Opioid-Induced "Likeability" and "Feeling Good" Are Not Associated With Return Visits to an ED Among Migraine Patients Administered IV Hydromorphone.

BACKGROUND

Parenteral opioids are used in more than 50% of emergency department (ED) visits for migraine. Use of opioids for migraine has been associated with subsequent ED visits, perhaps because of opioid-induced euphoria. In this study, we quantify the extent to which nontherapeutic effects of opioids influence migraine outcomes. We hypothesized that "feeling good" and medication likeability would in fact be associated with receipt of opioids (rather than relief of migraine pain) and that receipt of opioids (rather than relief of migraine pain) would be associated with return visits to the ED.

METHODS

During an ED-based clinical trial, migraine patients were randomized to receive hydromorphone 1 mg or prochlorperazine 10 mg + diphenhydramine 25 mg IV. Thirty minutes after medication administration, we asked, (1) How much did you like the medication you received? and (2) How good did the medication make you feel? Participants were asked to provide answers on a 0-10 scale. We also determined 0-10 pain scores at baseline and 1 hour and number of return visits for headache during the subsequent month.

RESULTS

Sixty-three patients received prochlorperazine and 64 hydromorphone. Prochlorperazine pain scores improved by 6.8 (SD: 2.6), hydromorphone by 4.7 (SD: 3.3) (95%CI for difference of 2.1: 1.0, 3.2). On the 0-10 likeability scale, prochlorperazine patients reported a mean of 7.2 (SD: 2.8), hydromorphone 6.9 (SD: 2.9) (95% CI for difference of 0.3: -0.7, 1.3). On the 0-10 feeling good scale, prochlorperazine patients reported a mean of 7.5 (SD: 2.3), hydromorphone 6.8 (SD: 2.8) (95%CI: for difference of 0.7: -0.2, 1.6). In the hydromorphone group, 8/57 (14%, 95%CI: 7, 26%) returned to the ED vs 5/63 (8%, 95%CI: 3,18%) in the prochlorperazine group. In regression modeling, feeling good was independently associated with pain relief (P < .01) but not with medication received (P = .67) or return visits (P = .12). Similarly, medication likeability was independently associated with pain relief (P < .01) but not medication received (P = .12) or return visits (P = .16).

CONCLUSION

We did not detect an association between hydromorphone and medication likeability, feeling good, or return visits to the ED. Headache relief was associated with medication likeability and feeling good.