Volume overload is a predictor of mortality in dialysis patients.

However, the fluid status of patients with chronic kidney disease (CKD)
but not yet on dialysis has not been accurately characterized. We used
the Body Composition Monitor, a multifrequency bioimpedance device,
to measure the level of overhydration in CKD patients, focusing on the
association between overhydration and cardiovascular disease risk
factors. Overhydration was the difference between the amount of
extracellular water measured by the Body Composition Monitor and the
amount of water predicted under healthy euvolemic conditions. Volume
overload was defined as an overhydration value at and above the 90th
percentile for the normal population. Of the 338 patients with stages 3
5 CKD, only 48% were euvolemic. Patients with volume overload were
found to use significantly more antihypertensive medications and
diuretics but had higher systolic blood pressures and an increased
arterial stiffness than patients without volume overload. In a
multivariate analysis, male sex, diabetes, pre-existing cardiovascular
disease, systolic blood pressure, serum albumin, TNF-, and proteinuria
were independently all associated with overhydration. Thus, volume
overload is strongly associated with both traditional and novel risk
factors for cardiovascular disease. Bioimpedance devices may aid in
clinical assessment by helping to identify a high-risk group with volume
overload among stages 35 CKD patients.

Keywords:

chronic kidney disease; inflammation; malnutrition; overhydration;

volume overload

Chronic kidney disease (CKD) substantially increases the risks of death and
cardiovascular disease (CVD) and the use of specialized health care. 1 Although
traditional Framingham risk factors for CVD are more prevalent in patients with
CKD than in the general population, these risk factors do not fully account for the
accelerated progression of CVD in CKD patients. 2 Therefore, many recent studies
have focused on the novel risk factors such as malnutrition, inflammation, and
volume overload in the CKD population. Volume overload is related to CVD 3, 4and
is a predictor of outcome in hemodialysis and peritoneal dialysis
patients.5, 6Although a large body of experimental evidence on fluid status has
been collected for dialysis patients, only a limited number of studies have been
conducted in CKD patients not yet on dialysis. 7 Furthermore, the fluid status of
predialysis CKD patients has not been characterized using a valid method. The
prevalence of volume overload during the earlier stages of CKD is unclear and its
significance has not been elucidated.

The clinical assessment of fluid status is relatively difficult, because physical

signs of edema are of limited value in diagnosing excess intravascular
volume.8Ultrasonic evaluation of the diameter of the inferior vena cava can be
used to assess intravascular volume (preload) but not tissue
hydration.9 Interpatient and interoperator variability and the presence of diastolic
dysfunction or right-sided failure also limit the use of this
technique.10, 11 Biomarkers such as brain natriuretic peptide (BNP) and N-terminal
pro-brain natriuretic peptide (NT-proBNP) can reflect changes in the fluid status
but are also influenced by CVD, and they can be accumulated in CKD
patients.12 The most direct and accurate method involves isotope dilution, but
the use of this method is limited to the research environment. Bioimpedance
spectroscopy is a simple and effective approach for the assessment of fluid
status.13, 14 The Body Composition Monitor (BCM, Fresenius Medical Care, Bad
Homburg, Germany) is a bedside bioimpedance spectroscopy device for clinical
use. The accuracy of fluid status and body composition measurements has been
validated against available gold standard reference methods, 15, 16 and the device
has been used to monitor patients receiving hemodialysis 17, 18 or peritoneal
dialysis.19, 20

We hypothesized that volume overload develops early during the course of CKD
and may contribute significantly to the development of CVD. 21 The primary
objectives of this study were to determine the fluid status in a representative
sample of CKD patients using the BCM device, and the measured fluid status was
compared with that of an age- and sex-matched healthy cohort. 22 We also sought
to identify CVD risk factors associated with volume overload.

Topof page

RESULTS

Patient characteristics

After the exclusion criteria were applied, 338 clinically stable patients (233 men
and 105 women; mean age 65.713.5 years) were enrolled in the study. All
patients had moderate-to-severe CKD (mean estimated glomerular filtration rate
(eGFR) 28.7ml/minper1.73m2; 151 in stage 3, 108 in stage 4, and 79 in stage
5). In this population, 45.3% were diabetic (n=153) and 23.4% had CVD (n=79)
(coronary artery disease (n=38), congestive heart failure (n=29), and/or
cerebrovascular accident (n=25)). At least one type of antihypertensive drug was
taken by 83.7% of the patients (calcium-channel blockers 51.2%, renin
angiotensin system (RAS) blockers 59.2%), with a mean of 2.01.4 drugs
prescribed per patient. A total of 113 (33.4%) patients were receiving diuretic
treatment.

Prevalence of volume overload

The baseline characteristics for the patient groups divided on the basis of the
absence or presence of volume overload (defined as overhydration (OH) 7%)
are presented in Table 1. Overall, 52% (n=175) of the study population showed
evidence of volume overload (Figure 1). The patients in the two groups were
similar with regard to age, sex, and smoking history, but there were greater
numbers of patients with diabetes mellitus (DM) and CVD in the volume overload
group. The proportion of patients with volume overload receiving
antihypertensive agents and diuretics was higher. Patients with volume overload
were found to have a similar body mass index and fat tissue index but a
significantly lower lean tissue index compared with patients without volume
overload. In addition, there were important differences in the blood pressure
(BP), arterial stiffness, routine biochemical parameters, and inflammatory
markers between the groups. Patients with volume overload had significantly
higher systolic BP, brachial-ankle pulse wave velocity (baPWV), extracellular
water (ECW), ECW to total body water ratio (ECW/TBW), NT-proBNP, urine
protein-to-creatinine ratio (UPCR), interleukin-6 (IL-6), and tumor necrosis factor-
(TNF-) levels and significantly lower intracellular water (ICW), eGFR, serum
albumin, and hemoglobin levels. The results of the analysis were similar when
volume overload was defined as absolute OH 1.1L (Supplementary Table
S1 online).

Figure 1.

Distribution of overhydration (OH) in 338 chronic kidney disease

patients, ranging between 13.8% and 37.1% (8.38.6%; median
7.4%). The normal range for OH (7 to 7%), corresponding to the 10th to 90th
percentiles of the OH values, is derived from the normal population.

Full figure and legend (40K)

Table 1 - Comparisons of CKD patients with and without volume

overload according to the OH values.

Full table

Factors associated with OH

Correlations between OH and other variables in the overall sample are presented
in Figures 2, 3, 4. OH was positively and strongly correlated with ln NT-proBNP
(r2=0.292; Figure 2). A number of patients had high ln NT-proBNP levels despite
normohydration or even underhydration. These patients were most likely
patients with CVD or worse kidney function. Figure 2a illustrates the linear
regression of OH on ln NT-proBNP and reveals that, for each value of OH, patients
with CVD had a higher ln NT-proBNP than patients without CVD. Similar results
were observed for stages 4 and 5 CKD compared with stage 3 CKD (Figure 2b).

Figure 2.

Correlation between ln NT-proBNP and overhydration (OH).Linear

regression of OH on ln NT-proBNP (a) in chronic kidney disease (CKD) patients
with cardiovascular disease (CVD) (ln NT-proBNP=5.222+0.108 OH) and
without CVD (ln NT-proBNP=4.571+0.087 OH), as well as (b) in patients with
stage 3 CKD (ln NT-proBNP=4.100+0.087 OH) and stages 4 and 5 CKD (ln NT-
proBNP=5.247+0.095 OH). NT-proBNP, N-terminal pro-brain natriuretic
peptide.

Full figure and legend (84K)

Figure 3.

Factors associated with overhydration (OH). Univariate analysis of the

correlations of OH with the (a) systolic blood pressure (BP), (b) brachial-ankle
pulse wave velocity (baPWV), (c) ln urine protein-to-creatinine ratio (UPCR), and
(d) estimated glomerular filtration rate (eGFR).

Full figure and legend (109K)

Figure 4.

Malnutrition and inflammation associated with overhydration

(OH). Univariate analysis of the correlations of OH with (a) ln interleukin-6 (IL-6),
(b) ln tumor necrosis factor- (TNF-), (c) albumin, and (d) lean tissue index.

Full figure and legend (102K)

OH also correlated positively with systolic BP (r2=0.097; Figure 3a), baPWV

(r2=0.021; Figure 3b), and ln UPCR (r2=0.193; Figure 3c) and correlated
negatively with the eGFR (r2=0.023; Figure 3d). With regard to malnutrition
inflammation complex syndrome in CKD patients, OH was positively correlated
with ln IL-6 (r2=0.065; Figure 4a) and ln TNF- (r2=0.113; Figure 4b) and was
negatively correlated with serum albumin (r2=0.255; Figure 4c) and lean tissue
index (r2=0.038; Figure 4d). No association was found with high-sensitivity C-
reactive protein or the lipid profile.

Multivariate regression analysis included OH as the dependent variable and

several relevant demographic (age and sex), clinical (DM, CVD, systolic BP, and
diuretic use), and laboratory factors (eGFR, ln UPCR, serum albumin, and ln TNF-
) that were previously identified in univariate analyses as independent
variables. As shown in Table 2, the novel risk factors (serum albumin, ln TNF-,
and ln UPCR), as well as the traditional risk factors (DM, systolic BP, male sex,
and pre-existing CVD), were independently associated with OH (adjusted R2 of
the model=0.401).

Table 2 - Stepwise multivariate linear regression model identifying

determinants of OH.

Full table
Volume overload in relation to malnutrition, inflammation, and
atherosclerosis (MIA) syndrome

To better distinguish individuals with evidence of MIA syndrome, a previously

described composite variable combining the serum albumin level (<3.0g/dl), the
IL-6 level (3.0pg/ml or higher), and CVD was used to categorize the study
population into four subgroups: patients with none (MIA score=0), one (MIA
score=1), two (MIA score=2), or three (MIA score=3) of these comorbid
conditions.23 The presence of MIA has additive effects on the OH level (Figure 5).
The means and s.ds. of OH in the four subgroups (none (n=110), one (n=150),
two (n=70), three (n=8)) were 4.997.14, 8.097.96, 12.498.81, and
23.367.57% (P<0.001), respectively. As shown in Table 3, only OH was strongly
associated with all of the components of MIA syndrome, whereas measures of
kidney function (eGFR and UPCR) were not.

Figure 5.

The presence of malnutrition, inflammation, and atherosclerosis (MIA)

has additive effects on the overhydration (OH) levels. A composite
variable combining serum albumin (<3.0mg/dl), interleukin-6 ( 3.0g/ml), and
the presence of cardiovascular disease was used to categorize the study
population into four subgroups with none (MIA score=0), one (MIA score=1), two
(MIA score=2), or three (MIA score=3) of these comorbid conditions.

Full figure and legend (36K)

Table 3 - Multivariate logistic regression model showing the

associations of the eGFR, UPCR, and OH with malnutrition,
inflammation, and atherosclerosis syndrome.

Full table
Topof page

DISCUSSION

Our study demonstrates that volume overload is a common problem in

predialysis CKD patients22 and is comparable to that reported for patients
undergoing hemodialysis17, 18 or peritoneal dialysis.20 We found the BCM to be
most beneficial to detect excess fluid in CKD patients with occult volume
overload. In our study, about 20% of CKD patients at the outpatient nephrology
clinic had OH 7% but in the absence of clinically detectable edema. More than
half of the CKD patients in the present study retained at least 1.1L of fluid,
equivalent to the 90th percentile of the absolute OH in the healthy reference
population. Furthermore, almost 20% of patients had severe fluid overload
(defined as OH>15% or an excess of fluid of >2.5l). The study by Wizemann et
al.,17 which included 269 chronic hemodialysis patients, showed that this degree
of OH is an independent predictor of mortality, second only to the presence of
DM. In the present study, OH was associated with important risk factors,
including male sex, diabetes, pre-existing CVD, systolic BP, serum albumin, TNF-
, and proteinuria, suggesting an explanation for the increased risk.

Volume overload has been recognized as an important contributor to an adverse

prognosis and an effect modifier in CKD patients. 24 However, it is not known
whether a biomarker can identify this complication in CKD patients, thus allowing
earlier intervention to improve fluid control. BNP is a peptide hormone released
primarily from the cardiac ventricles in response to increased left ventricular wall
stress. There is ample evidence that the elevation of NT-pro-BNP may partly
reflect volume overload in addition to being a biomarker of ventricular
dysfunction related to background CVD.25 Nevertheless, its diagnostic value has
been considered to be limited in CKD patients because renal dysfunction itself
may affect BNP levels, and the interpretation of its clinical significance should
take renal status into consideration.26 We found that, although there was a strong
association between NT-proBNP and OH, the values for a number of patients did
not comply with this correlation. In our study, patients with CVD had significantly
higher NT-pro-BNP levels than patients without CVD at the same degree of OH
(Figure 2a). Similarly, patients with stages 4 and 5 CKD had significantly higher
NT-pro-BNP levels than patients with stage 3 CKD at the same degree of OH
(Figure 2b). The differential diagnosis of volume overload or congestive heart
failure among CKD patients is often not clinically straightforward. Determining
the optimal cutoff values for NT-pro-BNP, adjusted for the eGFR, for the detection
of CVD across various OH levels in CKD patients requires further investigations.

In CKD patients, reduced glomerular filtration of sodium, activation of the RAS,

and superimposed CVD lead to sodium and water retention. Volume expansion is
a major cause of hypertension in CKD patients, and nearly all CKD patients are
hypertensive upon the initiation of dialysis.27 Hypertension causes ventricular
hypertrophy and increased arterial stiffness, both of which are associated with a
higher mortality among CKD patients. Nevertheless, there might be effects
beyond traditional risk factors that link fluid retention to cardiovascular death.
Previous studies have demonstrated that CKD is not merely an independent risk
factor for CVD but is also associated with increased levels of inflammatory
biomarkers. In our study, there was a strong association between malnutrition
inflammation complex syndrome and volume overload. However, it is impossible
to determine whether malnutritioninflammation complex syndrome is a
consequence or cause of volume overload. Thus, our data may be viewed as
hypothesis-generating data. We found that the TNF- level was elevated in CKD
patients with volume overload. Interestingly, previous studies have shown that
the TNF- level was also elevated in patients with congestive heart failure and
that there is a direct relationship between the level of TNF- and the severity of
disease.28 Patients with CKD have similarities to heart failure patients in that both
populations frequently retain fluid and have excessively high CVD mortality. The
mechanisms by which fluid retention influences cardiovascular survival in CKD
patients may be similar to those in congestive heart failure patients. 5 It has been
previously shown that TNF- mRNA and protein are rapidly expressed in the
hearts of animal models subjected to significant volume overload. 29 Moreover,
both basic and clinical studies strongly support the hypothesis that the
myocardial expression of TNF- is an important step in the pathophysiological
pathway leading to progressive cardiac dilatation and failure. 30 Inflammation is a
pivotal process in the progression of atherosclerosis. Our data are consistent with
previous experimental observations that volume overload is a process that
involves immune activation.

In our study cohort, the use of diuretic agents was associated with volume
overload. We believe this observation is hampered by bias by indication, and
thus the prevalence of volume overload among CKD patients may be even
higher. Another interesting finding was that male gender was an independent
predictor for OH, corroborating previous findings. 31 It is not clear from this study
whether this is related to lower compliance to fluid restriction in male patients or
to other factors. In the present study, OH had a strong and positive association
with proteinuria. In contrast, there was only a modest inverse association with
the GFR, and this association disappeared in the multivariate analysis. This
finding suggests that kidney dysfunction may be associated with fluid retention
only when proteinuria is present. The GFR is one component of the excretory
function of the kidneys but is widely accepted as the best overall index of kidney
function. Although the GFR and proteinuria share many similar correlates in CKD
patients, significant differences emerged regarding the direction and magnitude
of these correlations. Some characteristics, such as fluid retention, may have a
unique correlation with proteinuria and contribute to proteinurias independent
relationship to adverse outcomes.32 Thus, the risk associations of the GFR and
proteinuria appear to be largely independent of one another.

Our study has a number of limitations. As with any observational study, we

cannot establish the causality of the relationships between volume overload and
CVD risk factors. It is unknown whether OH is predictive of disease progression or
future CVD in CKD patients. Longitudinal studies are currently underway to
establish such a relationship. Another limitation was the lack of data on the
dietary fluid intake. Patients can be overhydrated because of excessive sodium
and water intake despite having substantial residual renal function. Therefore,
dietary noncompliance should also be considered when managing volume
overload in CKD patients. Studies focusing more attention on dietary fluid and
salt intake are warranted. Finally, the BCM is unable to distinguish between
intravascular and extravascular ECW. In our study, we excluded patients with
active inflammatory or infectious diseases, which are often associated with
vascular leak and tissue edema. Then, to test the hypothesis that excess fluid
may not be equally distributed into the intravascular space in CKD patients with
a low serum albumin level (decreased plasma oncotic pressure), the study
population was categorized into those with albumin levels below or above the
median (3.6g/dl). OH remained positively and strongly correlated with NT-proBNP
(r=0.428, P<0.001), a marker of intravascular volume expansion, in subjects with
hypoalbuminemia. Classically, the edema formation of nephrosis was considered
to be secondary to increased sodium retention from intravascular volume
depletion due to a low plasma oncotic pressure, referred to as the underfill
hypothesis. However, many recent studies have supported the conceptually
opposite idea, the so-called overfill hypothesis of primary sodium retention. We
proposed that volume overload, as detected by the BCM, is associated with an
increase in intravascular volume, although it does not increase in proportion with
the interstitial compartment in some patients.33Therefore, normalizing the OH in
these patients may adversely impact intravascular volume. The risk to benefit
ratio of normalizing the OH should be tested in carefully designed clinical trials. 34

In conclusion, volume overload is not uncommon in CKD patients and exhibits a

significant association with both traditional and non-traditional CVD risk factors.
OH could serve as a useful adjunct to risk stratification for CVD in CKD. For CKD
patients, close attention to volume control has the potential to influence the
rates of volume-related morbidity and mortality. The BCM is an attractive clinical
tool for assessing volume overload from the earlier stages of CKD. Using the BCM
may help identify among stages 35 CKD patients a high-risk group that may
benefit from more intensive treatment. Further studies will determine whether
effective strategies guided by OH levels will lead to improved outcomes in CKD
patients with volume overload.

Topof page

MATERIALS AND METHODS

Study design and patients

This study was a cross-sectional study conducted in the affiliated hospital of the
Buddhist Tzu Chi University. This study complied with the Declaration of Helsinki
and was approved by the Institutional Review Board. All participants gave their
written informed consent. Over a 16-month period (from September 2011 to
December 2012), 395 consecutive prevalent Taiwanese CKD patients seen in the
outpatient clinics were assessed for eligibility for inclusion if they were >18 years
of age. All patients received multidisciplinary CKD care, focusing on dietary salt
and protein restriction and nephrotoxin avoidance. CKD was defined on the basis
of two eGFR values determined using a simplified Modifications of Diet in Renal
Disease equation35 and separated by an interval of >3 months. eGFR values<60
ml/minper1.73m2 were considered indicative of CKD, and the subjects were
categorized according to Kidney Disease Outcomes Quality Initiative criteria:
CKD stage 3, eGFR 5930ml/minper1.73m2; CKD stage 4, eGFR 2915ml/min
per1.73m2; and CKD stage 5, eGFR<15ml/minper1.73m2. Patients were
evaluated during a routine clinical visit. Patients were excluded if they had a
cardiac pacemaker (n=6) or metallic implants (n=14), were amputees (n=4), or
were pregnant. Patients were also excluded if they had malignancies (n=3),
chronic inflammatory diseases (n=3), renal failure due to reversible reasons
(n=5), or clinical conditions affecting body composition, such as liver cirrhosis
(n=12), active infectious disease (n=6), or any acute CV event (n=4) during the
3 months before screening for inclusion.

For all patients, a thorough medical history was taken at the time of study
enrollment. The definition of CVD included coronary artery disease, as
documented by coronary angiography or a history of myocardial infarction, class
III to IV congestive heart failure, or cerebrovascular accident. DM was assumed to
be present in patients who reported the current or past use of insulin and/or oral
hypoglycemic agents. Hypertension was defined on the basis of either a BP
140/90mmHg or current treatment for hypertension with a prescription
medication. The use of seven therapeutic drug classes-blockers, -blockers,
calcium-channel blockers, diuretics, RAS blockers, direct vasodilators, and other
centrally acting drugswas recorded. It was possible for each patient to be
taking more than one antihypertensive medication or an antihypertensive
medication containing more than one active antihypertensive ingredient. In both
cases, the patient was counted once for each antihypertensive drug class used.

Laboratory measurements

All samples were taken from patients who had fasted overnight. The plasma
levels of IL-6, TNF- (R&D Systems, Minneapolis, MN), and NT-proBNP (Roche
Diagnostics, Indianapolis, IN) were measured using the commercially available
enzyme-linked immunosorbent assay kits according to the manufacturers
instructions. The albumin level was determined using a bromocresol purple
assay. Proteinuria was expressed as UPCR determined using the first morning
void. Arterial stiffness was assessed by measuring the baPWV using a VP-1000
analyzer (Colin Corporation, Komaki, Japan) as previously described. 36

BCM measurements

Measurements of the fluid status were performed by a single well-trained nurse,

using a portable whole-body bioimpedance spectroscopy device, the BCM. The
BCM measures the electrical responses at 50 different frequencies between 5
and 1000kHz. Input variables include the patients height, weight, age, and sex.
Electrodes were attached to the hand and foot on the nondominant side of the
body, after the patient had been in recumbent position for at least 5min. ECW,
ICW, and TBW were determined from the measured impedance data, calculated
by equations of Moissl et al.15 Only one BCM measurement was performed for
each individual patient, because this method had good interobserver and
intraobserver reproducibility.37

In previous studies using bioimpedance, the ratios of ECW/ICW and ECW/TBW

were often used as measures of the fluid status. In contrast, fluid status assessed
by the BCM is represented by OH, which is derived from the impedance data
based on a three-compartment model, developed by Chamney et al.38 The three
compartments are lean tissue mass, adipose tissue mass, and OH. OH is the
difference between the amount of ECW in the tissue actually detected by the
BCM and the amount of water present in tissue predicted using physiological
models under normal (euvolemic) conditions. Therefore, the OH value obtained
from the BCM can be compared directly with that of the normal population. 22 OH
values were further normalized to the ECW and expressed as a percentage of the
ECW in this study. All values of OH were compared with those obtained from an
age- and sex-matched healthy population. Fluid overload was defined as an OH
value 7% (or an absolute OH 1.1L), corresponding to the value of 90th
percentile for the reference cohort when the fluid status was measured with the
same technology.22 The same cutoff threshold for the definition of volume
overload has been validated in a study involving 350 Taiwanese healthy controls
(unpublished data).

Statistical analysis

All variables were expressed as the frequency and percentage for categorical
data and as the meanss.ds. or medians and interquartile ranges for continuous
data with or without a normal distribution, respectively. The baseline
characteristics of the two study subgroups with OH<7% and 7% were
compared using a 2 test for categorical variables and the Students t-test or the
MannWhitney U-test for continuous variables. The OH values of different groups
were compared with one-way analysis of variance. Univariate correlations
between OH and potential explanatory variables were assessed by Pearsons
correlation analyses. Forward stepwise multivariate regression analysis was used
to identify variables that were independently associated with OH. Age, sex, and
clinically relevant variables with a P-value 0.1 in the univariate analysis were
fitted. We did not adjust for NT-proBNP to avoid its potential collinearity with
other variables. We also examined the associations of the eGFR, UPCR, and OH
with components of MIA syndrome using multivariate logistic regression models,
with adjustments for age, sex, DM, systolic BP, RAS blocker use, and statin use. A
two-tailed P-value <0.05 was considered statistically significant. All statistical
analyses were performed using the computer software Statistical Package for the
Social Sciences, version 20.0 (SPSS, Chicago, IL).

Topof page

DISCLOSURE

All the authors declared no competing interests.

Topof page
REFERENCES

1. Go AS, Chertow GM, Fan D et al. Chronic kidney disease and the risks of
death, cardiovascular events, and hospitalization. N Engl J Med 2004; 351:
12961305. | Article | PubMed | ISI | CAS |

2. Longenecker JC, Coresh J, Powe NR et al. Traditional cardiovascular disease

risk factors in dialysis patients compared with the general population: the
CHOICE Study. J Am Soc Nephrol 2002; 13: 1918
1927. | Article | PubMed | ISI |

3. Demirci MS, Demirci C, Ozdogan O et al. Relations between malnutrition-

inflammation-atherosclerosis and volume status. The usefulness of
bioimpedance analysis in peritoneal dialysis patients. Nephrol Dial
Transplant 2011; 26: 17081716. | Article | PubMed |

4. Wang AY, Lam CW, Yu CM et al. N-terminal pro-brain natriuretic peptide: an

independent risk predictor of cardiovascular congestion, mortality, and
adverse cardiovascular outcomes in chronic peritoneal dialysis patients. J
Am Soc Nephrol 2007; 18: 321330. | Article | PubMed | CAS |

5. Kalantar-Zadeh K, Regidor DL, Kovesdy CP et al. Fluid retention is

associated with cardiovascular mortality in patients undergoing long-term
hemodialysis. Circulation 2009; 119: 671679. | Article | PubMed |

6. Agarwal R. Hypervolemia is associated with increased mortality among

hemodialysis patients. Hypertension 2010; 56: 512
517. | Article | PubMed | CAS |

7. Bellizzi V, Scalfi L, Terracciano V et al. Early changes in bioelectrical

estimates of body composition in chronic kidney disease. J Am Soc
Nephrol2006; 17: 14811487. | Article | PubMed |

8. Agarwal R, Andersen MJ, Pratt JH. On the importance of pedal edema in

hemodialysis patients. Clin J Am Soc Nephrol 2008; 3: 153
158. | Article | PubMed | CAS |

9. Kraemer M, Rode C, Wizemann V. Detection limit of methods to assess

fluid status changes in dialysis patients. Kidney Int 2006; 69: 1609
1620. | Article | PubMed |

10.Jaeger JQ, Mehta RL. Assessment of dry weight in hemodialysis: an

overview. J Am Soc Nephrol 1999; 10: 392403. | PubMed | ISI | CAS |

11.Moreno FL, Hagan AD, Holmen JR et al. Evaluation of size and dynamics of
the inferior vena cava as an index of right-sided cardiac function. Am J
Cardiol 1984; 53: 579585. | Article | PubMed |

12.Wang AY, Lai KN. Use of cardiac biomarkers in end-stage renal disease. J
Am Soc Nephrol 2008; 19: 16431652. | Article | PubMed | ISI | CAS |
13.Matthie JR. Bioimpedance measurements of human body composition:
critical analysis and outlook. Expert Rev Med Devices 2008; 5: 239
261. | Article | PubMed |

14.Jaffrin MY, Morel H. Body fluid volumes measurements by impedance: a

review of bioimpedance spectroscopy (BIS) and bioimpedance analysis
(BIA) methods. Med Eng Phys 2008; 30: 12571269. | Article | PubMed |

15.Moissl UM, Wabel P, Chamney PW et al. Body fluid volume determination

via body composition spectroscopy in health and disease. Physiol
Meas2006; 27: 921933. | Article | PubMed |

16.Wabel P, Chamney P, Moissl U et al. Importance of whole-body

bioimpedance spectroscopy for the management of fluid balance. Blood
Purif 2009; 27: 7580. | Article | PubMed |

17.Wizemann V, Wabel P, Chamney P et al. The mortality risk of

overhydration in haemodialysis patients. Nephrol Dial
Transplant 2009; 24: 15741579. | Article | PubMed |

18.Machek P, Jirka T, Moissl U et al. Guided optimization of fluid status in

haemodialysis patients. Nephrol Dial Transplant 2010; 25: 538
544. | Article | PubMed |

19.Devolder I, Verleysen A, Vijt D et al. Body composition, hydration, and

related parameters in hemodialysis versus peritoneal dialysis
patients.Perit Dial Int 2010; 30: 208214. | Article | PubMed |

20.Van Biesen W, Williams JD, Covic AC et alon behalf of the EuroBCM study
group. Fluid status in peritoneal dialysis patients: the European Body
Composition Monitoring (EuroBCM) Study cohort. PLoS One 2011; 6:
e17148. | Article | PubMed | CAS |

21.Pecoits-Filho R, Goncalves S, Barberato SH et al. Impact of residual renal

function on volume status in chronic renal failure. Blood Purif 2004; 22:
285292. | Article | PubMed |

22.Wieskotten S, Heinke S, Wabel P et al. Bioimpedance-based identification

of malnutrition using fuzzy logic. Physiol Meas 2008; 29: 639
654. | Article | PubMed |

23.Liu Y, Coresh J, Eustace JA et al. Association between cholesterol level and

mortality in dialysis patients: role of inflammation and
malnutrition. JAMA2004; 291: 451459. | Article | PubMed | ISI | CAS |

24.Hung SC, Lin YP, Huang HL et al. Aldosterone and mortality in

hemodialysis patients: role of volume overload. PLoS One 2013; 8:
e57511. | Article | PubMed |

25.Mak GS, DeMaria A, Clopton P et al. Utility of B-natriuretic peptide in the

evaluation of left ventricular diastolic function: comparison with tissue
 Doppler imaging recordings. Am Heart J 2004; 148: 895
902. | Article | PubMed | CAS |

26.Takami Y, Horio T, Iwashima Y et al. Diagnostic and prognostic value of

plasma brain natriuretic peptide in non-dialysis-dependent CRF. Am J
Kidney Dis 2004; 44: 420428. | Article | PubMed | ISI | CAS |

27.Collins AJ, Foley RN, Herzog C et al. United States Renal Data System 2008
Annual Data Report. Am J Kidney Dis 2009; 53: S1S374.

28.Torre-Amione G, Kapadia S, Benedict C et al. Proinflammatory cytokine

levels in patients with depressed left ventricular ejection fraction. a report
from the Studies of Left Ventricular Dysfunction (SOLVD). J Am Coll
Cardiol1996; 27: 12011206. | Article | PubMed | CAS |

29.Kapadia SR, Oral H, Lee J et al. Hemodynamic regulation of tumor necrosis

factor-alpha gene and protein expression in adult feline myocardium. Circ
Res 1997; 81: 187195. | Article | PubMed | ISI | CAS |

30.Feldman AM, Combes A, Wagner D et al. The role of tumor necrosis factor
in the pathophysiology of heart failure. J Am Coll Cardiol 2000; 35: 537
544. | Article | PubMed | ISI | CAS |

31.Wabel P, Moissl U, Chamney P et al. Towards improved cardiovascular

management: the necessity of combining blood pressure and fluid
overload. Nephrol Dial Transplant 2008; 23: 2965
2971. | Article | PubMed |

32.Anand IS, Bishu K, Rector TS et al. Proteinuria, chronic kidney disease, and
the effect of an angiotensin receptor blocker in addition to an angiotensin-
converting enzyme inhibitor in patients with moderate to severe heart
failure. Circulation 2009; 120: 15771584. | Article | PubMed | CAS |

33.Deschnes G, Feraille E, Doucet A. Mechanisms of oedema in nephrotic

syndrome: old theories and new ideas. Nephrol Dial Transplant 2003; 28:
454456.

34.John B, Tan BK, Dainty S et al. Plasma volume, albumin, and fluid status in
peritoneal dialysis patients. Clin J Am Soc Nephrol 2010; 5: 1463
1470. | Article | PubMed | CAS |

35.Levey AS, Bosch JP, Lewis JB et al. A more accurate method to estimate
glomerular filtration rate from serum creatinine: a new prediction
equation. Modification of diet in Renal Disease Study Group. Ann Intern
Med 1999; 130: 461470. | Article | PubMed | ISI | CAS |

36.Yamashina A, Tomiyama H, Takeda K et al. Validity, reproducibility, and

clinical significance of noninvasive brachial-ankle pulse wave velocity
measurement. Hypertens Res 2002; 25: 359364. | Article | PubMed | ISI |
37.Wabel P, Chamney P, Moissl U. Reproducibility of bioimpedance
spectroscopy for the assessment of body composition and dry weight. J
Am Soc Nephrol 2007; 18: A255. | Article |

38.Chamney PW, Wabel P, Moissl UM et al. A whole-body model to distinguish

excess fluid from the hydration of major body tissues. Am J Clin
Nutr 2007;85: 8089. | PubMed | CAS |