Why we Should Consider Not Using Morphine in Acute Heart Failure

Background/Introduction: Acute heart failure is a common diagnosis encountered among patients presenting to the Emergency Department with complaints of shortness of breath. The emergency treatment of these patients has traditionally focused on alleviation of their symptoms of breathlessness and anxiety in addition to optimization of hemodynamics and rapid reduction in both preload and afterload. The treatment of associated symptoms has often included the administration of morphine, which has been posited to have both beneficial physiologic (vasodilation, reduction of preload) and central nervous system (reduction of breathlessness, anxiety, and pain) effects. However, recent experimental and experiential data have pointed to morphine’s potential for effecting negative physiological and CNS responses, thereby raising the possibility of increasing patient morbidity and/or mortality. Additionally, no large randomized controlled trials have been conducted to study the potential risks and benefits of morphine administration in patients presenting with acute heart failure. Despite these factors, a contingent of Emergency Physicians continue to routinely use morphine in the treatment of patients presenting with acute heart failure.

Article Being Reviewed:  Miro, O. et al. Morphine Use in the ED and Outcomes of Patients with Acute Heart Failure: A Propensity Score-Matching Analysis Based on the EAHFE Registry. Chest.2017 Oct; 152(4): 821-832. PMID: 28411112

What They Did:

  • Multicenter, observational, propensity matched cohort study using a pre-existing data bank (Epidemiology of Acute Heart Failure in the Emergency Department Registry, also referred to as the EAHFE)
    • 6516 patients presenting to one of 34 EDs in Spain during two separate recruitment periods (2 months each in 2011 and 2014).
    • Patients were diagnosed clinically with acute heart failure, and then entered into the EAHFE database at the time of their presentation to the ED.
  • Patients were assigned to “morphine” and “without morphine” cohorts based on the use of intravenous morphine within the first three hours after initiation of treatment in the Emergency Department.
  • A total of 46 variables deemed “potentially related to prognosis” were collected and analyzed. A multivariate logistic regression analysis was then performed on each cohort, and a sub-group of 275 propensity score-matched (PS-matched) patients was taken from each cohort based on the presence of 24 discordant variables. Statistical analysis of these PS-matched subgroups was then performed.
  • All treatment decisions – including intervention and ultimate disposition – were made by the Emergency Department physician at the time of patient’s presentation to the ED, and without regard to the study protocols and/or inclusion criteria.
  • Patients with acute ST-elevation myocardial infarction who developed AHF were not included in this study. Additionally, approximately 2% of patients meeting inclusion criteria for this study declined to participate.


  • Primary Outcome – 30 day all-cause mortality
  • Secondary Outcomes – Mortality at 3, 7, and 14 days. In-hospital mortality. Length of hospital stay.


  • Among all enrolled patients, 416 (6%) received morphine in the Emergency Department, while 6100 (94%) did not.
  • In the cohort of all patients 30-day mortality was higher among patients receiving morphine (26.7% vs 8% in the no morphine group.)
  • Within the propensity-matched cohorts, a higher 30-day mortality was also noted among patients receiving morphine (20% vs. 12.7% OR 1.66, CI = 1.09-2.54; p = .017)
  • Mortality was higher among patients receiving morphine at all examined intermediate times, although the most significant risk increase was noted at 3 days (mortality 8.0% vs. 2.5%). The authors posit that this may be due to the pharmacokinetics of morphine administration.
  • No difference was noted on in-hospital mortality or length of hospital stay


  • IV morphine use during the initial three hours after ED arrival was used as the benchmark for inclusion, a criterion that makes sense from a pharmacodynamic standpoint.
  • All subgroup analyses of the primary outcomes were planned a priori
  • Independent factors associated with the primary end point of 30 day mortality were assessed in a logistic regression model to ensure a minimized bias of analysis
  • All patients presenting to the Emergency Department with AHF were included in the study population – not just those who were subsequently admitted to the hospital. The study population includes six patients who died in the ED prior to hospital admission.
  • Principal investigators at each center double-checked patients entered into database to assure that criteria for “acute heart failure” were met.
  • PIs at each center were blinded as to the treatment received in the Emergency Department, thereby limiting the possibilities for lead bias.
  • B-type Natriuretic Peptide and/or echocardiogram data was utilized (when available – 92% of cases) to confirm the ED diagnosis of acute heart failure. It should be noted, however, that the clinical utility of BNP values in establishing the diagnosis of acute heart failure has been questioned.
  • Results were gathered from a large data pool (6516 patients from multiple tertiary and community EDs.)
  • Patients were propensity-matched across multiple variables that could potentially affect morbidity and mortality, and the study reported data trends from both the absolute and propensity-matched populations. PS matching minimizes confounding factors by indication and is a proposed solution to overcome immortal time bias (from the patient’s enrollment into a cohort to study drug administration)
  • Treatment decisions were made in real-time by Emergency Department clinicians, and inclusion in the trial did not alter treatment in any way.
  • The study examined a patient-centered outcome (mortality.)


  • Patients in the “morphine” group had higher rates of ischemic heart disease, cerebrovascular disease, peripheral artery disease, dementia, a worse functional status, and higher NYHA class of heart failure than those in the “without morphine” group, meaning that this was a patient population with higher comorbid disease. This has the potential to bias the results of mortality toward the morphine cohort.
  • Similarly, patients in the “morphine” group were more likely to receive intravenous medications, IV nitrates, vasopressors, and ventilator support. As such, it appears that opiates were more frequently used in the setting of acute severe presentations. Therefore, the increased mortality noted this subgroup cannot be determined to result solely from treatment with opiates, as opposed to a consequence of more severe disease.
  • Only a small number of patients (6%) were initially included in the treatment (morphine) arm, though after propensity score matching, an equal number (275) of patients were enrolled in both the treatment and non-treatment groups.
  • Though patients were enrolled into the database by individual clinicians caring for them, the subsequent review of those patients included by each site’s primary investigator introduces the possibility of selection bias.
  • Patients found to have ST-elevation myocardial infarction who subsequently developed acute heart failure were not included in this study. Although this represented only 3% of AHF cases encountered, and is not likely to affect overall outcomes of this study, its exclusion does limit the applicability of results to this subgroup of patients.
  • All patients included had some degree of renal dysfunction at baseline (mean CrCl = 58 ml/min)
  • The authors concede that the use of the Propensity Score may have hidden cofounders that were unaccounted for in the analyses.
  • No consideration was given as to patients’ resuscitation (DNR/DNI) status, and to whether or not morphine was used in palliative care cases.
  • The majority of patients did not receive therapy consistent with the established Emergency Department standard of care for acute heart failure. Only 16% received intravenous nitrates, while only 6% received non-invasive positive pressure ventilation. Approximately 4% of the patients required intubation and mechanical ventilation. Whether this variance from the established standard of care was intentional or was driven by other factors was not addressed
  • No consideration was given to the dose of morphine received by the treatment group.As such, it is possible (though unlikely) that the included patients received doses of morphine different than those typically offered to acute heart failure patients. Additionally, failure to quantify the doses received by patients in the treatment arm impedes the ability to determine whether a dose-dependent relationship exists with regard to morphine’s deleterious effects.


  • Morphine has been utilized by generations of physicians for the adjunctive treatment of the anxiety and dyspnea associated with acute pulmonary edema, and experimental data from as far back as the 1950’s has been used to justify its use. Apologists proclaim that morphine increases vasodilation, reduces preload, and facilitates improved oxygen delivery through dilation of the pulmonary artery. However, newer in-vitro and in-vivo studies have served to counterbalance and contradict many of these long-held truths. Riggs et al. described morphine-induced myocardial depression as long ago as 1986 5, while a mounting body of evidence has shown that morphine’s long-touted ability to reduce “breathlessness” is actually due to its propensity to induce respiratory depression6– a consequence that can be catastrophic in a patient who presents to the ED already dyspneic and fatigued. Morphine’s histaminergic effect has also come under intense scrutiny and caution, as the histamine release associated with morphine administration can trigger hypotension, systemic vasodilation, pruritis, and nausea – each of which may necessitate additional therapy to manage these symptoms and their consequences.
  • The European Society of Cardiology guidelines on the treatment of heart failure recommend against the routine use of opiates 3, while the American Heart Association recommends opiate use in heart failure be limited to the palliative care of patients with end-stage HF and severe respiratory distress 4.
  • Additional studies have also shown an association between the use of morphine in acute heart failure and an increase in mortality. Peacock et al reported data from the ADHERE registry and demonstrated that the use of morphine in patients with acute heart failure is an independent predictor of increased in-hospital mortality7. Similarly, a 2011 study from Israel found that the use of morphine in acute heart failure was independently associated with in-hospital death in a multivariate analysis (OR 2 with 95% CI 1.1 – 3.5.)8This study by Miro and colleagues provides similar data, though it is somewhat unique in examining Emergency Department patients and interventions specifically.
  • Despite the growing body of scientific evidence advocating against the routine use of opiates – particularly morphine – in patients with acute decompensated heart failure, a small cadre of Emergency Physicians still advocate its use, often continuing to cite the “justifications” offered above. While this may be due, at least in part, to the fact that “old habits die hard,” it does highlight the need for further patient-centered and prospective data in order to fully elucidate what role, if any, opiate administration should play in the treatment of acute heart failure.

Author Conclusions: “The use of IV Morphine in patients with acute heart failure is associated with an increase in 30-day mortality.”

Clinical Take-Home Point: Due to the retrospective nature  and methodological limitations of this study, strong conclusions cannot be made because of the potential for hidden confounders. This study does, however, add to the growing body of evidence against the use of morphine in the emergency department for patients with acute heart failure. Based on the results of this trial, patients presenting to the Emergency Department with Acute Heart Failure should not be routinely treated with morphine, as it is associated with a higher rate of mortality. Appropriate treatment of acute heart failure in the ED should focus on the optimization of hemodynamics, the initiation of appropriate respiratory support, assessment for the presence of concomitant acute coronary syndrome, and the rapid administration of medications aimed at reducing cardiac preload and afterload. Alternative medications and methods should be utilized for the symptomatic management of breathlessness, anxiety, and pain in these patients. 

Guest Post By:

James C. Fletcher, MD, FACEP
Clinical Assistant Professor
Department of Emergency Medicine
Department of Internal Medicine, Division of General Internal Medicine
Brody School of Medicine at East Carolina University
Greenville, NC
Twitter: @jfletchermd


  1. Miro, O. et al. Morphine Use in the ED and Outcomes of Patients with Acute Heart Failure: A Propensity Score-Matching Analysis Based on the EAHFE Registry. Chest. 2017 Oct; 152(4): 821-832. PMID: 28411112
  2. Dominguez-Rodriguez, A and Abreu-Gonzalez, P. A critical appraisal of the morphine in the acute pulmonary edema: real or real uncertain? J Thorac Dis. 2017 Jul;9(7):1802-1805. PMID: 28839968
  3. Agewall, S. Morphine in Acute Heart Failure. J Thorac Dis. 2017 Jul; 9(7): 1851–1854. PMID: 28839982
  4. Yancy CW, Jessup, M, Bozkurt B, et al. 2016 ACC/AHA/HFSA focused update on new pharmacological therapy for heart failure: an update of the 2013 ACCF/AHA guideline for the management of heart failure : a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America. J Am Coll Cardiol. 2016;68(13):1476-1488. PMID: 28455343
  5. Riggs TO, Yano Y, Vargish T. Morphine depression of myocardial function. Circ Shock. 1986;19(1):31-8. PMID: 3719915
  6. Radke JB, Owen KP, Sutter ME, et al. The effects of opioids on the lung. Clin Rev Allergy Immunol 2014;46:54-64.). PMID: 23636734
  7. Peacock WF, Hollander JE, Diercks DB, et al. Morphine and outcomes in acute decompensated heart failure: and ADHERE analysis. Emerg Med J. 2008;25:205-9. PMID: 18356349
  8. Iakobishvili Z, Cohen E, Garty M, et al. Use of intravenous morphine for acute decompensated heart failure in patients with and without acute coronary syndromes. Acute Card Care. 2011;13:76-80. PMID: 21627393

For More on this Topic Checkout: 

Post Peer Reviewed By: Anand Swaminathan, MD (Twitter: @EMSwami) and Salim R. Rezaie, MD (Twitter: @srrezaie)

Cite this article as: Salim Rezaie, "Why we Should Consider Not Using Morphine in Acute Heart Failure", REBEL EM blog, August 2, 2018. Available at: https://rebelem.com/why-we-should-consider-not-using-morphine-in-acute-heart-failure/.

Like this article?

Share on Facebook
Share on Twitter
Share on Linkdin
Share via Email

Want to support rebelem?