14. Kolesnyk I, Dekker FW, Boeschoten EW et al. Time-dependent
reasons for peritoneal dialysis technique failure and mortality.
Peritoneal Dial Int J Int Soc Peritoneal Dial 2010; 30: 170–177
11. Baillod RA, Comty Chr M, Crockeit R et al. Experience with
regular haemodialysis in the home. EDTA 1966; 3: 126
12. Pauly RP, Maximova K, Coppens J et al. Patient and technique
survival among a Canadian Multicenter Nocturnal Home Hemodialysis Cohort. Clin J Am Soc Nephrol 2010; 5: 1815–1820
13. NHSBT-www.organdonation.nhs.uk/statistics
Received for publication: 12.2.2013; Accepted in revised form: 21.5.2013
Nephrol Dial Transplant (2013) 28: 2620–2628
doi: 10.1093/ndt/gft296
A multicentric, international matched pair analysis of body
composition in peritoneal dialysis versus haemodialysis patients
Wim van Biesen1,
Kathleen Claes2,
ORIGINAL ARTICLE
Renal Division, Ghent University Hospital, Ghent, Belgium,
2
Department of Nephrology and Renal Transplantation, University
Hospitals Leuven, Leuven, Belgium,
Adrian Covic3,
3
Nephrology Department, Gr. T. Popa University of Medicine and
Pharmacy, Iasi, Romania,
4
Stanley Fan ,
4
Monika Lichodziejewska-Niemierko5,
Department of Renal Medicine and Transplantation, The Royal
London Hospital, London, UK,
Volker Schoder6,
Christian Verger
1
5
Fresenius Nephrocare PD Unit, Department of Nephrology
Transplantology and Internal Medicine, Medical University, Gdańsk,
7
Poland,
and Peter Wabel6
6
Fresenius Medical Care Deutschland GmbH, Bad Homburg,
Germany and
7
Keywords: body composition, bioimpedance, haemodialysis,
peritoneal dialysis, volume status
Correspondence and offprint requests to:
Wim van Biesen;
E-mail: [email protected]
Results. Four hundred and ninety-one matched pairs (55.2%
males, median age 60.0 years) were included. The body mass
index (BMI, PD = 26.5 ± 4.7 versus HD = 25.9 ± 4.6 kg/m²,
P = 0.18 in males and 27.4 ± 5.8 versus 27.5 ± 6.6 kg/m²,
P = 0.75 in females) and fat tissue index (males: 11.5 ± 5.3
versus 11.4 ± 5.4 kg/m², P = 0.90, females: 14.8 ± 6.7 versus
15.4 ± 7.2 kg/m², P = 0.30) were not different in PD versus HD
patients, whereas the lean tissue index (LTI) was higher in
PD versus HD patients (males: 14.5 ± 3.4 versus 13.7 ±
3.1 kg/m², P = 0.001, females: 12.6 ± 3.3 versus 11.5 ± 2.6 kg/m²,
P < 0.0001). VO/extracellular water (ECW) was not different
between PD versus just before the HD treatment (males:
10.8 ± 12.1 versus 9.2 ± 10.2%, P = 0.09; females: 6.5 ± 10.8
versus 7.7 ± 9.4%, P = 0.19). The relative TAVO was higher in
A B S T R AC T
Background. Volume status, lean and fat tissue are gaining
interest as prognostic predictors in patients on dialysis. Comparative data in peritoneal dialysis (PD) versus haemodialysis
(HD) patients are lacking.
Methods. In a cohort of PD (EuroBCM) and HD (Euclid database) patients, matched for country, gender, age and dialysis
vintage, body composition was assessed by bioimpedance
spectroscopy (BCM, Fresenius Medical Care). Time-averaged
volume overload (TAVO) was defined as the mean of pre- and
post-dialysis volume overload (VO), and relative (%) (TA)VO
as (TA)VO/ECV.
© The Author 2013. Published by Oxford University Press on
behalf of ERA-EDTA. All rights reserved.
University Hospital René Dubos, Pontoise, France
2620
compare body composition using bioimpedance spectroscopy
in a large international cohort of PD and HD patients. We
hypothesized that differences in body composition between
subgroups can potentially lead to a better understanding of the
differences in mortality between subgroups of patients on the
two dialysis modalities.
PD versus HD (10.8 ± 12.1% versus 3.2 ± 11.2%, and
6.5 ± 10.8% versus 1.2 ± 10.9%, both P < 0.0001).
Conclusions. The LTI was impaired, and this was more in
males versus females, but was better preserved on PD versus
HD, whereas fat tissue index (FTI) was increased, but not
different between PD and HD. Volume overload was more
present in PD versus HD when TAVO, but not when predialysis volume status, was used as a reference.
M AT E R I A L S A N D M E T H O D S
There is increasing interest in objective evaluation of volume
status as an adequacy parameter in patients on renal replacement therapy (RRT) [1, 2]. Volume overload has been linked
to heart failure [3], left ventricular hypertrophy [4, 5], sleep
apnoea [6] and mortality [7]. It is also associated with hypoalbuminaemia [8], another negative predictor of long-term
outcome. On the other hand, volume deficiency is related to
decreased quality of life (cramps, dizziness, fatigue...), and is
also thought to be associated with an accelerated loss of
residual renal function [9]. In addition, further reducing
volume status in some hypertensive RRT patients—on the
basis that hypertension is generally considered as a marker of
volume overload [10]—may lead to severe diastolic hypotension, decreased coronary perfusion [11] and even sudden
cardiac death, particularly in those normovolaemic patients in
which increased blood pressure (BP) values are mainly caused
by increased vascular stiffness [1, 12]. Adequate assessment
and management of volume status are critically important,
and clinical parameters including blood pressure are not
reliable enough to guide treatment decisions. There is thus a
need for tools that evaluate volume status in the routine clinical setting, and different methods have been proposed and investigated during the last decade. Some of these methods
measure the circulating volume, and its effect on cardiac preload (e.g. chest X-ray, inferior vena cava diameter and collapsibility, lung interstitium ultrasound [13, 14]), other methods
evaluate more the capillary refilling capacity during haemodialysis (HD) [15, 16], or are based on biochemical markers of atrial
dilatation, difficult to interpret in anuric patients (ANP, BNP,
pro-BNP, NT-proBNP) [17]. Using multi-frequency biospectroscopy allows us to estimate ECW and total body water, and thus
indirectly also intracellular water, using mathematical modelling
[18]. In this way, not only the volume status, but also lean body
weight and fat mass can be derived [19], allowing in fact a bedside analysis of body composition. This is of great interest, as
there is compelling evidence for an association between volume
status, inflammation and nutritional status [20].
Long-term outcomes in peritoneal dialysis (PD) and HD
have been reported to be equal, with a slightly better outcome
for PD patients during the first years [21, 22]. This prognosis
has been reported despite suggestions that PD patients are
much more volume overloaded than HD patients [23] and
possibly more obese (due to absorption of the intraperitoneal
glucose). So far, no comparative data from large cohorts on
body composition in PD versus HD are available. This observational, hypothesis generating cohort study was set up to
BCM measurement
All body composition measurements were performed with
the BCM (Body-Composition-Monitor; Fresenius Medical
Care), a well-validated tool [24, 25]. The BCM measures the
impedance spectroscopy at 50 different frequencies between
5 and 1 MHz. Electrodes were attached to one hand and one
foot at the ipsilateral side with the patient in a recumbent
position. In the PD patients, measurements were timed as described previously [1]. All HD patients were measured directly
before an HD session.
Volume overload (VO) was derived from the impedance
data based on a physiological tissue model [18, 19]. This
model separates the patient’s body into three compartments—
volume overload, normohydrated lean tissue and normohydrated fat tissue. Absolute volume overload represents the
difference between the measured amount of extracellular water
2621
Body composition in HD versus PD
ORIGINAL ARTICLE
Patients
The data of the PD patients were collected in the EuroBCM
study [1]. This was a cross-sectional, observational, multicentre trial in 28 centres in six European countries (Belgium,
France, Poland, Romania, UK and Switzerland). PD patients
from Belgium and Switzerland were excluded in the current
analysis as no data from HD patients were available from these
countries. In each selected centre, all prevalent patients on PD
were assessed for eligibility for inclusion ( prevalent crosssectional cohort approach). This trial has been registered at
the Cochrane Renal Group trials registry (http://www.
cochrane-renal.org) under the number CRG110800153.
The HD patients used for the matching of the PD patients
were extracted from a large clinical database (EuCliD—
Fresenius Medical Care, NephroCare) with more than 20 000
patients with BCM measurements available. The country
matched patients (5236 HD patients available) served as a
pool of controls for the matching for age, gender, dialysis
vintage and country. The representativeness of the final HD
sample (n = 491) to the country-matched HD patients
(n = 5236) was assessed by comparing age, gender distribution
and dialysis vintage.
In a previous study, a Caucasian population of 1247 male
and female subjects was collected, which was used as a presumed healthy reference population [18]. The subjects were
equally distributed over the complete age range. They represent a ‘presumably healthy reference’ population—but no
detailed clinical questionnaire or additional laboratory tests
were performed. The age-dependent median of the body composition parameters of this population is used as a reference in
this analysis.
INTRODUCTION
ORIGINAL ARTICLE
and the amount of water expected in normohydrated tissue
conditions [19]. The absolute volume overload was then
normated to the measured extracellular volume (VO/ECV), a
parameter denoted as ‘relative volume overload’ (relative VO),
expressed in percentage, to allow a more objective comparison
between patients with different anthropometric features. To
incorporate the change in volume overload with time between
the dialysis sessions in the HD patients, we also defined a ‘time
averaged volume’ (TAVO) as the mean of the pre-and postdialytic volume overload, where the post-dialytic volume overload was calculated as the pre-dialysis volume overload minus
the intradialytic ultrafiltration. Analogous to the body mass
index (BMI), the lean tissue index (LTI) and fat tissue index
(FTI) are calculated by dividing the measured normohydrated
lean or fat tissue mass, respectively, by the squared height of
the patient.
analysis. Results of the subgroup analysis are presented in the
Supplementary material.
Statistical analysis
For all continuous variables, mean and standard deviation
are given unless otherwise noted (e.g. median and min/max
for skewed distributions). Statistical comparisons between PD
and HD patients are based on paired t-tests using pairs as resulted from matching, for all tests where pairing was not feasible, two-sample t-tests were used.
For correlations analysis, Pearson’s r was calculated. In a
multiple linear regression, the simultaneous influence of
several variables on VO was examined. All analyses were done
with SAS V 9.2
R E S U LT S
Procedure for matched pairing
For matching of cases and controls, a publicly available SAS
macro (%gmatch, Biostatistics department, MAYO Clinic) was
used, which was adapted to the specific conditions of our
study.
For each PD patient, a corresponding HD patient was
identified from the same country of the same gender and with
a difference in age and dialysis vintage of less than 3 years and
6 months, respectively. For 41 patients, no suitable control was
available, for the other 491 patients one or more controls were
identified. In case of more than one control fulfilling the
matching criteria, a random selection was applied.
The same procedure was applied to the subgroup of
patients with confirmed diabetes (PD: n = 136, HD: n = 702).
To enable a sufficient number of matches, the criteria were
relaxed to age difference <5 years and vintage difference <12
months; 126 matched pairs were found for this subgroup
Demographic and clinical data are presented in Table 1, separately for the PD and HD cohorts. As expected, because of the
matching procedure, no differences in age or dialysis vintage
were observed between the PD and HD cohorts. There was no
difference in systolic BP between PD and HD patients, but
diastolic BP was higher in PD patients. The majority of PD
patients were treated by continuous ambulatory peritoneal
dialysis (CAPD) (75.9%), whereas the majority of HD patients
were treated by standard HD (84.1%) and only a minority by
online haemodiafiltration.
When comparing HD patients from the matched cohort
(n = 491) with the HD patients from the database who were not
used for matching, it became evident that the patients used
for the matching have comparable age (59.2 ± 13.7 versus
58.1 ± 15.0 years), a comparable intradialytic weight loss
(2.10 ± 1.05 versus 2.07 ± 1.06 kg) and a comparable BMI
Table 1. Description of the matched PD and HD population
PD (n = 491)
HD (n = 491)
Gender male (%)
271 (55.2%)
271 (55.2%)
1.0
Age (years)
59.2 ± 13.7
59.2 ± 13.7
0.25
Vintage (months) (min–max)
23.9 (1.0–178.5)
24.2 (0.5–183.2)
0.33
France (patients/centres)
39/5
39/3
UK (patients/centres)
161/2
161/23
Poland (patients/centres)
74/5
74/21
Romania (patients/centres)
217/9
217/16
Confirmed Diabetes
133 (27.1%)
78 (15.9%)
Treatment Information
373 CAPD/118 APD
412 HD/78 oHDF/1 Single-needle
Systolic BP (mmHg)
138 ± 26
140 ± 23
0.32
Diastolic BP (mmHg)
80 ± 15
76 ± 15
<0.0001
HD (g/dL)
11.2 ± 1.6
11.5 ± 1.6 (n = 390)
Where applicable the mean and the standard deviation are shown.
2622
W. van Biesen et al.
P
<0.0001
0.0016
(26.7 ± 5.6 versus 26.1 ± 5.3 kg/m²), but a lower dialysis vintage
(24.2 (0.5–183.2) versus 38.4 (0.0–467.3) months, respectively.
Body composition
Table 2 represents the body composition parameters, separately for the PD and HD population, and also segregated by
gender. There was no difference in BMI or FTI between PD
and HD patients. The LTI was higher in PD versus HD
patients, both in males and in females, and all results were
similar in the subgroup with confirmed diabetes (see Supplementary material).
We analysed whether body composition or the difference
in body composition between PD and HD was associated with
either age or dialysis vintage (Table 3), but no firm associations were observed. Most noteworthy is the absence of an
association between dialysis vintage and the difference
between PD and HD in FTI and VO, see Table 3. The body
composition in terms of BMI, LTI and FTI in the PD and HD
cohorts is represented in Figure 2, separately for males and
females. The comparison between the PD and HD patients
and the healthy reference population is shown in Figure 3.
Volume
The relative volume overload was not different between PD
versus HD patients. In the male cohort (relative), TAVO was
higher in PD versus HD patients. In the subgroup with confirmed diabetes, male PD patients presented a lower relative
VO as HD patients—otherwise all other results were confirmed, see Supplementary material.
VO was in a multivariate model independently predicted
by male gender, PD as dialysis modality and confirmed diabetes ( p of the model < 0.0001), but still a lot of variability in
VO remains unexplained by our model (R² = 0.10) and may
thus be attributed to other causes.
The absolute (l) and relative (% of ECW) volume overload
for PD patients, HD patients before dialysis and time-averaged
fluid overload in HD patients are represented in Figure 1, separately for males and females. As can be observed, the lower
mean VO in HD patients was achieved by creating a severe
dehydration in a substantial (62.5% < 0 L and even 37.5% <−1
L) part of patients after the dialysis session, resulting in a negative TAVO. In contrast, in the PD patients, a negative VO was
only rarely observed (20.8% <0 L and 7.5% <−1 L).
DISCUSSION
In this international multi-centric matched pair cohort study,
we found that fluid overload (VO/ECW) was equally present
ORIGINAL ARTICLE
F I G U R E 1 : Comparison of the volume status of PD and HD patients separated by gender. On the left-hand side the data from male subjects,
on the right-hand side female subjects. PD, peritoneal dialysis patients; HD pre, HD patients measured before the HD treatment; HD TAVO,
time-averaged volume overload of the HD patients. In the upper row, the absolute volume status, in the lower row the volume status relative
to the ECV is presented.
2623
Body composition in HD versus PD
Table 2. Body composition in the HD and PD patients
Male
Female
PD (n = 271)
HD (n = 271)
Weight (pre) (kg)
77.5 ± 15.2
75.2 ± 15.9
BMI (kg/m²)
26.5 ± 4.7
FTI (kg/m²)
LTI (kg/m²)
P
PD (n = 220)
HD (n = 220)
P
0.06
69.1 ± 15.1
69.0 ± 16.9
0.92
26.0 ± 4.6
0.18
27.4 ± 5.8
27.5 ± 6.6
0.75
11.5 ± 5.3
11.4 ± 5.4
0.90
14.8 ± 6.7
15.4 ± 7.2
0.30
14.5 ± 3.4
13.7 ± 3.1
0.001
12.6 ± 3.3
11.5 ± 2.6
<0.0001
VO (L)
2.3 ± 2.7
1.7 ± 1.9
0.005
1.1 ± 1.9
1.2 ± 1.5
0.71
TAVO (L)
2.3 ± 2.7
0.6 ± 1.9
<0.0001
1.1 ± 1.9
0.2 ± 1.5
<0.0001
Rel. VO (%) (VO/ECV)
10.8 ± 12.1
9.2 ± 10.2
0.09
6.5 ± 10.8
7.7 ± 9.4
0.19
Rel. TAVO (%) (TAVO/ECV)
10.8 ± 12.1
3.2 ± 11.2
<0.0001
6.5 ± 10.8
1.3 ± 10.6
<0.0001
ECV (L)
19.3 ± 3.6
17.8 ± 3.1
<0.0001
15.6 ± 3.0
14.9 ± 2.8
0.01
ECV/height (L/cm)
0.11 ± 0.02
0.10 ± 0.02
<0.0001
0.10 ± 0.02
0.09 ± 0.02
0.01
ORIGINAL ARTICLE
Weight (pre), predialysis weight for the HD patients; BMI, body mass index; FTI, fat tissue index; LTI, lean tissue index; VO, volume
overload; TAVO, Time-averaged volume overload; ECV, extracellular volume. For all parameters, the mean and the standard deviation are
shown.
Table 3. Correlation table between the mean value of the pair, the difference between the pair
and the mean age and the mean dialysis vintage of the pair (diff and mean refer to within the
PD/HD pair)
Correlation coefficient
with mean pair age (years)
Correlation coefficient with
mean pair dialysis vintage (log scale)
Mean pair BMI (kg/m²)
0.16
−0.006
Mean pair FTI (kg/m²)
0.31
0.04
Mean pair LTI (kg/m²)
−0.39
−0.09
Mean pair TAVO (L)
−0.01
−0.03
0.02
−0.02
−0.16
0.12
Difference pair FTI (kg/m²)
0.02
0.02
Difference pair LTI (kg/m²)
−0.05
−0.06
Difference pair TAVO (L)
−0.03
0.06
Difference pair TAVO/ECW (%)
−0.004
0.07
Difference pair albumin (n = 401) (g/L)
−0.07
0.06
Mean pair TAVO/ECW (%)
Mean pair albumin (n = 401) (g/L)
in PD and in HD patients just before dialysis. The TAVO was,
however, significantly lower in the HD patients. Irrespective of
dialysis modality, male gender and confirmed diabetes were
independently associated with fluid overload. In contrast to
general belief, BMI and fat tissue mass were comparable
between PD and HD patients, while the lean tissue mass was
higher in the PD patients (independent of the diabetic state).
Lean tissue mass seems to be impaired, while fat tissue is
increased in dialysis patients when compared with the general
population
For a long time, the emphasis of ‘adequate dialysis’ [26] was
placed on small solute clearance, mostly expressed as Kt/V.
Evidence is compiling that increasing small solute clearance
does not lead to improved outcomes [27, 28]. In HD, it
appears that increasing ‘t’ leads to improved solute removal
even when Kt/V remains equal [29]. It is well conceivable that
2624
W. van Biesen et al.
the FTI and the LTI.
increasing ‘t’ also allows better maintenance of volume status,
as it allows slower ultrafiltration, itself being associated with
improved outcomes [5]. In the field of PD, better volume
removal [30], not improved small solute clearance [27], was
associated with improved outcomes. Head-to-head comparisons of PD versus HD with regard to volume status are scanty
[23, 31]. All recent publications which aim at comparing the
volume status have limitations with respect to the analysed
patient numbers (n < 100), and significant differences in the
age and dialysis vintage of compared patient populations. The
current paper is the first to present a matched pair analysis of a
large PD and HD population.
In the matched pair analysis of this large cohort, VO in PD
patients was comparable with that of HD patients just before
the dialysis session, but was higher when the TAVO was considered. However, this came at the expense of dehydration of a
substantial part of HD patients. This observation might well
explain why residual renal function is less well preserved in
HD when compared with PD patients. Irrespective of treatment modality, male patients and diabetics were more volume
overloaded. For diabetics, this was attributed to their increased
thirst due to residual hyperglycaemia, leading to increased
volume intake. Also in other PD cohorts, diabetes has been
associated with VO [32]. In HD cohorts, diabetic patients tend
to gain more weight in between dialysis sessions, and this is
independent of their salt intake, leading to hyponatraemia in
the pre-dialysis setting [33]. Although the finding of higher
VO in males is consistent in most cohorts [1, 32], the underlying reasons remain unclear. As VO is associated with increased mortality, the higher volume overload in males versus
females might explain to some extent the reported higher mortality in male dialysis patients, irrespective of their treatment
modality [21]. Most authors attribute this higher VO in males
to a lesser compliance of male patients to the dietary restrictions, although formal analysis fails to corroborate this
hypothesis [34].
Despite the higher VO in PD patients, there were no differences in systolic BP. However, it has been demonstrated before
that, although there is a statistically significant association
between fluid overload and systolic BP, this association does
not hold true for an important percentage of patients, due to
confounding by vascular stiffness and congestive heart failure
[1], as can also be seen in Figures 1 and 2 in the Supplementary material.
Of note, PD patients tended to have a higher diastolic BP,
and thus a lower pulse pressure, when compared with HD
patients. This might be a representation of the lower degree of
2625
Body composition in HD versus PD
ORIGINAL ARTICLE
F I G U R E 2 : Comparative analysis of the body composition of HD and PD patients separated by gender. Shown is the body mass index (BMI),
F I G U R E 3 : The body composition of PD and HD patients in our study compared with a healthy reference population. Peritoneal patients are
ORIGINAL ARTICLE
indicated by the open triangles, whereas HD patients are shown with the full dots. The full line represents the age-dependent median functions
for LTI and FTI as calculated by a sliding-window filtering technique based on the measured BCM data of a healthy reference population with
2071 subjects.
vascular stiffness that has been observed in PD versus HD
patients [35].
There is a widespread belief that the glucose absorbed
during the PD dwell leads to an increase in weight and fat
mass. BMI was not higher in PD when compared with HD
patients in this cohort of age, gender and dialysis vintage
matched patients. We also did not observe an association
between BMI or FTI and dialysis vintage which is another
important fact to note against intuitive expectations. Our data
demonstrates that fat mass is increased both in PD and in HD
patients when compared with a healthy reference population
not on dialysis, but that there is no difference in PD versus
HD patients. On the other hand, it is clear from our data that
patients on dialysis have a reduced lean tissue mass when compared with the general population not on dialysis, an effect
that is much more mitigated, however, in patients on PD.
It can be speculated that the presence of an additional caloric
load because of the absorbed glucose protects the PD patients
from using muscle protein as an energy source. Our data
support the idea that HD patients are severely protein and
calorie malnourished, whereas PD patients seem to be relatively protected from catabolism through their more appropriate caloric load associated with the continuous glucose
absorption. Also, PD patients have a better correction of their
metabolic acidosis when compared with HD patients, due the
continuous infusion of buffer, and growing evidence suggests
that correction of acidosis leads to improvement of muscle
mass and nutritional status [36, 37]. The data in Table 3
reveals that PD patients do not accumulate more VO and fat
mass in the time they are on the PD therapy when compared
with the HD patients. In this respect, there is no difference
between the dialysis modalities.
The data indicate that male patients are more susceptible to
VO and additionally to the loss in lean tissue mass when compared with the healthy reference population. These differences
in the development of body composition between male and
female subjects open a wide field of future research to analyse
if genetic, hormonal or psychological effects are explanations
for these observations.
Limitations
No detailed clinical questionnaire or additional laboratory
tests were available from the healthy reference population.
If anything, deviation between the results of the cohorts on
RRT and the truly healthy population is thus underestimated
in this approach. Only limited information was available on
the comorbid conditions of the patients. No information on
the clinical presence of oedema was collected during this
study. This is mainly due to the origin of the HD patients data,
with only limited availability of detailed background patient
information in this cohort. We performed a subgroup analysis
in the patients with confirmed diabetes—all major findings
were confirmed. In the future, it would be important to take
into account also the presence of cardiac failure and residual
renal function, although both these parameters cannot be considered mere ‘confounders’ as they can both result or cause
VO and malnutrition. This hypothesis generating matched
pair analysis of PD and HD patients opens the field for future
studies, as uncertainty remains about the causal relations
between the observed results.
2626
W. van Biesen et al.
S U P P L E M E N TA R Y D ATA
Supplementary data are available online at http://ndt.oxfordjournals.org.
C O N F L I C T O F I N T E R E S T S TAT E M E N T
P.W. and V.S. are employees of Fresenius Medical Care,
Germany. Adrian Covic and Monika Lichodziejewska-Niemierko
are employed at dialysis units of Fresenius Medical Care. W.V.
B., K.C., S.F. and C.V. received travel grants from Fresenius
Medical Care, Baxter and Gambro on different occasions.
REFERENCES
1. Van Biesen W, Williams JD, Covic AC et al. Fluid status in peritoneal dialysis patients: the European Body Composition Monitoring (EuroBCM) study cohort. PloS One 2011; 6: e17148
2. Dou Y, Zhu F, Kotanko P. Assessment of extracellular fluid
volume and fluid status in hemodialysis patients: current status
and technical advances. Semin Dial 2012; 25: 377–387
3. Hayashi SY, Seeberger A, Lind B et al. A single session of haemodialysis improves left ventricular synchronicity in patients with
end-stage renal disease: a pilot tissue synchronization imaging
study. Nephrol Dial Transplant 2008; 23: 3622–3628
4. Wang AY, Lam CW, Yu CM et al. Troponin T, left ventricular
mass, and function are excellent predictors of cardiovascular congestion in peritoneal dialysis. Kidney Int 2006; 70: 444–452
5. Agarwal R, Bouldin JM, Light RP et al. Probing dry-weight improves left ventricular mass index. Am J Nephrol 2011; 33:
373–380
6. Elias RM, Bradley TD, Kasai T et al. Rostral overnight fluid shift
in end-stage renal disease: relationship with obstructive sleep
apnea. Nephrol Dial Transplant 2012; 27: 1569–1573
7. 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: 671–679
2627
Body composition in HD versus PD
ORIGINAL ARTICLE
8. 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
9. Van Biesen W, Lameire N, Verbeke F et al. Residual renal function and volume status in peritoneal dialysis patients: a conflict of
interest? J Nephrol 2008; 21: 299–304
10. Agarwal R. The controversies of diagnosing and treating hypertension among hemodialysis patients. Semin Dial 2012; 25:
370–376
11. McIntyre CW. Haemodialysis-induced myocardial stunning in
chronic kidney disease—a new aspect of cardiovascular disease.
Blood Purif 2010; 29: 105–110
12. 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
13. Mallamaci F, Benedetto FA, Tripepi R et al. Detection of pulmonary congestion by chest ultrasound in dialysis patients. JACC
Cardiovasc Imaging 2010; 3: 586–594
14. Cheriex EC, Leunissen KM, Janssen JH et al. Echography of the
inferior vena cava is a simple and reliable tool for estimation of
‘dry weight’ in haemodialysis patients. Nephrol Dial Transplant
1989; 4: 563–568
15. Gabrielli D, Krystal B, Katzarski K et al. Improved intradialytic
stability during haemodialysis with blood volume-controlled ultrafiltration. J Nephrol 2009; 22: 232–240
16. Rodriguez HJ, Domenici R, Diroll A et al. Assessment of dry
weight by monitoring changes in blood volume during hemodialysis using Crit-Line. Kidney Int 2005; 68: 854–861
17. Yashiro M, Kamata T, Yamadori N et al. Evaluation of markers
to estimate volume status in hemodialysis patients: atrial natriuretic peptide, inferior vena cava diameter, blood volume changes
and filtration coefficients of microvasculature. Ther Apheresis
Dial 2007; 11: 131–137
18. Moissl UM, Wabel P, Chamney PW et al. Body fluid volume
determination via body composition spectroscopy in health and
disease. Physiol Meas 2006; 27: 921–933
19. 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: 80–89
20. Jacobs LH, van de Kerkhof JJ, Mingels AM et al. Inflammation, overhydration and cardiac biomarkers in haemodialysis
patients: a longitudinal study. Nephrol Dial Transplant 2010; 25:
243–248
21. Noordzij M, Jager KJ. Survival comparisons between haemodialysis and peritoneal dialysis. Nephrol Dial Transplant 2012;
27: 3385–3387
22. Lukowsky LR, Mehrotra R, Kheifets L et al. Comparing mortality
of peritoneal and hemodialysis patients in the first 2 years of
dialysis therapy: a marginal structural model analysis. Clin J Am
Soc Nephrol 2013; 8: 619–628
23. 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: 208–214
24. Dou Y, Liu L, Cheng X et al. Comparison of bioimpedance
methods for estimating total body water and intracellular water
changes during hemodialysis. Nephrol Dial Transplant 2011; 26:
3319–3324
Conclusion
Volume overload is a frequent problem in PD and HD
patients. It seems that PD patients have a higher TAVO when
compared with HD patients. However, VO immediately before
the HD session was equal to PD, and the lower TAVO is thus
obtained at the expense of a certain period of underhydration
immediately after the HD session in a substantial part of the
HD patients. Both PD and HD patients have increased fat
mass and decreased lean tissue mass when compared with a
healthy reference population. These changes in body composition seem to be triggered by the underlying uraemic state,
and not by the treatment modality itself. Most importantly PD
patients do not present an increased BMI or fat mass when
compared with HD patients, despite the glucose overload in
PD patients.
32. Davison SN, Jhangri GS, Jindal K et al. Comparison of volume
overload with cycler-assisted versus continuous ambulatory peritoneal dialysis. Clin J Am Soc Nephrol 2009; 4: 1044–1050
33. Ramdeen G, Tzamaloukas AH, Malhotra D et al. Estimates of interdialytic sodium and water intake based on the balance principle: differences between nondiabetic and diabetic subjects on
hemodialysis. ASAIO J 1998; 44: 812–817
34. Lindberg M, Wikstrom B, Lindberg P. Subgroups of haemodialysis patients in relation to fluid intake restrictions: a cluster analytical approach. J Clin Nurs 2010; 19: 2997–3005
35. Verbeke F, Van Biesen W, Honkanen E et al. Prognostic value of
aortic stiffness and calcification for cardiovascular events and
mortality in dialysis patients: outcome of the calcification
outcome in renal disease (CORD) study. Clin J Am Soc Nephrol
2011; 6: 153–159
36. Workeneh BT, Mitch WE. Review of muscle wasting associated
with chronic kidney disease. Am J Clin Nutr 2010; 91:
1128S–1132S
37. Movilli E, Viola BF, Camerini C et al. Correction of metabolic
acidosis on serum albumin and protein catabolism in hemodialysis patients. J Ren Nutr 2009; 19: 172–177
ORIGINAL ARTICLE
25. Wabel P, Chamney P, Moissl U et al. Importance of whole-body
bioimpedance spectroscopy for the management of fluid balance.
Blood Purif 2009; 27: 75–80
26. Vanholder R, Dhondt AM, Van Biesen W. Is Kt/V urea a satisfactory measure for dosing the newer dialysis regimens? Seminars
Dial 2001; 14: 12–14
27. Paniagua R, Amato D, Vonesh E et al. Effects of increased peritoneal clearances on mortality rates in peritoneal dialysis:
ADEMEX, a prospective, randomized, controlled trial. J Am Soc
Nephrol 2002; 13: 1307–1320
28. Cheung AK, Levin NW, Greene T et al. Effects of high-flux
hemodialysis on clinical outcomes: results of the HEMO study.
J Am Soc Nephrol 2003; 14: 3251–3263
29. Eloot S, Van Biesen W, Dhondt A et al. Impact of hemodialysis
duration on the removal of uremic retention solutes. Kidney Int
2008; 73: 765–770
30. Paniagua R, Amato D, Mujais S et al. Predictive value of brain natriuretic peptides in patients on peritoneal dialysis: results from
the ADEMEX trial. Clin J Am Soc Nephrol 2008; 3: 407–415
31. Yao YH, Fu CH, Ho SJ et al. Peritoneal dialysis as compared with
hemodialysis is associated with higher overhydration but noninferior blood pressure control and heart function. Blood Purif
2012; 34: 40–47
Received for publication: 16.3.2013; Accepted in revised form: 26.5.2013
Nephrol Dial Transplant (2013) 28: 2628–2636
doi: 10.1093/ndt/gft276
Advance Access publication 16 July 2013
Longer interdialytic interval and cause-specific hospitalization
in children receiving chronic dialysis
1
Tamar Springel1,
USA,
Benjamin Laskin2,
2
2
Justine Shults ,
Ron Keren
Department of Pediatrics, Cooper University Hospital, Camden, NJ,
Department of Pediatrics, The Children’s Hospital of Philadelphia,
Philadelphia, PA, USA,
3
2,3
Center for Pediatric Clinical Effectiveness, The Children’s Hospital
of Philadelphia, Philadelphia, PA, USA and
and Susan Furth2,4
4
Department of Biostatistics and Epidemiology, University of
Pennsylvania School of Medicine, Philadelphia, PA, USA
Keywords: admission, end-stage renal disease
Correspondence and offprint requests to:
Tamar Springel; E-mail: [email protected]
hemodialysis (HD). This association has not been previously
demonstrated in children. We hypothesized that the risk of
hospitalization for hypertension (HTN), fluid overload or electrolyte abnormalities would be increased on the days following
a longer interdialytic interval in children.
A B S T R AC T
Background. Previous studies have demonstrated a relationship between longer interdialytic intervals and hospitalization
for cardiovascular causes in adults maintained on
© The Author 2013. Published by Oxford University Press on
behalf of ERA-EDTA. All rights reserved.
2628