Treatment of the pruritus of cholestasis: From unbearable lightness to substance



Background & Aims

To assess the efficacy and safety of naltrexone for the short and long term treatment of pruritus of cholestasis.


Twenty patients with pruritus and cholestasis were included. A baseline pruritus score was obtained over 1 week. Patients were then randomized to receive 50 mg/day of naltrexone or placebo for 2 weeks. Subsequently, a 1-week washout period ensued and patients were crossed over to the other therapy for 2 additional weeks. Pruritus was assessed daily with a visual analogue scale (VAS) from 0 to 10. Patients whose pruritus decreased >50% of basal with naltrexone received naltrexone 50 mg/day for 2 additional months.


Mean basal VAS was similar in both groups. VAS showed greater and more significant changes with naltrexone than with placebo (P < .0003). In nine out of 20 patients (45%) receiving naltrexone, pruritus decreased >50% compared to basal value, including five whose pruritus disappeared completely. No significant changes were observed in serum biochemistry. Most of the adverse events that occurred during the first 48 h of naltrexone therapy were consistent with opioid withdrawal-like phenomena and spontaneously disappeared 2 days after starting treatment.


Naltrexone can be considered as an alternative option to treat pruritus of cholestasis. In the current study, side effects were transient and did not require specific medication.

Controlled trials have suggested that opiate antagonist therapy may be effective for the treatment of the symptoms of cholestasis. The oral opiate antagonist naltrexone in particular has started to enter into routine clinical use for amelioration of cholestatic itch. Attention regarding the side effects of opiate antagonist therapy has, to date, largely focused on an opiate withdrawal–type reaction (which can be controlled effectively by titrated therapy introduction regimens). Here we describe three cases of a further clinically important side effect: loss of control of pain resulting from other pathologies, which in each case necessitated the withdrawal of hitherto clinically effective opiate antagonist therapy. Of the 14 patients treated by our unit with opiate antagonist agents for the control of cholestatic symptoms, 13 (93%) showed resolution of or significant improvement in symptoms. Of the 13 patients showing a clinical response, seven (54%) subsequently had to discontinue therapy because of side effects (including the three patients with uncontrolled pain). It is our experience that in the routine clinical setting, opiate antagonists are highly effective for the treatment of cholestatic symptoms. In practice, however, their usefulness is limited by their side effect profile.

Terg R, Coronel E, Sorda J, Munoz AE, Findor J. Efficacy and safety of oral naltrexone treatment for pruritus of cholestasis, a crossover, double blind, placebo-controlled study. J Hepatol 2002;37:717–722. (Reprinted with permission from Elsevier.) McRae CA, Prince MI, Hudson M, Day CP, James OF, Jones DE. Pain as a complication of use of opiate antagonists for symptom control in cholestasis. Gastroenterology 2003;125:591–6. (Reprinted with permission.)


Two articles concerning the use of opiate antagonists for the treatment of the pruritus of cholestasis have been published in the last several months.1, 2

The first publication concerns a randomized, double-blind, placebo-controlled crossover study of naltrexone for the treatment of the pruritus of cholestasis conducted in Argentina in a heterogeneous group of 20 patients.1 The duration of each treatment phase was 2 weeks, with a 1-week washout period between them. The administration of naltrexone was associated with a significant decrease (<50%) in the mean visual analogue score of pruritus as assessed subjectively by a visual analogue scale in 45% of the patients. In contrast, the administration of placebo had no effect on the mean pruritus visual analogue score. Side effects included insomnia, nausea, and vomiting. Some of these (e.g., nausea and vomiting) occurred in association with both naltrexone and placebo, but they were 1.5 to 8 times more frequent during treatment with the drug than during treatment with the placebo. Two patients dropped out of the study on the third day of treatment and one patient did not wish to continue taking naltrexone beyond 48 hours because of persistent side effects. At the end of the controlled double-blind study phase, the nine patients who had responded to naltrexone with a decrease in pruritus were treated for 2 additional months. Of the nine, five patients consistently experienced relief of the pruritus on the drug, while the other four either had persistent side effects or failed to experience consistent relief and did not continue to take the drug. The authors state that most side effects occurred during the first 24 hours of treatment and disappeared spontaneously. The occurrence of an opiate withdrawal–like reaction immediately after the initiation of drug therapy, and its disappearance over the first few hours of treatment is consistent with our experience in treating patients with the opiate antagonists nalmefene (prescribed as oral preparation) and naloxone (to a lesser degree than with the oral preparations) in clinical trials of patients with cholestasis and pruritus.3–6 The ameliorating effect of an opiate antagonist on the pruritus of cholestasis was confirmed by Terg et al.1 in Argentina, a population of patients that might be different from that studied in England,7 the United States,3–6, 8 and The Netherlands9 in which opiate antagonists were also associated with amelioration of the pruritus. The limitation of this study is that the methodology used was subjective.10 Considering, however, that opiate antagonists have been tested in studies that applied objective methodology,3–6 the study of Terg et al. supports further a role of opiate antagonists for the treatment of the pruritus of cholestasis.11

The second publication is a report of three cases and a review of the work that has provided a rationale for the use of opiate antagonists for the treatment of the pruritus of cholestasis.2 The authors report that three patients who were being treated satisfactorily with opiate antagonists for the pruritus of cholestasis experienced pain associated with normally painful conditions. The endogenous opioid system is a survival system whose function includes endogenous analgesia.12, 13 Accordingly, the resurgence of pain secondary to postherpetic neuralgia after it had been adequately controlled with an antidepressant (case 1), the appearance of acute pain associated with hepatomegaly of uncertain etiology not responding to acetaminophen (which is consistent with the behavior of visceral pain), and the difficulties controlling joint pains in polyarthralgia considered to be uric acid mediated with nonsteroidal anti-inflammatory drugs do not seem surprising in the face of treatment with opiate antagonists. Furthermore, pain associated with inflammatory conditions—such as those experienced by the patients reported by McRae et al.2 while on naltrexone—may be explained by the enhanced sensitivity to noxious stimulus that results from inflammation, and that can be accentuated by opiate antagonists.14, 15 According to their report, half of the patients that were treated with naltrexone for the pruritus of cholestasis in their unit had the drug discontinued because of side effects, but 93% of the patients treated experienced relief of their pruritus while on the opiate antagonist.2 The authors state that because of the side effects, the use of opiate antagonists to treat patients with the pruritus of cholestasis should not be generalized2; a more balanced recommendation seems in order given the reported success in using opiate antagonists to treat the pruritus by these and other investigators.1, 3–9

The importance of opiate antagonists as a therapeutic alternative for the treatment of the pruritus of cholestasis includes the identification, for the first time, of a system that participates in the mediation of the pruritus, the provision of specific (i.e., opiate antagonists) and not empirical treatment and the definition of a line of research in an area of marked relevance to the quality of life of patients with liver disease.

Activation of the endogenous opioid system is associated with analgesia, as is the administration of opiate drugs.16, 17 Morphine, which elicits its effect by binding to opioid receptors, is one of the most frequently used drugs in the management of acute pain. Clinical observations in human beings, behavioral studies in animals, and data from basic neurophysiology experiments suggest that activation of the endogenous opioid system also results in pruritus and scratching, as follows: (1) the central administration (e.g., intrathecally) of morphine is associated with pruritus in human beings, which can be effectively treated with opiate antagonists,18, 19, 20–22 (2) the central administration of morphine and the opiate agonist ligand D-ala2-MePhe4,gly-ol5- enkephalin is associated with scratching behavior in laboratory animals,23–26 and (3) the stimulation of some neurons in the superficial dorsal horn of rats by morphine facilitates neuronal response to peripheral stimuli by histamine, a pruritogenic substance in human beings.27

In cholestasis, there is evidence to suggest that central opioidergic neurotransmission is increased.5–7, 28, 29 That the increased opioidergic neurotransmission contributes, at least in part, to the pruritus associated with this syndrome is supported by the amelioration of pruritus by opiate antagonists.1, 3–9

An opiate withdrawal reaction is one that results from the chronic use of substances that bind to opiate receptors (e.g., morphine, heroin). The reaction that patients with cholestasis and pruritus can experience upon the administration of an opiate antagonist suggests that in cholestasis the opioidergic tone is increased. Some of the symptoms associated with the classic opiate withdrawal reaction (e.g., increased body temperature, respiratory rate, and pulse, weight loss, yawning, lacrimation, rhinorrhea) were not reported by the patients with cholestasis who participated in the studies we conducted3–6 or by the patients reported by Thornton and Loswosky.7 Furthermore, hallucinations and depersonalization, two manifestations that are not part of the classic opiate withdrawal reaction, were experienced by the patients treated with nalmefene.5–7 The reaction described by some of the patients with cholestasis who received opiate antagonists shares characteristics with the classic opiate withdrawal reaction, but it is not a classic reaction. Several points should be considered in this respect: (1) the classic opiate withdrawal reaction has been described in the prolonged use of morphine and heroin (morphine is an alkaloid and not a peptide as are endogenous opioids), (2) the “endogenous opioid” withdrawal reaction in human beings is not well described, (3) morphine exerts its pharmacological effects primarily by binding to the mu type of opioid receptor, so it is possible that the side effects associated with nalmefene and naltrexone in patients with cholestasis are being mediated by their binding to another receptor type, such as the kappa opioid receptor (this possibility exists because the binding affinity of nalmefene and naltrexone for the kappa or delta receptors is much higher than that of naloxone), and (4) in liver disease, more than one neurotransmitter system may be altered.

In our experience, the patients who reported the most intense unpleasant mental symptoms on nalmefene were the patients who experienced the most marked decrease in their pruritus.5, 6 This result suggests that both the withdrawal-like reaction and the pruritus result from the degree of alterations in the endogenous opioid system—that is, it may be that the more altered the system is, the greater the possibility of “withdrawal” and the greater the alleviation of the pruritus.

The recognition that the endogenous opioid and, possibly, other systems of neurotransmission are altered in cholestasis has defined a line of research. Along this line of research, pruritus is no longer considered a symptom derived only from the skin but as a manifestation of liver disease that may be at least in part centrally mediated and/or as having neurophysiological implications. This line of thinking has and will continue to define the research agenda on pruritus for the next several years.

The ideal antipruritic medication should be safe and should treat pruritus specifically. The use of naloxone in hospitalized patients appears to be safe if used as indicated and under controlled conditions, but it is not practical for chronic use in the treatment of the pruritus of cholestasis. The experience with the use of naltrexone in clinical medicine is limited; there are concerns about the accumulation of its metabolites in patients with decompensated liver disease—a population in which pruritus is not a common complication—although the available data tend to suggest that the safety profile is acceptable.30 There are other problems with the use of opiate antagonists for the pruritus of cholestasis, however. First, the potential for an opiate withdrawal–like reaction, although transient, can discourage patients and physicians from the implementation of that type of treatment; this reaction can be avoided or minimized by introducing the opiate antagonists at ultra-low doses in a controlled situation, as published.28 Second, tolerance may develop. Patients can experience a “breakthrough” in the perception of their pruritus after an initial amelioration, which can be controlled again by an increase in the dose of the antagonists in many instances.5, 6 We have tentatively attributed the breakthrough phenomenon to opiate antagonist–induced changes in opioid receptor density in cholestatic patients, in whom an up-regulation of opioid receptors would occur as a result of treatment with an antagonist. As a result of the up-regulation, the presumed opioid pruritogen would bind to an increased number of receptors, and one of the biological effects would be recurrence of pruritus. Third, it is not ideal to block the endogenous opioid system chronically. Clearly, therefore, research on the pruritus of cholestasis is necessary.

The goals of research in the pruritus of cholestasis are substantial and include the identification of the pruritogens, the understanding of neurophysiological and pathophysiological mechanisms that mediate the sensation of itch and the resulting protective scratching behavior, and the subjects per se. As it relates to the subjects, it is very intriguing that not all patients with cholestasis complain of pruritus, even with the same degree of cholestasis, as expressed by levels of alkaline phosphatase, gamma glutamyl transpeptidase, and bile acids. Also, patients with very little cholestasis as reflected by those biochemical markers can have maddening pruritus. Potential explanations for this clinical observation include the need for a factor to be present in addition to liver disease for patients to perceive pruritus and the individual threshold required to experience that sensation.

Pruritus is a sensation. Accordingly, there is a rationale to explore neurotransmission through systems that mediate sensory stimuli. The traditional empirical management of the pruritus of cholestasis switched for the first time to specific management when opiate antagonists emerged as therapeutic alternatives for this type of pruritus. The use of opiate antagonists to treat the pruritus of cholestasis has a rationale that is supported by data obtained from controlled clinical trials that applied objective methodology and included well-defined end points—not by clinical anecdotes or reports from “expert committees.”31

The standards of pruritus research have changed. The need to use objective methodology, which was raised by investigators in the field of pruritus years ago32, 33 and by hepatologists recently,34 has been met.35 In contrast to other symptoms, the sensation of pruritus invariably results in scratching behavior36; accordingly, changes in this behavior as a result of an investigational intervention define the end points in clinical trials. Instruments to record scratching behavior independently from gross body movements have been developed35 and adapted37, 38 making it possible to design studies that yield objective, analyzable, and interpretable data.