OUP user menu

Opiate antagonist therapy for the pruritus of cholestasis: the avoidance of opioid withdrawal‐like reactions

E.A. JONES, J. NEUBERGER, N.V. BERGASA
DOI: http://dx.doi.org/10.1093/qjmed/95.8.547 547-552 First published online: 1 August 2002

Abstract

Increased opioidergic neurotransmission in the brain appears to contribute to the pruritus that complicates cholestasis and certain non‐cholestatic chronic liver diseases. Opiate antagonists have been shown to decrease scratching activity in patients with the pruritus of cholestasis. Initiation of oral administration of an orally bioavailable opiate antagonist may precipitate a florid opioid‐withdrawal‐like reaction in patients with pruritus complicating cholestasis. Such reactions can be minimized, or avoided completely, by cautiously infusing naloxone before giving small oral doses of an orally bioavailable opiate antagonist. The infusion rate of naloxone should initially be very low; it should be increased gradually and stopped when a rate known to be associated with opioid antagonist effects has been attained. Oral therapy with an opiate antagonist can then be initiated.

Introduction

Generalized pruritus, which may affect particularly the palms of the hands and the soles of the feet, is a common, and often distressing, complication of cholestasis and certain non‐cholestatic chronic liver diseases. The pruritus of cholestasis may limit normal activities, cause sleep deprivation and, when intractable, may lead to suicidal ideation.1 Examination of the skin may be normal or may reveal the consequences of scratching activity, such as excoriations or prurigo nodularis. Traditional therapies for the pruritus of cholestasis are empiric, lacking a sound scientific rationale based on an understanding of pathogenic mechanisms; they tend to have inconsistent effects and are usually only partially efficacious, at best.2

During the last decade, increased opioidergic neurotransmission (tone) in the brain has been postulated to be a component of the pathophysiology of cholestasis, which contributes to the pruritus of cholestasis.3,,4 For this hypothesis to be valid, it would be necessary to demonstrate that: (i) increased interaction between opioid agonists and opioid receptors in the brain is a mechanism of pruritus; (ii) increased opioidergic neurotransmission in the brain occurs in cholestasis; and (iii) opiate antagonists ameliorate the pruritus of cholestasis.

The pruritogenic effects of opioid agonists

Pruritus is a well‐recognized side‐effect of morphine.5 The intracisternal injection of morphine (0.2–0.5 mg/kg) into cats precipitates violent scratching activity, which lasts for up to 90 min.6 Furthermore, the injection of morphine (1–10 µg) or the opioid agonist (D‐Ala2N–Me‐Phe4, Gly5‐ol)‐enkephalin, but not saline, into the medullary dorsal horn of monkeys (Macaca fasicularis) induces dose‐dependent scratching activity, which is reversed or prevented by the opiate antagonist naloxone.7 Thus, opioid agonists, which increase opioidergic tone, induce opioid receptor‐mediated scratching activity of central origin.

Opioidergic tone in cholestasis

Several observations are consistent with opioidergic tone being increased in cholestasis: (i) oral administration of the experimental opiate antagonist nalmefene to patients with cholestasis due to primary biliary cirrhosis (PBC), who have not been treated with opiates, induces a transient reaction, the features of which resemble the withdrawal reaction of opiate addiction (see below);8 (ii) a rat model of acute cholestasis, but not a rat model of acute hepatocellular necrosis, exhibits decreased perception of pain (analgesia, antinociception), that is stereoselectively reversed by naloxone, and, consequently, is opioid receptor‐mediated;9 (iii) in a rat model of acute cholestasis, total plasma opioid activity is increased, and in the model and in patients with chronic cholestasis, plasma concentrations of individual endogenous opioid agonists are elevated;8,,10 (iv) in a rat model of acute cholestasis, µ‐opioid receptors in the brain are down‐regulated,11 a phenomenon that could arise as a consequence of changes in opioid receptor dynamics in response to increased availability of endogenous opioid agonists at opioid receptors; (v) plasma extracts from patients with PBC and pruritus, but not from patients with PBC without pruritus, induce naloxone‐reversible facial scratching activity when injected into the medullary dorsal horn of monkeys,12 suggesting that plasma of patients with PBC and pruritus contains one or more substances that induce central opioid‐receptor‐mediated scratching activity. Taken together, these five observations strongly suggest that central opioidergic tone is increased in the syndrome of cholestasis, possibly due to increased availability of endogenous opioid agonists at opioid receptors in the brain. The source of such increased levels of endogenous opioid peptides in cholestasis has not been established, but may be the cholestatic liver itself.13,,14

The effect of opiate antagonists on the pruritus of cholestasis

Because increased central opioidergic tone is a mechanism of pruritus, and central opioidergic tone appears to be increased in cholestasis, increased central opioidergic tone may contribute to the pruritus of cholestasis. If this is so, opiate antagonists may ameliorate the pruritus of cholestasis.

The first report of the effect of an opiate antagonist on the pruritus of cholestasis was published in 1979. A subcutaneous injection of naloxone, but not an injection of saline, was followed by an apparently dramatic amelioration of intractable pruritus in a patient with PBC.15 In a subsequent randomized, double‐blind, placebo‐controlled study involving 20 patients with cholestatic liver disease, intravenous naloxone appeared to modulate the perception of itch.16 Moreover, Thornton and Losowsky reported in 1988 that oral administration of nalmefene to nine patients with PBC was associated with apparently substantial ameliorations of pruritus, which appeared to be sustained in eight of the patients throughout a 6‐month course of the drug.8 The subjective observations made in these three studies suggested that opiate antagonists may ameliorate the pruritus of cholestasis. This inference was subsequently confirmed by the results of controlled clinical trials, the design of which included an objective quantitative primary efficacy endpoint,17–,20 specifically, 24‐h measurements of scratching activity, independent of arm or hand movements.21

In a single blinded trial, scratching activity was less in each of eight patients with pruritus due to PBC during intravenous infusions of naloxone than during placebo infusions; mean scratching activity was 50% less during naloxone infusions than during placebo infusions (p<0.001).17 This finding was confirmed in a randomized, double‐blind, placebo‐controlled trial involving 29 patients with pruritus complicating chronic cholestasis.18 In this trial, the mean of a subjective visual analogue score of pruritus was significantly less during naloxone infusions than during placebo infusions.18

In contrast to naloxone, nalmefene is an orally bioavailable opiate antagonist. It also has a more potent action at opioid receptors and a longer plasma half‐life than naloxone.19 Nalmefene is currently classified as an experimental opiate antagonist; it has not yet been licensed for clinical use. An open‐label trial of oral nalmefene therapy for the pruritus of cholestasis indicated that scratching activity was less during nalmefene therapy than before the drug was administered in 12 out of 14 patients.19 Furthermore, in a randomized, blinded, controlled trial, both scratching activity and a subjective visual analogue score of pruritus were significantly less during oral nalmefene therapy than when nalmefene was not being administered.20 Naltrexone is another orally bioavailable opiate antagonist, which, unlike nalmefene, is widely available and is licensed as an opiate antagonist for clinical use. In two controlled trials, in which a subjective score of pruritus was used as the primary endpoint, oral naltrexone therapy appeared to be associated with less pruritus than placebo treatment in patients with chronic cholestasis.22,,23

Thus, naloxone infusions may have a place in the emergency treatment of severe exacerbations of the pruritus of cholestasis,17,,18 and orally bioavailable opiate antagonists, such as nalmefene or naltrexone, may have a place in its long‐term management.8,19,20,22,,23

Opiate antagonist‐induced opioid withdrawal‐like reactions in patients with cholestasis

In their seminal 1988 paper, Thornton and Losowsky reported the consistent and abrupt precipitation of a reaction in patients with PBC following oral administration of a small dose (5 mg) of nalmefene. Features of the reaction included anorexia, nausea, colicky abdominal pains, tremor, insomnia, perspiration, increase in blood pressure, decrease in pulse, cool skin, mood change and hallucinations. The reaction was always transient, usually subsiding spontaneously after 2 or 3 days despite continued administration of the drug. Such a reaction does not occur when nalmefene is administered in high doses (e.g. 300 mg) to healthy subjects.8 The features of the reaction closely resemble the withdrawal reaction of opiate addiction. The occurrence of such a reaction in patients with a chronic cholestatic liver disease, who had not ingested opiates, by itself strongly suggests that central opioidergic tone is increased in cholestasis.

Symptoms and signs compatible with opioid withdrawal‐like phenomena have been reported in other studies in which opiate antagonists have been administered to patients with cholestasis.18–20,22–,24 In one of the two trials of naloxone infusions, one patient developed anxiety, restlessness and an increase in diastolic blood pressure during an initial naloxone infusion and also, to a lesser extent, during a subsequent naloxone infusion. Three other patients in this study developed ill‐defined anxiety during naloxone infusions.18 In the nalmefene trials, symptoms that some patients developed during nalmefene therapy included the perception of pins and needles, anxiety and depression, abdominal cramps and nausea, insomnia, difficulties in visual focusing, dizziness, chronic goose‐bumps, depersonalization, mental ‘fuzziness’, anorexia, nightmares, changes in mood, generalized discomfort and chest tightness. These symptoms did not prevent continued administration of the drug. They were always transient, they responded to dose reduction and they did not recur with subsequent dose escalation.19,,20 In the naltrexone trials, symptoms associated with naltrexone therapy in some patients included general malaise, nausea and vomiting, dizziness, flushing, drowsiness, headache, nightmares, tremor, abdominal cramps, dry mouth, and night sweats. These symptoms were also transient, usually subsiding within three days despite continued drug administration.22,,23

In general, opioid withdrawal‐like phenomena have been less overt and troublesome in cholestatic patients following the initiation of a naloxone infusion (0.2 µg/kg/min),17,,18 than following the initiation of oral therapy with an orally bioavailable opiate antagonist (e.g. nalmefene 5 mg or naltrexone 25 mg.8,19,20,22–,24 The severity of an opiate antagonist‐induced opioid withdrawal‐like reaction in a cholestatic patient may depend on the extent to which opioidergic tone is increased, the potency of the opiate antagonist and opioid receptor occupancy by the antagonist (a function of dose).

The opioid withdrawal‐like reactions that may occur following the initiation of opiate antagonist therapy in cholestatic patients do not appear to be dangerous, but they can be distressing, or even alarming, for patients, relatives and attending medical staff. The question arises whether an opiate antagonist‐induced opoid withdrawal‐like reaction can be avoided when initiating opiate antagonist therapy in patients with the pruritus of cholestasis.

Optimizing the initiation of oral opiate antagonist therapy in cholestatic patients

Three approaches to preventing or minimizing opioid withdrawal‐like reactions when initiating oral opiate antagonist therapy in cholestatic patients have been adopted.

The first of these, described in Thornton and Losowsky's original report,8 was to co‐administer clonidine with nalmefene. Clonidine ameliorates the symptoms of opiate withdrawal reactions in drug addicts.25 The reaction of cholestatic patients to nalmefene was so severe that patients were admitted to hospital for initiation of nalmefene therapy. The dose of clonidine was 100 mg three times a day for the first three days, followed by a decreasing dose schedule culminating in discontinuation of the drug on day 7. Co‐administration of clonidine appeared to decrease the severity of opioid withdrawal‐like phenomena, but it did not prevent their occurrence.

The second approach was to start therapy with a very small dose of an orally‐administered opiate antagonist, and to double the dose, say every 2 or 3 days, until a satisfactory therapeutic response was obtained. This approach may well have decreased the incidence or intensity of opioid withdrawal‐like phenomena, but it did not prevent their occurrence in the two trials of oral nalmefene for the pruritus of cholestasis.19,,20 Furthermore, this approach is not always effective in preventing or minimizing opioid withdrawal‐like reactions when initiating oral naltrexone therapy in cholestatic patients. For example, in the management of a 35‐year‐old woman with pruritus complicating PBC, it was decided to start oral naltrexone therapy with a very low dose to prevent or minimize an opioid withdrawal‐like reaction. However, naltrexone 12.5 mg (i.e. a quarter of a 50 mg tablet) precipitated a florid opioid withdrawal‐like reaction in this patient. The reaction included hallucinations, irrational behaviour and intense pruritus of the tongue, and lasted for 2.5 h. Four weeks later, naltrexone 2 mg precipitated a similar reaction in the same patient. The second reaction lasted 1.25 h.24 In the two controlled trials of naltrexone for the pruritus of cholestasis, adequate precautions to avoid or minimize opioid withdrawal‐like phenomena were not taken.22,,23

A third approach was adopted to try and initiate oral naltrexone therapy in the 35‐year old patient with PBC mentioned above.24 About 6 months after her reaction to the oral administration of naltrexone 2 mg, she was admitted to hospital for an intravenous infusion of naloxone. The initial infusion rate was very low (0.002 µg/kg/min). From 1 to 3 h after the start of the infusion, she experienced mild itching of the tongue and generalized paraesthesiae. The infusion rate was gradually increased over 5 days to one known to be associated with opioid antagonist effects (0.2 µg/kg/min).17 Oral naltrexone was then introduced at a dose of 12.5 mg twice daily. There was no reaction, and the naloxone infusion was discontinued. Throughout the next 12 months, she took 25 mg naltrexone daily and was completely free from pruritus.24 This approach to initiating oral naltrexone therapy in patients with the pruritus of cholestasis has been applied successfully to eight other patients with persistent pruritus complicating liver disease (Table 1). Maintenance doses have varied from 25 mg to 250 mg daily.

When applying this approach, after a predetermined period of infusing naloxone, the infusion rate can be doubled if no new symptoms or signs have developed. On the other hand, if a new symptom or sign develops, the infusion rate is not increased further until that symptom or sign has resolved. The approach adopted so far may well have been too cautious. With increasing clinical experience, it may be possible to shorten the duration of the naloxone infusion substantially. Indeed, it may be possible to initiate oral naltrexone therapy in cholestatic patients using this approach in a hospital day‐care setting.

View this table:
Table 1

Use of naloxone infusion in initiation of oral opiate antagonist therapy in nine patients with pruritus complicating liver disease

DiagnosisAge (years)SexDuration of naloxone infusion*ReactionInitial naltrexone dosageOutcome**Source
Primary biliary cirrhosis35F5 daysMild itching of tongue and paraesthesiae 1–3 h after starting infusion12.5 mg twice dailyPruritus‐freeReference 24
Primary biliary cirrhosis49F3 daysNone12.5 mg thrice dailyPruritus‐freeReference 26
Chronic hepatitis B56M7 daysNone25 mg twice dailyPruritus‐freeEAJ
Hepatic sarcoidosis31F29 hNone12.5 mg thrice dailyPruritus not troublesomeEAJ
Benign recurrent intrahepatic cholestasis54M5 daysNone12.5 mg twice dailyPruritus not troublesomeEAJ
Chronic hepatitis C53F29 hNone12.5 mg thrice dailyPruritus‐freeEAJ
Primary biliary cirrhosis56F22 hNone12.5 mg thrice dailyPruritus‐freeEAJ
Chronic hepatitis C66M48 hNone25 mg thrice dailyPruritus‐freeEAJ
Chronic hepatitis C73M36 hNone12.5 mg thrice dailyPruritus‐freeEAJ
  • *Initial infusion rate 0.002 µg/kg/min. Final infusion rate 0.2 µg/kg/min. The infusion rate was not increased while patients were sleeping. **On maintenance dose of naltrexone. EAJ, one of the authors, evaluated and treated specific patients at the Academic Medical Center, Amsterdam, as indicated in the last column.

Conclusions

Evidence supporting the hypotheses that increased opioid‐mediated neurotransmission in the brain is a mechanism of pruritus and that central opioidergic tone is increased in cholestasis, provide a rationale for treating the pruritus of cholestasis with opiate antagonists.

In controlled trials, opiate antagonists (intravenously administered naloxone or orally administered nalmefene) have been shown to reduce scratching activity by patients with the pruritus of cholestasis.

Intravenous naloxone has potential for emergency treatment of exacerbations of the pruritus of cholestasis, whereas orally bioavailable opiate antagonists, such as nalmefene or naltrexone, have potential in the long‐term management of the pruritus of cholestasis. Opiate antagonist therapy is having a positive impact on the management of the pruritus of cholestasis. For example, it is already clear that liver transplantation for intractable pruritus is contraindicated until after an adequate trial of opiate antagonist therapy.26

Initiation of treatment with an orally administered opiate antagonist may precipitate an opioid withdrawal‐like reaction in cholestatic patients.

Such reactions can probably be avoided by starting opiate antagonist therapy with an intravenous infusion of naloxone. The infusion rate should initially be very low. It should gradually be increased to a rate known to be associated with opioid antagonist effects. The infusion can then be stopped, and small oral doses of an orally bioavailable opiate antagonist can be given. The oral dosage should be gradually increased until an optimal therapeutic response is obtained.

Footnotes

  • Address correspondence to Dr E.A. Jones, Department of Gastrointestinal and Liver Diseases, Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam‐ZO, The Netherlands. e‐mail: e.a.jones{at}amc.uva.nl

References

View Abstract