Bradycardia Associated With Intravenous Methadone Administered for Sedation in a Patient With Acute Respiratory Distress Syndrome

Veena Karir, Pharm.D.


Pharmacotherapy. 2002;22(9) 

In This Article


Methadone is an excellent option for sedating patients in the intensive care unit who have received the maximum dosage of shorter-acting opioids because of their incomplete cross-tolerance to other µ-receptor agonists. Initial administration of methadone results in a rapid onset of analgesia with accumulation over time; however, given its long elimination half-life and large volume of distribution, the amount of methadone necessary to maintain adequate pain control and sedation decreases over time. For this reason, rapid titrations of a continuous methadone infusion should be performed with caution because plasma concentrations may not reach steady state for several days.[9,11] Furthermore, individual pharmacodynamic differences in half-life and clearance of methadone vary widely. A retrospective review measuring clearance of methadone in 185 patients with cancer pain revealed a mean methadone clearance of 186 ml/minute (range 23-2100 ml/min) and a mean elimination half-life of 31.8 hours (range 4.2-130 hrs).[11]

Pharmacokinetic models have attempted to correlate measured methadone concentrations to predicted levels as well as patient-specific analgesia. A study of 22 patients with severe pain secondary to cancer or burn rehabilitation, or of nonmalignant origin, determined that measured and predicted levels of administered methadone correlated closely (r2=0.9704).[12] A similar study of patients with cancer pain who were receiving continuous methadone infusions measured both concentration and patient response using visual analogue scales. A direct correlation was found between the level of methadone administered and pain relief.[13] However, the application of such a model in the intensive care unit is not useful because patients are usually heavily sedated, thus pain scores cannot be obtained. Also, interpatient variability would make it difficult to use predicted models to correlate with measured levels and sufficient levels of sedation and analgesia.

The use of low tidal volumes with permissive hypercapnia may require patients to be heavily sedated to tolerate mechanical ventilation. The patient described in this case report received intravenous methadone 160 mg/hour based on his strong respiratory drive, ventilator-patient dyssynchrony, and worsening oxygenation. However, high-dose intravenous methadone as a single agent has been administered without adverse effects in patients with cancer -- 1900 mg/24 hours in one, 44 mg/hour in another.[8,9] I believe the rapid titration of methadone in my patient from 90 mg/hour on hospital day 39 to 160 mg/hour on day 41 correlated more closely temporally to his bradycardic episode than any other agent he was taking at the time; however, other potentially cardiotoxic agents must be considered. Of note, methadone is metabolized by way of the cytochrome P450 (CYP) pathway, predominantly by the 3A4 isoenzyme, where N-demethylation of methadone occurs. Erythromycin and midazolam are both substrates and inhibitors of the CYP3A4 pathway and potentially can decrease clearance of methadone.[14,15]

Cardiac effects of histamine are mediated by H1 and H2 receptors. Specifically, H2 receptors promote inotropy, and H2 antagonists such as cimetidine and ranitidine have been implicated with sinus bradycardia through this mechanism.[16] Patients with ranitidine-induced bradycardia have experienced symptoms with a single injection of ranitidine, on rechallenge, with oral administration, and even after several weeks of therapy.[17,18] In my patient, intravenous ranitidine for stress ulcer prophylaxis was begun at hospital admission and was continued through the resolution of his bradycardia, making ranitidine an unlikely cause of this adverse event.

Of three patients with metoclopramide-associated total heart block and dysrhythmias, one had jaundice, one was predisposed to dysrhythmias, and one had atrial fibrillation treated with digoxin.[19,20] In my patient, metoclopramide was started on hospital day 6 and, as with his ranitidine, was continued through the resolution of his bradycardia. Thus, metoclopramide was also an unlikely cause of the adverse event.

More frequently than ranitidine and metoclopramide, erythromycin has been linked with cardiotoxicity, specifically QT prolongation with torsades de pointes.[21,22] Erythromycin acts more like a class III antiarrhythmic by blocking potassium reuptake and prolonging the action potential.[22,23] Bradycardia and nodal blockade have been rarely associated with erythromycin.[24] Recurrent hypotension and nodal bradycardia were associated with intravenous erythromycin administration in a 73-year-old woman; however, all episodes occurred during treatment with erythromycin.[24] In fact, most reports link intravenous erythromycin and cardiotoxicity with the time of drug infusion, indicating a temporal relationship.[22,23,24,25,26]

A small prospective study investigated QT prolongation and cardiac arrhythmias induced by intravenous erythromycin in seven critically ill patients.[25] The study demonstrated a positive correlation between rapid infusion of erythromycin and QT prolongation. However, in a prospective study of 44 patients,[26] QT prolongation occurred despite slow administration of erythromycin.

My patient received intravenous erythromycin 250 mg 4 times/day on hospital days 38-42 (for 15 doses) to improve his gastrointestinal motility. The patient remained hemodynamically stable throughout his bradycardia, which never progressed to torsades de pointes, the most commonly associated arrhythmia. No temporal correlation was noted between the time of erythromycin administration and the bradycardic episode, making erythromycin a less likely cause of the adverse event. Yet, given the indication and the literature linking erythromycin with cardiotoxicity, the drug was discontinued.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.