To investigate the efficacy and safety of combination analgesic products containing low-dose codeine (up to 30 mg/dose) for pain.
Electronic databases were used to identify eligible placebo-controlled, randomized controlled trials (RCTs). Two authors extracted data and assessed the risk of bias. Data were pooled using a random-effects model with the strength of evidence assessed using Grading of Recommendations Assessment, Development and Evaluation. The primary outcome was immediate pain relief (3 hours post administration) on a 0 to 100 pain scale.
Ten RCTs were eligible. There is low-quality evidence (4 RCTs, n=211 participants) that a single dose of a combination analgesic product (with an nonsteroidal anti-inflammatory) containing low-dose codeine (15 to 30 mg) provides small pain relief for acute dental pain (mean difference [MD], −12.7; 95% confidence interval [CI], −18.5 to −6.9) and moderate-quality evidence (1 RCT, n=93) of small pain relief for post-episiotomy pain and orthopedic surgery pain (MD,, −10.0; 95% CI, −19.0 to −1.0 and MD, −11.0; 95% CI, −20.7 to −1.3), respectively. There is low-quality evidence (1 RCT, n=80) that a multiple-dose regimen provides small pain relief for acute pain following photorefractive keratectomy (MD, −16.0; 95% CI, −24.5 to −7.5) and moderate-quality evidence of moderate pain relief for certain chronic pain conditions: for hip osteoarthritis (MD, −19.0; 95% CI, −31.2 to −6.8) and for temporomandibular joint pain (MD, −26.0; 95% CI, −44.5 to −7.5). Two studies reported a higher incidence of drowsiness in the treatment group compared with the placebo group (relative risk, 8.50; 95% CI, 1.96, 36.8 and 19.3; 95% CI, 1.2-306.5, respectively).
There is low to moderate level evidence that combination analgesic products containing low-dose codeine provide small to moderate pain relief for acute and chronic pain conditions in the immediate short term with limited trial data on use beyond 24 hours. Further research examining regular use of these medicines is needed with more emphasis on measuring potential harmful effects.
*Faculty of Medicine and Health, School of Public Health
‡Faculty of Medicine and Health, School of Pharmacy, University of Sydney
†Institute for Musculoskeletal Health, Sydney Local Health District, Sydney, NSW, Australia
Supported by the Therapeutic Goods Administration, Canberra, Australia.
The authors declare no conflict of interest.
Reprints: Christina Abdel Shaheed, PhD, Level 10 North, King George V Building, Royal Prince Alfred Hospital, P.O. Box M 179, Missenden Road, Sydney, NSW 2050, Australia (e-mail: email@example.com).
Received August 31, 2018
Received in revised form June 23, 2019
Accepted June 30, 2019