A randomized, double-masked, clinical study of the efficacy of four nonsteroidal anti-inflammatory drugs in pain control after excimer laser photorefractive keratectomy

Objective: This study assessed the efficacy of 4 nonsteroidal anti-inflamma tory drugs (NSAIDs) after excimer laser photorefractive keratectomy (PRK). Background: Inadequate control of pain after PRK surgery can be a severe so urce of distress to patients and can interfere with their willingness to un dergo a second PRK procedure. Methods: This randomized, double-masked, placebo-controlled clinical study was conducted in 125 patients. Four NSAIDs (diclofenac, flurbiprofen, ketor olac, and indomethacin) were tested against a placebo group (artificial tea rs). Pain levels after PRK were quantified using Present Pain Intensity (PP I) and Pain Rating Indices based on rank values (PRI[R]) scores, both of wh ich were calculated using patient responses to a modified McGill Pain Quest ionnaire. The PRI(R) consisted of 4 subscales-sensory (S), affective (A), e valuative (E), and miscellaneous (M)-as well as a total score (T). Results: Three hours after PRK, no differences in PPI scores were found bet ween the ketorolac, diclofenac, and indomethacin groups, whereas placebo wa s significantly less effective than the NSAIDs. Patients who received flurb iprofen reported PPI scores that were significantly lower (P < 0.001) than those of patients who received diclofenac and indomethacin, but PPI scores in the flurbiprofen and the ketorolac groups did not differ significantly. Twenty-four hours after surgery, patients treated with flurbiprofen, ketoro lac, and diclofenac reported the lowest PPI scores compared with those trea ted with indomethacin and placebo (P < 0.001). Moreover, flurbiprofen-treat ed patients also had the lowest PRI(R)T scores (P < 0.001). When the pain r ating index was examined by subclass, a significantly lower PRI(R)S score w as detected in the flurbiprofen group at 24 hours (P < 0.001). The PRI(R)A score was significantly higher in the placebo and indomethacin groups compa red with the other groups (P < 0.001). At the 48- and 72-hour time points, flurbiprofen-treated patients again reported significantly lower PPI and PR I(R)T scores (P < 0.001 for both) in pairwise comparisons with the other tr eatment groups. The number of patients who self-administered additional ora l analgesics did not differ significantly between the groups. However. the mean number of analgesic tablets used was significantly higher in the place bo group than in any NSAID group (P < 0.001). The ketorolac group had the l argest number of patients complaining of itching (P < 0.043). No other subj ective symptoms were significantly different across groups. Finally, ail NS AIDs, except flurbiprofen, prolonged the mean reepithelialization period sl ightly (P < 0.001). Conclusion: Flurbiprofen appeared to be the most effective NSAID for the tr eatment of pain, even at 24 hours after surgery when pain was at a maximum.