No patient looked sicker or scared me more when I was a new resident than the 6 a.m. acute pulmonary edema patient who rolled in gasping desperately for air. Now that I'm older, better at crushing my fear and anxiety into a teeny-tiny ball to be released later at a wildly inappropriate time, and have seen the wonders that a little BiPAP and a lot of intravenous nitroglycerin can do, I'm less likely to hit the panic button when I see a patient with APE pop up on the board.
Unfortunately, that fear is now sometimes replaced by rage, and it's a rage caused by a consultant or EM trainee ordering intravenous morphine as part of the treatment plan. I know that EM providers are busy people, so if you're reading this on shift, I'll save you the trouble of tackling the rest of the article: Do not give morphine to patients with acute pulmonary edema.
I should start by saying that in this discussion we're really talking about patients presenting with acute decompensated heart failure (ADHF), and that we don't have any good randomized control trials on this. What evidence there is, however, is startling enough that it should give even the most cavalier cardiologist and egotistical emergency physician cause for concern.
The most compelling and certainly the most cited article on morphine for APE was performed by W. Frank Peacock IV, MD. (Emerg Med J 2008;25:205.) Before looking at the paper, we need to acknowledge how awesome his name is. This was a retrospective analysis of the Acute Decompensated Heart Failure National Registry (ADHERE) that compared the records of 20,782 hospitalizations for APE where the patient received IV morphine with 126,580 hospitalizations where the patient was not given morphine. The quick summary is that the patients who received IV morphine did worse in essentially every way, but I'll summarize the high points here:
- Need for intubation: 15.4 percent in the morphine group vs. 2.8 percent in the no-morphine group (NNH=8).
- Need for ICU admission: 38.7 percent in the morphine group vs. 14.4 percent in the no-morphine group (NNH=5).
- Mortality: 13.0 percent in the morphine group vs. 2.4 percent in the no-morphine group (NNH=10).
It's true that the patients who received morphine did have some baseline differences. The authors noted “a higher prevalence of rest dyspnoea, congestion on chest radiography, rales, and raised troponin occurred in the morphine group.” They performed risk adjustments for pretty much every conceivable underlying risk factor or patient characteristic to account for this. It didn't matter: Morphine was associated with worse outcomes and increased mortality (i.e., OR 4.84 (95% CI 4.52 to 5.18), p<0.001) in every mode of comparison.
Judd Hollander, MD, who is no methodological slouch, was the second author on this paper, and I'm going to name-drop him by way of vouching for the statistical analyses performed.
I would be remiss if I didn't stress that we can only draw limited conclusions for this type of retrospective study, noting a strong association between IV morphine administration in patients with APE and poor outcomes. This is different from claiming that morphine is what's causing the poor outcomes. That said, the strength of the association here is compelling, and there is a paucity of evidence demonstrating any benefit for using morphine for APE patients.
Where did this whole morphine for APE originate? I remember learning that it worked by reducing preload, to a lesser extent afterload, and that it decreased heart rate and lessened oxygen hunger. The evidence for this? Please let me refer you to a wonderful review article titled, appropriately enough, “Morphine in the Treatment of Acute Pulmonary Oedema — Why?” (Int J Cardiol 2016;202:870.)
The authors wrote, “The background of this belief is three trials from 1966 to 1976, where the groups consisted of 12 dogs, 12 patients with myocardial infarction (where nobody received morphine as the sole drug), and 13 patients with mild pulmonary oedema, respectively. ... [I]n 1994, a trial with vessels from dogs in vitro showed that morphine had a relaxing effect on the smooth muscles in both veins and arteries.”
We basically have a cohort of two dozen dogs, a dozen people with MI, and 13 patients with mild pulmonary edema forming the basis for using morphine in patients with APE. When balanced against the striking negative association found in the ADHERE trial, a study showing some canine vasodilation is a weak basis for putting our patients at risk.
What about older studies performed by well-known emergency physicians whom I love and trust, you ask? So glad you did. None other than the esteemed Jerry Hoffman, MD, published a study examining the efficacy of various pharmacological treatment regimens given by EMS to patients with presumed pulmonary edema. (Chest 1987;92:586.)
What did they find? Nitro seemed to be beneficial, and furosemide and morphine “may not add anything to its efficacy, and may be potentially deleterious in some of these patients.” On the downside, the study only enrolled 57 patients, 23 of which (40%) turned out not to have APE but rather things like pneumonia and COPD. We hold Dr. Hoffman in great regard, but this was not necessarily a game-changing paper.
Not to be outdone, fellow EM luminary, Alfred Sacchetti, MD, and colleagues undertook a chart review of 181 patients with APE, looking at associations between what kind of medications they received and how often they were intubated and admitted to the ICU. (Am J Emerg Med 1999;17:571.) And the results? IV morphine was associated with increased odds of being intubated (OR 5.04; p=0.001) and increased ICU admissions (OR 3.08; p=0.002). Again, this was a chart review of a relatively heterogeneous group, and the association between IV morphine and worse outcomes was pretty straightforward.
If you've stuck with me this far, the take-home point should be pretty clear, but just to summarize one more time: Do not give morphine to patients with acute pulmonary edema. Perhaps someday a study will give us cause to revisit this issue, but until then I'll hold the morphine rather than holding my breath.
Share this article on Twitter and Facebook.
Access the links in EMN by reading this on our website or in our free iPad app, both available at www.EM-News.com.
Comments? Write to us at firstname.lastname@example.org.Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.