Co-proxamol: withdrawal of a drug taken by millions (update) (summary)

Produced by the Royal College of Physicians of Edinburgh and Royal College of Physicians and Surgeons of Glasgow

Co-proxamol: withdrawal of a drug taken by millions (update)

Professor DN Bateman, Professor in Clinical Toxicology, Director, National Poisons Information Service, Edinburgh, Scottish Poisons Information Bureau, Royal Infirmary of Edinburgh, Edinburgh, Scotland

Summary

In this BTMH user-requested article Professor Nick Bateman provides an overview of the withdrawal of co-proxamol, a popular painkiller which was removed from the market in the UK following concerns regarding its safety. Why was it withdrawn, what were the risks and what alternative treatments are available?

Key Points

Co-proxamol is a combination of paracetamol and dextropropoxyphene.
Dextropropoxyphene, an opioid drug, rapidly causes potentially fatal abnormal heart rhythms and respiratory depression when taken in overdose.
Death following overdose is ten times more common with co-proxamol than with other comparable analgesics.
The withdrawal of co-proxamol has resulted in a significant reduction in overdose deaths in Scotland.
Analgesics such as paracetamol alone or other compound analgesics are generally as effective as co-proxamol.
Paracetamol and codeine used as separate medicines may be more effective than using them in a compound pill.

Declaration of interests: No conflict of interests declared

On 31 January 2005, the UK’s Committee on Safety of Medicines (CSM) issued a letter to all prescribers indicating that the analgesic co-proxamol was to be withdrawn from the market. The letter was accompanied by further advice available on the Medicine and Healthcare products Regulatory Agency (MHRA) website (http://www.mhra.gov.uk).

Co-proxamol had been in use in the UK for many years and this announcement surprised some and angered a few, including doctors who had been prescribing the drug and patients who had been receiving it.

In his letter to healthcare professionals, Professor Gordon Duff, Chairman of the CSM, indicated this was not a new concern, and indeed the author of this article was involved in attempts twenty years ago to limit the availability of co-proxamol. At that time, in 1985, the CSM published a warning to doctors about the potential dangers of co-proxamol in overdose, and its relative lack of effectiveness as an analgesic. Despite this, prescription numbers did not fall.

It may be relevant that in recent years the Department of Health began to set targets for ‘health gain’, one of which was suicide reduction. One approach that has been taken to try to reduce successful suicide is the restriction of paracetamol pack size. Analysis of the impact of this change has thrown further light on the true toxicity of co-proxamol when taken in overdose. This is the issue that underlies the recent licensing action.

What is co-proxamol?

Co-proxamol is a combination tablet including the analgesic paracetamol in a dose of 325 mg and the weakly acting opioid dextropropoxyphene in a dose of 32.5 mg. It is difficult to compare directly the effects of dextropropoxyphene as an analgesic with other opioids, but estimates suggest that this dose of dextropropoxyphene is equivalent to about 3 mg of morphine at best. One of the difficulties is that there is no evidence that this combination of dextropropoxyphene and paracetamol is actually more effective as an analgesic than paracetamol alone at its standard dose of 1 g. Indeed, it is quite difficult to understand the rationale for the ingredient strengths in a co-proxamol tablet.

Why is co-proxamol so much more toxic in overdose?

Risk of death in overdose is related to the time-course of effect and the toxicity of the product. In most cases patients reaching hospital will survive. It has been known for over 20 years that co-proxamol is unique amongst paracetamol-containing analgesics in that it can cause death within 1 to 2 hours. So rapid is this that the majority of deaths (~80%) with co-proxamol occur before patients reach hospital. Indeed, could patients reach hospital, it is highly likely they would be able to be resuscitated. The reasons the patients die are still not entirely clear. It seems likely that the true explanation lies in the cardiotoxicity of the dextropropoxyphene combined with a respiratory depressant effect, also caused by the drug’s opioid action.

Dextropropoxyphene is an unusual opioid, in that it has a propensity to have actions on the heart which are not associated with its opioid analgesic effects. These actions are dose-related, and the abnormality caused is prolongation of the QRS complex on the ECG, an effect normally associated with drugs that block sodium channels in the heart. This effect is not seen at the normal therapeutic dose. It is, however, seen when patients take an excess dose and is then likely to be associated with severe cardiac arrhythmias. This may be more probable in patients who co-ingest additional respiratory depressants such as alcohol. The lack of oxygen that results, and the acidosis that follows, is likely to increase the risk of cardiac arrhythmia.

In practice, it is often difficult to establish the exact cause of death, but what is clear is that mortality with co-proxamol taken in overdose is around ten times that of an overdose of an alternative paracetamol-opioid combination tablet, such as co-codamol or co-dydramol.

Whose pills?

A fact probably not widely appreciated by the general public is that very many patients taking overdoses do not in fact use their own pills but those prescribed for a member of the same household. Thus, restricting prescriptions just to patients who are deemed at low-risk of self-harm may not be an effective way of preventing death from co-proxamol. This effect, combined with the tendency of opioid drugs to depress respiration, seems the likely reason for the much higher mortality rate in overdose.

Why withdraw the product?

Co-proxamol had been in use for over 30 years in the UK, and assessing benefit and risk in older drugs may be a problem. Traditionally, medicines licensing did not involve any routine assessment of the significance of a drug taken in overdose. The principle issues considered were effectiveness and safety in therapeutic doses. From this respect, co-proxamol seemed fine. Patients taking the drug as a painkiller are unlikely to develop problems. The difficulty is that co-proxamol taken in overdose is a different matter.

Isn’t it a good analgesic?

It seems to this author very likely that if co-proxamol had been shown to be a more effective analgesic than paracetamol alone it would have been much more difficult for the MHRA to institute a full withdrawal. There is certainly an impression amongst prescribing physicians that co-proxamol does, in some patients, produce a degree of euphoria, which is not seen with other opioid products. The evidence base behind this statement is absent, but many patients do feel that co-proxamol ‘suits them’ whereas other opioids do not. A key issue is that there is really no ideal ‘mid-range’ analgesic presently on the market. Paracetamol in clinical trials is extremely effective, but because it is available over the counter this affects patients’ perceptions of its effectiveness. An important factor is patient expectation, as clinical trials show that placebos produce effective pain relief in approximately 40% of patients receiving them. Furthermore, the prescribing physician, in particular his or her personality, also alters pain perception.

What happened after the withdrawal?

Patients on co-proxamol were, to a large extent, withdrawn from it, and an alternative analgesic was substituted. The result was a rapid reduction in prescription and a significant fall in deaths related to the drug. Equally important, toxicologists were unable to show any increase in mortality from other drugs, and suicide rates fell in proportion to those for co-proxamol. While this data requires confirmation in England, and further follow-up in Scotland, the message is very encouraging.

What about patients still on co-proxamol?

The MHRA suggested options for patients who had been taking co-proxamol for some time. In the author’s opinion, for most patients the appropriate analgesic substitute is paracetamol at the normal therapeutic dose of 1 g four times daily. Withdrawal from opiates is a problem, and in patients who have been receiving co-proxamol over a period of time, it seems sensible to tail off the dose over a few weeks or months by gradually substituting paracetamol for co-proxamol as individual tablets. For some patients this may be quite difficult to achieve, and with these individuals it may be worth considering an alternative combination analgesic.

One difficulty is that both codeine and dihydrocodeine are metabolised in the liver by the enzyme CYP2D6, which displays a polymorphism in the community: about 8% of the population does not possess adequate concentrations of enzyme to metabolise these compounds into their active metabolite. In the case of codeine this is morphine, and in dihydrocodeine it is dihydromorphine. In general, combination analgesics are less satisfactory, because the doses of opioids in them are often too high if taken in a full regimen of two tablets four times a day. The author’s preference is still to use separate preparations, and titrate according to the patient’s response. This does require them to take more tablets but often produces a better therapeutic effect, particularly in the elderly where accumulation of morphine-6-glucoronide, the active metabolite of morphine itself, may accumulate and contribute to adverse effects.

It may also be worth noting in patients with neuralgic-type pain who felt co-proxamol helped, that dextropropxyphene is a sodium channel blocker; in some patients this property may have contributed to its efficacy. Amitryptiline at low dose and carbamazepine are two other drugs worth considering in patients who find co-proxamol withdrawal a challenge.

Behind The Medical Headlines