Res omnes venena sunt, dosis sola facit venenum
Overwhelmed by the sheer number of patients included, one is tempted to accept the conclusions made Peacock et al. in their recent paper  on the perceived risks of morphine in acute decompensated heart failure (ADHF).
However, in several ways their analysis is difficult to reconcile with published data and observations in the palliative care setting. It is surprising that in the rationale given for the administration of morphine, symptomatic improvement as perceived by the patient does not seem to be an issue of relevance. In the context of dyspnoea of cancer this is an important positive endpoint (reviewed e.g. by Ben-Aharon ), according to self rating by patients on a numeric or visual rating scale. Abernethy et al. demonstrated that the application of sustained-release morphine at low dosage provided significant symptomatic improvement in refractory dyspnoea in patients suffering from chronic obstructive pulmonary disease . Morphine in the situation described in the study of Peacock et al. is usually being used for symptom management and not as a first-line treatment of ADHF. Concerning dyspnoea there was a significant difference at the timepoint of initial clinical presentation, suggesting a higher burden of symptoms for the morphine-group. However, due to the study design, no data on this issue were available. Therefore no conclusion can be drawn regarding this outcome parameter.
In comparable settings there are numerous studies [e.g. 4,5] including patients with diseases other than cancer [3,6,7], which show that with cautious dosage respiratory depression is negligible or completely absent. Certainly these populations are different from patients with ADHF, but may be comparable in vulnerability.
More disturbing is however a statement that the authors are repeating in different phrases, that intravenous morphine “may cause….ventilatory depression”. They underline this by reporting that “more adverse outcomes occurred in the morphine cohort, including higher rates of mechanical ventilation”, to conclude that there is “a marked increase in the association between mechanical ventilation and morphine use”. This line of arguments can hardly be followed, as there are no supporting data presented on the indication for morphine use (e.g. symptom control vs. induction of anaesthesia), the dose being administered or even the timing of morphine injection and onset of “adverse events”. Additionally there seems not to be a standardized documentation of so called “adverse events”. To underline their conclusion another retrospective study by Sacchetti is being cited. However, this author follows the same line of arguments, that the use of morphine resulted in higher frequency in the need of mechanical ventilation, not taking into account that morphine might have been used for induction of anaesthesia or had to be used due to the symptom intensity. The effect of morphine on dyspnoea had not been assessed in this study, too. In contrast to Peacock et al. palliative care literature provides evidence, that morphine given for pain relief does not affect prognosis [as reviewed in Sykes and Thorns 8].
Are there possible explanations (other than differing underlying conditions) as to why these contradictions might arise? Three limitations of the present study are striking.
First, the authors themselves raise the issue that the timing of administration was not provided, which precludes any meaningful analysis of the context of administration.
Second, no data on morphine doses were recorded. Some recommendations in the literature concerning pulmonary edema (e.g. morphine “given in 2- to 4 mg i.v. boluses…”, but at what intervals and how many boluses ?) might result in a dose intensity beyond the recommendations for symptom control in opioid-naïve patients. Maybe the phrase “Res omnes venena sunt, dosis sola facit venenum” (Paracelsus: “All things are poisonous, it is the dose that makes the poison”) explains part of the problem.
However a third observation might perturb the results even more. Table 8 shows an odds ratio for mortality of 5.77-10.7 (adjusted and unadjusted) in patients without mechanical ventilation. Assuming proper monitoring and excluding unattended deaths, this allows in my opinion some inference about the intentions of physicians. If there were a reasonable chance of survival, death would rarely occur without an attempt to use maximal ICU-therapy, including intubation and mechanical ventilation. It is reasonable to assume that these odds ratios indicate that in a considerable proportion of patients morphine was used as a symptomatic treatment or sedative, and that most often these unfortunate patients were considered poor candidates for artificial ventilation in the first place.
One has to keep in mind that including large numbers of patients is not a means to eliminate bias from studies. Any of the explanations mentioned might in theory cause sufficient bias to eliminate any suspicion against morphine in this setting. On the other hand the principle of providing the optimal relief while doing as little harm as possible requires more reliable data for a sound ethical debate.
1. Peacock WF, Hollander JE, Diercks DB, Lopatin M, Fonarow G, Emerman CL. Morphine and outcomes in acute decompensated heart failure: an ADHERE analysis. Emerg. Med. J. 2008; 25: 205-209.
2. Ben Aharon I, Gafter-Gvili A, Paul M, Leibovici L, Stemmer SM. Interventions for alleviating cancer-related dyspnea: a systematic review. J. Clin. Oncol. 2008; 26: 2396-2404.
3. Abernethy AP, Currow DC, Frith P, Fazekas BS, McHugh A, Bui C. Randomised, double blind, placebo controlled crossover trial of sustained release morphine for the management of refractory dyspnoea. BMJ 2003; 327: 523-528.
4. Clemens KE, Klaschik E. Symptomatic therapy of dyspnea with strong opioids and its effect on ventilation in palliative care patients. J. Pain Symptom. Manage. 2007; 33: 473-481.
5. Clemens KE, Klaschik E. Effect of hydromorphone on ventilation in palliative care patients with dyspnea. Support. Care Cancer 2008; 16: 93- 99.
6. Clemens KE, Klaschik E. Morphine in the management of dyspnoea in ALS. A pilot study. Eur. J. Neurol. 2008; 15: 445-450.
7. Johnson MJ, McDonagh TA, Harkness A, McKay SE, Dargie HJ. Morphine for the relief of breathlessness in patients with chronic heart failure--a pilot study. Eur. J. Heart Fail. 2002; 4: 753-756.
8. Sykes N, Thorns A. The use of opioids and sedatives at the end of life. Lancet Oncol. 2003; 4: 312-318.
9. Harrisons Principles of internal Medicine, 17th edn. McGraw-Hill Professional, 2008.
Opiates should be avoided in acute decompensated heart failure
We were pleased to read Peacock et al’s paper warning of the dangers of morphine therapy in acute decompensated heart failure.  This retrospective review was the largest yet performed, with nearly 150000 patients in the database. The most concerning finding was an increased risk of mortality with an odds ratio of 5, despite extensive risk adjustment. Their conclusion that a prospective trial is needed is correct, but does not go nearly far enough. The implication is that current practice is acceptable until proven otherwise.
Handbooks for doctors  and prehospital staff  still advocate use of morphine (or diamorphine) in acute cardiogenic pulmonary oedema. The most recent edition of a popular emergency medicine handbook states that opioids should only be given to those with chest pain or in distress , and it is encouraging that in Peacock’s paper only 14% of patients received morphine. A retrospective study showing an association between morphine use and mortality is not proof of causation, but there is minimal evidence of any benefit from using morphine. A review from 2004  found no evidence supporting the use of morphine, and two smaller retrospective studies suggesting an association with harm.
In view of the extremely serious consequences of using morphine suggested by the best evidence available, the routine use of morphine in acute cardiogenic pulmonary oedema should be discontinued immediately. We should not wait until a randomised trial demonstrates harm before we stop: the onus should be on a prospective study to prove benefit.
 Peacock WF, Hollander JE, Diercks DB et al. Morphine and outcomes in acute decompensated heart failure: an ADHERE analysis. Emerg Med J 2008;25:205-9.
 Longmore M, Wilkinson IB, Turmezei T, Cheung CK. Oxford Handbook of Clinical Medicine. 7th ed. Oxford: OUP; 2007. p. 787.
 Greaves I, Porter K. Oxford Handbook of Pre-Hospital Care. Oxford: OUP; 2007.
 Wyatt JP, Illingworth RN, Graham CA et al. Oxford Handbook of Emergency Medicine. 3rd ed. Oxford: OUP; 2006.
 Graham CA. Pharmacological therapy of acute cardiogenic pulmonary oedema in the emergency department. Emerg Med Austral 2004; 16:47-50.
Competing interests: Colin Graham is a co-author for the Oxford Handbook of Emergency Medicine (reference 4 above).
This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of EMJ.
View free sample issue >>
Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.
Poll now closed
What impact do you think co-locating primary care services with or within EDs has on ED visits:
Decreased ED visits: 55% Increased ED visits: 28% Has no effect: 17%