www.kidney-international.
org
A single weekly Kt/Vurea target for peritoneal
dialysis patients does not provide an equal
dialysis dose for all
Sally El-Kateb1, Sivakumar Sridharan2, Ken Farrington2,3, Stanley Fan4 and Andrew Davenport1
1
UCL Centre for Nephrology, Royal Free Hospital, University College London Medical School, London, United Kingdom; 2Renal Unit, Lister
Hospital, Stevenage SG1 4AB, United Kingdom; 3University of Hertfordshire, Hatfield, United Kingdom; and 4Barts Health NHS Trust,
London, United Kingdom
Dialysis adequacy is traditionally based on urea clearance,
adjusted for total body volume (Kt/Vurea), and clinical
guidelines recommend a Kt/Vurea target for peritoneal
dialysis. We wished to determine whether adjusting dialysis
dose by resting and total energy expenditure would alter
the delivered dialysis dose. The resting and total energy
expenditures were determined by equations based on
doubly labeled isotopic water studies and adjusted Kturea
for resting energy expenditure and total energy
expenditure in 148 peritoneal dialysis patients (mean age,
60.6 years; 97 male [65.5%]; 54 diabetic [36.5%]). The mean
resting energy expenditure was 1534 kcal/d, and the total
energy expenditure was 1974 kcal/day. Using a weekly
target Kt/V of 1.7, Kt was calculated using V measured by
bioimpedance and the significantly associated (r [ 0.67)
Watson equation for total body water. Adjusting Kt for
resting energy expenditure showed a reduced delivered
dialysis dose (ml/kcal per day) for women versus men (5.5
vs. 6.2), age under versus over 65 years (5.6 vs. 6.4),
weight <65 versus >80 kg (5.8 vs. 6.1), low versus high
comorbidity (5.9 vs. 6.2), all of which were significant.
Adjusting for the total energy expenditure showed
significantly reduced dosing for those employed versus not
employed (4.3 vs. 4.8), a low versus high frailty score
(4.5 vs. 5.0) and nondiabetic versus diabetic (4.6 vs. 4.9).
Thus, the current paradigm for a single target Kt/Vurea for
all peritoneal dialysis patients does not take into account
energy expenditure and metabolic rate and may lead to
lowered dialysis delivery for the younger, more active
female patient.
Kidney International (2016) -, -–-; http://dx.doi.org/10.1016/
j.kint.2016.07.027
KEYWORDS: body surface area; Kt/Vurea; peritoneal dialysis; resting energy
expenditure; total body water; total energy expenditure
Copyright ª 2016, International Society of Nephrology. Published by
Elsevier Inc. All rights reserved.
Correspondence: Andrew Davenport, UCL Centre for Nephrology, Royal Free
Hospital, University College London Medical School, Rowland Hill Street,
London NW3 2PF, United Kingdom. E-mail: andrewdavenport@nhs.net
Received 29 April 2016; revised 11 July 2016; accepted 14 July 2016
Kidney International (2016) -, -–-
clinical investigation
M
ore than 3 million patients with end-stage kidney
disease are currently treated by dialysis worldwide,
with w300,000 treated by peritoneal dialysis (PD).
As with hemodialysis, there are clinical guidelines recommend-
ing that patients receive a minimal amount of dialysis based on
urea clearance.1 These urea-based clearance targets are derived
from observational studies.2 However, prospective studies
comparing different peritoneal dialysis regimens designed to
achieve different urea clearance targets consistently failed to
demonstrate any advantage for greater urea clearance in terms
of patient morbidity or mortality.3–5 Indeed, PD technique and
patient survival have been linked to preservation of residual
renal function6 rather than PD urea clearance.7
The amount of urea clearance (Kt/Vurea) for dialysis patients
is currently based on the volume of distribution of urea, total body
water (TBW) derived from anthropomorphic measurements.8
However, TBW varies with body composition, as some tissues
such as muscle contain more water than fat,9 and also varies
between racial groups10 and patients with diabetes and other
comorbidities.11 As such, for the same Kt/Vurea, the delivered
urea clearance has been suggested to differ among patients.12
Rather than dosing the amount of dialysis required on urea
clearance based on volume of distribution, an alternative
approach based on metabolic activity has been proposed.13
Urea is generated as a by-product of intracellular nitrogen
metabolism. Total body metabolic activity is a composite of
resting metabolic rate and that due to physical activity. Pre-
vious studies in PD patients have concentrated on measuring
resting energy expenditure (REE).14,15 but this underestimates
total energy expenditure (TEE), by excluding that due to
activity energy expenditure.
We recently validated an assessment of TEE and REE in
dialysis patients using a patient self-reported questionnaire
and doubly labeled isotopic water.16,17 To establish whether
there is a difference in the amount of dialysis delivered for a
fixed Kt/Vurea target, we calculated urea clearance adjusted
for energy expenditure to determine whether some groups of
patients would be at a disadvantage under current clinical
guideline recommendations.
RESULTS
We studied 148 adult PD patients with a mean calculated REE
of 1534
241 kcal/d and TEE 1974
414 kcal/d (Table 1).
1
clinical investigation S El-Kateb et al.: Resting and total energy expenditure in peritoneal dialysis patients

Table 1 | Patient demographic characteristics, peritoneal Table 2 | Estimates of daily REE and TEE in patients according
dialysis prescription, results of peritoneal dialysis adequacy, to age, comorbidity, frailty, and ethnicity groupings
and transport status in all patients and those with
Variable REE, kcal/d TEE, kcal/d
contemporaneous bioimpedance measurements
Male 1597
217 2029
423
Variable Total Cohort Bioimpedance Group Female 1412
2401 1868
3772
N 148 118 Age <65 yr 1646
209 2173
392
Male, % 65.5 63.6 Age >65 yr 1408
2111 1750
3141
Age, yr 60.6
17.5 59.5
18.2 Nondiabetic 1522
233 2021
435
Weight, kg 73.6
16.7 73.1
16.6 Diabetic 1556
254 1893
3662
Body surface area, m2 1.86
0.24 1.85
0.24 Employed 1577
237 2305
511
White 43.2 42.4 Unemployed 1523
242 1890
3401
African/Afro-Caribbean 24.3 21.1 Low comorbidity 1532
245 2012
441
South Asian 27.7 29.7 High comorbidity 1539
231 1862
300
East Asian 5.4 6.8 Low frailty score 1533
227 2049
453
Employed 20.3 22.9 High frailty score 1535
256 1894
3532
Dialysis vintage, mo 9.1 (3.5–25.2) 9.4 (3.8–25.5) Weight <64 kg 1305
151 1706
306
Comorbidity score 4.0 (0–6.0) 4.0 (0–6.0) Weight 64–80 kg 1514
1421 1973
4141
Heart disease, % 19.7 19.7 Weight >80 kg 1775
1591 2233
3391
Myocardial infarction, % 10.2 9.4 PNA <60 g/d 1450
214 1826
317
Diabetes mellitus, % 32.4 29.7 PNA >60 g/d 1622
2291 2133
4381
Frailty 3.0 (3.0–4.0) 3.0 (3.0–4.0) Non-Asian 1561
225 2060
462
Hemoglobin, g/l 109.9
4.8 110.5
4.5 Asian 1522
243 1866
3592
Serum albumin, g/l 36.5
5.5 36.6
5.6 REE, energy expenditure; TEE, total energy expenditure; PNA, protein nitrogen
C-reactive protein, mg/l 5.0 (2.0–16.8) 5.0 (2.0–15.0) appearance.
Serum glucose, mmol/l 5.9 (4.9–8.5) 5.7 (4.9–5.1) Daily PNA g/d. Results expressed as mean
SD.
IFCC, mmol/mg 38.4 (33.3–51.4) 36.2 (32.7–47.5)
1
P < 0.01 comparing groups, adjusted for multiple comparisons (Bonferroni
Serum cholesterol, mmol/l 4.47
1.44 4.50
1.50 method).
2
P < 0.05, adjusted for multiple comparisons (Bonferroni method).
Serum urea, mmol/l 18.4
6.1 18.5
5.9
Serum creatinine, mmol/l 698 (523–871) 696 (525–909)
Peritoneal cycler, % 85.5 83.9
Icodextrin use, % 75.7 81.2 P < 0.001 for men), BSA was relatively greater at lower TBW
Icodextrin volume, l/d 1.8 (0.5–2.0) 1.8 (1.0–2.0) volumes and relatively lower at higher TBW volumes. TBW
22.7–23.0 glucose use, % 57.4 54.2 had also measured by bioimpedance at the time of adequacy
22.7–23.0 glucose, l/d 2.5 (0–5.0) 2.9 (0–6)
Previous peritonitis episodes 0 (0–1) 0 (0–1) testing in 118 of the patients (79.7%) (Table 1). There was
Total weekly Kt/Vurea 2.1 (1.7–2.6) 2.1 (1.7–2.60) no statistically significant difference in TBW: Watson equa-
Weekly urinary Kt/Vurea 0.8 (0.3–1.3) 0.9 (0.3–1.4) tion, 40.3
6.1 versus bioimpedance, 40.6
3.4 L; mean
Weekly peritoneal Kt/Vurea 1.2 (0.9–1.6) 1.2 (0.8–1.6)
difference on Bland-Altman analysis, 0.72 L (Figure 1).
Total creatinine cleared 67.4 (55.8–84.7) 70.1 (55.8–86.7)
per week/1.73 m2 Although the mean difference for women was 1.43 L and
Urine creatinine cleared 29.6 (13.4–58.4) 28.9 (12.7–59.4) that for men was 0.31 L, the 95% limits of agreement were
per week/1.73 m2, L broad; 9.20 to 10.16 L for women, and from 11.04 to
Peritoneal creatinine cleared 35.6 (23.2–45.2) 37.3 (23.5–49.3)
per week/1.73 m2, L
11.66 L for men. There were positive correlations between
Urine <100 ml/d, % 12.2 15.3 BSA and both REE and TEE (r ¼ 0.92, P < 0.001 and r ¼
Urine volume, ml/d 946 (450–1249) 940 (448–1408) 0.59, P < 0.001, respectively) and also between TBW and
4-hr dialysate/plasma creatinine 0.71
0.11 0.73
0.11 both REE and TEE (r ¼ 0.85, P < 0.001 and r ¼ 0.66,
24-hr ultrafiltrate, ml 566 (200–908) 536 (192–899)
Protein nitrogen appearance, 0.89
0.26 0.89
0.25
P < 0.001, respectively).
g/kg per day Clinical guidelines have recommended a minimal weekly
IFCC ¼ International Federation of Clinical Chemists. Kt/Vurea of 1.7. We then calculated Kt values for a weekly Kt/
Values shown as number, mean
SD, median (interquartile range), and percentage. Vurea of 1.7 using both Watson equation and bioimpedance
estimates of TBW. These Kt values were then adjusted by BSA,
Twenty-five percent were classified as high comorbidity18 and REE, and TEE. There was a positive relationship between
48% as frail.19 Male patients were heavier than female patients TBW and Kt/TEE (Figure 2). The results of the adjusted Kt
and had a greater REE and TEE (Table 2). Patients who were dialysis dosing are shown in Table 3 and Figure 3 for different
employed, those with greater weight, and greater protein ni- patient groups. For the same prescribed dialysis dose
trogen appearance (PNA) had a higher TEE (Table 2), (Kturea), women, younger patients, patients who were
whereas those with greater frailty and comorbidity, those who employed, and patients weighing less (Figure 3) received less
were diabetic, and those who were Asian tended to have a dialysis than men, older patients, unemployed patients, and
lower TEE. heavier patients (Table 3). In addition, patients with fewer
As previous studies have suggested that Kt be adjusted for comorbidities and less frailty, nondiabetic patients, and pa-
body surface area (BSA), we compared Watson TBW with tients of non-Asian races also tended to receive less dialysis
BSA. Although there was a strong association between TBW than those with more comorbidities and those who were
and BSA (r2 ¼ 0.99, P < 0.001 for women and r2 ¼ 0.83, diabetic, frail, and of Asian ethnicity.

2 Kidney International (2016) -, -–-

S El-Kateb et al.: Resting and total energy expenditure in peritoneal dialysis patients
Figure 1 | Bland-Altman analysis of total body water (TBW)
measured by bioimpedance or calculated by the Watson
equation. Mean difference: 0.72 L (95% limits of agreement 9.2
to þ10.7 L).
Multivariable analysis showed that sex was a significant
predictor of Kt/BSA (Table 4). Sex and age were significant
predictors of Kt/REE. For Kt/TEE, sex, age, and employment
were common predictive factors irrespective of whether Kt
was derived using TBW calculated by the Watson equation or
bioimpedance. Both high comorbidity and diabetes were
additional predictive factors for TEE, adjusted using the
Watson formula for TBW (Table 4).
DISCUSSION
Traditionally, the target amount of dialysis for patients with
end-stage kidney failure has been based on Kt/Vurea
adjusted for TBW volume. However, multiple prospective
trials have failed to show an association between greater PD
urea clearance and survival.3,4,7 Cellular metabolism, in
particular protein turnover, generates waste products that
accumulate in patients with end-stage kidney failure. As
these azotemic toxins are generated by cellular metabolism,
it has been suggested that the amount of dialysis required
for patients should be based on metabolic rate rather than
urea clearance.2 Studies to date have concentrated on
measuring resting metabolic rate,14,15 but this overlooks
physical activity and, as such, potentially underestimates
energy expenditure. We used equations based on patient
Figure 2 | Relationship between Watson total body water (TBW)
for men and women and total urea clearance adjusted for total
energy expenditure (TEE) ml/kcal per day.
Kidney International (2016) -, -–-
clinical investigation
self-reported physical activity questionnaires, which have
been validated using doubly labeled isotopic water,16 to es-
timate REE and TEE. As expected, energy expenditure was
associated with body weight, male sex, and younger age
group.20 Patients with higher REE and TEE had a higher
PNA rate due to increased urinary and peritoneal urea losses.
However, we also noted that although REE was similar, TEE
was lower with increasing frailty and comorbidity, in
particular in patients with diabetes and in unemployed pa-
tients, compared with those patients with less frailty and
lower comorbidity scores who were not diabetic or those
who were employed. We also found that patients of an Asian
background had lower TEE compared with white and Afri-
can/Afro-Caribbean patients. This is in keeping with previ-
ous observations of lower energy expenditure, particularly in
South Asian patients, and this has been suggested to be due
to differences in terms of body composition related to brown
fat tissue stores.21
We then compared the delivered dialysis dose assuming
that all patients received the minimum weekly KtVurea target
as recommended by clinical practice guidelines,1 using Kt
calculated by both the Watson formula8 and TBW as
measured by bioimpedance.22 Although we found no signif-
icant difference between TBW by either method, the mean
bias was greater for bioimpedance-measured TBW than that
estimated using the Watson equation. However, the 95%
limits of agreement were broad,9 suggesting that at the
individual patient level, TBW should preferably be measured.
Previous reports have shown major differences between
TBW derived by the Watson formula and TBW derived by
bioimpedance were observed with obese patients with a body
mass index >35; in our study group, <2% had a body mass
index of this level.
We then adjusted the delivered dialysis dose by both BSA,
which is relatively greater for patients with lower TBW and
relatively lower for those with greater TBW and also for both
REE and TEE. Adjusting Kt for BSA, which has been advo-
cated for hemodialysis patients,23 we found that this resulted
in a lower dose being delivered to women and those with a
high PNA rate and lower body weight. Adjusting for REE,
female sex, younger age, lower weight, lower PNA along
with frailty and comorbidity scores, and ethnicities other than
Asian, all received relatively less delivered dialysis. When Kt
was adjusted for TEE, then women, younger patients, and
those weighing less, those who were employed, and those with
less frailty, in particular those without diabetes, all would
receive lower delivered dialysis dosing compared with men,
heavier patients, those unemployed, those more frail, co-
morbid patients, and those with diabetes.
Previous studies targeting a dialysis dose defined by a
weekly KtVurea for PD patients have not shown an advantage
of 1 target compared with another.3,4 Our study shows that
achieving the same urea clearance does not equate to the same
delivered dose of dialysis and, as such, potentially explains
why prospective studies have failed to show a significant
benefit of 1 KtVurea target for all patients.
3
clinical investigation S El-Kateb et al.: Resting and total energy expenditure in peritoneal dialysis patients

Table 3 | Comparison of a fixed total weekly Kt/V of 1.7 (urea clearance l/m2 per day or ml/kcal per day) adjusted for BSA, REE,
and TEE for peritoneal dialysis patients comparing sexes, age, diabetic/nondiabetic, employment status, comorbidity, weight,
ethnicity, high and low frailty, comorbidity, PNA rate, employed, unemployed, and ethnicity (Asian vs. other races)
Variable Kt/BSA Kt/REEW Kt/TEEW Kt/REEBIA Kt/TEEBIA
Male 5.13
0.36 6.15
9.61 4.96
0.71 6.23
0.62 4.93
0.70
Female 4.42
0.401 5.50
0.411 4.23
0.651 5.64
0.641 4.27
0.711
Age <65 yr 4.83
0.46 5.58
0.55 4.29
0.53 5.93
0.73 4.52
0.81
Age >65 yr 4.95
0.42 6.38
0.491 5.18
0.611 6.12
0.62 4.93
0.651
Diabetic 4.96
0.45 6.03
0.58 5.00
0.691 5.92
0.65 4.93
0.692
Nondiabetic 4.84
0.46 5.90
0.66 4.53
0.82 6.06
0.71 4.57
0.78
High frailty 4.91
0.47 6.06
0.66 4.96
0.75 6.01
0.61 4.90
0.71
Low frailty 4.86
0.43 5.85
0.602 4.46
0.781 6.01
0.75 4.54
0.781
High comorbidity 4.99
0.43 6.19
0.63 5.14
0.66 5.91
0.63 4.90
0.63
Low comorbidity 4.85
0.45 5.87
0.621 4.55
0.801 6.05
0.71 4.62
0.81
Unemployed 4.89
0.43 5.99
0.61 4.87
0.72 5.96
0.73 4.81
0.75
Employed 4.89
0.54 5.82
0.71 4.07
0.821 6.19
0.51 4.31
0.721
High PNA rate 4.79
0.40 5.91
0.62 4.75
0.78 5.77
0.70 4.56
0.70
Low PNA rate 4.95
0.482 5.98
0.68 4.62
0.87 6.23
0.592 4.81
0.81
Asian 4.87
0.46 5.97
0.65 4.67
0.85 6.10
0.65 4.71
0.76
Other 4.87
0.42 5.90
0.60 4.79
0.69 5.79
0.762 4.64
0.80
BIA, bioimpedance; BSA, body surface area; DM, diabetic; REE, resting energy expenditure; TEE, total energy expenditure; W, Watson formula.
Volume (V) was estimated by the Watson formula or measured by BIA.
1
P < 0.01 after Bonferroni post hoc correction for multiple testing.
2
P < 0.05 after Bonferroni post hoc correction for multiple testing.

There are a number of limitations that should be consid-
ered. We used solely a weekly target Kt/Vurea, whereas some
guidelines additionally recommend liters of creatinine cleared
as an additional target for PD patients, and there may be
differences between these targets depending on use of PD
cyclers, and the amount of residual renal function.1,24 We
adjusted Kt using both the Watson equation and bio-
impedance. There were some differences between these
methods. The Watson equation was established using a
healthy population, whereas PD patients have increased
TBW.25 As such, bioimpedance measurements are preferable,
but ideally should be measured with the peritoneal dialysate
drained.26 In addition, we calculated PNA rates using equa-
tions developed in a previous era, when patients were pre-
dominantly treated by continuous ambulatory PD and
glucose-only dialysates,27 and although these are often used
as a surrogate for dietary protein intake, these estimates may
not be as reliable for patients with increasing comorbidity,
and we did not formally estimate dietary protein intake.
Although we accepted that using Kt/Vurea for dialysis dosing
has some limitations,28 more recent observational studies have
suggested an advantage for adjusting Kt for BSA.23 We found
that adjusting for BSA detected a difference between men and
women and in relation to body weight and PNA. However,
adjusting for TEE additionally demonstrated that younger and
more fit patients, employed patients, and patients with less
comorbidity received a relatively lower delivered dialysis dose
compared with older, more frail, comorbid, and diabetic pa-
tients. Although we chose to investigate the effect of a weekly
target Kt/Vurea of 1.7, our findings would be equally applicable
to any set Kt/V target applied to patients. Therefore, we suggest
that a single Kt/Vurea target dose is not applicable to all patients,
and the dose of dialysis should be increased for those who are
more physically active with greater TEE. On the other hand, the
results of our study should not be misinterpreted to imply that
some patient groups require less dialysis treatment. Our results
generate a hypothesis that requires formal testing to determine
whether increasing the minimum target dose of dialysis in some
groups of PD patients improves patient outcomes.

Patients and methods

Figure 3 | Adjusted daily urea clearance according to body
weight. Fixed weekly Kt of 1.7 urea adjusted for body surface area
(BSA) and resting energy expenditure (REE) and total energy
expenditure (TEE) using Watson (W) total body water or
bioimpedance (BIA) measured total body water. *P < 0.05 and
**P < 0.01 versus weight <64 kg after Bonferroni correction.
Adult patients with end-stage kidney disease established on
PD were recruited from University College London partner
hospitals when attending for outpatient assessments of PD
adequacy. Corresponding spent dialysate effluent, 24-hour
urine collections, and serum samples were analyzed by stan-
dard methods, and the weekly dialysis dose was calculated as
Kt/Vurea. The PNA rate was estimated using the Bergström
equation and normalized for body weight (g/kg per day).27
Patient demographic characteristics were obtained from

4 Kidney International (2016) -, -–-

S El-Kateb et al.: Resting and total energy expenditure in peritoneal dialysis patients clinical investigation

Table 4 | Multivariable step backward models for weekly Kt expenditure in patients with chronic kidney disease.16 Phys-
adjusted for BSA, REE, TEE using both total body water ical activity data were determined by each reported activity
calculated by Watson equation and measured by BIA, being assigned a metabolic equivalent of task (MET) value
unstandardized b, SE, standardized b, and 95% CL according to the Compendium of Physical Activities.31 The
Standardized equations for calculating REE and TEE are detailed in the
Variable b SE b b t 95% CL P Value Supplementary Appendix.
Kturea/BSA UK clinical guidelines recommend a minimum weekly Kt/
Male 0.70 0.05 0.77 13.5 0.6, 0.87 <0.001 Vurea of 1.7.1,32 Hence, in order to compare minimum
Kturea/REEW dialysis targets using alternative scaling parameters, weekly Kt
Male 0.58 0.08 0.44 7.5 0.43, 0.74 <0.001
Age, yr 0.02 0.01 0.54 9.3 0.2, 0.25 <0.001 was calculated as Kt ¼ 1.7 multiplied by V. Corresponding
Kturea/REEBIA target values of Kt/BSA, Kt/REE, and Kt/TEE were calculated
PNA rate 0.01 0.01 0.37 4.4 0.01, 0.02 <0.001 by dividing daily Kt by the respective parameters.
Male 0.39 0.12 0.28 3.2 0.15, 0.63 0.002
Ethical approval of the study was granted by the UK Na-
Age, yr 0.01 0.01 0.19 2.3 0.01, 0.01 0.025
Kturea/TEEW tional Research Ethics Committee–Essex, and the study was
Male 0.59 0.10 0.35 6.1 0.40, 0.79 <0.001 registered in UK Clinical Research Network portfolio number
Age, yr 0.02 0.01 0.44 7.5 0.02, 0.013 <0.001 14018. All patients provided written informed consent in
Unemployed 0.52 0.11 0.26 4.5 0.29, 0.74 <0.001
High comorbidity 0.30 0.11 0.16 2.7 0.01, 0.52 0.009
keeping with the Declaration of Helsinki.
Diabetic 0.21 0.10 0.12 2.0 0.01, 0.41 0.045
Kturea/TEEBIA Statistical analysis
Male 0.51 0.13 0.33 3.8 0.25, 0.77 <0.001
Statistical analysis was performed using the Student t test or
Age, yr 0.01 0.01 0.32 7.5 0.01, 0.02 <0.001
Unemployed 0.44 0.15 0.24 3.0 0.15, 0.74 0.004 Mann-Whitney U test, analysis of variance, the Kruskal-Wallis
BIA, bioimpedance; BSA, body surface area; CL, confidence limits; PNA, protein ni-
test with appropriate post hoc correction, the Pearson or
trogen accumulation; REE, resting energy expenditure; TEE, total energy expenditure; Spearman test for univariate correlation (GraphPad Prism,
W, Watson equation. version 6.0, San Diego, CA), and step backward linear
BSA model, r2 0.60 adjusted, 0.59; model adjusted for REEW, r20.60; adjusted 0.59; model
adjusted for REEBIA r20.42; adjusted, 0.37. Model adjusted for TEEW r20.42, adjusted 0.3, regression of variables on univariate analysis of P < 0.1 and
and adjusted for TEEBIA r20.35, adjusted 0.33. Sex (female vs. male). those considered to be clinically relevant, with log trans-
formation of variables that were not normally distributed and
computerized hospital records, and comorbidity was deter- removal of variables that were not statistically significant
mined using self-administered comorbidity grading18 based unless they improved model fit. Models were checked for
on medical conditions and complications including diabetes collinearity using SPSS, version 22 (SPSS Inc., Chicago, IL)
mellitus (as defined by WHO criteria), cardiac disease, res- and the Bland-Altman comparison (Analyse-It Software,
piratory disease, liver disease, arthritis, depression, malig- version 3.0, Leeds, UK). Data are presented as the mean

nancy, and a frailty score previously reported in patients with SD, median (interquartile range), mean and 95% confidence
chronic kidney disease.19 We defined a high comorbidity limits (CL), or percentage.
score as $4.0 and a high frailty score $4.0, in keeping with
previous studies.18,19 DISCLOSURE
TBW was calculated using the Watson equation.8 In All the authors declared no competing interests.
addition, in 118 of the patients, contemporaneous measure-
ments of TBW made with bioimpedance (InBody 720, ACKNOWLEDGMENTS
InBody, Seoul, South Korea; Body Composition Monitor, The study was funded by a grant from the British Renal Society. SE-K
Fresenius, Bad Homberg, Germany), which had been per- was awarded an International Society for Nephrology fellowship.
formed in a standardized manner,28,29 were available for re-
view. Bioimpedance measurements made by the Body SUPPLEMENTARY MATERIAL
Composition Monitor and InBody were standardized using Supplementary Appendix. Resting energy expenditure (REE) was
previously derived equations.30 BSA was calculated using the estimated from a newer novel predictive equation that was derived
Gehan and George equation, as recommended by the Euro- and validated in a cohort of hemodialysis patients.18
pean Best Clinical Practice guidelines.24 Supplementary material is linked to the online version of the paper at
Physical activity data were obtained using the Recent www.kidney-international.org.
Physical Activity Questionnaire,16 which collects information
about both activity and the time spent performing activities REFERENCES
1. Woodrow G, Davies SJ. Peritoneal Dialysis (PD) (Guidelines PD 3.1 – 3.3).
over the preceding 4 weeks, including activities performed at Available at: http://www.renal.org/guidelines/modules/peritoneal-
home and work and during leisure time. The Recent Physical dialysis-in-ckd#sthash.Br67xjah.dpuf. Accessed July 25, 2016.
Activity Questionnaire has been validated against a doubly 2. Jansen MA, Termorshuizen F, Korevaar JC, et al. Predictors of survival in
anuric peritoneal dialysis patients. Kidney Int. 2005;68:1199–1205.
labeled water technique in the general population and has 3. Paniagua R, Amato D, Vonesh E, et al.; Mexican Nephrology Collaborative
been shown to be a reliable tool for estimation of energy Study Group. Effects of increased peritoneal clearances on mortality

Kidney International (2016) -, -–- 5

clinical investigation S El-Kateb et al.: Resting and total energy expenditure in peritoneal dialysis patients
rates in peritoneal dialysis: ADEMEX, a prospective, randomized,
controlled trial. J Am Soc Nephrol. 2002;13:1307–1320.
4. Lo WK, Ho YW, Li CS, et al. Effect of Kt/V on survival and clinical outcome
in CAPD patients in a randomized prospective study. Kidney Int. 2003;64:
649–656.
5. Paniagua R, Amato D, Vonesh E, et al.; Mexican Nephrology Collaborative
Study Group. Health-related quality of life predicts outcomes but is not
affected by peritoneal clearance: the ADEMEX trial. Kidney Int. 2005;67:
1093–1104.
6. Churchill DN, Taylor DW, Keshaviah PR. Adequacy of dialysis and
nutrition in continuous peritoneal dialysis: association with clinical
outcome. J Am Soc Nephrol. 1996;7:198–207.
7. Bargman JM, Thorpe KE, Churchill DN; CANUSA Peritoneal Dialysis Study
Group. Relative contribution of residual renal function and peritoneal
clearance to adequacy of dialysis: a reanalysis of the CANUSA study. J Am
Soc Nephrol. 2001;12:2158–2162.
8. Watson PE, Watson ID, Batt RD. Total body water volume for adult males
and females estimated from simple anthropometric measurements. Am J
Clin Nutr. 1980;33:27–39.
9. Davenport A. Differences in prescribed Kt/V and delivered haemodialysis
dose–why obesity makes a difference to survival for haemodialysis
patients when using a ’one size fits all’ Kt/V target. Nephrol Dial
Transplant. 2013;28(suppl 4):iv219–iv223.
10. Davenport A, Hussain Sayed R, Fan S. The effect of racial origin on total
body water volume in peritoneal dialysis patients. Clin J Am Soc Nephrol.
2011;6:2492–2498.
11. Davenport A, Willicombe MK. Does diabetes mellitus predispose to
increased fluid overload in peritoneal dialysis patients? Nephron Clin
Pract. 2010;114:c60–c66.
12. Spalding EM, Chandna SM, Davenport A, Farrington K. Kt/V
underestimates the haemodialysis dose in women and small men.
Kidney Int. 2008;74:348–355.
13. Daugirdas JT, Levin NW, Kotanko P, et al. Comparison of proposed
alternative methods for rescaling dialysis dose: resting energy
expenditure, high metabolic rate organ mass, liver size, and body surface
area. Semin Dial. 2008;21:377–384.
14. Bazanelli AP, Kamimura MA, da Silva CB, et al. Resting energy expenditure
in peritoneal dialysis patients. Perit Dial Int. 2006;26:697–704.
15. Wang AY, Sea MM, Tang N, et al. Resting energy expenditure and
subsequent mortality risk in peritoneal dialysis patients. J Am Soc
Nephrol. 2004;15:3134–3143.
16. Sridharan S, Wong J, Vilar E, Farrington K. Comparison of energy
estimates in chronic kidney disease using doubly-labelled water. J Hum
Nutr Diet. 2016;29:59–66.
6 Kidney International (2016) -, -–-
17. Vilar E, Machado A, Garrett A, et al. Disease-specific predictive formulas for
energy expenditure in the dialysis population. J Ren Nutr. 2016;29:59–66.
18. Sridharan S, Berdeprado J, Vilar E, et al. A self-report comorbidity
questionnaire for haemodialysis patients. BMC Nephrol. 2014;15:134.
19. Rockwood K, Song X, MacKnight C, et al. A global clinical measure of
fitness and frailty in elderly people. CMAJ. 2005;173:489–495.
20. Finkel T. The metabolic regulation of aging. Nat Med. 2015;21:1416–1423.
21. Boon MR, Bakker LE, van der Linden RA, et al. High prevalence of
cardiovascular disease in South Asians: central role for brown adipose
tissue? Crit Rev Clin Lab Sci. 2015;52:150–157.
22. Davies SJ, Davenport A. The role of bioimpedance and biomarkers in
helping to aid clinical decision-making of volume assessments in dialysis
patients. Kidney Int. 2014;86:489–496.
23. Ramirez SP, Kapke A, Port FK, et al. Dialysis dose scaled to body surface
area and size-adjusted, sex-specific patient mortality. Clin J Am Soc
Nephrol. 2012;7:1977–1987.
24. Dombros N, Dratwa M, Feriani M, et al.; EBPG Expert Group on Peritoneal
Dialysis. European best practice guidelines for peritoneal dialysis. 7
Adequacy of peritoneal dialysis. Nephrol Dial Transplant. 2005;20(suppl
9):ix21–ix23.
25. Papakrivopoulou E, Booth J, Pinney J, Davenport A. Comparison of
volume status in asymptomatic haemodialysis and peritoneal dialysis
outpatients. Nephron Extra. 2012;2:48–54.
26. Fan S, Sayed RH, Davenport A. Extracellular volume expansion in
peritoneal dialysis patients. Int J Artif Organs. 2012;35:338–345.
27. Bergström J, Heimbürger O, Lindholm B. Calculation of the protein
equivalent of total nitrogen appearance from urea appearance. Which
formulas should be used? Perit Dial Int. 1998;18:467–473.
28. Davenport A. Effect of intra-abdominal dialysate on bioimpedance-derived
fluid volume status and body composition measurements in peritoneal
dialysis patients. Perit Dial Int. 2013;33:578–579.
29. Fürstenberg A, Davenport A. Assessment of body composition in
peritoneal dialysis patients using bioelectrical impedance and dual-
energy x-ray absorptiometry. Am J Nephrol. 2011;33:150–156.
30. McCafferty K, Fan S, Davenport A. Extracellular volume expansion,
measured by multi-frequency bioimpedance, does not help preserve
residual renal function in peritoneal dialysis patients. Kidney Int. 2014;85:
151–1557.
31. Ainsworth BE, Haskell WL, Herrmann SD, et al. 2011 Compendium of
Physical Activities: a second update of codes and MET values. Med Sci
Sports Exerc. 2011;43:1575–1581.
32. Daugirdas JT. Kt/V (and especially its modifications) remains a useful
measure of haemodialysis dose. Kidney Int. 2015;88:466–473.