sinaiem.org

We continue to make our way through the recent pain management papers, once again turning to the August Annals (Vol 48, No 2). Chang, Gallagher et al. strike back with a second analgesia piece in this issue — from now on, Montefiore will be simply be known as the House of Pain. The paper’s called Safety and Efficacy of Hydromorphone as an Analgesic Alternative to Morphine in Acute Pain: An RCT (if you’re logged into the Sinai library, full text is here). It’s full of provocative hypotheses, good study technique, and fun historical trivia… more below!

While some may doubt it, surveys and chart reviews still suggest pain is not well treated in the ED — even 0.1 mg / kg of morphine (resulting in 10 mg in the heavy folks) probably isn’t sufficient. Yet docs are loathe to order 10 mg off the bat. Surprisingly, though, docs are less averse in ordering 1-1.5 mg of hydromorphone (dilaudid)  in the same situation — and that’s approximately an equianalgesic dose. Why should this be? It could be psychological — the number is smaller and less set-in-stone, and thus, docs are more comfortable with it. Who knows, but Gallagher and company use this lack of inhibition as the basis for their research question: Can hydromorphone be first-line for pain?

The author’s intro begins with a retrospective on dilaudid. It’s a semi-synthetic opioid that predates the synthetic meperidine (demerol) by ten years, debuting in 1920. There’s more from Wikipedia. In general, Cochrane reviews of pain management give short shrift to hydromorphone — there are only a few good trials, featuring limited uses and fixed doses.

So, this trial is a prospective random double-blind one-center trial of consecutively presenting adults, 21-65, with less than 7 days of acute pain severe enough to warrant opioids, as per an emergency physician (EP). Exclusion criteria included allergy, hypotension, intoxication, recent opioids, MAO inhibitor use, or chronic pain disorders such as sickle cell or fibromyalgia. Patient’s weights were determined by patient estimate, which could make things interesting. Max dosing was 10 mg morphine (0.1 mg / kg) or 1.5 mg dilaudid (0.015 mg / kg)  so heavy (100 kg +) patients were excluded. Consent was obtained, and then a single dose of opioid was given. The ratio they chose as equianalgesic was 6.67 : 1, based on Goodman & Gilman — but other sources vary. They made up the dilaudid to look as voluminous as the morphine, for blinding.

The study’s primary outcome was between-group difference in numeric pain scale rating from 0 to 30 minutes. A secondary goal was change in pain at 5 and 120 minutes. Other secondary measures were side effects, and need for more meds.

Of the 252 patients assessed, 18 refused and 36 were excluded, leaving 99 to get morphine and 99 to get hydromorphone. By chance, the hydromorphone group’s initial pain scores were higher. And, this being New York City, the median pain was 10 (new city motto: Even Our Pain is Bigger).

Other results: the mean change in pain at 5 and at 30 minutes was -5.5 points with dilaudid, and -4.1 points with morphine. Mean pain was similar at 2 hrs. There were no statistically significant differences.

In a subgroup analysis, in those patients with an initial pain of 10, hydromorphone made for a greater decrease at 5 and 30 minutes compared to morphine sulfate (a drop of 5.9, versus 3.7, with 95% CI being a difference between 3.2 and 1.0) . The authors interpret this as consistent with the lipophilic character of hydromorphone. There was no difference in hypotension, bradypnea, desats, or nausea / vomiting. Morphine patients had more itchiness (6, vs. 0 for hydromorphone). No one needed naloxone, and many patients needed more meds (no statistical difference between the groups).

Some limits they point out: Maybe they got the equianalgesic dose wrong: they used 6.67:1 but other sources vary from 2-12 ! Maybe Goodman & Gilman really underestimate hydromorphone’s potency, and as a consequence, the folks at Montefiore were giving heftier doses of narcotic than they expected. Also, their timepoints — they went with 30 minutes so as not to miss any opioid effects, even though 15 minutes might have been more clinically appropriate.

I’d also like to add that patients estimating their weight is a bad idea, though I’m not sure which side the data would be skewed towards. I wish initial pain was equal in the arms and at least some effort was made to quantify and balance other potential pain-ameliorating interventions — did both groups get equal amount of fluids? Or reglan? Or nitro? What was causing this pain, anyway — were all the morphine-recipients suffering from mesenteric ischemia, whereas the dilaudid group had appys?

At any rate, the authors are hesitant to conclude that the difference between pain drops of 5.5 (dilaudid) and 4.1 (morphine) means much — they’re just happy to say that dilaudid at 0.015 mg / kg can be a first-line agent for pain control, making it more likely that patients will get the analgesia they need. And, bonus, the price is about the same — $1 for 10 mg morphine or for 1.5 mg of dilaudid.

Has this changed my practice? Well, yes — I’m now more aware of that pro-dilaudid bias, but also more comfortable with it now. I must say, I’m more likely to give it — especially in those patients with 10 out of 10 pain.

Posted on Sunday, November 5th, 2006 at 7:09 am by Nick. Filed under Pain Management, Journal Club.
You may post a comment.