Analgesic efficacy of non-steroidal anti-inflammatory drugs in experimental pain in humans.

1. The aim of this study was to establish a simple and reliable experimental pain model that could distinguish the analgesic effects of non-steroidal anti-inflammatory drug (NSAID) treatment from placebo in human volunteers. 2. The reproducibility and reliability over time of subject pain ratings was compared using cutaneous electrical stimuli applied to either the thenar eminence or the ear lobe at varying intensities and modes. Subjects were asked to respond firstly, when the stimulus became clearly sharp and painful ('first pain') and secondly, when the sensation became deep and burning and no further increase in stimulus intensity could be tolerated ('second pain'). 3. Constant voltage stimuli were found to be more reproducible than constant current stimuli. Both phasic (intermittent) and tonic (continuous) stimulation modalities produced 'first' and 'second pain' sensations. The latter sensation was more reproducible, and was perceived as a burning pain which is akin to clinical pain. 4. Analgesics from the NSAID class were found to attenuate reliably only 'second pain' sensations. The analgesic effects of ibuprofen (ibuprofen vs placebo: 0.12 +/- 0.09 vs 0.02 +/- 0.07 volt h(-1), P = 0.03; 95% confidence interval for differences (CI): 0.03-0.18) and diflunisal (diflunisal vs placebo: 0.29 +/- 0.40 vs 0.005 +/- 0.27 volt h(-1), P = 0.0001; CI: 0.168-0.407), respectively, could be distinguished from placebo.