To evaluate the efficacy and safety of paracetamol as an analgesic drug in a range of pain conditions.
STUDY DESIGN: Systematic review of the analgesic effect of paracetamol in randomized, placebo-controlled trials. Conduct of systematic reviews was assessed using AMSTAR-2; confidence in effect estimates (quality of evidence) was assessed according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) criteria.
Data sources: MEDLINE, EMBASE, PsycINFO, Cochrane Database of Systematic Reviews; systematic reviews published 1 January 2010 to 30 April 2020.
Data synthesis: We extracted pain and adverse event outcomes from 36 systematic reviews evaluating the efficacy of paracetamol in 44 pain situations. Persistent pain outcomes were expressed as mean differences (MDs; standardized 0-10 scale); dichotomous outcomes were expressed as risk ratios (RR). There is high-quality evidence that paracetamol is effective in patients with knee or hip osteoarthritis (MD, -0.3 points; 95% CI, -0.6 to -0.1 points) and after craniotomy (MD, -0.8 points; 95% CI, -1.4 to -0.2 points); moderate-quality evidence for its effect on tension-type headache (2-hour pain free: RR, 1.3; 95% CI, 1.1-1.4) and perineal pain shortly after delivery (pain relief 50 % of patients: RR, 2.4; 95% CI, 1.5-3.8). There was high-quality evidence that paracetamol was ineffective in relieving acute low back pain (MD, 0.2 points; 95% CI, -0.1 to 0.4 points). Evidence for efficacy in other conditions was of low or very low quality. Adverse event rates were generally similar in those who received placebo or paracetamol, except that transient increases in blood liver enzyme levels during repeated paracetamol administration were more frequent in patients with spinal pain (RR, 3.8; 95% CI, 1.9-7.4).
CONCLUSIONS: For most cases, the evidence on the effectiveness of paracetamol is insufficient to draw firm conclusions. Evidence for its efficacy in four settings was moderate to strong, and there was strong evidence that paracetamol was ineffective in reducing acute low back pain. Investigations to evaluate more typical dosing regimens are needed.