Electrophysiologic properties of propoxyphene and norpropoxyphene in canine cardiac conducting tissues in vitro and in vivo

Propoxyphene and the risk of out-of-hospital death

PURPOSE

The opioid analgesic propoxyphene was withdrawn from the US market in 2010, motivated by concerns regarding fatality in overdose and adverse cardiac effects, including prolongation of the QT interval. These concerns were based on case reports, summary vital statistics, and surrogate endpoint studies.

METHODS

Using the linked Tennessee Medicaid database (1992-2007), we conducted a retrospective cohort study that compared risk of sudden cardiac, medication toxicity, and total out-of-hospital death for propoxyphene users with that for comparable nonusers of any prescribed opioid analgesic and users of hydrocodone, an opioid with similar indications. Cohort members had 1,873,500 propoxyphene prescriptions, 1,873,500 matched nonuser control periods, and 936,750 matched hydrocodone prescriptions.

RESULTS

Current propoxyphene users had no increased risk for sudden cardiac death (versus nonusers: hazard ratio [HR] = 1.00 [0.81-1.23]; versus current hydrocodone users: HR = 0.91 [0.68-1.21]) but did have increased risk for medication toxicity deaths (versus nonusers: HR = 1.85 [1.07-3.19], p = 0.027; versus current hydrocodone users: HR = 2.10 [0.87-5.10], p = 0.100). Because toxicity deaths were a small proportion of study deaths, total out-of-hospital mortality differed by less than 10% between the study groups and was not significantly elevated for propoxyphene (versus nonusers: HR = 1.09 [0.95-1.25]; versus current hydrocodone users: HR = 1.06 [0.87-1.29] ).

CONCLUSIONS

Our findings support the concern that propoxyphene has greater toxicity in overdose but do not provide evidence that it increases the risk of sudden cardiac death.

A systematic review of the use of opioid medication for those with moderate to severe cancer pain and renal impairment: a European Palliative Care Research Collaborative opioid guidelines project

BACKGROUND

Opioid use in patients with renal impairment can lead to increased adverse effects. Opioids differ in their effect in renal impairment in both efficacy and tolerability. This systematic literature review forms the basis of guidelines for opioid use in renal impairment and cancer pain as part of the European Palliative Care Research Collaborative's opioid guidelines project.

OBJECTIVE

The objective of this study was to identify and assess the quality of evidence for the safe and effective use of opioids for the relief of cancer pain in patients with renal impairment and to produce guidelines.

SEARCH STRATEGY

The Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials, MedLine, EMBASE and CINAHL were systematically searched in addition to hand searching of relevant journals.

SELECTION CRITERIA

Studies were included if they reported a clinical outcome relevant to the use of selected opioids in cancer-related pain and renal impairment. The selected opioids were morphine, diamorphine, codeine, dextropropoxyphene, dihydrocodeine, oxycodone, hydromorphone, buprenorphine, tramadol, alfentanil, fentanyl, sufentanil, remifentanil, pethidine and methadone. No direct comparator was required for inclusion. Studies assessing the long-term efficacy of opioids during dialysis were excluded.

DATA COLLECTION AND ANALYSIS

This is a narrative systematic review and no meta-analysis was performed. The Grading of RECOMMENDATIONS Assessment, Development and Evaluation (GRADE) approach was used to assess the quality of the studies and to formulate guidelines.

MAIN RESULTS

Fifteen original articles were identified. Eight prospective and seven retrospective clinical studies were identified but no randomized controlled trials. No results were found for diamorphine, codeine, dihydrocodeine, buprenorphine, tramadol, dextropropoxyphene, methadone or remifentanil.

CONCLUSIONS

All of the studies identified have a significant risk of bias inherent in the study methodology and there is additional significant risk of publication bias. Overall evidence is of very low quality. The direct clinical evidence in cancer-related pain and renal impairment is insufficient to allow formulation of guidelines but is suggestive of significant differences in risk between opioids.

RECOMMENDATIONS

RECOMMENDATIONS regarding opioid use in renal impairment and cancer pain are made on the basis of pharmacokinetic data, extrapolation from non-cancer pain studies and from clinical experience. The risk of opioid use in renal impairment is stratified according to the activity of opioid metabolites, potential for accumulation and reports of successful or harmful use. Fentanyl, alfentanil and methadone are identified, with caveats, as the least likely to cause harm when used appropriately. Morphine may be associated with toxicity in patients with renal impairment. Unwanted side effects with morphine may be satisfactorily dealt with by either increasing the dosing interval or reducing the 24 hour dose or by switching to an alternative opioid.

Arrhythmias associated with drug toxicity

Cardiac arrhythmias may result from poisoning by a variety of agents. Although some cardiac disturbances are of limited concern, others occur because of the specific cardiac properties of the ingested drugs, and require specific antidotes. Clinicians must understand the resultant pathophysiology so that therapy is appropriate. This article focuses on drugs that cause ventricular arrhythmias and bradyarrhythmias.

Methadone block of K+ current in squid giant fiber lobe neurons

Voltage-dependent ionic currents were recorded from squid giant fiber lobe neurons using the whole-cell patch-clamp technique. When applied to the bathing solution, methadone was found to block IK, I Na and I Ca. Both I Na and I Ca were reduced without apparent change in kinetics and exhibited IC(50)'s of 50-100 and 250-500 mu M, respectively, at +10 mV. In contrast, IK was reduced in a time-dependent manner that is well fit by a simple model of open channel block (K(D)= 32+/- or 2 mu M, +60 mV, 10 degrees Celsius). The mechanism of I(K) block was examined in detail and involves a direct action of methadone, a tertiary amine, on K channels rather than an opioid receptor-mediated pathway. The kinetics of I(K) block resemble those reported for internally applied long chain quaternary ammonium (QA) compounds; and recovery from I(K) block is QA-like in its slow time course and strong dependence on holding potential. A quaternary derivative of methadone (N-methyl-methadone) only reproduced the effects of methadone on I(K) when included in the pipette solution; this compound was without effect when applied externally. I(K) block thus appears to involve diffusion of methadone into the cytoplasm and occlusion of the open K channel at the internal QA blocking site by the protonated form of the drug. This proposed mode of action is supported by the pH and voltage dependence of block as well as by the observation that high external K+ speeds the rate of drug dissociation. In addition, the effect of methadone on I(K) evoked during prolonged (300 ms) depolarizations suggests that methadone block may interfere with endogenous K+ channel inactivation. The effects of temperature, methadone stereoisomers, and the methadone-like drugs propoxyphene and nor-propoxyphene on IK block were examined. Methadone was also found to block I(K) in GH3 cells and in chick myoblasts.

Adverse cardiac manifestations following dextropropoxyphene overdose: can naloxone be helpful?

Dextropropoxyphene overdose may be complicated by serious cardiovascular manifestations, including conduction abnormalities and collapse. We report two patients in whom cardiac toxicity developed. Cardiovascular depression seemed to be improved after naloxone infusion in these two cases. Possible mechanisms are briefly discussed.

Interaction between ethanol and opioids in a protozoan assay

Ethanol excess combined with opioids can be fatal due to their toxic interaction, yet the nature of the interaction is little documented. Since ethanol and some opioids have membrane stabilizing activity, the present study used a protozoan motility model to test the possibility that ethanol may interact with some opioids on this basis. The EC50 in motility reduction for ethanol, dextropropoxyphene, methadone and pethidine was 522.0 +/- 36.7 mM, 0.59 +/- 0.08 mM, 0.40 +/- 0.09 mM and 4.57 +/- 0.36 mM, respectively. When ethanol was combined with one of the three drugs in equitoxic doses at a ratio of 0.5:0.5, the predicted/observed EC50 values for ethanol-dextropropoxyphene, ethanol-methadone and ethanol-pethidine were 1.37, 1.11 and 1.00, each being close to unity, indicating an additive interaction. The interaction between ethanol and dextropropoxyphene was further explored at 0.25:0.75 and 0.75:0.25 equitoxic dose ratios, with the predicted/observed EC50 values of 0.98 and 0.97, also showing an additive interaction. This suggests that a non-specific interaction between ethanol and opioids may also take place in vivo, which could cause increased toxicity over and above the involvement of opioid receptors. Information from this study should aid understanding of the mechanism of interactions in human poisoning by agents with membrane stabilizing activity.

Marked QRS complex abnormalities and sodium channel blockade by propoxyphene reversed with lidocaine

The opiate analgesic propoxyphene produces cardiac toxicity when taken in overdose. We recently observed a patient with propoxyphene overdose in whom marked QRS widening was reversed by lidocaine. The reversal is apparently paradoxical as both agents block the inward sodium current (INa). We examined possible mechanisms of the reversal by measuring INa in rabbit atrial myocytes during exposure to propoxyphene and the combination of propoxyphene and lidocaine (60 and 80 microM, respectively). Propoxyphene caused use-dependent block of INa during pulse train stimulation. Block recovered slowly with time constants of 20.8 +/- 3.9 s. Block during lidocaine exposure recovered with time constants of 2-3 s. During exposure to the mixture, block recovered as a double exponential. The half time for recovery during exposure to the mixture was 1.6 +/- .9 s compared with a half-time of 14.3 +/- 2.9 s during exposure to propoxyphene alone. During pulse train stimulation, less steady-state block was observed during exposure to the mixture than during exposure to propoxyphene alone when the interval between pulses was greater than 0.95 s. Both drugs compete for a common receptor during the polarizing phase. The more rapid dissociation of lidocaine during the recovery period leads to less block during the mixture than during exposure to propoxyphene alone. The experiments suggest a mechanism for reversal of the cardiac toxicity of drugs which have slow unbinding kinetics.

Dextropropoxyphene overdose. Epidemiology, clinical presentation and management

This paper comprehensively reviews the worldwide situation regarding acute overdosage of dextropropoxyphene (propoxyphene). The changing epidemiology of this type of poisoning over the last 20 years is described with discussion of concurrent trends and, in particular, the effects of different preventive measures adopted in various countries. The clinical pharmacology of dextropropoxyphene relevant to the clinical toxic effects resulting from acute overdosage is described, and the management is detailed. In particular, the importance of early diagnosis and treatment is stressed in view of the potentially lethal complications that may suddenly occur with this poisoning. Recommendations for the correct use of the specific narcotic antagonist, naloxone, are made, together with other intensive supportive measures. As dextropropoxyphene is frequently taken together with other toxic agents, the concomitant effects of alcohol and sedative drugs are described and the treatment of paracetamol (acetaminophen) in combination with dextropropoxyphene is emphasised. The most effective preventive measures for the future are suggested, but caution is advised regarding the prescription for 'at risk' patients of alternative analgesics, which may be no safer in overdosage.

The effects of naloxone on central hemodynamics and myocardial metabolism in experimental propoxyphene-induced circulatory shock

The courses of the hemodynamic and cardiometabolic effects of naloxone were evaluated in propoxyphene-induced shock in eight pentobarbital-anesthetized pigs. Circulatory shock was induced by an infusion of propoxyphene chloride 15 mg . min-1 i.v. At shock, i.e. MAP less than 60 mmHg and/or CI less than 2.0 l . min-1 . m-2, naloxone was administered at 0.75, 1.5 and 3.0 mg . kg-1 with an interval between increments of 8 min. The propoxyphene infusion of 15 mg . min-1 was continued throughout the study. Following the injection of naloxone 0.75 mg . kg-1, increases were observed (% of baseline value) in MAP (41%), i.e. deficit to baseline 59%, HR (66%), CI (67%) and SVI (108%), whereas MPAP and MPAOP were unchanged. dP/dt increased (34%). In the coronary circulation naloxone initiated the following changes: CSF increased (69%) as did MVO2 (48%) with unchanged MO2-extraction, but CVR decreased further (36%). The maximum effects of naloxone were registered 2-3 min after 0.75 mg . kg-1. Following 1.5 and 3.0 mg . kg-1, no changes in hemodynamics were observed other than those caused by progressing propoxyphene intoxication.

The effects of cardiac pacing on central hemodynamics in experimental propoxyphene-induced cardiac failure

Coronary sinus pacing was evaluated in 10 pigs during propoxyphene-induced cardiac failure. From baseline, propoxyphene chloride 15 mg . min-1 was infused until circulatory shock developed. Cardiac pacing was evaluated at different dose levels expressed as % of the shock dose of propoxyphene: at intoxication levels below 50% of the shock dose, cardiac pacing improved cardiac performance. At dose levels above 50% of the shock dose cardiac performance deteriorated further during pacing. The results are consistent with a severe negative inotropic effect of propoxyphene in overdose.

Ethanol does not increase lethality due to propoxyphene in rats

Male Wistar rats (210-330 g) were used as test animals. Propoxyphene (175 mg/kg) and ethanol (2 g/kg) were administered by gastric intubation, naloxone (2 mg/kg) by subcutaneous injection. Four groups, each consisting of 19 rats received either of the following drug treatments: Propoxyphene; ethanol + propoxyphene; naloxone + propoxyphene; and naloxone + ethanol + propoxyphene. The drugs were given in the sequence mentioned at the beginning of the experiment. Naloxone was also given 45 and 90 min later. Mortality was reduced to 42% in the group that received ethanol and propoxyphene as compared to 73% in the group that received propoxyphene only. Naloxone protected against lethality in both groups. A rise in the propoxyphene/norpropoxyphene (P/N) ratio due to an increase in the absolute concentrations of propoxyphene and a decrease in the absolute levels of norpropoxyphene in blood, brain and heart tissues was observed in the ethanol + propoxyphene group, compared to the propoxyphene group. Although these pharmacokinetic data indicate impaired propoxyphene metabolism in the presence of ethanol, ethanol did not enhance propoxyphene induced lethality. This is also contrary to suggestions from previous studies. Our results demonstrate that at least in one species and at one dose ratio (ethanol/propoxyphene) ethanol might reduce the lethality caused by propoxyphene alone. This suggests antagonism between the two drugs, probably in the central nervous system.

Circulatory shock following intravenous propoxyphene poisoning. An experimental study of cardiac function and metabolism in pentobarbital-anesthetized pigs

The effects of continuously administered intravenous propoxyphene chloride (15 mg X min-1) on ECG, systemic, pulmonary and coronary circulations and myocardial oxygenation were investigated in eight pentobarbital-anesthetized pigs. Circulatory shock, defined as a systolic blood pressure below 60 mmHg (8.0 kPa) and a cardiac output of approximately 2.0 l X min-1 X m-2, occurred after 675 to 2025 mg propoxyphene chloride. At the time when shock occurred plasma concentrations of propoxyphene ranged from 9.6 to 15.3 micrograms X ml-1 which is within the range of the lethal concentration observed in man. Statistically significant decreases were observed for the following variables: maximum rate of rise of left ventricular pressure dP/dt (-90%), mean arterial pressure (-73%), heart rate (-46%), cardiac index (-58%), stroke volume index (-22%), left ventricular stroke work index (-85%), right ventricular stroke work index (-63%) and systemic vascular resistance (-50%). Mean pulmonary arteriolar occlusion pressure increased (+42%), whereas mean right atrial pressure and pulmonary vascular resistance remained unchanged. The arteriovenous oxygen difference increased (+53%) and total body oxygen consumption decreased (-35%). The following coronary variables decreased: coronary sinus blood flow (-57%), coronary vascular resistance (-65%), myocardial oxygen consumption (-68%), myocardial oxygen extraction (-26%) and myocardial lactate extraction (-28%). Prolongation of the ECG PQ and QRS intervals were recorded shortly before shock appeared, and all animals were in sinus rhythm till the last minute before death. The results indicate that intravenously administered propoxyphene besides being a powerful negative inotropic and chronotropic agent, is also a potent systemic and coronary vasodilator.

Dextropropoxyphene kinetics after single and repeated oral doses in man

The kinetics of dextropropoxyphene (DP) and its main metabolite norpropoxyphene (NP) were studied in 6 healthy male subjects after a single oral dose of 195 mg DP HCl, and during and after 12 daily single oral doses of 195 mg DP HCl. The kinetics varied up to five-fold between individuals after the single dose, the apparent mean elimination half-life (t1/2) was 16 h for DP and 29 h for NP. The mean apparent overall plasma clearance (CL) for DP was 2.61/min. There was no systematic difference in DP clearance between the single and multiple doses, but the accuracy of individual predictions from single to multiple doses was poor, probably because of imprecise determinations of the AUC and t1/2 in the single dose experiments. The individual correlation between single and multiple dose kinetics was good for NP, although the predicted plasma levels during steady state were significantly higher than the observed levels (mean AUCss/AUCsd: 0.81). There was no sign of saturation kinetics on repeated administration. In fact, autoinduction, resulting in significantly lower plasma concentrations after treatment for 1 week was found for NP and was indicated for DP. On discontinuing DP after 12 days of treatment, the apparent mean t1/2 of DP was 23 h and of NP 25 h.

Severe, acute propoxyphene overdose treated with dopamine

Six patients suffering from cardiovascular failure due to acute propoxyphene overdose, were treated with dopamine infusion in doses of 2-17 microgram/kg/min. All patients responded with increased systolic arterial blood pressure, increased urinary output and decreased central venous pressure. The two most severely poisoned patients showed no increase in heart rate in spite of dopamine infusion in positively chronotropic doses. In two out of three patients the initially abnormal ECGs normalised during treatment. It is concluded that dopamine seems suitable for reversal of propoxyphene induced circulatory failure.

Lack of evidence of increased lethality due to propoxyphene overdose in the presence of ethanol in male Wistar rats

The primary purpose of the present investigation was to evaluate if the presence of ethanol increased lethality induced by propoxyphene. A secondary aim was to study the effect of naloxone on propoxyphene lethality alone, and on the concomitant administration of propoxyphene and ethanol. Male Wistar rats (210-330 g) were used as test animals. Propoxyphene (175 mg/kg) and ethanol (2 g/kg) were administered by gastric intubation, naloxone (2 mg/kg) by subcutaneous injection. Four groups, each consisting of 19 rats, received either of the following drug treatments: Propoxyphene, ethanol + propoxyphene, naloxone + propoxyphene, and naloxone + ethanol + propoxyphene respectively. The drugs were given in the sequence mentioned at the beginning of the experiment. Naloxone was also given 45 and 90 min later. Mortality was reduced to 42% in the group that received ethanol and propoxyphene compared to 73% in the group that received propoxyphene only. Naloxone protected against lethality in both groups. Some animals died despite naloxone administration, possibly due to a nonopioid cardiotoxic effect of propoxyphene or its metabolite. An increase in the propoxyphene/norpropoxyphene (P/N) ratio due to an increase in the absolute concentrations of propoxyphene and a decrease in the absolute levels of norpropoxyphene in blood, brain, and heart tissues was observed in the ethanol + propoxyphene group, compared to the propoxyphene group. In the animals which died, the highest P/N ratio was observed in brain tissue and the lowest in heart muscle. Despite the pharmacokinetic data obtained in this investigation indicating impaired propoxyphene metabolism in the presence of ethanol, ethanol did not enhance propoxyphene-induced lethality.(ABSTRACT TRUNCATED AT 250 WORDS)

Dextropropoxyphene poisoning: problems with interpretation of analytical data

Difficulties in the estimation and interpretation of dextropropoxyphene and norpropoxyphene analyses have added to the controversy concerning the toxicity of these compounds. Poor performance in external proficiency assessment trials reflects the unsatisfactory state of many dextropropoxyphene and norpropoxyphene analyses in general use. There is substantial overlap between blood concentrations of dextropropoxyphene and norpropoxyphene encountered in chronic therapeutic dose and those associated with acute toxicity. As dextropropoxyphene and norpropoxyphene are often taken in overdose with other drugs, their blood concentrations must be interpreted in the context of careful identification and quantitation of such agents. The present availability of accurate methods for measurement of parent drug and metabolite should now make anecdotal reports unsupported by analytical data entirely superfluous.

Left bundle branch block associated with propoxyphene hydrochloride poisoning

Described is the case of a patient who developed left bundle branch block following acute propoxyphene hydrochloride overdosage. The left bundle branch block was transient and associated with no permanent sequelae. Previously documented cardiac abnormalities, specific narcotic antagonist therapy, and animal studies correlating cardiac toxicity of propoxyphene hydrochloride with its potent local anesthetic action are discussed.

Failure of naloxone to reverse the cardiotoxocity of Distalgesic overdose

A case of severe poisoning with dextropropoxyphene and paracetamol (Distalgesic) in which the cardiotoxic effects of dextropropoxyphene were not reversed by naloxone is reported. A probable reason for this failure is suggested with reference to the differing pharmacological actions of dextropropoxyphene on cardiac conducting tissue and the central nervous system.