Lesson of the Week: Fatal methadone overdose
BMJ 1996; 313 doi: https://doi.org/10.1136/bmj.313.7055.481 (Published 24 August 1996) Cite this as: BMJ 1996;313:481
- T J Hendra, consultant physiciana,
- S P Gerrish, consultant anaesthetista,
- A R W Forrest, consultant chemical pathologista
- a Directorate of Medicine, Intensive Care Unit and Department of Clinical Chemistry, Royal Hallamshire Hospital, Sheffield S10 2JF
- Correspondence to: Dr T J Hendra, Department of Geriatric Medicine, Royal Hallamshire Hospital, Sheffield S10 2JF.
Methadone is a synthetic opioid with potent analgesic effects often used for the detoxification or maintenance of an opiate addict. It differs from morphine in that it has an exceptionally prolonged duration of action with a half life averaging 25 hours, although durations of up to 52 hours have been reported during long term maintenance treatment.1 The variation in metabolism may be responsible for the irregular and unpredictable clinical course of patients who have taken overdoses of this agent.
Methadone overdose can follow an unpredictable course in non- tolerant patients, who are at risk of sudden death
Case report
A 22 year old man presented to the accident and emergency department at 1400 hours having taken an overdose of 420 mg of methadone three hours earlier; he had obtained the drug illicitly. He was a known drug misuser who had epilepsy, treated with sodium valproate. He had been treated before for depression with amitriptyline and fluoxetine. A heavy cigarette smoker with a high alcohol intake, he had taken four overdoses within the past two months.
On arrival he was drowsy and refused to submit to regular nursing observations. On leaving the hospital, he collapsed on the pavement while smoking a cigarette. After being brought back into the department, unconscious and unresponsive to pain, he was given three 0.4 mg boluses of intravenous naloxone and 0.4 mg flumazenil, with a prompt improvement in his level of consciousness. He was transferred to the admissions ward, where he cooperated poorly with monitoring by the nursing staff and hindered the use of a pulse oximeter by frequently getting out of bed and wandering out of the ward. At 1830 hours he had an episode of altered consciousness and responded to 0.4 mg naloxone within two minutes. At 2130 hours a naloxone infusion of 0.3 mg/hour was started following another episode of unconsciousness associated with an oximeter oxygen saturation of 22%.
He was transferred to the intensive care unit, where he remained haemodynamically stable but had a fluctuating level of consciousness, varying from wide awake to rousable on request. The naloxone infusion of 0.3 mg/hour was continued, and a bolus injection of 0.2 mg given at 2345 hours greatly improved his level of consciousness. Blood was taken for toxicological analysis and oxygen administered via a face mask.
The following morning the naloxone infusion was temporarily stopped to assess whether it was still required and he got up to smoke a cigarette under nursing supervision. After he became sleepy, but was still rousable and able to communicate, the naloxone infusion was restarted at a rate of 0.2 mg/hour with good effect. He was seen by the on call psychiatric team and a decision was made to discharge him back to the general medical ward if stable. Before leaving the intensive care unit at 1800 hours the serum toxicology results were reported as methadone 353 μg/l, temazepam <50 μg/l, desipramine <50 μg/l, diazepam 813 μg/l, and nordiazepam 990 μg/l. The therapeutic ranges given were up to 250 μg/l for diazepam and up to 1000 μg/l for methadone. The laboratory also commented that while serum methadone concentrations of up to 1000 μg/l were found in patients receiving methadone maintenance treatment, death following overdose might be associated with a concentration as low as 200 μg/l in blood. He was transferred awake and alert to a general medical ward with instructions for the naloxone infusion to finish later that evening.
On the medical ward he was mobile and talking to the nursing staff until 0200 hours, when he had a cup of hot chocolate before retiring to bed. He was not being monitored by an oximeter at this time. The naloxone infusion was slow to complete, eventually finishing at 0330. He was observed to be sleeping and breathing satisfactorily at 0400 but was found dead in bed at 0655.
At necropsy there were no clinically relevant physical findings: his methadone concentration was 822 μg/l in femoral blood and 7 μg/l in stomach contents. Diazepam and nordiazepam blood concentrations were 103 μg/l and 165 μg/l, respectively. The pathologist's opinion of the cause of death was methadone overdose.
Comment
This young man took an opiate overdose and initially responded to the specific antagonist naloxone but then died unexpectedly 44 hours later within a few hours of the naloxone infusion being completed. Although methadone overdose can require high doses of naloxone, in this patient the quantities needed were not large either as bolus or infusion. Current recommendations for naloxone infusion are to administer two thirds of the initial reversal bolus hourly, with half of the reversal dose being administered as a bolus 15 minutes after the infusion is started. Infusions can usually be tapered after several hours but may be needed for much longer for long acting drugs.1
The toxicological data are consistent with death from methadone toxicity becoming manifest after discontinuation of the naloxone infusion. The benzodiazepine concentrations had fallen significantly after death and are unlikely to have contributed to his death. The methadone concentrations in plasma before death, and blood after death, were in the range potentially associated with severe toxicity in a person who does not take methadone regularly. An antemortem plasma methadone concentration of 353 μg/l would convert to a whole blood concentration of about 460 μg/l, which is significantly less than that found in the blood after death. Although there may be a slight increase in blood methadone concentration after death, due to tissue redistribution, this is unlikely to account for the significant increase in this case. The possibilities are either that absorption of methadone from the gut was not complete at the time that the antemortem sample was obtained or that the patient took further methadone while in hospital after transfer from the intensive care unit. The low concentration of methadone in the stomach contents after death makes the latter hypothesis unlikely, although this possibility cannot be totally excluded.
The delay between stopping the naloxone infusion and death may be related to the rate of methadone absorption from the gut. There is anecdotal evidence that deaths from methadone overdose have occurred shortly after eating food, so it is possible that the chocolate he drank at 0200 hours stimulated his gut, leading to increased absorption of methadone while he slept.
The clinical usefulness of methadone lies in its ability to suppress the symptoms of heroin withdrawal, a starting dose being in the region of 36-40 mg/day, with an experienced addict taking up to 80-100 mg daily. Methadone is well absorbed orally, producing peak plasma concentrations within two to four hours and peak clinical effects within one to two hours. The drug concentrations correlate well with respiratory depression and miosis, with the clinical effects lasting more than 72 hours in some subjects after a single oral dose.2 As little as 40-50 mg may produce coma and respiratory depression in a non-tolerant adult.3 As the half life of methadone varies between 25 and 52 hours, overdose can lead to a protracted clinical course.
Methadone toxicity is more likely in those trying the drug for the first time before tolerance has had time to develop.4 Deaths related to methadone are increasing,5 with patients starting on maintenance programmes being at particular risk.6 Because the sedative and depressant effects of methadone do not develop immediately, it is easy to take a large amount without being aware that harm is being done. One addict's daily maintenance dose of 50-100 mg can cause life threatening poisoning in a non-tolerant adult.7 Although plasma methadone concentrations have been recommended as a routine way of monitoring treatment,8 interpretation must take into account whether the individual has recently started treatment or is stabilised on a maintenance dose.
We recommend that patients with methadone overdose are admitted to hospital empirically for 72 hours, during which time their oxygen saturation should be continuously monitored with an oximeter and prompt referral made to intensive care if they suffer altered consciousness or hypoxia. Regardless of the duration of hospital stay, we also suggest that patients should be monitored with an oximeter for a further 24 hours after discontinuation of naloxone infusion.