Review of: Rikard C, O Meara P, McGrail M, et al: A randomized controlled trial of intranasal fentanyl vs intravenous morphine for analgesia in the prehospital setting. American Journal of Emergency Medicine. 25(8), 911-917, 2007.
The Science
This study from Australia attempted to compare pain relief of intranasal fentanyl (INF) with that of intravenous morphine (IVM). The study was randomized and open label, which means patients who met the criteria were randomized to either INF or IVM. Of course, both the patients and the paramedics knew what they were getting because they had to establish an IV for the morphine.
They measured the reduction in pain using the Verbal Rating Score (VRS) we re all familiar with on a 10-point scale and also documented the occurrence of adverse effects. INF patients received repeated doses of 60ug of fentanyl while the IVM received 2.5-5.0 mg morphine. In Australia it s routine to use methoxyflurane (Penthrane), which is delivered via an inhaler and has properties similar to nitrous oxide. In this study almost a third of the patients received methoxyflurane in addition to INF or IVM.
The VRS for both groups was similar, although the VRS for patients with back pain was significantly higher than for any other group. Both the INF and IVM groups included patients with chest pain and these were given nitroglycerin as well as the analgesics.
The researchers found no overall significant difference in pain reduction between intranasal fentanyl and intravenous morphine. Furthermore, they documented almost twice as many adverse side effects in the INF group than the IVM one, though this did not reach statistical significance.
The Street
At first blush, one could use this paper to dispel the use of intranasal fentanyl in the pre-hospital setting. However, one must read the study in its entirety to fully grasp the conclusions drawn.
In the first paragraph of the paper the authors state that their intention was to enroll 200 patients in each group to have an 80-percent chance of detecting a one-point reduction in the VRS between the two groups. They only enrolled half that number. Therefore, any conclusion regarding the relative benefits of the two drugs is not statistically sound.
The addition of methoxyflurane to the study confuses the results and further weakens the findings despite the fact that the researchers found that the patients who received methoxyflurane in addition with either INF or IVM had less reduction in pain. I have to wonder if those patients didn t have methoxyflurane provided before the medics decided to enter them into the study and give the other drugs. If this is true, then the time to drug delivery was increased, and the opportunity for repeat doses and detection of a difference in VRS would be reduced.
Including patients with possible cardiac chest pain further confuses the results. In fact, recent evidence indicates that opiates may be harmful to patients having a heart attack.
The documentation of more adverse side effects in the INF compared to the IVM is concerning, but further analysis reveals that the side effects were reported by the medics and not by a trained research observer — or better yet, the patients themselves. Studies that allow medics this level of subjective reporting are prone to selection bias where for whatever reason the medics may or may not prefer the study drug or treatment.
In conclusion, this study at least shows that someone is trying to examine the vital issue of prehospital pain management. A more elegant study, such as starting an IV on all patients and providing them a squirt of either NS or fentanyl up their nose compared to either NS or morphine IV, would provide us a far better understanding of the difference between the two agents.
