To evaluate the efficacy of non-steroidal anti-inflammatory drugs (NSAIDs), compared with placebo or analgesic agents, in the treatment of acute renal colic.

MEDLINE was searched from 1966 to 1992, and EMBASE from 1975 to 1992, using the keywords 'anti-inflammatory agent', 'colic', 'kidney disease' and 'ureteral disease'. The Family Medicine Library Index (1988 to 1991), the documentation services of pharmaceutical companies, and the library of the emergency department of the Centre Hospitalier de l'Universite Laval, were also consulted. References of retrieved articles were examined.

Study designs of evaluations included in the review

RCTs comparing a NSAID with a placebo or an analgesic agent in the emergency treatment of acute renal colic.

Specific interventions included in the review

Randomised controlled trials (RCT) comparing a NSAID with a placebo or an analgesic agent in the emergency treatment of acute renal colic, and which were written in either French or English.

Four trials compared idomethacin or diclofenac with placebo; 15 compared diclofenac, indomethacin or other NSAIDs with an analgesic agent; 1 trial used both diclofenac and indomethacin; and one article included 2 trials evaluating diclofenac. With the exception of 1 study, diclofenac was administered intramuscularly and indomethacin intravenously.

Participants included in the review

The authors do not provide details of the patients.

Outcomes assessed in the review

The number of patients who experienced pain relief 20 to 30 minutes after drug administration was assessed.

How were decisions on the relevance of primary studies made?

The studies were assessed independently by four evaluators using a detailed checklist, adapted from Chalmers et al. (see Other Publications of Related Interest no.1) and Sackett et al. (see Other Publications of Related Interest no.2), with final scoring by group consensus.

Validity was assessed by a 20-item checklist, adapted from Chalmers et al. (see Other Publications of Related Interest no.1) and Sackett et al. (see Other Publications of Related Interest no.2), with each item scored on a scale of 0 to 2. Elements evaluated were selection process, attrition of patients, description of randomisation, description of treatments, measurement of outcomes, statistics, group comparability and description of results. Study validity was independently assessed by four evaluators using a 20-item checklist; final assessment was by group consensus.

The data were extracted by four independent evaluators using a 20-item checklist; final assessment was by group consensus.

How were the studies combined?

The results of pain relief 20 to 30 minutes after drug administration were pooled using the Mantel-Haenszel method for 3 distinct group of studies: NSAIDs versus placebo, NSAIDs versus analgesic agents for partial pain relief, and NSAIDs versus analgesic agents for complete pain relief.

For each study, the relative risk (RR) was calculated and the 95% confidence interval (CI) of the RR used. This measure of treatment effect was calculated by dividing the proportion of patients relieved from pain in the treatment group (NSAID) by the proportion of patients relieved from pain in the control group (placebo or analgesic agent).

How were differences between studies investigated?

Differences between the studies were assessed using the Breslow-Day test for homogeneity, and by examining graphical display of the proportions of patients with pain relief.

The 19 articles presented 20 studies, mostly comparing parenteral diclofenac or indomethacin (18 out of 20) with placebo (n=4) or analgesic agents (n=16), most of which were narcotic agents. Sample size varied from 9 to 118 patients per study group, with an average of 44 plus or minus 28 (standard deviation) patients.

Statistically-significant heterogeneity was found between analgesic groups for both partial and complete pain relief, but was not found between the placebo groups.

The results of pain relief 20 to 30 minutes after drug administration were: NSAIDs versus placebo, RR 2.34, (95% CI: 1.79, 3.07); NSAIDs versus analgesic agents for partial pain relief, RR 1.07 (95% CI: 1.02, 1.12); and NSAIDs versus analgesic agents for complete pain relief, RR 1.19 (95% CI: 1.03, 1.37).

Sensitivity analysis: the NSAIDs were clinically and statistically superior to the analgesic agents for total pain relief (RR 1.52, 95% CI: 1.23, 1.89) when 2 studies, using lower doses of diclofenac (50 versus 75 mg) and having the lowest methodology quality scores, were excluded.

Parenteral NSAIDs are more effective than placebo, and as effective as analgesic agents, in the treatment of acute renal colic.

This is a well-written and methodologically-rigorous review. A comprehensive search strategy was employed, although unpublished studies were not sought for inclusion.

The authors acknowledge that the review may possibly suffer from publication bias since only published positive studies were included and unpublished negative trials were excluded. Thus, the small, significant positive overall treatment effects in the 'analgesic' studies might merely be a biased overestimation. In addition, as noted by the authors, findings should be interpreted with caution due to between-study heterogeneity.

1. Chalmers TC, Smith H, Blackburn B, Silverman B, Schroeder B, Reitman D, et al. A method for assessing the quality of a randomized control trial. Control Clin Trials 1981;2:31-49. 2. Sackett DL, Haynes RB, Tugwell P. Clinical epidemiology. A basic science for clinical medicine. Boston (MA): Little Brown; 1985.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.