Article Text
PDF
Abstract
Background Cholestatic itch is caused by intrahepatic liver diseases, such as primary biliary cirrhosis and extrahepatic obstruction of the biliary tree, often caused by tumours. The pathophysiology of cholestatic itch is complex and no single treatment has proved definitive. Naltrexone is an opioid receptor antagonist, which reduces central opioidergic tone, believed to be raised in patients with cholestatic pruritus.
Aim To review and assess the efficacy of oral naltrexone for the treatment of cholestatic itch.
Methods Search of electronic databases, grey literature, clinical trials registries and handsearching for studies including naltrexone for cholestatic itch. Full papers were obtained if relevant and studies graded.
Results Thirteen papers were included in the analysis, including three randomised controlled trials, one controlled clinical trial, one open-label pilot study, seven case reports and one retrospective notes review. All studies found naltrexone to be effective in relieving pruritus. In all five studies performing statistical analysis, naltrexone significantly reduced pruritus compared with baseline. 37% of patients reported side effects, notably opioid withdrawal-type reactions and recurrence of previous pain, from all pathologies.
Conclusions Oral naltrexone therapy helps relieve cholestatic itch and although it should be used with caution in patients using exogenous opioids for analgesia, it should be considered when treating refractory pruritus in patients with end-stage liver disease.
Statistics from Altmetric.com
Introduction
Itch is defined as an unpleasant sensation of the skin or mucous membranes associated with a desire to scratch or rub1 2 the affected area. Pruritus generally refers to a pathological condition in which itch is intense, generalised and triggers repeated scratching in an attempt to relieve the discomfort, consequently causing breaks in the skin, inflammation and/or bleeding.3 4 Secondary skin lesions as a result of the itch-scratch cycle are excoriations and scars. Chronic itch becomes more intense at night5
Pruritus is a common and poorly understood symptom of many localised and systemic disorders,3 with a frequency rate of 6% for general palliative settings, affecting 5%–24% of patients with incurable cancer diagnoses.6 7 Pruritus can be very distressing, leading to sleep deprivation, depression, impaired quality of life and even suicidal ideation.3 8 Pruritus is challenging to manage, with limited evidence available on treatment options.9
The pathophysiology of pruritus is complex and poorly understood.3 9 Pruritus is initiated by the activation of slow-conducting unmyelinated C-fibres with free nerve endings near the dermal-epidermal junction by either physical or chemical stimuli.3 Impulses are conveyed to the dorsal horn of the spinal cord, relayed to the contralateral spinothalamic tract to the thalamus and somatosensory cortex, where it is identified as itch.4 9 Physical stimuli which induce pruritus include pressure and thermal stimulation.10 Chemical stimuli include amines (histamine, serotonin), neuropeptides, proteases, bradykinin, opioids, growth factors and cytokines, as shown in figure 1.3 4 11 Though both pain and itch have the same chemical messengers that excite C-fibres, itch neurons are located more superficially and are more sensitive to pruritogenic substances than pain receptors.3
Factors mediating itch.49 DRG, dorsal root ganglion.
Pathological itch can be classified as primary (idiopathic), seen in more than 70% of patients with itch, or secondary, that is, itch secondary to various dermatological or systemic diseases.
Cholestasis refers to a reduction in bile flow, due to extrahepatic causes, such as obstruction of the biliary tree by tumour, or intrahepatic causes, such as the liver diseases primary sclerosing cholangitis (PSC), primary biliary cirrhosis (PBC) and chronic hepatitis.12 PBC is a chronic, autoimmune, cholestatic liver disease characterised by the slow destruction of small intrahepatic bile ducts, impaired biliary secretion and stasis of toxic endogenous bile acids within the liver with progression to liver fibrosis and cirrhosis.13 The prevalence of pruritus in PBC may be as high as 69%.14 PSC is a chronic cholestatic liver disease caused by diffuse inflammation and fibrosis that can involve the entire biliary tree. PSC is progressive and can lead to biliary cirrhosis, portal hypertension and hepatic failure. In symptomatic patients, median survival from presentation to death or liver transplantation is about 12 years. It is a premalignant condition, and the majority of deaths are from malignancy, particularly cholangiocarcinoma or colon cancer.15 Cholestasis may occur in the palliative population due to gallstones unrelated to the terminal diagnosis, drugs, PBC and PSC, as well as obstruction from primary or secondary tumours involving the pancreas and biliary tree.16 Pruritus in hepatic disease can be severe, and, when persistent, is an indication for liver transplantation.17 Cholestatic pruritus particularly affects the palms of the hands and soles of the feet, although itch may be generalised and described as ‘lying on a bed of cactus’, ‘pins and needles’ and, unlike other causes of pruritus, is not relieved by scratching.17 18
The pathophysiology of pruritus in liver disease has not been clearly delineated.19 Theories are shown in table 1.
Pathophysiology of cholestatic itch
Itch is an intrinsically subjective perception, which makes it difficult to quantify. However, in the clinical setting, the severity of pruritus has been found to be best assessed using numerical scoring systems or Visual Analogue Scales (VAS).20–22 Scratching activity, defined as a behavioural consequence to pruritus, can be quantitated. However, the severity of the perception of the pruritus of cholestasis does not always correlate closely with scratching activity.20
There are several treatments available for cholestatic pruritus, as shown intable 2The poor understanding of the mechanisms underlying pruritus has ensured that no single treatment has proved definitive.21 The evidence base for all reported treatments is variable leading to difficulty in identifying an effective treatment.
Eligibility criteria
Treatment aims of cholestatic itch and examples of medication used
The aim of this limited systematic review is to review and assess the effectiveness of oral naltrexone for the treatment of cholestatic itch.
Treatment of cholestatic itch is challenging, especially in palliative patients, in whom certain interventions, such as biliary stenting or drainage procedures may be inappropriate, and in whom polypharmacy is common.
Methods
Information sources and search strategy
For this review, a search of the electronic databases, including MEDLINE (1950 to February 2016), EMBASE (1980 to February 2016), CINAHL (1981 to February 2016) and the Cochrane central register of Controlled Trials (CENTRAL) in March 2016, was performed. A language restriction was not imposed on studies. See online supplemental appendix 1 for the MEDLINE search strategy. Database searching took place on 29 February 2016. Search terms adapted for the Ovid interface were (1) Naltrexone OR naltrexone AND (2) cholestatic itch OR Pruritis OR pruritis OR itch OR cholestasis. A search of clinical trials registers, the WHO International Clinical Trials Registry Platform (http://www.who.int/ictrp/en/), ClinicalTrials.gov, Current Controlled Trials (www.controlled-trials.com/) and the EU Clinical Trials Register (clinicaltrialsregister.eu) for this review on 29 March 2016 was also performed, using the search term ‘naltrexone’. Handsearching complemented the electronic searches (using reference lists of included studies, relevant chapters, review articles, National Institute for Health and Care Excellence guidance). Grey literature was sought by using Care Search, an Australian palliative care knowledge network with information of studies in process on 29 March 2016.23 An updated literature review was performed on 4 June 2020 and no new studies or case reports were identified. The eligibility criteria are shown in Table 3.
Supplemental material
Data collection and analysis
For this review, records were imported from each of the above electronic databases into a Word document, where titles, keywords and abstracts were inspected for relevance. If it was not possible to accept or reject an abstract with certainty, the full text of the article was obtained for further evaluation. The flow of studies through the review is shown in Figure 2.
Study selection
A scoping search revealed few randomised controlled trials (RCT), so all trials were evaluated, including non-analytical studies. Outcome measures were chosen to assess efficacy and safety of the treatment.
Data extraction and analysis
A descriptive summary of studies has been given and simple statistics are provided for key outcomes. Studies were graded according to the SIGN system for level of evidence. It was planned to assess studies for treatment effect, specifying numbers needed to treat for an additional beneficial outcome and numbers needed to treat for an additional harmful outcome.
Flow of studies through review.
Results
Characteristics of included studies
Full texts of 36 papers were reviewed, of which 13 were included in the analysis. Three were RCTs, one was a single-blind self-controlled cross-over trial, four were case reports of adults, one was a non-blinded open-label pilot study, one was a retrospective observational review of children and three were case reports of children. Studies are from the USA (n=3),24–26 the UK (n=3),27–29 Argentina (n=1),30 The Netherlands (n=2),31 32 Iran (n=1),33 Colombia (n=1),34 Israel (n=1)35 and India (n=1).36
Adult patients were recruited from hepatology or gastroenterology outpatient clinics and oncology inpatient wards. Children were recruited from paediatric dermatology outpatient clinics, paediatric hepatology or gastroenterology outpatient clinics and Paediatric inpatient wards.
An overview of study characteristics is shown in online supplemental table S1.
Supplemental material
Quality assessment
All studies were graded according to the (Scottish Intercollegiate Network) SIGN system for level of evidence, as shown in Table 4.37 Eight of 13 studies were evidence level 3, that is, non-analytical studies, mostly case reports. Two studies were high-quality RCTs, both double blinded.30 31 Jain et al’s study was also a RCT, but as this has only been published as a poster, there was not sufficient detail to ascertain whether the study was blinded, or for risk of bias to be excluded, therefore it has been graded as 1-.36 Communication with the study author has not resulted in further study details being established. Mansour-Ghanaei et al’s study is graded as 2++, as it is a self-controlled clinical trial, in which 34 patients received placebo for 1 week, followed by naltrexone for 1 week.33 The study is not randomised and is only single-blinded, therefore, there is some low risk of bias. Carson et al’s study is graded as 2−, as this open-label pilot study is limited by its small size (n=5), with no control, randomisation or blinding.26
Quality assessment, according to sign grading system37
The five analytical studies were subjected to a more rigorous quality assessment process, using the Quality Assessment Tool for quantitative studies,38 as shown in table 5. Three studies scored strong, two RCTs,30 31 and a controlled clinical trial.33 Jain et al’s RCT scored only moderately, losing points due to potential confounders, that is, lack of information as to differences between the three patient groups in the study.36 Should this study be submitted as a full paper, these details can be clarified, and the paper should be regraded. Carson et al’s small (n=5) study is rated weak, as the study is not blinded, and there was no control or placebo group.26
Quality assessment for the five studies graded 1 and 2 according to sign guidelines38
Number of patients
In 13 studies, a total of 118 patients (20 children, 98 adults) received oral naltrexone for cholestatic itch. The most common diagnosis in adults was PBC, n=32 (33% of adult patients), followed by acute cholestatic viral hepatitis (n=25, 26%) and cirrhosis (n=12, 12%).
Dosing: adults
The most common dose of oral Naltrexone used in adults was 50 mg once daily,26 29–31 33 although in two studies,30 31 25 mg two times per day was given on day 1. Other studies used 25 mg two times per day,34 12.5 mg once a day,32 12.5 mg two times per day or 25 mg once a day,28 and 12.5 mg three times a day.29 In four studies, naltrexone dose was gradually escalated over time (‘few days’ to 2, 4 and 6 weeks later), to 50 mg daily,34 to 25 mg two times per day or 50 mg two times per day,28 to 150 mg daily29 and 400 mg daily.32
Naloxone infusion precommencement of naltrexone oral therapy in adults
In three studies,28 29 34 a naloxone intravenous infusion was given prior to commencing oral naltrexone treatment, for 2 hours,29 24 hours34 or 5 days.28 With regard to timing of commencement of oral naltrexone therapy, one study states29 ‘the infusion was then stopped and oral therapy with naltrexone started’, and it is assumed that the other studies28 34 used a similar approach.
Children
Of 20 children included in the review, 18 (90%) had Alagille’s syndrome, a genetic disorder associated with characteristic facies, liver, heart, bones, eye and kidney abnormalities. Cholestatic liver disease characterised by paucity of intrahepatic bile ducts is the hallmark of the syndrome.35 One child had PSC and one congenital biliary atresia, a childhood liver disease in which one or more bile ducts are abnormally narrow, blocked or absent.39
The dose of naltrexone used in children varied between studies: from 0.02 mg/kg/day increasing gradually to 0.08 mg/kg/day,35 0.25–0.5 mg/kg/day,27 2 mg/kg/day24 and 5 mg/kg/day.25 In the study using the 5 mg/kg/day dose, the child, aged 17 months, was taking a final dose of 25 mg two times per day, a dose seen in adult studies. One child received an intravenous infusion of naloxone (duration unclear), prior to commencing oral naltrexone.24
Study scoring of pruritus
In 9 of 13 studies, formal scoring systems were used, most commonly the VAS (0–10) (7 studies)26 28 30 32–35 ; VAS 0–100 (1 study)31 ; or, in one study, VAS>5 (42). Further details of the last scoring system have not been forthcoming on contacting the author. In four studies,24 25 27 29 formal scoring was not used, rather parental observations of scratching activity and patient reported symptoms. In one study, only one of three patients was scored.28
Effect of naltrexone on pruritus scores
In 5 of 13 studies,26 30 31 33 36 statistical tests were used and naltrexone was found to significantly improve pruritus when compared with baseline, however, in one trial,33 placebo also significantly improved pruritus compared with baseline. Of the nine studies using a scoring system for measurement of pruritus, all showed a reduction in score following oral naltrexone therapy.
Effect of naltrexone on fatigue, energy levels, quality of life
Three studies looked at additional symptoms. Carson found no significant difference in energy levels, but found quality of life was significantly improved (p=0.013) following treatment with naltrexone.26 McRae et al found a 4.5 point reduction in fatigue score for one patient, although no test of significance was performed.28 Wolfhagen et al found no significant difference in fatigue, but a significant reduction in sleep disturbance (p=0.0001) following oral naltrexone therapy.31
Side effects of naltrexone
Of 118, 44 patients (37%) had side effects, 39/98 (40%) adults and 5/20 (25%) children, most frequently opioid withdrawal type symptoms including nausea, vomiting, malaise, flushing, insomnia, dry mouth, lethargy, irritability, weakness, dizziness, sweating, headache, tremor, abdominal cramps and bowel frequency (n=38, 32%); and recurrence of pain, from previous pain syndromes including postherpetic neuralgia and polyarthritis (n=5, 4%). Three patients (3%) chose to reduce naltrexone dose due to side effects, although 1 patient returned to the original dose 7 days later, and 15 patients (13%) chose to stop naltrexone treatment altogether, although two patients chose to restart some days later with a dose reduction.
Long-term follow-up
Nine studies completed more than 1 week of follow-up on patients taking oral naltrexone. Patients in Wolfhagen et al’s study took naltrexone for 4 weeks.31 In Carson et al and Terg et al’s studies, patients who responded well to naltrexone in the initial treatment intervention (1 week and 2 weeks, respectively), could choose to continue treatment in an open-label trial, for up to 28 weeks or for an additional 8 weeks.26 30 In two studies, the phenomenon of tachyphylaxis, that is, loss of efficacy despite unchanged therapy, was noted.26 32 This has been reported to occur infrequently under long-term mu-opioid receptor (MOR) antagonists, with variable latency between 4 weeks and 9 months.40 41 The antipruritic effect can be re-established by increasing the dose, or by interrupting administration of naltrexone for ‘drug holidays’ of 2 days per week or for 2–3 weeks at a time.26 32 40
Discussion
Pruritus is a common and often disabling symptom of cholestatic liver disease. Existing surgical interventions such as biliary stenting may be limited by patient frailty and prognosis and may not be effective in relieving pruritus. Traditional therapies for the pruritus of cholestasis are empiric, tend to have inconsistent effects and are usually only partially efficacious at best.18 Naltrexone, an oral MOR antagonist, is one treatment aimed at reducing central opioidergic tone, believed to be increased in patients with the pruritus of cholestasis. There are no clinical trials of naltrexone in palliative patients, so the scope of this review widened to include patients, both adults and children, with cholestasis from other causes, both acute and chronic, with the assumption that findings from these studies may be transferable to the palliative population. More patients with non-malignant conditions are now seen by palliative care teams, and with advancing medical treatment, children who died of chronic liver diseases in childhood are now transitioning into adult services, including adult palliative care teams. Adult patients with chronic liver diseases of PBC and PSC included in these studies, carry a high risk of progressing to cirrhosis, liver failure, or hepatocellular cancer, which could see them becoming palliative in the future.
In all 13 studies, naltrexone was found to be effective in relieving itch, with pruritus scores. Decreasing an average of 5.35 points on a 10 point rating scale (2.6–10 points). In all five studies conducting a statistical analysis, naltrexone significantly improved pruritus compared with baseline.
Thirty-seven per cent of patients experienced side effects, mostly similar to the ‘cold turkey’ withdrawal reaction of opioid addiction, including malaise, nausea, vomiting and flushing.18 These symptoms usually subsided within 3 days despite continued drug administration, but resulted in 15 patients (13%) stopping naltrexone, and 3 adults reducing naltrexone dose. A small number of patients also withdrew from subsequent open-label trials. Opioid withdrawal-like phenomena appear less troublesome in cholestatic patients following a naloxone infusion than initiation of oral therapy.42 43 Therefore, approaches to minimising opioid withdrawal reactions when initiating naltrexone include: (1) Coadministration with clonidine28 29; (2) Giving a small naltrexone test dose, followed by gradually increasing doses until a therapeutic response is obtained, as used by five studies in this review; (3) Commencing an intravenous infusion of naloxone at a slow subtherapeutic rate, increasing the rate slowly until a therapeutic range is achieved, then switching to low dose naltrexone, as used by four studies in this review.29
The other notable side effect of naltrexone found in this review is loss of control of pain, from various pathologies, seen in five patients (4%). This is possibly the main reason why naltrexone has not been used more widely in the palliative population, as there is concern that the antipruritic effect may be associated with an increase in pain intensity, even if the patient is not receiving exogenous opioids. Zylicz et al described a patient with metastatic cancer achieving good relief of itch from naltrexone therapy but increased doses resulted in abdominal pain.32 McRae et al found the severity of the pain experienced was reduced significantly by discontinuation of naltrexone, suggesting that in the patients experiencing pain, endogenous opioid activity, which would in an individual not receiving opioid antagonist therapy make the painful stimulus tolerable, was being reduced, resulting in the expression of pain.28 In the palliative setting, many patients are prescribed opioids for analgesia and coprescription of an opioid antagonist such as naltrexone, which may reverse the analgesic effect or cause withdrawal symptoms, seems counterintuitive.
Patients with cholestatic itch who are nearing the end of life therefore pose a significant challenge in terms of palliation.19 A study by Nelson et al expanded on previous findings that cholestasis was associated with antinociception that was reversible with naloxone and naloxone methiodide (a peripherally restricted opioid antagonist). The study found that both central and peripheral opioid antagonists reduced the antinociceptive effects experienced by cholestatic mice. The authors concluded that the antinociception of cholestasis by endogeneous opioids occurred at the level of sensory nerve endings. Given the complex pathophysiology of pain and itch, they suggested these findings could have an impact on the management of pruritus in cholestasis (544 0).
One option could be subcutaneous methylnaltrexone, an MOR antagonist with reduced lipophilicity, which limits its effect to peripheral opioid receptors. Methylnaltrexone, therefore, potentially decreases pruritus without reversing opioid-mediated analgesia, and has been found effective in palliative patients with cholestatic itch taking regular hydromorphone for analgesia.19 Another option is a low-dose subcutaneous naloxone infusion as used with good effect in an elderly hospice inpatient with irreversible drug-induced cholestasis and following discontinuation of naltrexone in the cancer patient described by Zylicz et al.8 32 It has been suggested that the small dose and short half-life of naloxone are insufficient to reverse the effect of exogenously administered opioids, and that palliative care and hospice physicians should consider the use of opiate receptor antagonists including naltrexone when treating refractory pruritus in patients with end-stage liver disease.8 Future treatment options may include cellular and molecular targets involved in the itch neural pathway. Opioids can cause itch, and spinal opioid-induced itch is thought to occur as a result of pain inhibition. Liu et al reported that morphine-induced scratching is abolished in mice lacking either gastrin-releasing peptide (GRP) receptor or the MOR.45 Sun and Chen suggested that GRP was the neurotransmitter released by sensory neurons to initiate itch-related signals.46 However, Mishra and Hoon failed to find GRP outside the spinal cord, indicating that GRP is not the primary trigger.47 Mishra and Hoon searched for the molecule that encodes the sensation of itch by screening genes in sensory neurons that are activated by pain, heat, touch and itch and found a protein called natriuretic polypeptide b (Nppb)which was expressed in only one neuron subset.48 Mutant mice lacking Nppb did not respond to itch-inducing compounds but responded normally to heat and pain, and when Nppb was injected into the mice’s necks, they scratched furiously. However, GRP is still involved in the itch response. Injecting GRP into mice lacking either Nppb or its receptor caused scratching, and mice in which GRP receptors were inhibited did not exhibit scratching behaviour, even with spinal cord injection of Nppb. This suggests GRP-releasing neurons are downstream of Nppb in the transmission of the itch sensation. The neural pathways for itch in humans are similar to those in mice, thus this research introduces a new target for treatment of itch in the clinical setting.
Limitations
Although the search for this review was not limited to papers published in English, one study in German could not be translated in time, hence its being excluded. Only one investigator decided which papers to include but the protocol and methods were checked by another researcher (JK). This review included studies of both adults and children, which added interest, but as three of the children’s studies were case reports and the fourth a retrospective notes review, they added little in terms of quality of evidence, and the review could have focused solely on adults. There are few controlled trials of naltrexone, and none in the terminally ill, which means that results are assumed to be transferable to the palliative population. As this was a limited systematic review, a meta-analysis was not undertaken.
Conclusion
Oral naltrexone therapy is helpful in relieving pruritus in adults and children with cholestatic liver diseases, including a small number of patients with cancer. Just over a third experienced side effects secondary to naltrexone, however, notably opioid withdrawal and recurrence of previous pain, however, these were largely transient and only 13% of patients chose to stop treatment altogether. In patients in whom first or second line anti-pruritic agents have failed, oral naltrexone is worth trialling, even in palliative patients taking regular analgesia, prior to considering more expensive and invasive agents such as subcutaneous naloxone or methylnaltrexone.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors FLM-B planned and conducted and wrote the review. JK, Cicely Saunders Institute, Kings College London, proof read the article and suggested edits, changes and an updated literature review.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.