Naltrexone Does Not Relieve Uremic Pruritus: Results of a Randomized, Double-Blind, Placebo-Controlled Crossover Study

Transcription

1 J Am Soc Nephrol 11: , 2000 Naltrexone Does Not Relieve Uremic Pruritus: Results of a Randomized, Double-Blind, Placebo-Controlled Crossover Study CHRISTIANE PAULI-MAGNUS,* GERD MIKUS, DOMINIK M. ALSCHER,* TILLMANN KIRSCHNER, WILFRIED NAGEL, NADJA GUGELER, TEUT RISLER, ELKE D. BERGER, ULRICH KUHLMANN,* and THOMAS METTANG* *Department of Internal Medicine, Division of Nephrology, Robert-Bosch-Hospital; Dr. Margarete Fischer- Bosch-Institute of Clinical Pharmacology; Dialysis Center Stuttgart, and Dialysis Center Dürrlewang, Stuttgart, Germany; and Department of Internal Medicine, Division of Nephrology, University of Tübingen, Tübingen, Germany. Abstract. Improvement of uremic pruritus was reported under short-term administration of the -receptor antagonists naltrexone and naloxone. The aim of the present study was to confirm the efficacy and safety of the oral -receptor antagonist naltrexone during a 4-wk treatment period in patients on hemodialysis and peritoneal dialysis. A placebo-controlled, doubleblind crossover study of uremic patients with persistent, treatment-resistant pruritus was performed. Of 422 patients screened between December 1997 and June 1998, 93 suffered from pruritus and 23 were eligible for the study. Patients were started either with a 4-wk naltrexone sequence (50 mg/d) or matched placebo. This was followed by a 7-d washout, and patients continued with a 4-wk sequence of the alternate medication. Pruritus intensity was scored daily by a visual analogue scale (VAS) and weekly by a detailed score assessing scratching activity, distribution of pruritus, and frequency of pruritusrelated sleep disturbance. Sixteen of 23 patients completed the study. During the naltrexone period, pruritus decreased by 29.2% (95% confidence interval [CI], 18.7 to 39.6) on the VAS and by 17.6% (95% CI, 4.2 to 31.1) on the detailed score. In comparison, pruritus decreased by 16.9% (95% CI, 6.8 to 26.9) on the VAS and by 22.3% (95% CI, 9.3 to 35.2) on the detailed score during the placebo period. The difference between the naltrexone and the placebo treatment period was not statistically significant. Nine of 23 patients complained of gastrointestinal disturbances during the naltrexone period compared with only one of 23 patients during the placebo period (P 0.05). These results show that treatment of uremic pruritus with naltrexone is ineffective. In addition, a high incidence of adverse effects was observed during treatment with naltrexone. Among the dermatologic abnormalities in uremia, pruritus is one of the most common and tormenting symptoms, which is reported to occur in up to 90% of uremic patients (1 2). Although various factors have been considered to be involved in the pathogenesis of uremic pruritus (3 7), the pathophysiology of this symptom remains unknown. Therefore, treatment has been mainly empirical, and the efficacy of therapies is often insufficient to provide adequate relief of pruritus (8 10). In 1985 the first case report describing successful treatment of uremic pruritus by intravenous administration of the opiateantagonist naloxone was published (11). Recently, a placebocontrolled clinical trial showed that administration of the oral -receptor antagonist naltrexone is associated with a significant decrease in pruritus perception in all of the treated patients Received January 26, Accepted August 19, Correspondence to Dr. Christiane Pauli-Magnus, Department of Internal Medicine, Division of Nephrology, Robert-Bosch-Hospital, Auerbachstrasse 110, Stuttgart, Germany. Phone: ; Fax: ; thomas.mettang@rbk.de / Journal of the American Society of Nephrology Copyright 2000 by the American Society of Nephrology with severe uremic pruritus (12). However, the number of patients studied was small and the treatment period (1 wk) was short. Furthermore, peritoneal dialysis patients were not included in that trial. Therefore, the aim of our study was to investigate the efficacy of oral naltrexone over a longer treatment period in a larger cohort of hemodialysis (HD) and peritoneal dialysis (PD) patients. Materials and Methods Patients Male or female patients ranging in age from 20 to 85 with endstage renal disease (ESRD) undergoing HD or PD treatment and who had substantial pruritus for more than 6 mo were considered for the study. Substantial pruritus was defined as persistent, treatment- resistant pruritus considerably impairing sleep or daytime activity. Treatment resistance was defined as no or only insufficient resolvement of pruritus under current antipruritic treatment regimens such as antipruritic lotions, antihistamines, ultraviolet therapy, oral activated charcoal, and electric needle acupuncture. Patients with a history of pruritus or dermatologic disease antedating renal failure, and those with skin disease other than the usual cutaneous findings of uremia such as xerosis or ecchymosis, and lesions due to scratching were excluded. Also, patients with systemic disease such as malignancy,

2 J Am Soc Nephrol 11: , 2000 Naltrexone Does Not Relieve Uremic Pruritus 515 Table 1. Reasons for not including 70 patients with uremic pruritus in the trial a Primary Exclusion Criteria No. of Patients % No severe pruritus or good relief under current antipruritic treatment Treatment with opiate analgesics Hemoglobin 10 g/dl Systemic disease (malignancy, cholestatic liver disease) Dermatologic disease Treatment with steroids Pruritus 6 mo Poor dialysis quality (Kt/V 1.2 on HD and 2.0 on PD) No consent Total a HD, hemodialysis; PD, peritoneal dialysis. cholestatic liver disease, or those under treatment with steroids or opiate analgesics were excluded. Out of 422 patients with ESRD who were stable on HD or PD and screened between December 1997 and June 1998 (378 HD patients and 44 PD patients), 93 had current pruritus (74 on HD and 19 on PD). Of these 93 patients, 70 did not fulfill inclusion criteria (Table 1); the remaining 23 patients were entered in the trial (18 on HD and five on PD). Previous antipruritic treatment of the 23 patients participating in the study is shown in Table 2. Study Design This was a randomized, double-blind, placebo-controlled crossover study in four dialysis centers in Southern Germany. The protocol was approved by the local ethics committee, and each patient gave informed written consent. Patients were assigned to either the naltrexone-placebo sequence (group A) or the placebo-naltrexone sequence (group B). Naltrexone and placebo treatment lasted for 4 wk each with a 7-d washout between the two treatment periods. Naltrexone therapy (Nemexin ; Du Pont Pharmaceutical) was started with a single morning dose of 50 mg of naltrexone hydrochloride or matching placebo. All other antipruritic treatment was terminated 1 wk before study entry. Plasma hemoglobin concentrations and serum concentrations of creatinine, urea, calcium, phosphate, alkaline phosphatase, bilirubin, and transaminase were measured in all patients at the beginning of each treatment sequence. Concentration of parathyroid hormone and dialysis adequacy expressed by Kt/V values were measured before the start of the study. All blood samples were taken immediately before dialysis in HD patients and at 8 a.m. in PD patients. The diet of the patients was not modified during the study. Compliance was assessed Table 2. Previous antipruritic treatment of the 23 patients participating in the trial Previous Antipruritic Treatment Patient No. Antipruritic lotions 1, 8, 10, 13 15, Antihistamines 1 6, 8 19, Ultraviolet therapy 1 4, 14, 16, 19, Oral activated charcoal 1 4, 8, 14, 19 Acupuncture 2 4, 5, 12 Other 3 4, 8, 10 by collecting the drug boxes at the end of each study period and by taking a blood sample for naltrexone measurement at a randomly chosen time during each treatment period. Pruritus Assessment Patients were evaluated at the beginning of the study and at the end of weeks 1, 2, and 4 of each study period. Each evaluation included a limited physical examination and the assessment of the severity of pruritus. The score was assessed by the same investigator for all patients and at each time of the study. For evaluation of itching, we modified a detailed score proposed by Duo (13,14). The severity, distribution, and frequency of pruritus-related sleep disturbance was monitored as follows. Severity. A slight itching sensation without necessity of scratching received 1 point, with necessity to scratch, but without excoriations 2 points, scratching accompanied by excoriation 4 points. Pruritus causing total restlessness received 5 points. Distribution. Itching at less than two locations received 1 point, more than two locations 2 points, and generalized itching 3 points. The score of severity and distribution of pruritus was recorded and multiplied separately for the morning and the afternoon, so that a maximum of 30 points could be achieved. Sleep Disturbance. Each episode of waking up because of itching received 2 points (maximum 10 points) and each scratching episode leading to excoriation during the night received 1 point (maximum 5 points). The score of sleep disturbance and the severity-distribution product must be added to assess the final score (maximum 45 points). Furthermore, patients were asked to estimate their itching intensity before study entry and then on a daily basis by marking a visual analogue scale (VAS) (0 no pruritus to 10 unbearable pruritus). In addition, particular attention was paid to any symptoms that could be attributed to an opioid withdrawal-like reaction such as gastrointestinal symptoms (nausea, vomiting), neurologic disorders (anxiety, depression, insomnia), and cardiovascular symptoms (palpitations, rise in BP) (15,16). Statistical Analyses The mean weekly VAS scores were calculated separately for week 1, week 2, and week 3 4 to ensure the comparability with the three different assessment time points of the detailed score (end of weeks 1, 2, and 4). The pruritus score at the beginning of the study and at the end of the 7-d washout period was set to 100% for the VAS and the

3 516 Journal of the American Society of Nephrology J Am Soc Nephrol 11: , 2000 detailed score, respectively. Results are given as percent of the initial pruritus score separately for VAS and the detailed score (mean and 95% confidence interval [CI]). Differences between naltrexone and placebo were tested for all patients and for different subgroups (patients with and without hyperparathyroidism and patients with and without hyperphosphatemia) by multifactorial ANOVA for repeated measurements (Statgraphics Plus, version 2.0; Manugistics, Inc., Rockville, MD). An intention-to-treat and a per-protocol analysis were carried out. Differences were considered significant at P Correlations between the severity of pruritus and the levels of parathyroid hormone and serum-phosphate were calculated by Spearman rank correlation. The McNemar test was used to assess differences between the two groups in the occurrence of adverse effects. The correlation between the two different pruritus scores was calculated by simple regression analysis. We included each subject several times for each of the time points and both treatment periods, after excluding a significant influence of the time points and the treatment period by multiple regression analysis. Results All patients were considered well-dialyzed with Kt/V values above 1.2 under a dialysis treatment regimen of 3 4to5 h/week for HD patients and weekly Kt/V values above 2 for PD patients. HD was performed with a polysulfone dialyzer, and all but one patient were on high-flux dialysis. Sterilization of filters and blood lines was performed using steam sterilization. None of the patients had severe anemia (hemoglobin 10 g/l). Six of 23 patients had evidence of hyperparathyroidism (parathyroid hormone 18 pmol/l), and 12 of 23 patients had evidence of hyperphosphatemia (serum phosphate 1.7 mmol/ L). Sixteen of 23 patients finished the study. Gastrointestinal side effects such as loss of appetite and nausea were observed in 10 of 23 patients, in nine patients during naltrexone treatment and in one patient during placebo treatment (P 0.05). Gastrointestinal side effects were followed by collapse in one patient and led to study dropout in four patients (three during naltrexone treatment and one during placebo period). Dropout of three more patients was caused by one case of lower limb amputation in a patient with diabetic gangrene that necessitated cessation of naltrexone therapy (during naltrexone period), one case of cerebral ischemia (during placebo period), and one case of renal transplantation. Dropout of three patients occurred immediately after the start of the study on day 1 (n 2) and day2(n 1), respectively; these patients were not considered for further analysis. Apart from gastrointestinal side effects, no other symptoms possibly related to an opiate withdrawal reaction were recorded. At study entry, the mean pruritus scores were 5.5 (95% CI, 4.2 to 6.8) on the VAS for group A and 6.5 (95% CI, 5.3 to 7.6) for group B. On the detailed score, mean initial pruritus was 17.7 (95% CI, 12.9 to 22.5) for group A and 16.8 (95% CI, 12.6 to 21) for group B. At the beginning of the second treatment period, pruritus scores were 4.1 (95% CI, 2.6 to 5.6) on the VAS for group A and 5.4 (95% CI, 4.0 to 6.9) for group B. The corresponding values on the detailed score were 14.2 (95% CI, 8.9 to 19.6) for group A and 8.9 (95% CI, 3.8 to 13.9) for group B (Table 3). There was a significant relationship between the two different scores used for pruritus assessment (P 0.01; r 0.66). Severity of pruritus was not correlated with the extent of hyperparathyroidism (Spearman rank correlation of 0.06 [95% CI, 0.48 to 0.38]) or hyperphosphatemia (Spearman rank correlation of 0.28 [95% CI, 0.16 to 0.63]). During the naltrexone period, pruritus decreased by 29.2% (95% CI, 18.7 to 39.6) on the VAS and by 17.6% (95% CI, 4.2 to 31.1) on the detailed score in the intention-to-treat-analysis (n 20). In comparison, pruritus decreased by 16.9% (95% CI, 6.8 to 26.9) on the VAS and by 22.3% (95% CI, 9.3 to 35.2) during the placebo period (Table 4 and Figures 1 and 2). The difference between the naltrexone and the placebo period was not statistically significant (P for the VAS and P 0.6 for the detailed score). This was also true when only the 16 perprotocol patients were considered for analysis (P 0.1 for the VAS and P 0.5 for the detailed score; data not shown). Subgroup analysis of patients with and without hyperparathyroidism and patients with and without hyperphosphatemia revealed no statistically significant difference with regard to the effectiveness of naltrexone treatment (Table 5). Overall compliance during the study was very good. Naltrexone was detectable in the serum of all patients during the verum period. The drug boxes were returned by all patients. The boxes were all empty, apart from one box with four tablets left over (during naltrexone period) and one box with two tablets left over (during placebo period). Discussion In the present trial, the effects of the oral -receptor antagonist naltrexone in the treatment of uremic pruritus were investigated in 23 patients who were stable on HD or PD treatment. Pruritus was assessed by two different but statistically correlated scores to allow a better and more complete assessment of the various characters of pruritus. Of a total of 422 patients with ESRD we screened for study entry, only 22% had current pruritus. This is noteworthy because epidemiologic data of the early 1990s suggest that uremic pruritus is present in up to 90% of patients with ESRD. Table 3. Mean pruritus scores at the beginning of the first and second treatment period for group A (naltrexone-placebo sequence) and group B (placebo-naltrexone sequence) a Scoring Method First Treatment Period Second Treatment Period VAS group A 5.5 (4.2 to 6.8) 4.1 (2.6 to 5.6) group B 6.5 (5.3 to 7.6) 5.4 (4.0 to 6.9) Detailed score group A 17.7 (12.9 to 22.5) 14.2 (8.9 to 19.6) group B 16.8 (12.6 to 21) 8.9 (3.8 to 13.9) a 95% confidence interval is given in parentheses. VAS, visual analogue scale.

4 J Am Soc Nephrol 11: , 2000 Naltrexone Does Not Relieve Uremic Pruritus 517 Table 4. Decrease of pruritus in percent of initial pruritus after week 1, week 2, and week 3 4, and over the 4-wk treatment period a Week 1 Week 2 Week 3 4 Weeks 1 to 4 Scoring Method Naltrexone Placebo Naltrexone Placebo Naltrexone Placebo Naltrexone Placebo 16.9 (6.8 to 26.9) P (18.7 to 39.6) 25.6 (8.2 to 43) 32.2 (14 to 50.5) 10.3 ( 7.1 to 27.7) 31.8 (13.5 to 50) 14.7 ( 2.7 to 32.1) VAS 23.5 (5.8 to 41.3) 22.3 (9.3 to 35.2) P (4.2 to 31.1) 37.5 (15.1 to 39.9) 17.9 ( 5.6 to 41.4) 17.1 ( 5.3 to 39.5) 15.4 ( 8.1 to 38.1) 12.1 ( 10.3 to 34.5) Detailed score 19.6 ( 3.2 to 42.5) a Results are given as mean (95% confidence interval). Figure 1. Pruritus given as percentage of initial pruritus on the visual analogue scale (VAS) during naltrexone and placebo periods (mean and 95% confidence interval). Figure 2. Pruritus given as percentage of initial pruritus on the detailed score during naltrexone and placebo periods (mean and 95% confidence interval). These epidemiologic changes are possibly due to the improvement in the management of dialysis patients that has occurred in the last decade in Germany. In our study, naltrexone treatment of uremic pruritus was ineffective. Although there was a tendency to a greater decrease of pruritus during naltrexone treatment on the visual analogue scale, the difference did not reach statistical significance (P 0.095). There was, however, no difference on the detailed score (P 0.6). Furthermore, we found a high incidence of naltrexone-related gastrointestinal side effects such as loss of appetite and nausea in up to 40% of the patients. In the absence of other opiate-withdrawal reactions such as cardiovascular or neurologic disorders, these adverse effects are more likely to be related to naltrexone itself than to an opiatewithdrawal reaction. By calculating pruritus relief as percent of initial pruritus independently of whether patients started with high or intermediate pruritus scores, these results even tend to overestimate the effects of naltrexone, because a decrease of 2 points on the interval scale means a greater percentage decrease for patients with lower initial pruritus score. By combining the daily VAS scores of weeks 3 and 4 for calculation of pruritus decrease, we would underestimate a possible naltrexone effect occurring at the end of the 4-wk

5 518 Journal of the American Society of Nephrology J Am Soc Nephrol 11: , 2000 Table 5. Decrease of pruritus over the 4-wk treatment period in percent of initial pruritus separately for patients with and without hyperparathyroidism and hyperphosphatemia, respectively a Condition VAS Detailed Score Naltrexone Placebo P Value Naltrexone Placebo P Value Hyperparathyroidism Yes (n 7) 16.1 (2.9 to 29.2) 15.5 (2.4 to 28.7) ( 25.8 to 14) 16.6 ( 3.2 to 36.5) 0.11 No (n 16) 36.1 (21.7 to 50.6) 18.1 (4.2 to 31.9) (12.7 to 46.9) 25.9 (9.5 to 42.2) 0.75 Hyperphosphatemia Yes (n 10) 14 (3.8 to 24.1) 22.8 (11.7 to 33.9) ( 8.9 to 29.4) 28.5 (7.6 to 49.5) 0.2 No (n 13) 38.7 (19.2 to 58.1) 18.3 ( 1.2 to 37.7) (0.6 to 44.8) 20 ( 2.1 to 42.1) 0.86 a Results are given as mean (95% confidence interval). treatment period. However, based on literature data, mean treatment effect is expected to occur in the first days after onset of naltrexone treatment, so these considerations would have a minor impact. The general tendency toward lower pruritus scores at the beginning of the second treatment period, regardless of whether patients started with the naltrexone or the placebo sequence, is most likely due to a study effect. The therapeutic use of opiate antagonists in patients with uremic pruritus was based on the assumption that endogenous opiate peptides may be involved in the pathogenesis of uremic pruritus. This pathogenetic concept was first developed for the treatment of cholestatic pruritus and is supported by the fact that opiate antagonists were successful in treatment of cholestatic pruritus (17,18). In contrast to cholestatic pruritus (15,19), there are only few data on changes in the opioid system in uremic patients. There is some evidence that plasma -endorphin is elevated in uremia (20,21), although these findings are inconsistent (22). Furthermore, a correlation between the severity of pruritus and the plasma levels of -endorphin could not be established (23). Data about uremiarelated changes in the central nervous system are completely lacking. Despite the limited evidence on involvement of the opioid system in the pathogenesis of uremic pruritus, Peer and coworkers found a nearly total remission of severe uremic pruritus in the first 48 h after onset of naltrexone treatment with only few side effects (12). This outcome is in sharp contrast to the results of our study and cannot be explained by differences in patient compliance, in naltrexone dosage, or study design, because both studies were randomized, placebo-controlled, double-blind crossover trials. As in the study of Peer, subjects included in our trial had long-lasting, treatment-resistant pruritus and no evidence of coexisting dermatologic disease. To exclude factors possibly aggravating uremic pruritus such as inadequate dialysis and anemia, only patients considered welldialyzed and with a hemoglobin level 10 g/l were included in our trial. We also included patients with evidence of hyperparathyroidism and hyperphosphatemia, as the pathogenetic role of these factors in uremic pruritus is controversial (24). However, to exclude a relevant influence of these factors on the effect of naltrexone treatment, we performed a subgroup analysis examining data separately for those with hyperparathyroidism or hyperphosphatemia and those without these laboratory findings. Naltrexone treatment was ineffective in all subgroups. In the study of Peer, patients had higher pruritus scores at study entry than in our trial. It is possible that naltrexone effectiveness depends on the severity of pruritus and is demonstrable only in the severest forms of uremic pruritus. This would mean that pathogenetic mechanisms involved in the etiology of uremic pruritus may be different depending on the severity of this symptom. In conclusion, our data suggest that naltrexone treatment of uremic pruritus is ineffective, expensive, and accompanied by a high incidence of gastrointestinal side effects. Furthermore, there is some evidence that the incidence and the severity of uremic pruritus have decreased during the past decade, possibly due to changes in the management of uremic patients. Additional studies are needed to provide new epidemiologic data about the distribution of uremic pruritus in other European countries and abroad. Analysis of regional differences in the distribution and the severity of uremic pruritus may provide new insights into the mechanisms involved in the pathogenesis of this syndrome and guide the development of new therapeutic strategies. Acknowledgments This work was supported by the Robert Bosch Foundation and the Khalil Foundation. References 1. Balaskas EV, Chu M, Uldall RP, Gupta A, Oreopoulos DG: Pruritus in continuous ambulatory peritoneal dialysis and hemodialysis patients. Perit Dial Int 1[Suppl 2]: , Mettang T, Fischer FP, Kuhlmann U: Urämischer Pruritus. Dtsch Med Wochenschr 12: , Nielsen T, Andersen KE, Kristiansen J: Pruritus and xerosis in patients with chronic renal failure. Dan Med Bull 27: , Massry SG, Popovtzer MM, Coburn JW, Makoff DL, Kleeman CR: Intractable pruritus as a manifestation of secondary hyperparathyroidism in uremia. N Engl J Med 279: , Thomas P, Hollinrake R, Lascelles R: The polyneuropathy of chronic renal failure. Brain 94: , 1971

6 J Am Soc Nephrol 11: , 2000 Naltrexone Does Not Relieve Uremic Pruritus Gill DS, Fonseca VA, Balliod R, Moorhead JF, Dandona P: Plasma histamine in patients with chronic renal failure and nephrotic syndrome. J Clin Pathol 44: , Bousquet J, Maurice F, Rivory P: Allergy in long-term hemodialysis. II. Allergic and atopic patterns of a population undergoing long-term hemodialysis. J Allergy Clin Immunol 81: , De Marchi S, Cecchin E, Villalta D, Sepiacci G, Santini G, Bartoli E: Relief of pruritus and decreases in plasma histamine concentrations during erythropoietin therapy in patients with uremia. N Engl J Med 326: , Hiroshige K, Kabashima N, Tagasugi M, Kuroiwa A: Optimal dialysis improves uremic pruritus. Lancet i: 215, Goyal S, Gittlen S, Schulman ES, Burke JF, Besarab A, Francos G: Elevated histamine levels in uremia: Effects of ketotifen in patients with uremic pruritus. Kidney Int 31: , Andersen LW, Friedberg M, Lokkegaard N: Naloxone in treatment of uremic pruritus: A case history. Clin Nephrol 21: , Peer G, Kivity S, Agami O: Randomized crossover trial of naltrexone in uremic pruritus. Lancet 348: , Duo LJ: Electrical needle therapy of uremic pruritus. Nephron 47: , Mettang T, Fritz P, Weber J, Machleidt C, Hübl E, Kuhlmann U: Uremic pruritus in patients on hemodialysis or continuous ambulatory peritoneal dialysis (CAPD): The role of plasma histamine and skin mast cells. Clin Nephrol 34: , Thornton JR, Losowsky MS: Opioid peptides and primary biliary cirrhosis. Br Med J 297: , Bergasa NV, Jones EA: The pruritus of cholestasis: Potential pathogenic and therapeutic implications of opioids. Gastroenterology 108: , Bergasa NV, Alling DW, Talbot TL: Effects of naloxone infusion in patients with the pruritus of cholestasis: A double-blind randomized controlled trial. Ann Intern Med 123: , Bergasa NV, Schmitt JM, Talbot TL: Open-label trial of oral nalmefene therapy for the pruritus of cholestasis. Hepatology 27: , Bergasa NV, Rothman RB, Vergalla J, Xu H, Swain MG, Jones EA: Central muopioid-receptors are down-regulated in a rat model of acute cholestasis. J Hepatol 15: , Aronin N, Krieger DT: Plasma immuno-reactive -endorphin is elevated in uremia. Clin Endocrinol 18: , Trelewicz P, Grzeszczak W, Drabczyk R: Serum beta-endorphin in non-dialysed and haemodialysed patients with chronic renal failure. Int Urol Nephrol 26: , Elias AN, Vaziri ND, Maksy M: Plasma beta-endorphin and beta-lipotropin in patients with end-stage-renal disease effects of hemodialysis. Nephron 43: , Mettang T, Fischer FP, Dollenbacher U, Kuhlmann U: Uremic pruritus is not related to beta-endorphin serum levels in hemodialysis. Nephrol Dial Transplant 13: , Cho YL, Liu HN, Huang TP, Tarng DC: Uremic pruritus: Roles of parathyroid hormone and substance P. J Am Acad Dermatol 36: , 1997