Factors Influencing Pruritus in Uremic Dialysis Patients

Wei-Yun Wang, Kwua-Yun Wang*, Pauling Chu**, Yue-Cune Chang***, Yu-Ying Tang*

Nursing department, Tri-Service General Hospital, Taipei

School of Nursing, National Defense Medical Center, Taipei*

Department of Nephrology, Tri- Service General Hospital, Taipei **

School of Mathematics and Graduate institute of Mathematics, University of Tamkang, Taiwan***

Correspondence author: Kwua-Yun Wang, 4F, No. 161, Min-Chuan E. Rd. Sec. 6, Taipei 114, Taiwan, ROC.

Tel: 886-2-87923100 ext. 18766.

Fax: 886-2-8792-3109.

E-mail:

國防醫學院護理系

抽印本索取:王桂芸教授

台北市內湖區民權東路6段161號4樓

電話:(02)87923100轉18766

傳真:(02)87923109

電子信箱:

Abstract

Background: Many research studies relate to uremic pruritus, but their results are inconsistent. The uremic pruritus rate in dialytic patients in Taiwan lies roughly between 55% and 90%. However, there are rare data available to show the uremic dialytic pruritus rate in Taiwan. Discussing the related factors and creating a prediction model was therefore an important issue.

Methods: This study used a cross-sectional study design. 110 patients were recruited from a medical center in Taiwan. Data were collected by structured questionnaires, which included demography, disease, dialysis, physiologic parameter, trait anxiety and pruritic scale. Data were analyzed logistic regression to construct the prediction model for uremic pruritus.

Results: Results showed that the pruritic incidence rate of uremic dialytic patients was 70.9%, for patients who received peritoneal dialysis was 72.4% and hemodialysis was 67.6%. The significant influencing factors for pruritic symptoms among uremic dialytic patients were the occupation pre-dialysis, types of renal disease, chloride, and skin humidity.

Conclusion: The skin humidity was the most relevant pruritic symptom. Therefore we can teach patients to use skin emollients to increase skin humidity and relieve itching. The optimal goal is to promote the quality of life of uremic dialytic patients.

Key words: uremic dialytic patients, pruritus, influencing factors.

Introduction

Dialytic therapy is the method used to maintain life in most uremic patients. According to the statistical results of the Taiwan Society of Nephrology,1 the dialytic population has increased 7% per year in Taiwan since 1994, that is, after the national health insurance scheme was instructed. There have been 41,675 dialytic people in Taiwan until 2005. The prevalence of dialysis is ranked second in the world, which severely threatens public health finances. Pruritus is a common distressing symptom affecting many long-term dialytic patients. The prevalence of uremic pruritus ranges between 40 and 90%.2-4 The symptoms of dry skin and itch in peritoneal dialytic patients ranges between 50 and 62%. However, the symptoms in hemodialytic patients ranges between 51 and 86%.5-6 The patients suffering from generalized itching ranges between 19 and 47%. Otherwise, the patients suffer from localized itching ranges between 17 and 61%.4, 6-8 The itching is strongest on back, the extremities, the area where the catheter is implanted, or the face and is associated with dry skin, xerosis and secondary infection.3, 9

Among the patients with itching of the skin, 36% became irritable, 8% became depressed and over 40% felt their emotions were affected, even causing sleep disorders.7, 8, 10 All of these data revealed that patients suffer greatly from uremic pruritus. Many research studies relate to uremic pruritus, but their results are inconsistent. Since there are no data showing predictors for uremic pruritus in dialytic patients, the purpose of this study was to identify the related factors of uremic pruritus in dialytic patients and to construct a predictive model.

Materials and Methods

Study design

The study was a cross-sectional design and adapted a structured questionnaire. 110 dialysis subjects were recruited from a medical center in Taipei, Taiwan between December 2006 and March 2007 by purposive sampling. Out of which 76 subjects were on peritoneal dialysis and 34 subjects on hemodialysis three times per week, four hours per time. Data were collected during the week in which we took blood samples for checking the physiological parameters of the patients. The inclusive criteria were as follows: 1) to be aged 18 years or older; 2) to be alert; 3) to be able to write or communicate in Chinese or Taiwanese; 4) to have been on dialytic therapy for more than six months; 5) not to suffer from mental disorder, other dermatologic disorder, systemic disease such as severe infection, hepatic failure, hematological disorder or biliary tract disorder. All patients signed a written consent form prior to the start of the study.

Measures

A comprehensive questionnaire was used to collect data. It included the basic characteristics of the patients (demographic, disease, dialytic, and trait anxiety), physiological parameters, and pruritic score.

In terms of demographic characteristics, age, sex, occupation before dialysis, current occupation and religion were collected. Disease characteristics included the types of renal disease and residual renal function. The pre-dialytic creatinine clearance rate (CCr) formulated by Gault,11 was used to estimate the residual renal function of patients. With regard to dialytic characteristics, these included the dialytic method, dialytic membrane and adequacy of dialysis. We took the formula of Kt/Vurea by Gotch and Sargent,12 published to estimate the adequacy of dialysis. Concerning anxiety, this was described as an unpleasant emotional state consisting of feelings of uncertainly.13 Spielberger14 indicated that it consisted of state anxiety and trait anxiety. The trait anxiety is the individualized anxiety level that changes with the situation. The Chinese language version of trait anxiety translated and revised by Chung and Long15 was used in this study. It included 7 items of negative questions and 13 items of positive questions. Each item was rated on a 4-point Likert scale, where a higher score indicated a higher anxious personality. Cronbach’s alpha in the present study was 0.88.

Hematology laboratory data and skin measurements were included in physiologic parameters. The laboratory data of creatinine, total calcium, free calcium, phosphate, calcium and phosphorus products (Ca*P), hematocrit, parathyroid hormone (PTH), blood urea nitrogen (BUN), chloride, triglyceride and magnesium were obtained from the medical record. In respect of skin measurement, we adapted the multi probe adaptor (MPA) made by Courage-Khazaka Electronic GmbH in Germany, to measure skin humidity and transepidermal water loss (TEWL). The MPA was corrected and checked regularly by professional engineers to ensure its precision.

Pruritus is a subjective experience of uncomfortable and unpleasant sensation, which elicits the desire to scratch.16 The definition of uremic pruritus by Duo17 is of itching bouts lasting more than 10 minutes, or if not, the total number of itching bouts had to be more than 20 times per half a day. For this study we translated and modified both Duo’s17 and Hung’s18 scales to evaluate severity, frequency, distribution and sleep disturbance during day or night by patients’ recall. In terms of pruritus severity, this was rated on a 4-point Likert scale (0=no pruritus, 1=itching without annoyance or necessity for scratching, 2=a few times of scratching, 3=frequent scratching, 4=scratching without relief of itching, or total restlessness). A maximum of 8 points can thus be given during the day (4 in the morning, 4 in the afternoon). For distribution, a maximum of 6 points can be given over the day, 3 in the morning and 3 in the afternoon. Without pruritus = 0 point; itching in one single location = 1 point; scattered itching = 2 points; generalized itching = 3 points. Frequency of pruritus was judged by the number of itching bouts and the duration of episodes. Every four short itching bouts (< 10 min) or one long bout (≧ 10 min) received 1 point and no pruritus received 0 point. Thus a maximum of 5 points can be given during the day. Sleep disturbance was judged by waking-up periods during the night for scratching. Waking up once because of itching scored 2 points. Thus a maximum of 14 points can be given during the night. The highest possible score was 38 points. In our study, we invited five nephrology, dermatology and nursing experts to check validity. The content validity index (CVI) of the scale was 1. The Cronbach’s alpha was 0.9 at the formal study.

Statistical analysis

For analysis of the variables the following tests were used: mean, standard deviation, frequency, percentage, chi-square test, student’s t test and multiple logistic regression model. The analysis was performed using SPSS 13.0 for Windows. P< .05 was considered statistically significant.

Results

Distribution of basic characteristics

The average age of 110 patients was 53.21±14.59 years (mean±SD), 53.06 years in the itch group and 53.56 years in the non-itch group respectively. The average pre-dialytic CCr was 5.74±1.62ml/min, 5.49 ml/min in the itch group and 6.34 ml/min in the non-itch group respectively. The average Kt/Vurea was 1.87±0.53, 1.86 in the itch group and 1.91 in the non-itch group respectively. The average trait anxiety inventory was 43.08±9.92, 43.07 in the itch group and 40.81 in the non-itch group respectively. Other characteristics are shown in Table 1.

Distribution of physiological parameters

The physiological parameters of creatinine, Ca*P, PTH, BUN, chloride, skin humidity, phosphate, and total calcium among the itch group were statistically significant compared with the non-itch group. Other parameters are shown in Table 2.

Distribution of pruritus in uremic dialytic patients

We used the definition by Zucker et al.8of an itching person, i.e. the appearance of an itch in a regular pattern during a period of six months. Of the 110 patients, 78 (70.9%) had pruritus. Of the 76 peritoneal dialytic patients, 55 (72.4%) had pruritus and of the 34 hemodialytic patients, 23 (67.6%) had prurirtus. The pruritic scale ranged from 0 to 34 (mean±SD: 7.75±7.17). For severity, the score in the afternoon (1.65) was higher than morning (1.45) and for frequency, the score in the afternoon (0.97) was significantly higher than morning (0.64) (t=-3.04,p<0.05). In addition, the average waking-up for scratching during the night was once.

The predictors for uremic pruritus

Univariate logistic regression analysis was performed to define the every variables used for itching and non-itching. The significant univariables included occupation before dialysis, religion, diabetic nephropathy, lupus nephritis, other renal disease, pre-dialytic CCr, creatinine, Ca*P, PTH, BUN, chloride and skin humidity. Then the multivariate analysis of logistic regression was performed. The predictors included occupation before dialysis (Odds=4.11, 95%C. I.: 1.06-15.94), lupus nephritis (Odds=0.08, 95%C. I.: 0.01-0.95), other renal disease (Odds=0.07, 95%C. I.: 0.01-0.70), chloride (Odds=0.87, 95%C. I.: 0.75-0.99) and skin humidity (Odds=0.94, 95%C. I.: 0.88-0.99). They accounted for 40% of the variance in uremic dialytic patients with pruritus (Table 3).

Discussion

After controlling the other variables, the multiple logistic regression revealed the significant factors to be occupation before dialysis, the types of renal disease, chloride, and skin humidity. Patients who worked before dialysis and with unknown renal disease, low chloride and low skin humidity easily incurred uremic pruritus. In terms of literature, from the Medline database 1987-2006, there were only two studies, Mistik et al.19 and Narita et al.,20 who used multiple logistic regression for data analysis. Most studies used univariate logistic regression. As a result, our discussion is divided into two parts with regard to statistics used.

In terms of multiple logistic regression, the types of renal disease, the odds of the occurrence of pruritus among lupus nephritis compared with the unknown renal disease, was significantly lower (Odds=0.08; 95%CI=0.01-0.95) and the odds of occurrence of pruritus among others renal diseases was also significantly lower (Odds=0.07; 95%CI=0.01-0.70). These findings were different from the results in the Mistik et al.’s study.19 The possible explanation for these findings might be that we took the unknown renal disease as a reference group, but Mistik et al.19 did not consider the effect of unknown renal disease. Otherwise, their study showed that the patients with liver disease were more likely to have pruritus, but our study had excluded patients with hepatic failure or biliary tract disorder. The reason was that we considered that patients with liver disease might confound the uremic pruritus. However, as the study by Narita et al.20 did not discuss the types of renal disease, we could not compare our results with that study. In addition, when skin humidity decreased by one unit, the odds of occurrence of pruritus were higher. However, the studies by Mistik et al.19 and Narita et al.20 did not discuss skin humidity in their regression model. Otherwise the calcium, phosphate, BUN and PTH were significant predictors in the model used by Narita et al.20 The results were not consistent with this study. The possible reason might be that the authors separated the variables into categorical variables and the variables as continuous in this study.

According to univariate analysis, we found the skin humidity in nonpruritic patients was significantly higher than in pruritic patients (see Table 2). The result demonstrated that the skin was more humid in nonprurituc patients with similar results in the studies by Robertson and Mueller3 and Szepietowski et al.4 All findings showed that the uremic patients might have stratum corneum, exogenous gland and sebaceous function impairment during the progress of disease, which in turn decreased the skin humidity and xerosis occurred. Xerosis might affect the terminal nerve on the skin surface or decreased the threshold of itching to induce uremic pruritus.3-4 The above results demonstrated that the patients with low skin humidity or electrolyte imbalance incurred pruritus easily.2, 4 In addition, the pre-dialytic CCr, Ca*P, BUN and PTH were significant in the univariate logistic regression in this study. When the pre-dialytic CCr decreased one unit, the odds of occurrence of pruritus was significantly higher (Odds=0.73; 95%CI=0.56-0.95). The reason might be that the worse residual renal function will retain metabolic materials in the body to induce pruritus. Similar results were achieved by Hiroshige and Kuroiwa,2 who showed that the severity of pruritus may reflect the residual renal function. In addition, the Ca*P, BUN and PTH in pruritic patients was significantly higher than non-pruritic patients. Similar results were also present in Ho et al.9 and Hiroshige et al.2 This showed that when BUN and PTH are retained in the body, patients with electrolyte imbalance might easily have pruritus.2, 4, 9

There were still several inconsistent results in the univariate analysis compared with the previous findings. Concerning dermatological characteristics, there was no statistically significant relationship between sex and pruritus, as suggested in previous studies.8, 21 However, the χ2 test of Mistik et al. 19 and Narita et al. 20 showed pruritus was more significant in males than in females. The possible explanation for these findings was that females usually regularly use emollients for skin care. Skin emollients of aqueous gel containing high water content could reduce itching, almost completely improving skin dryness in patients with mild uremic pruritus.22-23