Non-steroidal anti-inflammatory drugs (NSAIDs) for cancer pain

Clinical bottom line: NSAIDs are effective in relieving cancer pain. Single doses of an NSAID are effective at six hours, and multiple doses of NSAID are effective at ten to 14 days. This review did not attempt to compare the efficacy of different NSAIDs. For single doses, NSAIDs appear to be as effective as intramuscular morphine 5 mg to 10 mg, weak opioids or opioid combinations and the same NSAID given at higher doses. For multiple doses, there was no difference between NSAIDs and weak opioids or opioid combinations. No other comparisons were possible.

Upper gastrointestinal upset, dizziness and drowsiness were the most common adverse effects for single and multiple dose trials. In multiple dose trials, NSAIDs were associated with fewer adverse effects than intramuscular morphine 5 mg to 10 mg.

NSAIDs are widely used in the treatment of cancer-related pain. World Health Organisation guidelines recommend NSAIDs as the sole treatment of mild to moderate cancer pain, and in combination with opioids for moderate to severe pain.

Systematic review

Eisenberg E, Berkey CS, Carr DB, Mosteller F, Chalmers TC. Efficacy and safety of nonsteroidal antiinflammatory drugs for cancer pain: a meta-analysis. J Clin Oncol. 1994; 12: 2756-65.

Date review completed: 1992
Number of trials included: 25
Number of patients: 1545
Control group: placebo and other active
Main outcomes: pain intensity, pain relief, adverse effects

Inclusion criteria were randomised, double blind, controlled trials of non-steroidal anti-inflammatory drugs (NSAIDs) with or without combination opioid for cancer pain; full journal publication; English language.

Reviewers extracted data on dosing regimens, adverse effects and pain. Pain data were patient ratings of pain intensity or pain relief scores using visual analogue or other quantitative scales. From this, a peak pain intensity or pain relief difference was calculated and a summed pain intensity difference (SPID) or total pain relief (TOTPAR). These were converted to percentages (percent of maximum). Means were weighted by sample size.

Comparison groups were placebo, different dose of NSAID including recommended versus supramaximal dose, weak opioid/weak opioid combination and morphine. Weak opioid was defined as those doses/drugs appropriate for treating mild to moderate pain.

Findings

Most trials reported on various types of cancer pain or did not specify type of cancer.

Single dose trials - six hour outcome

Placebo versus NSAID comparisons: there were between eleven and 14 placebo comparisons for each measure, looking at up to six different NSAIDs. Doses were not stated. NSAIDs were superior to placebo at six hours on each of the four pain measures (percentage values of TOTPAR or SPID and of peak pain relief or peak pain intensity difference) (p<0.05). Percentages for these measures ranged from 31% to 60% for NSAIDs compared with 15% to 36% for placebo).

Weak opioid/weak opioid combination versus NSAID comparisons: there were eight to nine comparisons for two measures (peak pain intensity and SPID) looking at a range of drugs/doses. There were no statistical differences between groups on either measure.

Intramuscular morphine 5 to 10 mg versus NSAID comparisons: there were four to five comparisons for two measures (SPID and TOTPAR). There were no statistical differences between groups on either measure.

Other comparisons: no significant differences were found comparing aspirin with other NSAIDs or comparing different doses of the same NSAID, including recommended dose with supramaximal dose.

Multiple dose trials - ten to 14 day outcomes

Only weak opioid/weak opioid combination versus NSAID comparisons were possible. There were two comparisons: one of ketorolac 10 mg versus acetaminophen 600 mg + codeine 60 mg and one of ketorolac 10 mg versus pentazocine 50 mg. Neither showed a significant difference on peak pain relief or TOTPAR.

Adverse effects

Common adverse effects for single and multiple dose trials included upper gastrointestinal upset, dizziness and drowsiness. For single dose trials, when all adverse effects were pooled, there were no significant differences between NSAIDs and placebo or NSAIDs and intramuscular morphine 5 mg to 10 mg. However, there were significantly more adverse effects with weak opioid or opioid combinations than with NSAIDs (20 versus 14 episodes per 100 patients).