Clinical Study

Am J Nephrol 2001;21:479–485

Chronic Fatigue in Long-Term PeritonealDialysis Patients

Wei-Kuang Changa Kuan-Yu Hungb Jenq-Wen Huangc Kwan-Dun Wuc

Tun-Jun Tsaic

Departments of Internal Medicine, aJen-Ai Municipal Hospital, bFar Eastern Memorial Hospital,cNational Taiwan University Hospital, Taipei, Taiwan, ROC

Received: May 3, 2001Accepted: September 27, 2001

Tun-Jun Tsai, MD, PhDDepartment of Internal Medicine, National Taiwan University HospitalNo. 7, Chung-Shan South RoadTaipei 100, Taiwan (ROC)Tel. +886 2 23123456, ext. 3953, Fax +886 2 23222955, E-Mail paul@ha.mc.ntu.edu.tw

ABCFax + 41 61 306 12 34E-Mail karger@karger.chwww.karger.com

© 2001 S. Karger AG, Basel0250–8095/01/0216–0479$17.50/0

Accessible online at:www.karger.com/journals/ajn

Key WordsChronic fatigue W Peritoneal dialysis W Dialysis adequacy

AbstractObjective: Fatigue is a common symptom in long-termdialysis patients. This study investigated possible clinicalfactors which may cause the development of fatigue inpatients receiving peritoneal dialysis (PD). We also inves-tigated the relationship between total solute clearance(TSC) and fatigue symptoms in PD patients. Design: Across-sectional study design was used to compare theclinical characteristics among groups of PD patients clas-sified by different degrees of fatigue. The relationshipamong dialysis adequacy (including Kt/Vurea and weeklycreatinine clearance; Ccr), clinical characteristics and fa-tigue symptoms were also assessed. Setting: The PDunit of a major university teaching hospital in Taipei, Tai-wan. Patients: Consecutive patients who had receivedPD for a minimum duration of 4 months were recruitedfor participation in the study. Patients were excluded ifthey had a history of ischemic heart disease, severe heartfailure (NYHA function III or IV), malignant neoplasm,active infection, major psychiatric problems, chronic ob-structive pulmonary disease, or disturbed conscious-ness. Finally, a total of 64 patients, 31 of whom werereceiving continuous ambulatory peritoneal dialysis and33 who were receiving continuous cycling-assisted peri-toneal dialysis, were enrolled in the study. Methods:

Fatigue was evaluated using a specially designed ques-tionnaire that includes fourteen items. Patients weredivided into three groups according to their fatiguescores (FS): mild (FS, 0–3), moderate (FS, 4–8), andsevere (FS, 9–14) fatigue. The demographic data, dialysisvariables, and clinical parameters of patients were com-pared among these groups. The relationship betweenfatigue and TSC was also examined. Results: The FSwere correlated with serum intact parathyroid hormone(iPTH) level and total cholesterol concentration (p ! 0.05).A linear correlation was also noted between serum iPTHlevel and total cholesterol level. When the patients weredivided into an adequate- and an inadequate-dialysisgroup according to values of TSC, Kt/Vurea as well asweekly creatinine clearance, a significant correlation wasfound between weekly Ccr and FS. Conclusion: Thisstudy has demonstrated that dialysis adequacy playsa key role in the development chronic fatigue. In addi-tion, weekly Ccr was better correlated with fatigue thanKt/Vurea.

Copyright © 2001 S. Karger AG, Basel

Fatigue is a common symptom and troublesome prob-lem in long-term dialysis patients [1]. It is difficult toresolve, and may influence the quality of life. A variety offactors may contribute to the development of fatigue,including anemia, malignant neoplasm, hypothalamic-pituitary-adrenal dysfunction, malnutrition, severe heart

Dow

nloa

ded

by:

Uni

vers

ity o

f Brit

ish

Col

umbi

a

19

8.14

3.46

.65

- 2/

8/20

16 1

:18:

57 A

M

480 Am J Nephrol 2001;21:479–485 Chang/Hung/Huang/Wu/Tsai

disease, major depression and infection [2–4]. Postdialy-sis fatigue had been discussed in the hemodialysis popula-tion [5, 6]. Contributing factors to postdialysis fatigue,such as changes of blood pressure and osmotic disequilib-rium, are often absent in patients undergoing peritonealdialysis (PD). However, fatigue symptoms are still com-monly observed in patients undergoing long-term PD.Nevertheless the factors which may contribute to thedevelopment of fatigue in long-term PD patients remainundetermined. This study aimed to identify the possiblefactors that may cause the development of fatigue in long-term PD patients.

Materials and Methods

SubjectsOne hundred and seventeen patients treated at our PD unit were

recruited for participation in this cross-sectional study during theperiod from April 2000 to June 2000. Patients with the followingcomorbid conditions were excluded: severe heart failure (NYHAfunction III or IV), malignant neoplasm, active infection, major psy-chiatric problems (e.g. schizophrenia, major depression), chronicobstructive pulmonary disease, and disturbed consciousness. Pa-tients with a duration of PD less than 4 months were also excluded. Atotal of 66 patients who fulfilled the criteria were enrolled and com-pleted the questionnaire at the time of their monthly interview byone of our investigators. No intercurrent events, such as peritonitis,surgery, hospitalization, occurred in any of the patients within 1month before the administration of the questionnaire. Two patientswho were unable to cooperate in checking the peritoneal solute clear-ance were excluded from the analysis. A total of 64 patients (20 malesand 44 females), completed the questionnaire. In addition, 40healthy persons (25 males and 15 females) were selected as the con-trol group to complete the questionnaire.

Demographic data and data for dialysis variables of the patientswere collected by review of medical records. Primary renal diseasewas classified into five major categories: chronic glomerulonephritis,diabetes mellitus, hypertension, systemic lupus erythematosus, andothers. Comorbid conditions were defined by the number of individ-ual comorbid conditions presented within 10 years after the onset ofend-stage renal disease according to the definition of USRDS [7].

Clinical ParametersData on clinical characteristics were collected including age,

duration of PD, body height, body weight, body mass index, hemato-crit (Hct) levels, serum albumin, serum total calcium, serum phos-phate, serum total cholesterol, intact parathyroid hormone (iPTH,chemiluminescence immunoassay), residual renal function, Kt/Vurea,weekly creatinine clearance (Ccr), and normalized protein nitrogenappearance (nPNA). Data on laboratory parameters, which wereaveraged from the results of the last three monthly measurements,were collected for each patient. Kt/Vurea, weekly Ccr, and nPNA, werecalculated according to the standards of the NKF-DOQI guidelines[8]. Residual renal function was assessed by measuring the renalcomponent of Kt/Vurea as Krt/Vurea. The renal component of weeklyCcr was expressed as Crcr.

Table 1. Clinical parameters and FS of patients receiving PD (n =64)

Patients (CAPD/APD) 31 /33Male/female 20 /44Age, years 44.0B12.5 (22–74)PD duration, months 39.2B39.4 (4–216)Body height, cm 159.7B7.62Body weight, kg 55.1B10.5Body mass index (kg/m2) 21.5B3.4Primary renal disease

Chronic glomerulonephritis 39 (60.9%)Diabetic nephropathy 4 (6.3%)Hypertension 5 (7.8%)Lupus nephritis 6 (9.4%)Others 10 (15.6%)

Fatigue scoresPhysical FS 4.1B2.1Mental FS 1.7B1.1Total FS 5.8B2.7

Dialysis parametersKrt/Vurea 0.18B0.24 (0–1.07)Kt/Vurea (peritoneal + renal) 2.35B0.39 (1.59–3.44)Weekly CrCr, l/week/1.73 m2 7.49B10.4 (0–41.9)Weekly Ccr, l/week/1.73 m2 60.0B12.6 (31.4–93.2)nPNA, g/kg body weight/day 1.13B0.23 (0.70–1.67)

Peritoneal Equilibration TestHigh transport high, 2 (3.1%);

high average, 18 (28.1%) Low transport low, 6 (9.4%);

low average, 38 (59.4%)Laboratory data

iPTH (pg/ml) 358.4B291.2 (31.5–1407)Hemoglobin/Hct, % 9.0B1.2 (7.3–13.9)/

28.0B3.3 (21.1–41.7)Serum albumin, g/dl 3.7B0.4 (3.1–4.7)Serum total cholesterol, mg/dl 204.5B36.6 (126.3–319.7)Serum total calcium, mmol/l 2.45B0.16 (1.94–2.90)Serum phosphate, mg/dl 5.71B1.43 (2.93–10.47)Calcium phosphorus product 55.9B14.5 (26.7–101.5)

PD RegimenWe previously described the PD regimen used in this study [9].

Briefly, our patients on chronic ambulatory peritoneal dialysis(CAPD) received four 2-liter exchanges of dextrose solution daily.The dextrose solution used was mainly 1.5% Dianeal®. One or twonighttime exchanges of 2.5 or 4.25% Dianeal were used for patientsin need of ultrafiltration. An additional daily 2-liter exchange wasalso given if Kt/Vurea or weekly Ccr did not reach the expected levelsbased on the DOQI guidelines [8]. 2.5-liter Dianeal was not habitual-ly used for limits of peritoneal capacity. Patients on continuouscycling-assisted peritoneal dialysis (CCPD) received a 2-liter rapidexchange on the cycler every 2 h for 5 times (total 10 liters of Dianeal)during the night and one exchange of 2.5 or 4.25% Dianeal duringthe day. The nighttime exchanges included two bags of dextrose solu-tion composed of 5 liters of 1.5 or 2.5% Dianeal. Adjustment of theprogrammed dialysis prescription for nighttime and daytime ex-change was also performed based on convenience for the patients andthe levels of Kt/Vurea and weekly Ccr.

Dow

nloa

ded

by:

Uni

vers

ity o

f Brit

ish

Col

umbi

a

19

8.14

3.46

.65

- 2/

8/20

16 1

:18:

57 A

M

Age, years

Chronic Fatigue in PD Am J Nephrol 2001;21:479–485 481

Table 2. Comparison of clinical parametersin PD patients with different degrees offatigue

Mild(n = 26)

Moderate(n = 23)

Severe(n = 15)

p value

41.4B11.8 48.9B13.2 41.0B10.8 0.061Male/female 8/18 5/18 7/8 0.278Body height, m 1.60B0.071 1.58B0.081 1.61B0.079 0.570Body weight, kg 55.1B11.4 55.98B10.51 53.5B9.4 0.783Body mass index (kg/m2) 21.3B3.3 22.3B3.5 20.7B3.6 0.372Kt/Vurea 2.25B0.36 2.45B0.34 2.38B0.47 0.206Krt/Vurea 0.19B0.27 0.14B0.18 0.21B0.28 0.513Weekly Ccr, liters/kg/week 64.32B14.85 59.24B10.09 55.67B12.59 0.105Weekly CrCr, liters/kg/week 7.28B12.6 6.01B8.64 7.77B10.9 0.634NPNA 1.11B0.19 1.02B0.22 1.14B0.23 0.149IPTH 314.4B281.1 301.8B211.6 521.8B364.3 0.043Hematocrit, % 27.8B3.7 27.8B3.0 28.6B3.4 0.720Albumin 3.71B0.38 3.67B0.35 3.63B0.39 0.814Total cholesterol 201.0B31.5 221.8B42.8 183.9B20.2 0.005Sodium, mmol/l 136.7B4.1 135.6B4.3 135.6B4.2 0.576Potassium, mmol/l 4.2B0.7 3.9B0.7 3.9B0.9 0.258Magnesium, mmol/l 0.93B0.18 0.86B0.12 0.91B0.17 0.335Calcium, mmol/l 2.43B0.13 2.50B0.12 2.42B0.25 0.234Phosphate, mg/l 5.74B1.19 5.67B1.54 5.69B1.71 0.985Calcium phosphate product 55.80B12.12 56.34B14.49 55.51B18.70 0.984Duration of dialysis, months 30.60B28.94 41.53B45.39 50.62B44.52 0.280

Values are expressed as mean B standard deviation (SD).

Assessment of FatigueFatigue was evaluated using the questionnaire developed by

Chalder et al. [10] (Appendix). The questionnaire, which containsfourteen items, is a comprehensive and validated instrument forfatigue evaluation. There were eight items investigating physicalsymptoms and six items investigating mental symptoms. Each itemof the fatigue questionnaire was transformed to a scale of 0 or 1, witha higher score indicating more fatigue. The highest possible score fora patient, therefore, was 14. Forty healthy subjects were selectedfrom the general population for the assessment of fatigue with thesame questionnaire.

Study DesignAccording to the severity of fatigue, patients were divided into

three groups: mild, moderate and severe. The cut-off values in thesethree groups were: mild, fatigue scores (FS) of 0–3; moderate, FS of4–8, and severe, FS of 9–14. The relationship between total soluteclearance (TSC) and the severity of fatigue were also investigated.For this purpose, patients were then divided into adequate-dialysisand inadequate-dialysis groups, based on the results of TSC, usingeither Kt/Vurea or weekly Ccr, as we previously described [9]. Briefly,and according to the DOQI guidelines [8], the selected cutoff value ofKt/Vurea was 2.0 and 2.1 for CAPD and CCPD patients, respectively.The cutoff value of weekly Ccr was 60 liters/week/1.73 m2 for CAPDpatients and 63 liters/week/1.73 m2 for CCPD patients.

StatisticsValues are expressed as mean B standard deviation (SD). A p

value of less than 0.05 was considered statistically significant. One-way analysis of variance (ANOVA) was used to compare data on age,

body height, body weight, body mass index, duration of PD, renalcomponent of solute clearance (Ktr/V and weekly Crcr), total soluteclearance (Kt/Vurea and weekly Ccr), and clinical laboratory parame-ters of the patients. Student’s t test was used to compare the total FSof the patients between the adequate- and the inadequate-dialysisgroup. Linear regression analysis and curve regression analysis wereused to access the correlation between weekly Ccr and clinical param-eters, including total FS, iPTH, and total cholesterol concentration.

Results

The mean age of the control group was 39.6 B 10.7years, and the mean FS was 3.7 B 2.1. The mean age ofthe patients participating in this study was 44.0 B 12.5years, and among them 4 persons were older than 65years. The causes of primary renal disease were: chronicglomerulonephritis, 39 (60.9%) patients; diabetes melli-tus, 4 (6.3%) patients; hypertension, 5 (7.8%) patients;systemic lupus erythematosus, 6 (9.4%) patients; and oth-ers, 10 (15.6%) patients (table 1). The FS of PD patientswere: physical, 4.1 B 2.1; mental, 1.7 B 1.0 and total FS,5.8 B 2.7. In contrast, the FS of the normal populationwas: physical, 2.5 B 1.9; mental, 1.3 B 1.3; and totalscores of 3.7 B 2.1. The FS (5.8 B 2.7) of long-term PDpatients was significantly higher than that of the controlgroup (3.7 B 2.1) (p ! 0.05).

Dow

nloa

ded

by:

Uni

vers

ity o

f Brit

ish

Col

umbi

a

19

8.14

3.46

.65

- 2/

8/20

16 1

:18:

57 A

M

Physical

482 Am J Nephrol 2001;21:479–485 Chang/Hung/Huang/Wu/Tsai

Table 3. FS and dialysis adequacyFS Kt/Vurea

adequate(n = 49)

inadequate(n = 15)

p value Weekly Ccr

adequate(n = 26)

inadequate(n = 38)

p value

3.9B2.6 3.1B2.5 0.285 2.9B1.9 4.9B1.7 0.000Mental 1.7B1.6 1.3B1.2 0.369 1.2B1.1 2.0B1.2 0.005Total FS 5.6B3.9 4.3B3.5 0.270 4.2B2.3 6.9B4.2 0.000

Dialysis adequacy was determined based on both Kt/Vurea and weekly Ccr.Adequate Kt/Vurea: defined as Kt/Vurea 6 2.0 in CAPD and Kt/Vurea 6 2.1 in CCPD.Adequate weekly Ccr: defined as weekly Ccr 6 60 l/kg/week and weekly Ccr 6 63 l/kg/week

in CCPD.

According to their FS, patients were further dividedinto mild FS (total FS, 0–3; 26 patients), moderate FS (to-tal FS, 4–8; 23 patients) and severe FS (total FS, 9–14; 15patients) (table 2). Patients in the severe group had higherserum levels of iPTH and lower serum levels of total cho-lesterol (p ! 0.05). There was no significant differenceamong these groups in demographic parameters and dial-ysis variables, including PD modality, age, body height,body weight, body mass index, nPNA, Hct, albumin, cal-cium, phosphate, calcium phosphorus product, and dura-tion of PD. The mean levels of Kt/Vurea and weekly Ccr ofpatients from the three subgroups were not significantlydifferent.

In order to investigate the possible relationship be-tween the total clearance indices (Kt/Vurea and weekly Ccr)and the FS (table 3), patients were divided into adequate-and inadequate-dialysis groups by the TSC goals as rec-ommended in the DOQI guidelines [8]. If we divided thepatients into adequate and inadequate dialysis based onurea clearance (Kt/Vurea) alone, there were 49 patients inthe adequate-dialysis group and 15 patients in the inade-quate-dialysis group. As shown in table 3, no significantdifference of the FS (physical, mental or total FS) wasfound between the two groups. However, if patients weredivided into adequate- and inadequate-dialysis groupsbased on weekly Ccr, the FS was significantly lower in theadequate-dialysis group than in the inadequate-dialysisgroup. Other clinical parameters were not significantlydifferent between these two groups, neither by Kt/Vurea

nor weekly Ccr (data not shown).Linear regression analysis and curve regression were

used to access the relation between the following parame-ters: FS, serum iPTH level, total cholesterol concentra-tion, and weekly Ccr. A curve correlation was notedbetween the weekly Ccr and the total FS (fig. 1). A linearcorrelation was noted between serum iPTH level and total

cholesterol concentration (fig. 2). In contrast, weekly Ccr

was not significantly correlated with either iPTH or totalcholesterol (data not shown).

Discussion

The aim of optimal dialysis is to return the patients’lifestyle and life expectancy to what they would be if theyhad never developed renal disease. Fatigue is more com-mon in long-term dialysis patients than in the normalpopulation [1]. Many factors may contribute to the devel-opment of fatigue in uremic patients, including anemia[1–3, 11, 12], malnutrition [13, 14], cardiovascular dis-ease [1], active infection [4], electrolyte imbalance andhyperparathyroidism [15, 16].

Age and body size have been reported to influence thepresentation of fatigue [2, 3]. Elderly PD patients tend tosuffer from malnutrition and depressive mood, and there-fore might be more likely to suffer from fatigue syndrome.In this study, however, age did not have a significantimpact on the presentation of fatigue (table 2). A largerbody size tends to result in inadequate dialysis. However,Fried et al. [17] reported that neither body size nor bodyweight directly influenced the outcome of PD patients.Only the adequacy of dialysis correlated with the outcomeof PD patients. Our finding that body size was not statisti-cally correlated with FS appears to support their finding.

Anemia, as is well known, is an important contributingfactor to the presentation of fatigue [11, 12]. The use ofrecombinant human erythropoietin to correct the anemiastatus of dialysis patients has been shown to provide aparallel improvement in the quality of life [18]. Similarobservations were also made in healthy individuals [11]and in cancer patients [19, 20]. However, we did not finda correlation between Hct/hemoglobin level and fatigue in

Dow

nloa

ded

by:

Uni

vers

ity o

f Brit

ish

Col

umbi

a

19

8.14

3.46

.65

- 2/

8/20

16 1

:18:

57 A

M

Chronic Fatigue in PD Am J Nephrol 2001;21:479–485 483

Fig. 1. Relationship between weekly Ccr andthe total fatigue scores in PD patients.

Fig. 2. Relationship between total cholester-ol concentration and iPTH level PD pa-tients.

dialysis patients in the present study. Toy et al. [11]reported that anemia did not contribute to chronic fatiguein patients who achieved the target hemoglobin level of612 g/dl. In this study, only 2 patients (3.1%) achieved ahemoglobin level of 612 g/dl, and the hemoglobin/Hctlevel was not significantly different among the threefatigue groups (table 2). Therefore, we were unable to fur-ther define the relationship between anemia and the pre-sentation of fatigue in PD patients.

Malnutrition has also been considered as a possiblecontributing factor to fatigue [13, 14]. Total cholesterol,albumin and nPNA value were often used for evaluatingthe nutritional status in dialysis patients [21]. Brown et al.[22] reported that low cholesterol concentrations resultedin depressive symptoms in elderly people. Wells et al. [23]also found that the mood of patients was more depressiveafter changing to a low-fat diet, and that a depressivemood would influence subjects to feel more fatigue. In the

present study, total cholesterol concentrations were signif-icantly different among patients in the different fatiguegroups (p ! 0.05). Patients in the severe fatigue grouptended to have lower total cholesterol concentrations (ta-ble 2). However, the albumin and nPNA levels were notstatistically different among the three groups. There aretwo possible explanations for these observations. First,albumin concentration per se does not merely reflect theintake of protein, and many types of comorbidity as wellas peritoneal albumin loss would influence the albuminconcentration [24]. Second, the majority of our patientsreached the target level of albumin. No severe hypoalbu-minemia (albumin ^3.0 g/dl) was found in the PD pa-tient group, and only 6 patients (9.4%) had albumin levels^3.2 g/dl. A similar finding was also noted for nPNA,with only 5 patients (7.8%) having nPNA !0.8 g/kg/day.As most of our PD patients were not malnourished, serumalbumin concentration and nPNA seemed unlikely

Dow

nloa

ded

by:

Uni

vers

ity o

f Brit

ish

Col

umbi

a

19

8.14

3.46

.65

- 2/

8/20

16 1

:18:

57 A

M

484 Am J Nephrol 2001;21:479–485 Chang/Hung/Huang/Wu/Tsai

to have influenced the magnitude of fatigue in our pa-tients.

Massry et al. [25] reported that PTH is one of the ‘ure-mic toxins’. Joborn et al. [15] and Kristofersson et al. [16]also observed a decreased magnitude of fatigue and im-proved muscle strength after parathyroidectomy in pa-tients with primary hyperparathyroidism. In the presentstudy, the severe fatigue group had a higher level of iPTH.Ritz et al. [26] reviewed the literature and found that PTHinfluenced many systems, including the nervous system,hematopoietic system, myocardium, peripheral vessels,skeletal muscles, and the metabolism of carbohydrate andlipid. PTH acts as a ubiquitous cellular signaling systemvia intracellular calcium [27–29]. This intracellular sig-naling may influence muscle strength as well as the pre-sentation of chronic fatigue.

The importance of electrolytes for intracellular signal-ing and transmembrane potential is well known. There-fore, electrolyte imbalance may influence muscle contrac-tion and nerve conductivity; moreover, it may causefatigue. The variation of electrolyte levels that includesodium, potassium, magnesium, calcium and phosphate(table 1) is insignificant whatever the degree of fatigue is.Under these circumstances, we could not elucidate theinfluence of electrolyte on fatigue.

Recently, the NKF-DOQI guidelines have been recom-mended as a standard guideline for achieving adequatedialysis in PD [8]. The DOQI guidelines recommendusing both Kt/Vurea and weekly Ccr to evaluate dialysisadequacy. According to the DOQI guidelines, the selectedcutoff value of Kt/Vurea is 2.0 for CAPD patients and 2.1for CCPD patients. The cutoff value of weekly Ccr is60 liters/week/1.73 m2 for CAPD patients and 63 liters/week/1.73 m2 for CCPD patients [8]. According to thesecriteria, we divided our patients into adequate- and inade-quate-dialysis groups based on the finding for Kt/Vurea

and weekly Ccr. Patients classified by Ccr as having inade-quate dialysis had a higher FS than patients with inade-quate dialysis (table 3). However, no difference in FS wasfound when patients were classified according to Kt/Vurea.Some characteristics of our PD patients might explainthis discrepancy. First, urea is more easily cleared thancreatinine, and most of our patients were of the lowerperitoneal transport type (low, 9.4%; low average, 59.4%).Thus the desired level for Kt/Vurea was easier to achievethan that of the weekly Ccr in our patients. Second, ourpatients had almost no residual renal function. One thirdof our PD patients were anephric (n = 20, 31.2%), andanother third (n = 18, 28.1%) were nearly anephric (Krt/Vurea ^0.1). In the failing kidney, a large portion of uri-

nary creatinine content results from tubular secretionrather than glomerular filtration [30]. In contrast, urea isresorbed to a significant degree in the renal tubules, andresorption continues to be a major factor even in the fail-ing kidney. In anephric patients, creatinine clearance ismuch lower relative to urea clearance. In other words, cre-atinine clearance depends more on residual renal functionthan urea clearance. In the present study, linear regressionanalysis showed that patients who had more adequateweekly Ccr also had less fatigue (fig. 1).

In terms of ANOVA, there is no correlation betweenweekly Ccr and fatigue. We have applied a bivariate corre-lation to compare the natural logarithm of weekly Ccr andfatigue in order to clarify the said discrepancy. A statisti-cal correlation (p ! 0.05) between the natural logarithm ofthese two variables is noted. The discrepancy could havetwo explanations. First of all, according to the recommen-dations of the DOQI guidelines, different modalities ofPD demand different dosages of dialysis. The value ofweekly Ccr per se cannot determine whether a patient getsadequate dialysis or not. Thus we could not find a statisti-cal correlation between FS and the weekly Ccr. Secondly,the discrepancy could have resulted from a limited pool ofpatients. Since there were only 64 patients enrolled for thestudy; no statistical correlation was noted between the twovariables except in their logarithmic value.

A linear correlation was also noted between the iPTHlevel and total cholesterol concentration (fig. 2). Massry etal. [31] and Akmal et al. [32] reported that serum PTHmight influence the metabolism of lipids, especially thatof triglycerides, and result in hyperlipidemia. However,Schaefer et al. [33] and Navarro et al. [34] observed thatserum PTH levels were significantly correlated with lipidprofiles. In contrast, patients with higher iPTH levels inthe present study had less total cholesterol. Further studymay help us to clarify this issue.

In conclusion, 59% of our PD patients suffered fromchronic fatigue (23 in the moderate and 15 in the severegroup). The FS were statistically correlated with the week-ly Ccr. This study demonstrated two important findings.First, dialysis adequacy plays a key role in the presenta-tion of chronic fatigue. Second, weekly Ccr is better corre-lated with chronic fatigue than Kt/V.

Acknowledgements

The authors express their appreciation to the Ta-Tung KidneyFoundation and Mrs. Hsiu-Chin Lee Kidney Research Foundationfor grant support.

Dow

nloa

ded

by:

Uni

vers

ity o

f Brit

ish

Col

umbi

a

19

8.14

3.46

.65

- 2/

8/20

16 1

:18:

57 A

M

Chronic Fatigue in PD Am J Nephrol 2001;21:479–485 485

Appendix

The Fourteen Items of the Fatigue Scale

Physical symptoms1 Do you have problems with tiredness?2 Do you need to rest more?3 Do you feel sleepy or drowsy?4 Do you have problems starting things?5 Do you start things without difficulty but get weak as you go on?6 Are you lacking in energy?7 Do you have less strength in your muscles?8 Do you feel weak?

Mental symptoms9 Do you have a difficulty concentrating?

10 Do you have problems thinking clearly?11 Do you make slips of the tongue when speaking?12 Do you find it more difficult to find the correct word?13 How is your memory?14 Have you lost interest in the things you used to do?

References

1 Cardenas DD, Kutner NG: The problem offatigue in dialysis patients. Nephron 1982;30:336–340.

2 Lawrie SM, Manders DN, Geddes JR, PelosiAJ: A population-based incidence study ofchronic fatigue. Psychol Med 1997;27:343–353.

3 Wessely S, Chalder T, Hirsch S, PawlikowskaT, Wallace P, Wright DJ: The prevalence andmorbidity of chronic fatigue and chronic fa-tigue syndrome: A prospective primary carestudy. Am J Public Health 1997;87:1449–1455.

4 Wessely S, Chalder T, Hirsch S, PawlikowskaT, Wallace P, Wright DJ: Postinfectious fa-tigue: Prospective cohort study in primary care.Lancet 1995;345:1333–1338.

5 Sklar AH, Riesenberg LA, Silber AK, AhmedW, Ali A: Postdialysis fatigue. Am J KidneyDis 1996;28:732–736.

6 Sklar A, Newman N, Scott R, Semenyuk L,Schultz J, Fiacco V: Identification of factorsresponsible for postdialysis fatigue. Am J Kid-ney Dis 1999;34:464–470.

7 The USRDS 1997 Annual Data Report IV:The USRDS dialysis morbidity and mortalitystudy: Wave 2. Am J Kidney Dis 1997;30(suppl 1):S67–S85.

8 NKF-DOQI clinical practice guidelines forperitoneal dialysis adequacy. Am J Kidney Dis1997;30(suppl 2):S67–S136.

9 Chen YC, Hung KY, Kao TW, Tsai TJ, ChenWY: Relationship between dialysis adequacyand quality of life in long-term peritoneal dialy-sis patients. Perit Dial Int 2000;20:534–540.

10 Chalder T, Berelowitz G, Pawlikowska T,Watts L, Wessely S, Wright D, Wallace EP:Development of a fatigue scale. J PsychosomRes 1993;37:147–153.

11 Toy P, Feiner J, Viele MK, Waston J, Yeap H,Weiskopf RB: Fatigue during acute isovolemicanemia in healthy, resting humans. Transfu-sion 2000;40:457–460.

12 Moreno F, Aracil FJ, Perez R, Valderrabano F:Controlled study on the improvement of quali-ty of life in elderly hemodialysis patients aftercorrecting end-stage renal disease-related ane-mia with erythropoietin. Am J Kidney Dis1996;27:548–556.

13 Cushman KE: Symptom management: A com-prehensive approach to increasing nutritionalstatus in the cancer patient. Semin Oncol Nurs1986;2:30–35.

14 Grant M, Kravits K: Symptoms and their im-pact on nutrition. Semin Oncol Nurs 2000;16:113–121.

15 Joborn C, Rastad J, Stalberg E, Akerstrom G,Ljunghall S: Muscle function in patients withprimary hyperparathyroidism. Muscle Nerve1989;12:87–94.

16 Kristofersson A, Boström A, Söderberg T:Muscle strength is improved after parathyroid-ectomy in patients with primary hyperparathy-roidism. Br J Surg 1992;79:165–168.

17 Fried L, Bernardini J, Piraino B: Neither sizenor weight predicts survival in peritoneal dialy-sis patients. Perit Dial Int 1996;16:357–361.

18 Moreno F, Sanz-Guajardo D, Lopez-GomezJM, Jofre R, Valderrabano F: Increasing thehematocrit has a beneficial effect on quality oflife and is safe in selected hemodialysis pa-tients. Spanish Cooperative Renal PatientsQuality of Life Study Group of the SpanishSociety of Nephrology. J Am Soc Nephrol2000;11:335–342.

19 Thomas ML: Anemia and quality of life in can-cer patients. Impact of transfusion and erythro-poietin. Med Oncol 1998;15(suppl 1):S13–S18.

20 Itri LM: Optimal hemoglobin levels for cancerpatients. Semin Oncol 2000;27(suppl 4):S12–S15.

21 CANADA-USA (CANUSA) Peritoneal Dialy-sis Study Group: Adequacy of dialysis andnutrition in continuous peritoneal dialysis: As-sociation with clinical outcomes. J Am SocNephrol 1996;7:198–207.

22 Brown SL, Salive ME, Harris TB, SimonsickEM, Guralnik JM: Low cholesterol concentra-tions and severe depressive symptoms in elder-ly people. Br Med J 1994;308:1328–1332.

23 Wells AS, Read NW, Laughare JDE, AhluwaliaNS: Alternation in mood after changing to alow-fat diet. Br J Nutr 1998;79:23–30.

24 Oreopoulos DG: The optimization of contin-uous ambulatory peritoneal dialysis. KidneyInt 1999;55:1131–1149.

25 Massry SG: Is parathyroid hormone a uremictoxin? Nephron 1977;19:125–130.

26 Ritz E, Stefanski A, Rambausek M: The role ofthe parathyroid glands in the uremic syn-drome. Am J Kidney Dis 1995;26:808–813.

27 Akmal M, Massry SG: Role of parathyroid hor-mone in the decreased motor nerve conductionvelocity of chronic renal failure. Proc Soc ExpBiol Med 1990;195:202–207.

28 Smogorzewski M, Massry SG: Uremic cardio-myopathy: Role of parathyroid hormone. KidInt 1997;52(suppl 62):S12–S14.

29 Massry SG, Smogorzewski M: Mechanismsthrough which parathyroid hormone mediatesits deleterious effects on organ function in ure-mia. Semin Nephrol 1994;14:219–231.

30 Blake PG: Creatinine is the best molecule totarget adequacy of peritoneal dialysis. PeritDial Int 2000;20(suppl 2):S65–S69.

31 Massry SG, Akmal M: Lipid abnormalities,renal failure, and parathyroid hormone. Am JMed 1989;87:N42–N44.

32 Akmal M, Kasim SE, Soliman AR, Massry SG:Excess parathyroid hormone adversely affectslipid metabolism in chronic renal failure. KidInt 1990;37:854–858.

33 Schaefer K, Offermann G, von Herrath D,Schroter R, Stolzel R, Arntz HR: Failure toshow a correlation between serum parathyroidhormone, nerve conduction velocity and serumlipids in hemodialysis patients. Clin Nephrol1980;14:81–88.

34 Navarro JF, Teruel JL, Lasuncion MA, Mora-Fernandez C, Ortuno J: Relationship betweenserum parathyroid hormone levels and lipidprofile in hemodialysis patients. Evolution oflipid parameters after parathyroidectomy. ClinNephrol 1998;49:303–307.

Dow

nloa

ded

by:

Uni

vers

ity o

f Brit

ish

Col

umbi

a

19

8.14

3.46

.65

- 2/

8/20

16 1

:18:

57 A

M