Nutritional status affects quality of life in Hemodialysis (HEMO) Study patients at baseline☆☆☆

Article Outline

• Abstract
• Methods
  • Eligibility
  • Population
  • Predictors
  • Protein and energy
  • Kinetic modeling
  • Biochemical
  • Anthropometric
  • Appetite assessment
  • Other characteristics
  • Outcomes
  • Statistical methods
• Results
  • Association with demographics
  • Association of nutrition status indicators with quality of life
    • Relationship of PCS with nutritional indicators, without controlling for comorbidity
    • Relationship of PCS with nutritional indicators, with comorbidity adjustment
    • MCS scores
    • Sensitivity analysis for eKt/V
  • Coexisting disease
  • Clinical case scenarios
• Discussion
• References
• Copyright

Abstract 

Objective: To evaluate associations between frequently used indicators for assessing nutritional status and health-related quality of life in hemodialysis patients after controlling for demographics, comorbidity, and dialysis dose. Design: Survey of 1,387 hemodialysis patients enrolled at baseline in the Hemodialysis (HEMO) Study. Nutritional status indicators included dietary energy intake, equilibrated normalized protein catabolic rate (enPCR), serum creatinine (SCr), serum albumin (SAlb), body mass index (BMI), calf circumference, and appetite. Health-related quality of life was measured by the Medical Outcomes Study Short Form-36 (MOS-SF-36) summary measures: the Physical Component Scale (PCS) and Mental Component Scale (MCS). Setting: Fifteen clinical sites throughout the United States providing in-center hemodialysis. Results: The mean PCS score was 36.1 ± 10 SD, lower than normative data in healthy populations. PCS scores were lower among women, whites, and those with diabetes, severe comorbidities, and poor appetites. Appetite, dietary energy intake, SAlb, and SCr were strongly associated with PCS scores even after controlling for demographics and comorbidity. The sum of the parameter estimates for the effects of nutritional status on PCS was large, 7 points or more depending on the individual's nutritional status indicators. The mean MCS score was 49.7 ± 10.1 SD, similar to scores in healthy populations, but lower among those with severe comorbidities, poor appetites, advanced age, and more years on dialysis. Appetite, age, and years on dialysis were significantly associated with MCS after controlling for other demographics and comorbidity. Dialysis dose did not significantly alter these relationships. Conclusion: Easy-to-use indicators for assessing nutritional status (appetite, energy intake, SAlb, and SCr) together are strongly associated with health-related quality of life, even after controlling for comorbidities and dose of dialysis in hemodialysis patients, providing an additional reason for maximizing patients' nutritional status and health. © 2002 by the National Kidney Foundation, Inc.

POOR NUTRITIONAL STATUS is associated with a poor health-related quality of life and increased morbidity and mortality among chronic dialysis patients.¹, ², ³, ⁴, ⁵, ⁶, ⁷, ⁸, ⁹ Many factors may influence these outcomes, including demographic factors (age, sex, race, and years on dialysis), type and severity of comorbid diseases, glomerular filtration rate at which dialysis is initiated, frequency, dose of dialysis, and possibly membrane flux.

This cross-sectional study examined whether frequently used indicators of nutritional status were associated with health-related quality of life at baseline among patients in the Hemodialysis (HEMO) Study, a National Institutes of Health–sponsored multicenter, prospective, randomized clinical trial examining the effects of dialysis dose and membrane flux on morbidity and mortality of maintenance hemodialysis patients.¹⁰, ¹¹ Summary measures of health-related quality of life from the Medical Outcomes Study Short Form-36 (MOS-SF-36) were used.

Back to Article Outline

Methods 

Eligibility 

Adults in 15 clinical sites between the ages of 18 and 80 years, on chronic hemodialysis for ≥3 months, receiving in center hemodialysis 3 times per week, and with residual renal clearance of <1.5 mL/min were eligible to be randomized. Exclusions included severe malnutrition indicated by a predialysis serum albumin (SAlb) <2.6 g/dL (determined by nephelometry), current active malignancies requiring radiation or chemotherapy, symptomatic acquired immunodeficiency syndrome (AIDS), cirrhosis with encephalopathy, severe congestive heart failure, unstable or new-onset angina pectoris, chronic pulmonary disease, and current hospitalization. The trial was approved by the Institutional Review Board at each of the clinical centers, and informed (written) consent was obtained from all study participants.

Population 

This analysis included 1,387 baseline participants recruited between March 1995 and March 1999 (52% of all subjects later recruited into the study) with complete information at baseline before randomization for all of the variables reported here.

Predictors 

Indicators of nutrition, comorbid medical conditions, and health-related quality of life were assessed during baseline while the participants were being dialyzed at the Kt/V in their usual dialysis prescription and on their usual dialyzer membrane. The measurements used for the indicators of nutritional status in the HEMO study are described in greater detail elsewhere.¹², ¹³

Protein and energy 

Dietary protein and energy intakes (DPI and DEI) were estimated from 2-day diet diary-assisted recalls (1 on a dialysis day, and 1 on a nondialysis day) obtained from participants at each clinical center by a dietitian using standardized procedures described elsewhere.¹², ¹³ Energy and protein intakes were expressed per unit of adjusted body weight; if actual body weight was <90% or >120% of standard body weight (SBW) as determined by the National Health and Nutrition Examination Survey II (NHANES II) (age, sex, frame, and stature-specific population medians), energy and protein intakes were normalized to an adjusted body weight (ABW) by using the following formula: ABW = ([patient's actual weight − SBW] × 0.25) + SBW.¹⁴, ¹⁵

Kinetic modeling 

The kinetic modeling session closest to the date of the 2-day dietary recall was used to estimate equilibrated Kt/V (eKt/V), assuming that patients were in a stable state and not catabolic. The equilibrated normalized protein catabolic rate (enPCR) was calculated from the patient's urea generation rate using double-pool urea kinetic modeling.¹⁶, ¹⁷, ¹⁸ In the steady state, enPCR approximates DPI, and thus provides a biochemical indicator of protein intake.

Biochemical 

Serum albumin was obtained from predialysis blood samples and measured at the study's Central Biochemistry Laboratory (LifeChem Clinical Laboratories [now known as Spectra East], Rockleigh, NJ) by nephelometry. A predialysis serum creatinine (SCr) (not necessarily obtained on the kinetic modeling day) was assayed at each clinical center's laboratory, and the results were transmitted to the Data Coordinating Center.

Anthropometric 

Predialysis and postdialysis weights, heights, and calf circumferences (after dialysis) were obtained during the baseline period using standardized techniques by trained and certified study dietitians at the clinical centers, and body mass index (BMI) was calculated.¹², ¹³

Appetite assessment 

A patient-reported appetite and diet assessment tool (ADAT) was administered on the day the dietary record was taken.¹⁹ The ADAT is a short questionnaire that patients complete to rate their appetite.

Other characteristics 

Age, sex, race, years on dialysis (vintage), diabetes status (diabetic [either non–insulin dependent or insulin-dependent] or nondiabetic), membrane flux, and eKt/V were obtained from medical records at baseline. The Index of Co-existing Disease (ICED) score, a more complete estimate of the patient's comorbidity status, was obtained from review of the patient's morbidities in the medical record at baseline.²⁰ The ICED is a composite comorbidity coding system that classifies the presence and severity of diseases and the impact of the disease on physical function as normal, mild, moderate, or severe disease comorbidity, respectively, using scores of 0, 1, 2, or 3.²¹, ²²

Outcomes 

Health-related quality of life was measured by the MOS-SF-36, a subjective self-administered questionnaire that explores 8 of the most frequently used generic health concepts (physical functioning, role limitations caused by physical problems, bodily pain, general health perceptions, vitality, social function, role limitations caused by emotional problems, and general mental health).²³ A higher score indicates better function.

The 2 summary measures of MOS-SF-36 scales, the Physical Component scale (PCS), a measure of physical health, and the Mental Component scale (MCS), a measure of emotional function, were used to reduce the number of comparisons for this analysis. The PCS is a valid measure of physical health, and scores are available in various disease states. The MCS correlates most highly with the mental health, role-emotional, and social functioning subscales. Both the PCS and MCS scores have reliability estimates of 0.90 or more, and 95% confidence intervals around the 2 summary measures are low, approximately 6 to 7 points.²⁴ The scoring algorithms of Ware et al²⁴ were used to calculate PCS and MCS, with each score being normalized to a population mean of 50 with a standard deviation of 10.

Statistical methods 

Linear regression, t-tests, and analysis of variance were used, as appropriate, to describe the pairwise relationships of the PCS and MCS with the individual nutrition, demographic, and comorbidity indices. Three multivariate analyses assessed the association of PCS and MCS scores with nutritional factors and further evaluated the extent to which this association could be accounted for by the demographic and comorbidity indices. The nutrition factors selected for the multivariate model were deemed to be both of great clinical importance and amenable to dietary influences: dietary energy intake, SAlb, enPCR, SCr, BMI, calf circumference, and appetite assessment score. Dietary protein intake was not included in the multivariate model to avoid collinearity with enPCR. C-reactive protein was not included because it was not measured in a sufficient number of patients. Erythropoietin dose levels, which have a powerful influence on hematocrit levels and probably thereby on quality of life, were not available and so hematocrit was not included. In future studies, hematocrit data should be included as covariates when examining health-related quality of life.

In the simplest multivariate analysis, multiple linear regression was used to relate the nutritional factors to the PCS and MCS scores while controlling for demographics (age, race, sex, and years since initiation of dialysis). A more inclusive multivariate model controlled for these same predictor variables as well as the ICED score and diabetes status to determine whether the associations with the nutritional variables with the PCS and MCS scores persisted even after controlling for comorbidity. By comparing the simpler and more inclusive models, the effects of comorbidities on attenuating the results could be examined. A sensitivity analysis was also used in which baseline eKt/V was added to the predictor variables from the second model to assess whether adjustment for eKt/V had any further effect on the modeled relationships beyond those accounted for by controlling for demographic and comorbidity factors. We focus on the more complete models to assess final nutritional effects.

It was not possible to assume a priori that the relationships of the MCS and PCS scores with the nutritional indices were linear. Therefore, we used restricted cubic splines with knots at the 33rd and 66th percentiles of the continuous demographic and nutritional indices to test for the presence of nonlinear associations, which would indicate different relationships with the quality-of-life scores at different levels of the demographic and nutritional indices.²⁵ For most of the nutritional indices, simple linear models were adequate. We retained SCr, BMI, and age, the nonlinear models in the analysis of the PCS, because they provided significantly better fits than the linear models at P < .025.

We also investigated whether relationships of the PCS and MCS with the nutritional indicators differed between subgroups of patients defined by demographic and comorbidity factors by individually testing the significance of interaction terms between these prespecified pairs of demographic (age) or comorbidity (ICED) and nutritional indices selected from clinical considerations in the multivariate models for DEI, SAlb, enPCR, SCr, BMI, calf circumference, and appetite assessment. Unless stated otherwise, all hypothesis tests were regarded as statistically significant at P < .05 (2-sided).

Back to Article Outline

Results 

Table 1 shows that age and primary renal diagnosis of the 1,387 patients were comparable with the 1999 USRDS, but that more of the patients were women and blacks.²⁶

Table 1. Comparison of HEMO study participants at baseline and selected indices from the 1999 United States Renal Data System (USRDS) prevalent patients

Association with demographics 

Table 2 shows that men, blacks, and people with fewer comorbidities (as indicated by absence of diabetes and lower ICED scores) and better appetites had higher PCS scores.

Table 2. Mean MOS-SF-36 PCS and MCS scores in various subgroups of the HEMO study

Those with better appetites and lower ICED scores had higher MCS scores. For all participants, mean PCS scores were 11 points lower than the mean scores in the US population for healthy individuals. In contrast, the MCS scores in the HEMO study cohort were virtually identical to the US population means.²⁴

Association of nutrition status indicators with quality of life 

Relationship of PCS with nutritional indicators, without controlling for comorbidity 

Table 3 presents the results of the regression analyses relating PCS scores to nutritional measures with progressively more extensive levels of adjustment for potential confounding factors. When the PCS scores were related to each parameter individually (model 1), each of the nutrition, demographic (with the exception of years on dialysis), and comorbidity indicators were significant at the P < .05 level. In model 2, the nutrition indices were related to the PCS in a multivariate analysis that also controlled for the demographic factors. Appetite, SAlb, SCr, BMI, calf circumference, and age were all independently related to the PCS (P < .05). The relationships of the PCS with SAlb and calf circumference were both approximately linear (a 0.5-g/dL higher SAlb was associated with a 1.09 ± 0.39 higher PCS, and a 1-cm higher calf circumference was associated with a 0.21 ± 0.08 higher PCS, throughout the ranges of these variables). A very good/good/fair rating on the appetite assessment was associated with a 3.94 ± 0.90 higher PCS than a poor/very poor appetite rating, after adjustment for the other factors in the model. The relationships of SCr, BMI, and age were all significantly nonlinear. The association of the PCS with SCr was stronger at lower than at higher SCr levels (a 1-mg/dL higher SCr was associated with a 0.76 ± 0.19 higher mean PCS for a SCr level of 8 mg/dL, but with only a 0.14 ± 0.23 higher PCS for a SCr of 15 mg/dL; Table 3, model 2). The PCS was inversely related to BMI at lower and especially at higher BMI levels, but not for intermediate BMI levels between approximately 20 and 28 kg/m² (a 1-kg/m² higher BMI was associated with a 0.23 ± 0.09 lower PCS score when BMI = 32.1 kg/m² and a 0.04 ± 0.04 higher PCS when BMI = 24 kg/m² [P = not significant]). Those with a BMI in the obese range (BMI > 30) and especially those over a BMI of 35 had lower PCS scores than their lighter counterparts. After adjusting for nutritional indicators and other demographic variables, from approximately 25 until 50 years of age, PCS declined and then leveled off.

Table 3. PCS score parameter estimates with univariate model (model 1), multivariate model controlling for demographics and nutrition indicators (model 2), and multivariate model controlling for demographics, nutrition indicators, and comorbidity (model 3)

Relationship of PCS with nutritional indicators, with comorbidity adjustment 

Significant relationships between appetite, SAlb, SCr, BMI, age, and PCS largely persisted after adjusting for the comorbidity indices in addition to the other demographic and nutrition factors (Table 3, model 3). The effects of appetite were largest after adjustment for comorbidity. Patients with good or very good appetites had PCS scores that were on average 3.83 points higher than those of individuals with poor or very poor appetites compared with a 3.94-point difference without comorbidity adjustment. A 0.5-g/dL higher SAlb was associated with a 0.87-higher PCS with comorbidity adjustment, as compared with a 1.09 higher PCS without. With comorbidity adjustment, dietary energy intake was positively and significantly associated with the PCS score only among individuals with relatively lower comorbidity (eg, the ICED score was normal or mild). Among them, for an additional 10 kcal/kg/d of energy intake, the PCS increased on average by 1.31 points. However, among those with ICED scores that indicated moderate or severe morbidity, there was no significant association between dietary energy intakes and PCS. The ICED was also a significant independent predictor of PCS. An ICED score in the normal or mild range at the median dietary energy intake value was associated with a PCS score 2.67 points higher on average than moderate or severe ICED scores. Among the demographic variables, only age remained significantly associated with PCS in model 3.

MCS scores 

Table 4 shows that only indicators of appetite, BMI, calf circumference, age, and years on dialysis had statistically significant univariate relationships with the MCS score (model 1).

Table 4. MCS score parameter estimates with univariate model (model 1), multivariate model controlling for demographics and nutrition indicators (model 2), and multivariate model controlling for demographics, nutrition indicators, and comorbidity (model 3)

After adjusting for demographics, only calf circumference, age, and years on dialysis remained significantly associated with the MCS score (model 2). When comorbidity was also controlled for (model 3), age, years on dialysis, and appetite were significant. Patients with very good, good, or fair appetites had MCS scores that were on average 2.74 points higher than those with poor or very poor appetites (P = .0086). There were no significant interactions found in the final multivariate model for MCS.

Sensitivity analysis for eKt/V 

Because adequacy of dialysis may affect appetite, the possibility of confounding by delivered dialysis dose (eKt/V) was examined using a sensitivity analysis. Neither the analyses of the PCS nor the MCS scores were substantially altered when eKt/V was taken into account.

Coexisting disease 

Table 5 presents adjusted means of the PCS and MCS scores by diabetes status and ICED score after controlling for nutrition indicators and demographic variables.

Table 5. Adjusted means by diabetes status and index of coexisting disease (ICED) controlling for demographics

Diabetes was significantly associated with lower PCS scores, lower values for SAlb, SCr, dietary energy intake, and higher BMI. Patients with severe ICED scores had significantly worse mean PCS and MCS levels compared with those with normal or mild ICED scores (differences of 5.1 and 2.2, respectively). Those with normal or mild ICED scores also had significantly higher values for SAlb, enPCR, dietary energy intake, calf circumference, and SCr, but did not have higher BMIs than those with moderate and severe ICED scores.

Clinical case scenarios 

Clinical case scenarios show how nutritional factors might influence SF-36 scores, assuming these models are generalizable to clinical practice. Case 1 represents an obese patient (BMI, 32.1) with a good appetite and high levels (90th percentile from the HEMO study) of the nutrition-related biochemical indicators SCr, SAlb, and enPCR; PCS was 37.3 ± 1.7 SE, and MCS was 52.1 ± 2.0 SE. Case 2 represents a well-nourished individual at normal weight (BMI, 24) with good appetite and high biochemical indicators (90th percentile from the HEMO study); PCS was 37.0 ± 1.7 SE, and MCS was 51.7 ± 2.0 SE. Case 3 is an undernourished patient (BMI, 18.5) with a poor appetite and low biochemical indices (10th percentile from the HEMO study); PCS was 26.7 ± 2.0 SE, and MCS was 47.6 ± 2.3 SE. Note that individuals who were obese or of healthy weight, whose nutritional indicators (SAlb, SCr, enPCR) were high normal, and whose appetites were good had substantially higher PCS (>10 points) and MCS (>4 points) scores than undernourished patients.

Back to Article Outline

Discussion 

Several nutritional status indicators (appetite, dietary energy intake, SAlb, and SCr) were significantly associated with the physical aspects of health-related quality of life as measured by PCS, the summary score reflecting physical outcomes associated with underlying disease. Combined indicators of poor nutrition (poor appetite and low SAlb, SCr, and enPCR) were related to average decreases in the PCS of more than 10 points, whereas poor nutrition depressed MCS by an average of only 4 points even after controlling for demographic and comorbidity confounders. To put this in a clinical context, the effects were at least as large as those found from before epoietin alpha treatment to after treatment in Beusterian's study.²⁷ Merkus et al²⁸ found, in their cohort, that only one third of the differences in health-related quality of life could be explained by clinical factors (hemoglobin, cardiovascular disease, and PCR) and comorbidities. Our findings suggest that nutritional variables may explain additional variability in patients' quality of life.

Other studies have assessed the impacts of nutrition on quality of life by examining only a few indicators of nutritional status, but often without controlling for demographics and comorbidity.⁹, ²⁷, ²⁹, ³⁰, ³¹ We found that age was independently associated with at least 1 of the quality-of-life summary scores. Other studies have also found that age is independently associated with decreased physical functioning.²⁸, ²⁹, ³⁰, ³², ³³ When we modeled age so that it could assume different curvilinear forms over different portions of the outcome values, the negative effects associated with age on PCS were attenuated, on average, for patients older than 50 years. In contrast, a significant linear increase was observed between MCS and increasing age, a trend that was also noted by Wu et al.³³

Several studies, including ours, have also found that increased comorbidity is associated with worsened physical status.²⁹, ³⁰, ³³, ³⁴ Merkus et al²⁹ found that a higher number of comorbid conditions were associated with lower health-related quality of life, and Mingardi et al³⁰ found that the number of comorbid conditions were significantly associated with physical domains relating to quality of life. Few studies have considered dialysis duration as a factor. We found that patients who had been on dialysis for more months had higher MCS scores and that vintage was 1 of only 3 factors that were independently associated with MCS.

This study shows for the first time that increased appetite, as measured by the ADAT, is independently associated with substantially higher MCS and PCS scores. Appetite may serve either as a marker for a patient's poor underlying physical or mental condition and, thus, the need for other interventions, or poor appetite may cause wasting and deterioration of other aspects of the patient's physical and mental status and require nutritional intervention.¹⁹, ³⁵, ³⁶

Serum albumin levels were significantly associated with the PCS and MCS in DeOreo et al's study,³⁰ with the subscales on the SF-36 of physical and mental health in Mingardi et al's study,³¹ and with a mental health indicator, the Index of General Affect, in the study by Laws et al.⁹ We found, as did Diaz-Buxo et al³² and Merkus et al,³⁴ no independent association between SAlb and MCS and showed that increased SAlb was associated with better physical function, a subcomponent of the PCS. We found that PCS improved with increased SCr levels even after controlling for dialysis adequacy (eKt/V), although the size of the effects decreased with SCr ≥ 11 mg/dL, suggesting that they are related to somatic protein concentrations, muscle mass, and nutritional status. In contrast, Merkus et al³⁴ did not find that SCr was significantly associated with quality of life. Other studies did not evaluate it.⁹, ²⁷, ²⁸, ²⁹, ³⁰, ³¹, ³³, ³⁷

Increased DEI was associated with higher PCS only among those patients who were healthier with normal or mild ICED scores, suggesting the need for interventions that decrease or treat the severity and number of comorbidities as well as measures to increase DEI. The effects of SCr on PCS were most evident in the range of 4 to 12 mg/dL and then leveled off. BMIs higher than 30, the cut-off for obesity, and especially higher than 35, which indicates significant obesity, were associated with decreased PCS scores, underlining the importance of keeping weight at healthier levels for quality of life as well as for other reasons.

Hemodialysis patients, like others with chronic disease, have lower health-related quality of life in some, but not all, dimensions.³⁸ Compared with healthy age- and sex-matched controls, their physical health as measured by PCS is lower and more variable than their mental and emotional health as measured by the MCS.²⁷, ²⁹, ³⁰, ³¹, ³⁹, ⁴⁰ The mean PCS scores of our hemodialysis participants were 36.1 ± 10.0 SD, similar to values of 35.2 ± 9.9 SD observed by DeOreo et al,³¹ and 35.5 ± 11.3 SD in Beusterien et al's study²⁷ of comparable groups. Other studies of health-related quality of life using different measures have also reported associations with physical health in dialysis patients.²⁷, ³¹ Mean MCS scores were 49.7 ± 10.1 SD in our study, 47.9 ± 11.6 SD in DeOreo et al's study³¹ and 45.8 ± 11.6 SD in Beusterien et al's study,²⁷ all more comparable with 50 ± 10 SD in the healthy population.

Some dialysis patients are more at risk of poor health-related quality of life than others. In our study, women, those with poor appetites, and people with diabetes and many comorbidities all had lower average PCS scores, although only the relationship with ICED was statistically significant. Mingardi et al³⁰ also found that women and older people on hemodialysis reported lower health-related quality-of-life scores, especially on physical health scales. Our findings extend and clarify these observations.

In undernourished patients seen in clinical practice, several nutrition indicators that affect quality of life are usually affected simultaneously (see case scenarios). The effects of the indicators on quality of life are complex and difficult to predict, but there is good reason to suspect that nutritional interventions to achieve more normal levels would provide clinically significant benefits, with aggregate effects making a difference of 10 points or more, depending on the individual's profile of indicators.

The effects of different nutritional interventions on indicators of nutritional status within ESRD patients with multiple morbidities need further study. Caution is indicated in making interpretations and inferences from these results because of the biases inherent in cross-sectional studies and our measures. For example, those who were depressed or very ill might have low appetites and intakes, and low intakes might decrease their biochemical nutritional indicators. A survivorship bias caused by the exclusion of malnourished patients and those with very high BMIs may also be present. HEMO participants are a selected subgroup of patients, excluding those who are very large. Indicators such as SAlb and SCr are also affected by inflammation, and so the effects noted may not be exclusively caused by dietary intake. Poor appetite is often associated with infections and comorbid conditions as well as with poor dietary intake, reflecting the effects of both poor intake and poor health. Prospective data from the HEMO study will help to clarify the associations between changes in nutritional status, dialysis dose, and type of membrane on quality of life.

Back to Article Outline

References 

1. Acchiardo SR, Moore LW, Latour PA. Malnutrition as the main factor in morbidity and mortality of hemodialysis patients. Kidney Int. 1983;24:S199–S203
   • View In Article
2. Marckmann P. Nutritional status and mortality of patients in regular dialysis therapy. J Intern Med. 1989;226:429–432
   • View In Article
   • MEDLINE
   • CrossRef
3. Oksa H, Ahonen K, Paternack A, et al.  Malnutrition in hemodialysis patients. Scand J Urol Nephrol. 1991;25:157–161
   • View In Article
   • MEDLINE
   • CrossRef
4. Goldwasser P, Mittman N, Antignani A, et al.  Predictors of mortality in hemodialysis patients. J Am Soc Nephrol. 1993;3:1613–1622
   • View In Article
   • MEDLINE
5. Ikizler TA, Hakim R. Nutrition in end-stage renal disease. Kidney Int. 1996;50:343–357
   • View In Article
   • MEDLINE
   • CrossRef
6. Marcen R, Teruel JL, de la Cal A, et al.  The impact of malnutrition on morbidity and mortality in stable hemodialysis patients. Nephrol Dial Transplant. 1997;12:2324–2331
   • View In Article
   • MEDLINE
   • CrossRef
7. National Kidney Foundation . Kidney Disease Outcomes Quality Initiative Clinical Practice Guidelines for Nutrition in Chronic Renal Failure. Am J Kidney Dis. 2000;35(suppl 2):S1–S104
   • View In Article
   • Full Text
   • Full-Text PDF (13 KB)
   • CrossRef
8. Kloppenburg WD, Stegeman CA, Hooyschuur M, et al.  Assessing dialysis adequacy and dietary intake in the individual hemodialysis patient. Kidney Int. 1999;55:1961–1969
   • View In Article
   • MEDLINE
   • CrossRef
9. Laws RA, Tapsell LC, Kelly J. Nutritional status and its relationship to quality of life in a sample of chronic hemodialysis patients. J Renal Nutr. 2000;10:139–147
   • View In Article
10. Eknoyan G, Levey AS, Beck GJ, et al.  The Hemodialysis (HEMO) Study: rationale for the selection of interventions. Semin Dial. 1996;9:24–33
    • View In Article
    • CrossRef
11. HEMO Study Group (Prepared by Greene T, Beck GJ, Gassman JJ, et al) . Design and statistical issues of the Hemodialysis (HEMO) Study. Controlled Clinical Trials. 2000;21:502–525
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (205 KB)
    • CrossRef
12. HEMO Study Group (Prepared by Dwyer JT, Cunniff PJ, Maroni BJ, et al) . The Hemodialysis (HEMO) Pilot Study: nutrition program and participant characteristics at baseline. J Renal Nutr. 1998;8:11–20
    • View In Article
13. Leung J, Dwyer J, Miller J, et al.  The role of the dietitian in a multi-center clinical trial of dialysis therapy: the Hemodialysis (HEMO) Study. J Renal Nutr. 2001;11:101–108
    • View In Article
14. Frisancho AR. New standards of weight and body composition by frame size and height for assessment of nutritional status of adults and the elderly. Am J Clin Nutr. 1984;40:808–819
    • View In Article
    • MEDLINE
15. Wilkens K, Brouns-Schiro K. Suggested Guidelines for Nutritional Care of Renal Patients. Chicago, IL: American Dietetic Association; 1992;
    • View In Article
16. Sargent JA, Gotch F, Borah M, et al.  Urea kinetics: a guide to nutritional management of renal failure. Am J Clin Nutr. 1978;31:1696–1702
    • View In Article
    • MEDLINE
17. Goldstein DJ, Federico CB. Effects of urea kinetic modeling on the nutrition management of hemodialysis patients. J Am Diet Assoc. 1987;87:474–479
    • View In Article
    • MEDLINE
18. Depner TA, Daugirdas JT. Equations for normalized protein catabolic rate based on two-point modeling of hemodialysis urea kinetics. J Am Soc Nephrol. 1996;7:780–785
    • View In Article
    • MEDLINE
19. Burrowes JD, Powers SN, Cockram DB, et al.  Use of appetite and diet assessment tool in the pilot phase of a hemodialysis clinical trial: Mortality and Morbidity in Hemodialysis Study. J Renal Nutr. 1996;6:229–232
    • View In Article
20. Miskulin DC, Athienites NV, Yan G, et al.  Comorbidity assessment using the Index of Coexistent Diseases (ICED) in a multicenter clinical trial. Kidney Int. 2001;60:1498–1510
    • View In Article
    • MEDLINE
    • CrossRef
21. Greenfield S, Blanco D, Elashoff R, et al.  Development and testing of a new index of comorbidity. Clin Res. 1987;35:346A
    • View In Article
22. Greenfield S, Nelson E. Recent developments and future issues in the use of health status assessment measures in clinical settings. Med Care. 1992;30:MS23–MS41
    • View In Article
    • MEDLINE
    • CrossRef
23. Ware JE. SF-36 Health Survey: Manual and Interpretation Guide. Boston, MA: The Health Institute, New England Medical Center; 1993;
    • View In Article
24. Ware JE, Kosinski M, Keller SD. SF-36 Physical and Mental Health Summary Scales: A User's Manual. Boston, MA: The Health Institute, New England Medical Center; 1994;
    • View In Article
25. DeBoor C. A Practical Guide to Splines. New York, NY: Springer-Verlag; 1978;
    • View In Article
26. United States Renal Data System: USRDS . 1999 Annual Data Report. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases,; 1999;
    • View In Article
27. Beusterian KM, Nissenson AR, Port FK, et al.  The effect of recombinant human erythropoietin on functional health and well-being in chronic dialysis patients. J Am Soc Nephrol. 1996;7:763–773
    • View In Article
    • MEDLINE
28. Merkus MP, Jager KJ, Dekker FW, et al.  Physical symptoms and quality of life in patients on chronic dialysis: results of The Netherlands Cooperative Study on Adequacy of Dialysis (NECOSAD). Nephrol Dial Transplant. 1999;14:1163–1170
    • View In Article
    • MEDLINE
    • CrossRef
29. Merkus MP, Jager KJ, Dekker FW, et al.  Quality of life in patients on chronic dialysis: self-assessment 3 months after the start of treatment. Am J Kidney Dis. 1997;29:584–592
    • View In Article
    • Abstract
    • Full-Text PDF (737 KB)
    • CrossRef
30. Mingardi G, Cornalba L, Cortinovis E, et al.  Health related quality of life in dialysis patients. A report from an Italian study using the SF-36 Health Survey. Nephrol Dial Transplant. 1999;14:1503–1510
    • View In Article
    • MEDLINE
    • CrossRef
31. DeOreo PB. Hemodialysis patient-assessed functional health status predicts continued survival, hospitalization, and dialysis-attendance compliance. Am J Kidney Dis. 1997;30:204–212
    • View In Article
    • Abstract
    • Full-Text PDF (866 KB)
    • CrossRef
32. Diaz-Buxo JA, Lowrie EG, Lew NL, et al.  Quality-of-life evaluation using Short Form 36: comparison in hemodialysis and peritoneal dialysis patients. Am J Kidney Dis. 2000;35:293–300
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (553 KB)
    • CrossRef
33. Wu AW, Fink NE, Cagney KA, et al.  Developing a health-related quality-of-life measure for end-stage renal disease: the CHOICE health experience questionnaire. Am J Kidney Dis. 2001;37:11–21
    • View In Article
    • Abstract
    • Full-Text PDF (933 KB)
    • CrossRef
34. Merkus MP, Jager KJ, Dekker FW, et al.  Predictors of poor outcomes in chronic dialysis patients: the Netherlands Cooperative Study on the Adequacy of Dialysis. Am J Kidney Dis. 2000;35:69–79
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (60 KB)
    • CrossRef
35. Burrowes J, Bergen C, Cockram DB, et al.  Relation between subjective assessment of diet and appetite and various nutrition and dialysis parameters at baseline in the Hemodialysis (HEMO) Study. J Renal Nutr. 1997;7:115; (abstr)
    • View In Article
36. Dwyer JT, Chumlea WC, Frydrych A, et al.  Better nutritional status and lower comorbidity associated with quality of life at baseline in the HEMO Study. J Am Soc Nephrol. 1998;9:206A; (abstr)
    • View In Article
37. Merkus MP, Jager KJ, Dekker FW, et al.  Quality of life over time in dialysis: the Netherlands Cooperative Study on the Adequacy of Dialysis. Kidney Int. 1999;56:720–728
    • View In Article
    • MEDLINE
    • CrossRef
38. Cagney KA, Wu WA, Fink NE, et al.  Formal literature review of quality of life instruments used in end stage renal disease. Am J Kidney Dis. 2000;36:327–336
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (65 KB)
    • CrossRef
39. Meyer K, Paranandi L, Hays R, et al.  Quality of life in the HEMO Study: an interim report. J Am Soc Nephrol. 1997;8:204A; (abstr)
    • View In Article
40. Meyer K, Paranandi L, Hays R, et al.  Clinical correlates of baseline quality of life in the HEMO Study: an interim report. J Am Soc Nephrol. 1997;8:204A; (abstr)
    • View In Article