Adequacy

Peritoneal
CANADA-USA

of Dialysis and Nutrition in Continuous

Dialysis: Association with Clinical Outcomes1
(CANUSA)

Peritoneal

Principal
Investigators:
David
Taylor, Prakash
R. Keshaviah
Coordinating
Mary Louise
Clinical

Center
Beecroft

Center

Dialysis

N. Churchill,

Investigators:

Investigators:

D. Wayne

Kevin

Kailash

Study

E. Thorpe,

K. Jindal.

Stanley

S.A.
Fenton,
Joanne
M. Bargman,
Dimitrios
G.
Oreopoulos.
David
N. Churchill,
George
G. Wu, Susan
D. Lavoie,
Adrian
Fine,
Ellen
Burgess.
James
C.
Brandes,
Karl D. Nolph,
Barbara
F. Prowant.
Denis
Page,
Francis X. McCusker,
Brendan
P. Teehan,
Mrinal
K. Dasgupta,
Kelvin Bettcher,
Ralph Caruana
Baxter

Lee

Healthcare

Investigators:

George

Soc.

Nephrol.

1996;

7:198-207)

ABSTRACT
The objective
of the study presented
here was to
evaluate
the relationship
of adequacy
of dialysis and
nutritional
status to mortality,
technique
failure,
and
morbidity.

This

was

prospective

cohort

study

680 patients

were

enrolled.

Follow-up

was

termi-

nated December
31 1993. There were 90 deaths,
137
transplants,
and 1 18 technique
failures.
Fifteen withdrew from dialysis. Analysis of the patient and technique
survival
used the Cox proportional
hazards
model with adequacy
of dialysis and nutritional
status
as time-dependent
covariates.
The relative risk (PR) of
death
increased
with increased
age, insulin-dependent diabetes
mellitus,
cardiovascular
disease,
decreased
serum albumin
concentration
and worsened
nutritional
status (subjective
global
assessment
and
percentage
lean body mass). A decrease
of 0. 1 unit
Kt/V per week was associated
with a 5% increase
in
the RR of death; a decrease
of 5 L/1 .73 m2 creatinine
clearance
(CCr) per week was associated
with a 7%
Increase
In the RR of death. The RR oftechnique
failure
,

Received

May

Appendix

17, 1995. Accepted

for participating

November
invesligators

21, 1995.
and

affiliated

104&6673/0702.0198$03.00/0
Journal of the American
Society of Nephrology
copyright
1996 by the American
Society
of Nephrology

198

Clinical

trial,

urea,

creatinine,

multivariate,

pa-

ontinuous

ambulatory
peritoneal
dialysis
1 ,2) is the treatment
used
for approximately
1 4% of the worlds
dialysis
population
(3). With
adjustment
for comorbidity,
Burton
and Walls
(4) and
the Canadian
Organ
Replacement
Register
(5) reported
no difference
between
CAPD and hemodialysis
with
respect
to patient
survival.
However,
Maiorca
et
al.

(6)

found

better

and greater
reported
an
patients

For
ysis

treated

patients
times

survival

treated
increased
with
better

with

these

et

53.5
a!.

yr
(7)

diabetic

studies

longer

status

better
patient
survival
(8,9).
small
molecular
weight
sobutes
decreased
morbidity
( 10). For
CAPD,
some
data
suggest
an
improved
survival
and
better
( 1 1 1 2). There
are conflicting
relationship
between
estimates
and clinical
outcomes
( 13-18).
of

aged

hemodialysis.

nutritional

with

itations

patients

CAPD.

treated

and

for

with
CAPD
while
Held
risk of death
for older

are

dial-

associated

Increased
removal
of
is associated
with
patients
treated
with
association
between
nutritional
status
reports
regarding
the
of adequacy
of dialysis
The methodobogic
lim-

include

small

sample

size

with inadequate
statistical
power,
insensitive
clinical
outcomes,
and use of univariate
rather
than
multivanate
statistical
analysis
(19).
In this article,
we report
the results
of a multicenter
prospective
cohort
study
of 680 patients
commencing
continuous
peritoneal
dialysis
in 14 centers
in Canada and the United
States
between
September
1 1990
and December
3 1 1992,
with follow-up
until
December 3 1 1993.
The objectives
of the study
were:
(1 ) to
evaluate
the association
of adequacy
of dialysis
with
mortality.
technique
failure
and hospitalization;
and
(2) to evaluate
the association
of nutritional
status
with
those
same
variables.
.

organizations.

correspondence
to Dr. D. churchill,
McMaster
University, St. Josephs
50 charlton
Avenue
East, Hamilton,
Ontario,
canada
L8N 4A6.
3

Key Words:
tient survival

of

consecutive
patients commencing
continuous
peritoneal dialysis in 14 centers in Canada
and the United
States. Between September
1 1990 and December31,
1992,

was increased
with decreased
albumin
concentration and
decreased
CCr. Hospitalization
was increased
with decreased
serum albumin
concentration,
worsened
nutrition
according
to subjective
global
assessment
and decreased
CCr. A weekly
Kt/V of 2. 1 and a weekly
CCr of 70 L/ 1.73 m2 were
each
associated
with an expected
2-yr survival
of
78%.

(CAPD)

deVeber,

W. Henderson

(J, Am.

Group23

Hospital,

METHODS
The
order

clinical
centers
participating
of number
of patients
entered,

Volume

in the study,
were Victoria

Number

listed
in
General

1996

Churchill

Hospital
(Halifax,
Nova Scotia);
Toronto
Western
and Toronto
General
Divisions
ofThe
Toronto
Hospital
(Toronto,
Ontario);
St. Josephs
Hospital
(Hamilton.
Ontario):
Credit
Valley
Hospital (Mississaugua,
Ontario);
Ottawa
Civic Hospital
(Ottawa.

Ontario);

St. Boniface

Hospital

(Winnipeg.

Manitoba):

Foot-

hills Hospital
(Calgary,
Alberta);
Medical
College
of Wisconsin (Milwaukee,
Wisconsin);
University
of Missouri
Medical
Center
(Columbia,
Missouri);
Ottawa
General
Hospital
(Ottawa, Ontario):
Lankenau
Hospital
(Philadelphia,
Pennsylvania):
University
of Alberta
Hospital
(Edmonton.
Alberta):
Medical
College
of Georgia
(Augusta,
Georgia).
The Coordinating
Center
was the Department
of Clinical
Epidemiology
and Biostatistics,
McMaster
University,
Hamilton,
Ontario.
All patients
commencing
continuous
peritoneal
dialysis
for
the first time between
September
1 , 1990 and December
31,
1992 were eligible for the study. Exclusion
criteria
were: (1)
being
unlikely
to survive
for at least
6 months;
(2) elective
living donor kidney
transplant
within 6 months;
(3) planned
move
from
the study
center
within
6 months;
(4) positive
hepatitis
B or HIV serology;
(5) active
systemic
inflammatory
disease;
or (6) failure
to sign the informed
consent
form. The
dialysis
prescription
was that prescribed
by the individual
patients
physician.
Changes
were made
for clinical
indications
and
without
reference
to the data
recorded
at the
Coordinating
Center.

The
age.

demographic
sex,

score

race,

(20),

data

recorded

functional

underlying

status

renal

at enrollment
according

disease,

to the

included
Karnofsky

insulin-dependent

dia-

betes
mellitus
(IDDM).
and history
of cardiovascular
disease
(CVD). CVD was defined
as a history
of previous
myocardial
infarction,
angina,
amputation
for vascular
disease,
or Class
III through
IV congestive
heart
failure.
Estimates
of nutritional
status
were
obtained
at enrollment
and every 6 months
thereafter.
If there
was an acute
medical
problem.
this estimate
was obtained
1 month
after
resolution
ofthe
problem.
Serum
albumin
concentration
was
determined
by the bromcresol
green
method.
The subjective
global
assessment
(SGA) of nutritional
status
was
determined
by using
a modification
of the method
described
by
Baker
et a!. and Detsky
et al. (2 1 .22). This
technique
is
reliable
and valid for ESRD patients
treated
with CAPD (23).
Four
items
were scored
on a seven-point
Likard-type
scale,
with
lower values
representing
worse
nutritional
status.
These
values
were
weight
loss during
the past
6 months,
anorexia,
subcutaneous
fat, and muscle
mass.
These
four
items
were
given
subjective
weights
to produce
a global
assessment.
Scores
of 1 to 2 represented
severe
malnutrition;
3 to 5, moderate
to mild
malnutrition;
and
6 to 7,
normal
nutrition.
The protein
catabolic
rate (PCR) was determined
according
to Randerson
et at. (24) and normalized
to
standard
body
weight
(total
body
water/0.58).
Total
body
water
(V) was determined
from the formula
of Watson
et at.

(25).

Percentage

of lean

body

mass

1% LBM] was

determined

from creatinine
kinetics
(26).
Adequacy
of dialysis
was estimated
total weekly
Kt/V for urea,
total weekly

by measurement
creatinine
clearance

of

(CCr),

(f32M]. Peritoneal

Kt

and

serum

beta-2-microglobulin

was estimated
from 24-h
dialysate
urea
excretion
serum
urea concentration.
Renal
Kt was estimated
concurrent
24-h urine
urea
excretion.
Peritoneal
clearance
was estimated
from the 24-h dialysate
excretion
and
the serum
creatinine
concentration

completion

creatinine

excretion

was corrected
for glucose
Interference
by using
correction
factor
(27) determined
for each clinical
laboratory.
The renal
contribution
to creatinine

a validated
chemistry
clearance

Journal

of the

collection.

of the American

Dialysate

and the
from the
creatinine
creatinine
at the

Society

of Nephrology

et al

was estimated
as the average
of renal
creatinine
and urea
clearance
(CCr + Curea)/2
(28). The serum
f32M estimations
were performed
by a solid-phase
time-resolved
fluoroimmunoassay
(DELFIA;
Wallace
Canada.
Vaudreuil,
Quebec)
in
two reference
laboratories.
The clinical
outcomes
were
mortality,
technique
failure,
and hospitalization.
Technique
failure
was defined
as transfer to hemodialysis
or to conventional
intermittent
peritoneal
dialysis.
Hospitalizations
for vascular
access
surgery.
renal
allograft
nephrectomy,
elective
pretransplant
assessment,
and renal transplantation
were excluded
from the analysis
of
the effect
of adequacy
of dialysis
on hospitalization.
The
duration
of follow-up
and total number
of days
hospitalized

were used

to calculate

days

hospitalized

per month

of follow-

up.
Statistical
analysis
of patient
mortality
and
technique
failure
used Andersen
and Gills (29) extension
to the Cox
proportional
hazards
model
(30) with estimates
of nutritional
status
and adequacy
of dialysis
as time-dependent
covariates (31,32).
Events
(e.g., death
or technique
failure)
were
attributed
to the level of nutrition
or adequacy
of dialysis

recorded

at the

6-month

evaluation

preceding

the

event.

Transplantation,
recovery
of renal
function,
technique
failure and loss to follow-up
were censored
observations
for the
patient
survival
analysis.
Death,
transplantation.
recovery
of
renal
function,
and loss to follow-up
were censored
observations for the technique
survival
analysis.
All baseline
demographic
and clinical
variables
and baseline
serum
albumin
concentrations
were entered.
Backward
elimination
removed
nonsignificant
variables.
The three
estimates
of adequacy
of
dialysis
(Kt/V.
CCr, and (32M) were each added
separately
to
this
model.
Three
estimates
of nutritional
status
(NPCR.
SGA, and % LBM) were
each
added
separately
to a model
containing
the significant
baseline
demographic
variables,
serum
albumin
concentration,
and one estimate
of adequacy
of dialysis.
The likelihood
ratio
test
(3 1 .33) was
used
to
determine
whether
or not the addition
of a variable
to a model
added
significantly
to that model.
The validity
of the proportional
hazards
assumption
was considered
for all variables
remaining
in the final models
by examining
the Schoenfeld
residuals
(34).
To further
explore
the effect
of Kt/V
and CCr on patient
survival,
the expected
survival
time
was
determined
for
several
theoretical
Kt/V
and
CCr values.
Survival
curves

were
values
values
survival
values

were

constructed

from

the fitted

models

(3 1 ) for weekly

of 2.3, 2.1,
1.9, 1.7, and
1.5, and
of 95, 80, 70, 55, and 40 L/wk
per
curves
assume
no change
in the
over follow-up
time.
Peritoneal
and

considered

equivalent

with

decreases

Kt/V

for weekly
CCr
1 .73 m2. These
Kt/V
and
CCr
renal
clearance

in renal

function

compensated
by Increased
peritoneal
clearance.
A log-linear
model
was used to analyze
the hospitalization
data
(35). The dependent
variable
was the number
of days
hospitalized
per month
of follow-up.
The baseline
explanatory variables
were:
country
(Canada/United
States);
age at
enrollment;
sex: race (Caucasian
/ non-Caucasian);
Karnofsky score
(<80 or 80):
renal
disease:
IDDM;
and CVD. For
explanatory
variables
estimated
at 6-month
intervals
during
the study.
the weighted
average
was used.
These
variables

were

serum

albumin

concentration,

SGA.

NPCR.

% LBM,

Kt/V,
CCr, and serum
f32M.
The sample-size
calculation
was made
on the basis
of the
patient
survival
analysis.
For the multivariate
analysis.
approximately
ten events
(death)
per independent
variable
are
required
to produce
a model
of reasonable
accuracy
(36).
Given
the ten independent
variables
to be evaluated,
the

199

Adequacy

study

and

population

100 deaths
ment target
the

Nutrition

had

in CAPD

to be large

enough

to experience

about

during
the follow-up
period.
The patient
was 700 patients,
with follow-up
extended

requisite

number

of events

was

enrolluntil

TABLE 2. Adequacy

accumulated.

Kt/V (weekly)
Kt/V renal
Kt/V peritoneal

There
were 680 patients
enrolled
in the study.
There
were 97.9%
treated
with CAPD and 2. 1% with continuous
cycling
peritoneal
dialysis
(CCPD).
The mean
and median
daily
prescribed
instilled
volumes
were
7.7 and 8.0 L, respectively.
The mean
age was 54.3 yr.
with
a range
of 18 to 82 yr. The
distribution
of
demographic
and clinical
factors
is shown
In Table
1.
The mean
height
was 1 .66 m, mean
weight
was 67.8
kg, and mean
body
mass
index
was 24.6 kg per m2.
The baseline
values
for adequacy
of dialysis
and nutritional
status
are shown
in Table
2. By SGA assesswere
30 patients
(4.2%)
with
severe
mal364
(5 1 .2%)
with
mild
to
moderate
and
3 1 7 (44.6%)
considered
well flour-

and

clinical

2.38
0.71
1.67

Race
Caucasian
African American
Asian
Aboriginal
Education
<High school
High school
Technical
College/university
Karnofsky

Gender
Male
Female

Renal Disease
Diabetes
Glomerulonephritis
Nephrosclerosis
Renal vascular
Polycystic

kidney

disease
Other
Comorbidity
IDDM
CHF(111 orlV)
Ml
Angina
Amputation
IDDM,
failure;
0

200

insulin-dependent
Ml,

myocardlal

diabetes
infarction.

0.99

1.89

38.8
44.2

34.6
43.6

18.2
38.5

53.2
50.1

Serum 32M
(mg/I)
Serum albumin
(g/L)
NPCR (g/kg)
SGA (1-7)
% IBM

22.5

21.0

16.0

27.0

34.9

35.0

31.0

39.0

Kt/V.
protein

for urea;
catabolic

1.04
5.19
62.6

1.01
5.00

0.87
4.00

1.19
6.00

53.3

60.9

71.0

OCr. creatinine
clearance
rate; NPCR,
rate; SGA, subjective
global
assessment;

normalized
LBM, lean

mass.

factorsa
N

(%)

558
57
38
20

82.1
8.4
5.6
2.9

1.0

315
187
79
99

46.3
27.5
11.6
14.6

399

58.7

268

39.4

13

1.9

394

57.9

286

42.1

202
95
42

29.7
23.5
14.0
6.2

47

6.9

134

19.7

151

22.2

0.28

156

22.9

ble

99
159
19

14.6
23.4
2.8

progressive

160

disease

2.68

98.0

nutritional

m ellitus; CHF. congestive

mean

Kt/V, CCr, and

estimates

of

status
(Months)

680

N
Kt/V

Peritoneal
Renal
CCr (1.73 m2)
Peritoneal
Renal

12

525

18

321

24
78

166

2.38

2.25

2.10

2.02

1.99a

1.67
0.71

1.67
0.58

1 .68
0.41

1.66
0.39

1.70
0.28#{176}

83.0
44.2
38.8

74.7
44.6
30.1

68.3

65.7

61.6#{176}

46.4
21.9

45.3
20.4

473b

Serum albumin
(g/L)
NPCR (g/Kg/day)
SGA (0-7)

34.9

35.1

35.1

35.1

35.2

% IBM

62.6

1.03
5.96
63.3

0.97
6.00
63.5

0.96
6.00
63.1

14.3#{176}

Nutrition

Score

80-100
50-70
<50

1.96
0.30
1.37
63.8

Follow-Up

Other

2.28
0.64
1.61

Quartile

78.8

TABLE 3. Weekly
Factor

3rd

1st
Quartile

Median

83.0

body

1 Demographic

status

CCr (1/week
per
1.73 m2)
CCr renal
CCr peritoneal

ished.

TABLE

nutritional

Mean

RESULTS

ment,
there
nutrition,
malnutrition,

and

(baseline)a

a
b

1.04
5.19

601b

65.3

p < o.ooooi . baseline

versus 24 months
(paired
t test).
p < 0.005. baseline
versus 24 months (paired t test).

The changes
In adequacy
status
are shown
In Table
numbers
is partly
because
dialysis,
staggered

and
transplantation.
entry
of patients

of dialysis
and
3. The decrease
of death,
transfer
over

but
the

nutritional
in patient
to hemo-

also
study

reflects
period.

total weekly
Kt/V,
there
is a progressive
decrease
2.38 to 1 .99. This decrease
is entirely
the result
of the renal
component.
which
decreases
from
whereas

at

Kt/V,
component,
whereas
44.2
to
for

heart

0.99

1 .67

the

to

decrease

this
the
47.3

peritoneal

1 .70.

For
from

component

total
83.0

the
For

from
of loss
0.71 to

remains

weekly

CCr,

sta-

there

to 6 1 .6 L/ 1 .73

is a
m2.

is entirely
the result
of loss of the
which
decreases
from
38.8
to
peritoneal
CCr increases
slightly
L/wk
per
1 .73 m2. The
serum

Volume

Number

As

renal
14.3
from
j32M

1996

Churchill

Increased

from

22.5

cant
increase
concentration,
During
the

to

24.3

mg/L.

There

for

serum

albumin

of

1 0, 1 38

patient-

months,
there
were
90
deaths,
classified
as cardiovascular,
five
malignancy,
and
1 3 as various
137 transplants
and
1 18 transfers
ysis
and
seven
to IPD).
Among

of

patients,

264

dialysis
tiveby

on December
censored
at

continued
sons:

on

were

alive
3 1,
time.

that

7 1 other

patients);
tients);
tients);
renal

had

for

with

the

(8

following

dialysate

patients).

The

CVDb

2.09

1.33-3.28

1.93
0.94
0.94
0.75

1.14-3.28
0.90- 0.97

Furthermore,
the assumption
The relative
slightly

detected

(37,38).
risk (RR)

of death

papaof

for

two

different

the

models

no violations
variable

(Kt/V

albumin
( I 1 g/L)
I 0.1 units/wk)
f 1 unit)

0.90-0.99

0.66-0.85

Kt/V as estimate
nutritional
status.

of adequacy

CVD, cardiovascular

TABLE

5. Cox

of dialysis

and

SGA as the estimate

of

disease.

proportional

hazards

model#{176}

esti-

and

was

95%
Confidence

Risk

Interval

Age(peryear)

1.03

1.01-1.05

IDDM
CVD
Country(USA)

1 .49
2.12

0.92-2.42
1.35-3.34

(f

Serum albumin
CCr ( f 5 L/wk/1
SGA ( I 1 unIt)
as the

OCr

estimate

of

Relative
Mortality

Variable

no serious
assumption.

for each

(
(

(USA)

rea-

dialysis
( 15
center
( 13
and recovery

and
showed
hazards

Interval
1.01-1.05

Kt/V
SGA

The final
Cox proportional
hazards
model
results
are summarized
in Tables
4 and 5. In Table
4, Kt/V is
used
as the estimate
of adequacy,
whereas
CCr is
used
in Table 5. The Schoenfeld
residuals
were plotted
for each
variable
the
proportional
Harrells
z test

95%
Confidence

0.89-2.36

Mortality

time
of

Relative
Mortality

1.45

mates
of patient
survival,
technique
survival,
and the
probability
of being
alive
and
on CAPD/CCPD
are
shown
In Figure
1 . The probability
of patient
survival
to 24 months
is 78%; of technique
survival,
75%; and
of being
alive and receiving
CAPD treatment,
58%.

against
violation

model#{176}

1.03

Serum

(22

Kaplan-Meier

hazards

Age(peryear)
IDDM
Country

dis-

collection

elective
withdrawal
from
move
to a non-study
dialysis
loss to follow-up
( 13 patients);
function

been

proportional

Risk

which
68 were
as infection,
four
as
causes.
There
were
( 1 1 1 to hemodial-

Follow-up

4. Cox

Variable

the remaining
335
on continuous
peritoneal
1 993 and were administra-

patients

noncompliance

TABLE

is a signifi-

in SGA,
but
not
NPCR,
or % LBM.
follow-up
period

et al

.73 m2)

estimate

of nutritional

of adequacy

CCr).

Accordingly,
where
different,
both
values
are provided.
The SGA has
been
used
as the estimate
of
nutritional
status
rather
than
NPCR
or % LBM because
it was considered
more
credible
than
estimates
with
statistical
coupling
with
adequacy.
The PR of

2-Year

weekly

Kt/V

1.14-3.31
0.90-0.97

0.93

0.88- 0.98

0.75

0.66-0.85

of dialysis

and

SGA

patient
and

survival

CCr

(1/1.73

m2)
Survival

81

95

86

78

80

81

1.7

74
71

1.5

66

70
55
40

78
72
65

Survival

2.3
2.1
1.9

is
two

as the

according

CCr

Kt/V

death
for

1.95
0.94

status.

TABLE 6. Expected
to sustained

1 g/1)

1 .03 greater
for
patients
identical

being

a 1 -yr difference
for all other
the PR ofdeath

in
risk

age (I.e.,
factors.

1 yr older increases
by 3%). The
those
with
IDDM
was
1 .45 to 1 .49 whereas
those
with a history
ofCVD
had an RR of2.09
to 2.12.
Patients
in the United
States
had a RR ofdeath
1 .93 to
1 .95 greater
than
those
in Canada.
A 1 g/L-lower
serum
albumin
concentration
was associated
with an
6% increase
in the RR of death.
A 1 -U lower SGA score

0)
C
>

C/)
C
C.)

a.

12
Time

Figure

Journal

1 . Probabilities

of the

American

of patient

Society

18

24

in Months

and

technique

of Nephrology

survival.

RR

for

was

associated

with

a 25%

increase

In the

RR of death.

For two patients

who differed
only in weekly
Kt/V,
a
0. 1 lower value
was associated
with a 6% increase
in
the PR of death.
For a 5 L/wk
per 1 .73 m2 lower
CCr,
there
was
a 7% increase
In the
RR of death.
In
separate
models,
1%-bower
% LBM and
a 1 mg/Lgreater
serum
2M
were associated
with 3% and 2%
increases
in the RR of death,
respectively.

201

Adequacy

The

and

Nutrition

expected

in CAPD

patient

survival

was

estimated

for

several
sustained
weekly
Kt/V
and
CCr values.
For
weekly
Kt/Vvalues
of2.3,
2.1,
1.9,
1.7, and
1.5, the
expected
2-yr
survivals
were
81%,
78%,
74%,
71%,
and 66%.
For weekly
CCr values
of95,
80, 70, 55, and

40 L/ 1 732
the expected
8 1 % 78%
72%
and 65%
Table
6 and Figures
2 and
.

2-yr
survivals
. These
data
3.

were
86%,
are shown
in

CO

Technique
There
There

Failure
were

were

a.

1 18
36

exit-site

infection,

function.

There

transfers

to hemodialysis

transfers

because

and

were

five

of

because

or
peritonitis

of catheter

19 transfers

attributed

or
mal-

to mad-

equate
dialysis
( 1 2 transfers
because
of inadequate
clearance
and
7 because
of failure
of ultrafIltration).
Patient
and partner
fatigue
accounted
for two transfers
each;
social
causes
were
responsible
for eight

significantly
L/wk
per

to this model.
1 .73 m2 increase

0.99).
None
of
associated
with

the
other
technique

Figure

failure

increased

by

serum
decrease

albumin
in CCr.

failure.
The RR associated
0.95
(95%
CI, 0.92
to 0.98).
of adequacy
which
added
The RR associated
with
a 5

was

0.95

estimates
failure.

5%

with

concentration

(95%
of

The

PR

1 g/L

and

CI, 0.9 1 to

nutrition

was

of technique
decrease

with

In

L/wk

Hospitalization
There
were
cardiovascular
23.0%);

1 239 hospitalizations;
disease
( 189, 15.3%);

gastrointestinal

disease

the causes
peritonitis

( 120,

9.7%);

12

18

24

Time in Months

failures.
The remaining
46 were considered
to be the
results
of other
medical
causes.
Among
the baseline
demographic
and clinical
variables,
a decreased
serum
albumin
concentration
was associated
with
an
increased
risk of technique
with
a 1 g/L increase
was
CCr was
the only
estimate

0
N.

IPD.

were
(285,
other

infections
(78, 6.3%).
No other
category
accounted
for
more
than
5%. There
are
more
days
hospitalized
associated
with increased
age, female
sex, worse
func-

3. PredIcted

tional

status.

CVD,

concentration.

a strong

the
and

estimates
decreased

tal

days,

of patient

IDDM,

and

Malnutrition,

had

(Table

probability

association

with

of adequacy
(32M were

whereas

bower

as

by CCr.

serum

albumin

estimated

by the

hospitalization.

of dialysis,
associated

Kt/V

survival

was

less

with

SGA,

Among

increased
fewer

strongly

CCr
hospi-

associated

7).

About
50%
of the
The data
were
highly
usual
linear
regression
was

used.

dialysis.
Among

Among
CCr

and

and

the three
% LBM were

tive

within

patients
were
not hospitalized.
skewed
and Inappropriate
for
methods.
Log-linear
modeling

the
2M

three

estimates

were

significant;

estimates
significant;

of adequacy
Kt/V

of nutritional
NPCR was

of

was

not.

status,
not. The

SGA

final
model,
using
SGA as the nutrition
estimate
and
CCr
as the adequacy
estimate,
is shown
in Table
8. All
variables
were statistically
significant
(P < 0.05).
The
exponential
coefficient
is an expression
of the relative
length
of time hospitalized
compared
with the alternaindicates

the
that

variable.

For

patients

country,

in centers

the

in the

value

1.23

United

States

were hospitalized
on average
1 .23 times
as bong as a
comparable
patient
in Canada.
Each
additional
year
of age translates
into
1 .02 times
the time
spent
in
hospital.
Those
with Karnofsky
scores
<80 spent
1.63
times
the days hospitalized
compared
with those
with
scores
80.
Patients
with IDDM
and CVD spent
1.47
and 1 .22 as many
days hospitalized
as those
without

0)
C
>

IDDM

was

Cl)

talized
/32M,

8
a.

and

CVD,

associated

respectively.

with

0.82

For

times

SGA,

as

a 1 -U increase

many

as those
with
a score
1 U bower.
a value
1 mg/L
greater
was
associated

times
as many
greater
CCr was
days
hospitalized.

days
hospitalized.
associated
with

days

hospi-

For serum
with
1.02

A 5 L per week
a 0.99 times
as many

DISCUSSION
0

12

18

24

Time in Months

Figure

202

2. PredIcted

probability

of patient

survival

by Kt/V.

In

a prospective

with continuous
nutrition
and

cohort
peritoneal
adequacy

study

of patients

dialysis,
of dialysis

Volume

treated

the association
with clinical

Number

of
out-

1996

Churchill

TABLE 7. Days

hospitalized

per year

TABLE 8. Multiplicative
per month

factor

on days

hospitalized

Days

Factor

.
Variable

Country
Canada
USA
Age (yr)
<45

Country

Age (1 year)
Karnofsky
(<80)

10.1

ESRDb

18.7

(glomerulonephritis)
(hypertension-RVD)
(other)
IDDM (present)
CVD (present)
Serum
albumin
(1 g/L)
SGA (1 unit)
CCr (5 L/wk per 1.73 m2)

29.0
22.7
16.7

Karnofsky
80
<80

11.3

30.2
15.2

Present

26.2
15.5

Absent
Present
Serum

albumin

1.47

1.22
0.95
0.82
0.99

(g/L)
7.3
12.7

23.6
35.6

Malnutrition
(SGA)
None
Mild-moderate
Severe
CCr (L/wk/ 1.73 m2)

0)
C

13.3

----.

Cl)

24.8
71.0

C
G)

17.9

27.8

>2.1

18.1

1.7-2.1
<1.7

25.8

Number
Canada
USA

Time

19.7

80%
63%

14.6

has
been
techniques

evaluated
to control

by
for

using
multivariate
baseline
demographic

53.4
The
both

sta-

istics
tically

and clinical
variables.
Better
nutrition
was associated
with better
patient
survival
and fewer
days
hospitalized. A higher
dose
of dialysis.
inclusive
of residual
function,

was

better

associated

technique

hospitalized.
The increased
age, IDDM,
and

of the

Figure 4. Predicted

At Risk
59

189
26

10

12

18

24

343

81

probability

in Months

of patient

survival

by country.

26.2

>30
20-30
<20

was
unexpected.
curves
(Figure

491

18.7

f32M (mg/L)

The

USA

Kt/V (wk)

centers

canada

a.

14.4

>80
60-80
<60

Journal

0.58
0.80
0.73

31.7

>40
35-40
30-34
<30

in the

1.02
1.63

Absent
IDDM

expected.

(USA)

The e0&
Is the
multiplicative
factor
Indicating
the relative
time
hospitalized
compared
to the alternative
within the variable
(e.g.. for
USA. the alternative
is Canada).
For continuous
variables,
the alternatives
are 1 year younger
for age.
1 gIL less for serum albumin
concentration,
1 unit less for SGA. and 5 L/wk per 1 .73 m2 less for CCr.
b For ESRD, the alternative
is diabetes.

CVD

renal
survival,

1.23

18.6

Sex
Females
Males

tistical

e COEF#{176}

21.7

45-64
>64

comes

et al

with

survival,

better

and

RR of death
associated
history
of cardiovascular
increased

in the
The

4) show

American

RR

United

of death

States

Kapban-Meier

a 2-yr

Society

with

1 .93

patient

survival

of Nephrology

days

data

compared

with

of the patients
significant

are compared
differences
were

In Table
present

9. Statisfor race,

observed

in

this

study

are

consistent

with

data

patients

from Canada
(4) and the United
States
(7). The different survival
rates
are not explained
by the variables
entered
in this analysis.
The degree
to which
serum
albumin
concentration

to

is an

increased
disease
was

among

(RR

centers

mean
ages were 54.4 and
yr in Canada
and the United
States,
respectively.
serum
albumin
concentration
was
34.9
g/L
in
countries.
Demographic
and clinical
character-

body
surface
area,
Kt/V
(total
and
peritoneal),
and
CCr (peritoneaJ).
However,
all of these
variables
were
controlled
in the multivariate
analysis.
The survival

patient

fewer

for those
in the Canadian
in the U.S. centers.
The

1.95)

survival

probability

of

estimate

and
Cox
(40)
among
patients

of nutrition

is controversial

reported
excellent
with persistently

patient
low serum

(39).

Fine

survival
albumin

203

Adequacy

and

Nutrition

in CAPD

TABLE 9. Canada-USA

Comparison

Factor

(47)

Canada

USA

Gender

suggested

that

the

malized

and

by

dry

weight

59%

54%

41%

46%

85%

66%

4%

30%

IDDM

22%

25%

CVD
BSA (m2)#{176}

34%

42%

A 0. 1 difference

1.74

1.80

6%

2.2

Peritoneal#{176}

1.7

Renal

0.7

1.5
0.7

m2 difference
in
difference.
When
controlling

CCr

2.4

clinical

there

Race#{176}

Kt/V (wk)
Totala

CCr (1 .73 m2/wk)

Total
Peritoneala

83.8

Each
of the three
were
independently

78.4
41.2

44.7
39.1

Renal

37.2

0.05.

op<

values.

Struijk
et al. (4 1 ) found
no association
between
serum
albumin
concentration
and patient
survival.
The PR for death
was 0.89 for a 1 g/L-greater
serum
albumin
concentration
(95%
CI, 0.76
to 1.03)
compared
with
a PR of 0.94
(95%
CI, 0.90
to 0.97)
in
this study.
study
from
statistical

The
the

back of statistical
Netherlands
may

power

albumin
with

the serum
albumin
dated
with
mortality.

ization.
change

significance
in the
be related
to low
Teehan
et al. (43) found
that
concentration
less than
35
very poor survival.
We found

(36,42).

an average
serum
g/L was associated

concentration
technique

A difference
in the PR

to be strongly
assofailure
and hospital-

of 1 g/L was associated

of death,
a 5% change

technique
survival,
and
a 5% change
tabized.
The SGA is a valid
clinical
estimate

with a 6%
in the RR of
in days

hospi-

other

in the

patients.
PCR nor-

estimates

of

of adequacy
with
patient

Kt/V

RR

was

of death;

was

for

variables,

nutri-

of dialysis
survival.

associated

with

a 5-L/week

per

associated

baseline

with

7%

demographic

were

1.73

and

associations

between

improved
patient
survival
and increased
serum
albumm
concentration,
greater
% LBM,
higher
SGA
scores,
lower serum
2M
concentration,
greater
Kt/V,
and greater
CCr.
Several
sources
of bias
were
considered.
We had
censored
the 15 withdrawals
from
dialysis.
Analysis
with the withdrawals
treated
as patient
death
did not
change
the association
of estimates
of nutrition
and
adequacy
of dialysis
with patient
survival.
Events
in
any
mate

6-month
period
had
of adequacy
at the

been
attributed
beginning
of that

to the
period.

estiFor

events
values

occurring
later
in a period,
the Kt/V
and CCr
would
have been
systematically
bower than
the
value
at the beginning
of that
period.
A randomized
clinical
trial with constant
dialysis
dose would
provide
a more accurate
evaluation
of the association
between
adequacy
and
clinical
outcomes.
Estimates
of adequacy
and
estimates
of nutrition
were
treated
as
time-dependent
covariates.
If nutrition
were
significantly

mate

affected

by

adequacy

of mortality

Analysis
for

estimates
associated

in weekly

difference

with

baseline
of nutrition

catabolic

in CAPD
between

tion.
Normalization
using
standard
body
weight
(V/
0.58)
dId not discriminate
between
well and
poorly
nourished
patients
whereas
use of ideal
body weight
from National
Health
and Nutrition
Examination
Survey (NHANES)
tables
(48) did.

Female

African-American

protein

of nutrition
correlations

Male

Caucasian

normalized

rate is a flawed
marker
They reported
negative

for
estimates

variables

mortality

for

of dialysis,

adequacy

the

would

of nutrition

did

not

adequacy

the

those

risk

esti-

incorrect.

entered

change

from

risk

be

only

as

estimates

determined

of
by

us-

patients
(23). Young
and colleagues
used
SGA
In a study
of prevalent
CAPD
patients
and reported
that 8% were severely
malnourished,
32% had mild to

ing

moderate

founder,
but, in a cohort
study,
cannot
be evaluated.
The expected
2-yr survival
probabilities
(Table
6) are
based
on the assumption
that increases
in peritoneal
clearance
will compensate
for the loss
of residual
renal
function
and sustain
Kt/V
and CCr values
at a
given
level.
However,
the survival
probabilities
were

CAPD

for

malnutrition,

malnutrition

moderate

and

(44).

We

malnutrition

Another
estimated

estimate
by

red,

anthropometrics,

46).

Keshaviah

lation

between

of death.
NPCR was

204

and

patients.
with

or
(26)

not

estimates
lower.
associated

associated

no

evidence

and

mild

5 1 .2%

A 1 -U change
a 25% difference
is % LBM,
impedance.

creatinine
reported

bioelectrical

infrared,
whereas
were
systematically
ence
in % LBM was
RR

severe

4.2%

of nutrition
bioebectrical
et at.

showed

found

in

tiveby of incident
CAPD
7-point
scale is associated
RR of death.
be

60%

(26,45,

excellent

impedance

in this
in the

which
can
near
infra-

kinetics
an

to

respec-

corre-

and

near

from
creatinine
kinetics
We found
that a 1% differwith a 3% change
in the

with

mortality.

Harty

et at.

time-dependent
tionship
between
quality

of

care

the

covariates.
estimates

provided,

the

If there
were
a relaof adequacy
and the
latter

could

be

a con-

generated
by deaths
that
occurred
during
loss
of
residual
renal
function
without
a compensatory
increase
in peritoneal
clearance.
The assumption
that
an increase
in peritoneal
clearance
can clinically
compensate
for the
loss
of residual
renal
function
is

unproven.
Theoretical
weekly

Kt/V

Several
tween

studies
weekly

constructs
should

have
Kt/V

suggest
be

2.0

that

to

2.25

addressed
and
patient

the

Volume

the

target

(1,2,49,50,51).

relationship
survival
by

Number

beusing

1996

Churchill

univariate
analyses
(13,15,52,53).
found
no effect
of Kt/V
on patient
patients.
In a study
of 102 CAPD
over
1 2 months,
the mean
weekly
92

patients

for

those

who

who

survival

survived

died

was

( 15).

of >90%

for

2.0,

compared
with
1.7
et al. reported
a 5-yr
a weekly
Kt/V
>1.89,

Teehan

those

Blake
et a!. (13)
survival
among
76
patients
in Spain
Kt/V
value
for the

with

compared
with
<50%
for those
with
less than
1.89
(52). Lamelre
et al. (53) reported
the changes
in Kt/V
for 16 patients
who had survived
for 5 yr on CAPD.
There
was progressive
loss of renal
function
over 5 yr.
The peritoneal
contribution
remained
> 1 .9 per
wk.
These

clinical

studies

suggest

a target

weekly

Kt/V

value
of 1 .9 to 2. 1 . Unlike
Kt/V,
there
are no theoretical constructs
on which
to base
adequate
dialysis
according
to CCr. The recommendation
of a total CCr
40 to 50 L per 1 .73 m2 per week
is based
on clinical
experience

Chen

et
longer

with

(54).
al. have
duration

with

tality;
beyond

optimal
dialysis
may
be defined
as the dose
which
the incremental
clinical
benefit
does not
the added
cost or patient
burden.
Within
the
dose

range

of

is

within

the

constructs

increase

CAPD

32M

in serum

treatment

increased

2M

(56).

from

2.0

We

provide

as

Kt/V
of
considered

high

simple

for a patient

a daily

production

a decrease

in

crease
increased

associated
with an increase
The
only
independent

1500

mg

function

in serum
variables

per

wk

should

be

2M.
associated

technique
failure
were
serum
albumin
concentration
and CCr.
The former
could
reflect
malnutrition,
overhydration,
or both,
whereas
the latter
is an estimate
of
dose
of dialysis.
Among
the
factors
associated
with
more

days

hospitalized

were

increased

age,

worse

functional
status,
IDDM,
and CVD. Malnutrition,
estimated
by the SGA,
had
a very
strong
association
with an increase
in days hospitalized.
A similar
association
with decreased
serum
albumin
concentration
may

be

partly

creased

a reflection

CCr

was

of

associated

nutritional

with

status.

fewer

days

In-

hospital-

ized;
In
with
This

there
was
this
study,
a sustained
same
2-yr

no association
with
Kt/V.
the expected
2-yr
survival
associated
weekly
Kt/V
value
of 2. 1 was
78%.
survival
was
associated
with
a weekly

CCr

of 70 L/wk

per

level
tion,

of residual
these
Kt/V

taneously.

1 .73

m2.

Because

renal
function
and CCr doses

With

progressive

ofthe

boss

of

of 78%

64-yr-old
between
reported
patient
States

the

73%

for

a cohort

of 45

to

patients
commencing
dialysis
in Canada
1989 and 1993 (5) and a 66% 2-year
survival
in the United
States
(64). This difference
in
survival
times
between
Canada
and the United
is consistent

Adequate

Journal

exceeds

of the

dialysis

American

with

that

may

be defined

Society

observed

in

as the

of Nephrology

and

by

theoretical

sliglitly

dose

as

of 70
target.

optimize

is feasible

L/week

dialysis

with

per

dose

1 .73

include

m2
pre-

devices.

with

clearance
volumes

a body
For

The

optimal

surface

to 3.0
CCPD

fill

area

to 3.5 for
patients,

body
one

surface
can
in-

by adding
daytime
exchanges.
and frequency
of exchanges
for

declining

renal

vob-

of 1 . 73 m2

function

If
used

are

unac-

ceptable
to the patient,
or if therapy
cost is excessive,
transfer
to hemodialysis
should
be considered.
According
to the peak
urea
concentration
hypothesis.
a
weekly
continuous
peritoneal
dialysis
Kt/V
of 2. 1 is
equivalent
to a thrice
weekly
hemodialysis
Kt/V of 1.5
(5 1). A transfer
ity to achieve
a hemodialysis

to hemodialysis
on the basis
of Inabila weekly
Kt/V
of 2. 1 would
require
that
Kt/V
of 1 .5 be provided.

The
determination
dose will require
further
a cost-effective

of

manner

optimal
peritoneal
dialysis
study.
In order
to achieve
it in
that

is consistent

with

life, health-care
providers
and
together
to optimize
the dialysis

a high

Industry
prescrip-

CANUSA
Peritoneal
Affiliations

Dialysis

D.N.
Churchill,
D.W.
croft,
St. Josephs

Taylor,
Hospital.

Study

Group

renal

function,
achievement
of the Kt/V
target
may not be
associated
with simultaneous
achievement
of the CCr
target.
Further
study
is required
to determine
which
estimate
of adequacy
should
be used
clinically.
The study
population
had a mean
age of 54 yr; 30%
had diabetes
mebbitus
as the cause
for ESRD.
The 2-yr
survival

Kt/V

APPENDIX

populasimul-

residual

suggested

and

is 2.5 L. This
increases
area
>2.0
m2 (60).

quality
of
must
work
tion.

significant

in the study
were achieved

2.25

automated

to compensate

with

to

a dialysis

to

ume

to

higher

mor-

scription
of exchange
volumes
based
on body
surface
area
(60) and
the addition
of a nocturnal
exchange
with

renal

study,

and

associated
with better
clinical
2-yr
survival
of 78%
was
Kt/V value
of 2. 1 . This value

2. 1 or a CCr
a reasonable

over
24 months.
Chen
et a!. (56) also reported
a
peritoneal
dialytic
32M
removal
of 1 19 mg;
et at. reported
22 1 mg removed
per wk (57).

of 1000

to 24.3

this

of morbidity

higher
than
the 1.9
to 2.0 suggested
by clinical
studies
( 15,48,49).
A 2-yr
survival
of 78% is better
than
that currently
observed
In North
America
(5,59).
Pending
further
studies
to
define
optimal
dialysis.
it would
appear
reasonable
to

mg/L
weekly
Lysaght

residual

22.5

in

( 1 ,2,45,47)

that

Given

serum

an

level

higher
CCr values
were
outcomes.
The predicted
associated
with a weekly

Strategies

shown

acceptable

found

(58),

the

dated

justify
dialysis

an

et al

this

dose

study.

asso-

Hamilton,

P.R.

K.E.
Thorpe.
McMaster

M.L.
BeeUniversity,

Ontario.

Keshaviah.

Healthcare,
K.K. Jindal,

G. deVeber,
Round
Victoria

L.W.

Lake,
General

Henderson,

Illinois.
Hospital,

Baxter

Halifax,

Nova

Scotia.
S.S.A.

Fenton,

Toronto
G.G.

J.M.

Hospital,

Wu,

Bargman,

D.G.

University

Credit

Valley

Oreopoubos,

of Toronto.

Hospital,

The

Ontario.

Mississauga.

On-

University

of Ot-

tario.

S.D.
tawa,

Lavoie,
Ottawa,

Ottawa

Civic

Hospital,

Ontario.

205

Adequacy

and

A. Fine,
St.
Winnipeg.

Nutrition

Boniface
Manitoba.

in CAPD

Hospital.

E.

Burgess,
Foothills
Calgary,
Alberta.
J.C.
Brandes,
Medical
kee, Wisconsin.

K.D.
D.

Nolph,

B.F.

Hospital,

B.P.

of Calgary,

of Wisconsin.

12.

Milwau13.

Prowant,

McCusker,

of Manitoba,

University

College

University

ical Center,
Columbia,
Page,
Ottawa
General
tawa,
Ottawa,
Ontario.

F.X.

University

ofMissouri

Missouri.
Hospital,

Teehan,

University

Lankenau

mas
Jefferson
Medical
College,
sylvania.
M.K.
Dasgupta.
K. Bettcher,
Hospital.
University
of Alberta,
R. Caruana,
Medical
College
Georgia.

Medof Ot-

Hospital,

Tho-

Philadelphia,

1990:6:173-177.

15.

Penn-

University
Edmonton,
of Georgia.

of

Alberta
Alberta.
Augusta,

ACKNOWLEDGMENTS
This

study

Clinical

was

funded

Research

the

United

Father
ilton,

by

States

of

Sean
OSullivan
Ontario);
the

Calgary.

Hospital.

Winnipeg.
Alberta:

Halifax.

Nova

the

Kidney

and

Award)

by

America.

Manitoba):
Nephrolog,y

the

Scotia;

and

the

Healthcare

Edward
for

for

Vonesh

her

research

for

secretarial
assistants

Funding

was

their

skills.
in each

(Premiere

in Canada
also

Research

support

in this

consultation
Finally.

we

of the

express

clinical

Fund

Toronto.
and
study.
and
our
centers

by

in
the

Hospital.
Ham(St.
Boniface

the
Foothills
Hospital
Research
Fund
ofVictoria
Hospital.
Woodworth,

and

provided

(St. Josephs
of Manitoba

Dialysis

statistical

ofCanada

Healthcare

Research
Centre
Kidney
Foundation

General
Division
of the Toronto
Ms. Barbara
Leavitt.
Mr. Walker
Baxter

Foundation

Baxter

Foundation,
Hospital.
of the

Toronto

Ontario.
Dr. John

We thank
Moran
of

We

thank

Ms.

also
Deirdre

appreciation
for their

Dr.

Hobeck
to

the

invaluable

work.

REFERENCES
1 . Popovich
PP. MoncriefJW,
Decherd
JB. Bomar
JB, Pyle
WK: The definition
of a novel portable/wearable
equilibrium peritoneal
dialysis
technique
[Abstract].
ASAIO (Am
Soc Artif Intern
Organs)
Trans
1976;5:64A.
2. Popovich
RP, Moncrief
JW: Kinetic
modeling
of peritoneal transport.
Contrib
Nephrol
1979; 17:59-72.
3. Nolph
KD: Whats
new in peritoneal
dialysis-An
overview. Kidney
Int 1992:42ESuppl
381:5148-5152.
4. Burton
PR, Walls
J: Selection-adjusted
comparison
of
life expectancy
of patients
on continuous
ambulatory
peritoneal
dialysis.
hemodialysis
and
transplantation.
Lancet

5.
6.

7.

8.

9.
10.
1 1.

206

1987:

1: 1 1 15-1

14.

assessment
of continuous
ambulatory
peritoneal
dialysis patients.
ASAIO
(Am Soc Artif Intern
Organs)
Trans
1987:33:650-653.
Teehan
BP, Schleifer
CR, Brown
JM, Sigler
MH, Raimondo
J: Urea kinetic
analysis
and clinical
outcome
on
CAPD.
A five year
longitudinal
study.
Adv Pent
Dial
1990;6:
18 1-185.
Blake
PG. Sombolos
K, Abraham
G, et at.: Lack
of
correlation
between
urea
kinetic
indices
and
clinical
outcomes
in CAPD
patients.
Kidney
mt 199 1:39:700706.
Keshaviah
PR, Nolph
1W, Prowant
B, et at: Defining
adequacy
of CAPD
with
urea
kinetics.
Adv Pent
Dial

DeAlvaro
F, Bajo MA. Alvarez-Ude
F, et at.: Adequacy
of
peritoneal
dialysis:
Does KT/V have the same
predictive
value
as in HD? A multicenter
study.
Adv Pent
Dial
1992:8:93-97.
16. Brandes
JC, Piering
WF. Beres
JA, Blumenthal
55,
Fritsche
C: Clinical
outcome
of continuous
peritoneal
dialysis
predicted
by urea and creatinine
kinetics.
J Am
Soc Nephrol
1992:2:1430-1435.
1 7. Arkouche
W, Debawari
E, My H, et at.: Quantification
of
adequacy
of peritoneal
dialysis.
Pent
Dial
Int
1993:
l3ISuppl
21:S215-S218.
18. Tzamaloukas
AR, Murata
GH, Sena
P: Assessing
the
adequacy
of peritoneal
dialysis.
Pent
Dial
mt
1993:
13[Suppl
21:5236-5239.
19. Churchill
DN: Adequacy
of peritoneal
dialysis.
How
much
dialysis
do we need?
Kidney
Int 1994:46[Suppl
48]:S2-S6.
20. Karnofsky
DA. Burchenal
JH: The clinical
evaluation
of
chemotherapeutic
agents
in cancer.
In: McLeod
CM, Ed.
Evaluation
of Chemotherapeutic
Agents.
New York: Columbia
University
Press:
1949.
2 1 . Baker
JP, Detsky
AS, Wesson
DE, et al.: Nutritional
assessment.
A comparison
of clinical
judgement
and
objective
measurements.
N Engl J Med 1982:306:969972.
22. Detsky
AS, McLaughlin
JR. Baker
JP, et at.: What
is
subjective
global
assessment?
J Parenter
Enteral
Nutr
1987; 11:8-13.
23. Enia G. Sicuso
C, Alati G, Zoccali
C: Subjective
global
assessment
ofnutrition
in dialysis
patients.
Nephrol
Dial
Transplant
1993;8:
1094-1098.
24. Randerson
DH, Chapman
GV, Farrell
PC: Amino
acid
and
dietary
status
in CAPD
patients.
In: Atkins
RC,
Farrell
PC. Thomson
N, Eds. Peritoneal
Dialysis.
Edinburgh:
Churchill-Livingstone:
1981:180-191.
25. Watson
PE, Watson
ID, Batt RD: Total body water
volumes for adult
males
and females
estimated
from simple
anthropometric
measurements.
Am J Clin Nutr
1980:
33:27-39.
26. Keshaviah
PR, Nolph
KD, Moore
IlL, et at.: Lean body
mass
estimation
by creatinine
kinetics.
J Am Soc Neph-

1 19.

Canadian
Institute
for Health
Information:
Canadian
Organ
Replacement
Register,
1993 Annual
Report.
Ontario:
Don Mills:
1995.
MaJorca
R, Vonesh
EF, Cavalbi
P. et at: A multicenter,
selection
adjusted
comparison
of patient
and technique
survivals
on CAPD and hemodialysis.
Pent Dial mt i
i:
11:118-127.
Held
PJ,
Port
FK, Turenne
MN, et at.: Continuous
ambulatory
peritoneal
dialysis
and hemodialysis:
Comparison
of patient
mortality
with adjustment
for comorbid conditions.
Kidney
mt 1994:45:1163-1169.
Lowrie
EG, Lew NL: Death
risk in hemodialysis
patients:
The predictive
value
of commonly
measured
variables
and
an evaluation
of death
rate
differences
between
facilities.
Am J Kidney
Dis 1990:15:458-482.
Held PJ, Levin
NW, Bovbjerg
RR, Pauly
MV, Diamond
LH: Mortality
and duration
of hemodialysis
treatment.
JAMA
199 1:265:871-875.
Gotch
F, Sargent
JA: A mechanistic
analysis
of the
National
Cooperative
Dialysis
Study
(NCDS).
Kidney
Int
1985:28:526-534.
Fenton
SSA, Johnston
N, Delmore
T, et at.: Nutritional

rol 1994:4:1475-1485.

27.
28.
29.

Farrell
SC, Bailey
MP: Measurement
of creatinine
in
peritoneal
dialysis
fluid.
Ann
Clin
Biochem
1991:28:
624-625.
Nobph KD: Update
on peritoneal
dialysis
worldwide.
Pent
Dial Int 1993: 1 3ESuppl
21:15-19.
Andersen
PK, Gill 1W: Coxs regression
model
for counting processes:
A large sample
study.
Ann Stat
1982; 10:
1100-1120.

30.
3 1.
32.
33.
34.
35.
36.

Cox D: Regression
models
and life-tables.
J R Stat Soc
1972:34:187-201.
Cox DR, Oakes
D. Analysis
of Survival
Data.
London:
Chapman
and Hall: 1984.
Kalbfleisch
JD, Prentice
RL. The Statistical
Analysis
of
Failure
Time Data.
New York: Wiley;
1980.
Cox DR, Hinkley
DV. Theoretical
Statistics.
London:
Chapman
and Hall; 1974.
Schoenfeld
D: Partial
residuals
for the proportional
hazards regression
model.
Biometnika
1982:69:239-241.
McCullagh
P. NelderJA.
Generalized
Linear Models.
2nd
Ed. London:
Chapman
and Hall; 1989.
Harrell
FE Jr. Lee KL, Matchar
DB, Reichent
TA: Regression models
for prognostic
prediction:
Advantages,
prob-

Volume

Number

1996

Churchill

lems
37.
38.
39.

40.

41

42.

and

44.
45.

46.

47.
48.

solutions.

Cancer

Treat

Rep

1985;

S-PLUS
Guide
to Statistical
and
Mathematical
Analysis.
Version
3.2. Seattle:
StatSci,
A
Division
of Mathsoft
Inc.; 1993.
Harrell
F. The PHGLM
procedure.
SAS Supplemental
Library
Users
Guide.
Version
5 Edition.
Cary,
NC: SAS
Institute,
Inc.:
1986.
Heimburger
0. Bergstrom
J, Lindholm
B: Is serum
albumin
an index
of nutritional
status
in continuous
ambulatory
peritoneal
dialysis
patients?
Pent
Dial mt
1994; 14:108-114.
Fine A, Cox D: Modest
reduction
of serum
albumin
in
continuous
peritoneal
dialysis
patients
is common
and
of no apparent
clinical
consequence.
Am J Kidney
Dis
1992;20:50-54.
Struijk
DG, Krediet
RT, Koomen
GCM, Boeschoten
EW,
Arisz
L: The effect
of serum
albumin
at the start
of
continuous
ambulatory
peritoneal
dialysis
treatment
on
patient
survival.
Pent Dial Int 1994; 14:121-126.
Concato
J, Feinstein
AR, Holford
TR: The risk of determining
risk with multivariable
models.
Ann
Intern
Med
1993:1

43.

suggested

69: 107 1-1077.

Statistical
Sciences.

18:201-2

10.

Teehan
BP, Schleifer
CR, Brown
J: Urea kinetic
modeling
is an appropriate
assessment
of adequacy.
Semin
Dialysis
1992:5:189-192.
Young
GA, Kopple
JD, Lindholm
B, et at.: Nutritional
assessment
of CAPD
patients:
An international
study.
Am J Kidney
Dis 1991;17:462-471.
Bergia
R, Bellini
ME, Vabenti
M, et al.: Longitudinal
assessment
of body composition
in continuous
ambulatory peritoneal
dialysis
using
bioelectrical
impedance
and anthropometric
measurements.
Pent Dial bnt 1993;
l3lSuppl
21:512-514.
Schmidt
R, Dumler
F, Cruz C, Lubkowski
T, Kilates
C:
Improved
nutritional
follow-up
of peritoneal
dialysis
patients
with bioelectrical
Impedance.
Adv Pent Dial 1992:
8:157-159.
Hasty JC, Boulton
H, Curwell
J, et at.: The normalized
protein
catabolic
rate is a flawed
marker
of nutrition
in
CAPD patients.
Kidney
Int 1994:45:103-109.
Frisancho
AR: New standards
of weight
and body composition
by frame
size and
height
for assessment
of

Journal

of

the

American

Society

of

Nephrology

et al

nutritional
state
of adults
and the elderly.
Am J Clin
Nutr
1984:40:808-819.
49. Teehan
BP, Schleifer
CR, Sigler
MH, Gilgore
GS: A
quantitative
approach
to the CAPD
prescription.
Pent
Dial Bull 1985:5:152-156.
50. Keshaviah
PR, Nolph
KD, Van
Stone
JC: The peak
concentration
hypothesis:
A urea
kinetic
approach
to
comparing
the adequacy
of continuous
penitoneal
dialysis (CAPD) and hemodialysis.
Pent Dial mt 1989:9:257260.
5 1 . Keshaviah
PR: Adequacy
of CAPD: A quantitative
approach.
Kidney
Int 1992;42[Suppl
38l:S 1 60-S 164.
52. Teehan
BP, Schleifer
CR, Brown
J: Urea kinetic
modeling is an appropriate
assessment
of adequacy.
Semin
Dialysis
1992:5:189-192.
53. Lameire
NH, Vanholder
R, Veyt D, Lambert
M-C, Ringoir
5: A longitudinal
five year survey
of urea kinetic
parameters
in CAPD patients.
Kidney
Int 1992;42:426-432.
54. Twardowski
Z, Nolph
K: Penitoneal
dialysis:
How much
is enough?
Semin
Dial 1988:1:75-76.
55. Blake
PG. Izatt
5, Sombolos
K, Oreopoulos
DG: Total
weekly
creatinine
clearance
in CAPD-How
much
do
patients
get and does it correlate
with outcomes?
lAbstract]
Pent Dial Int 1992; l2lSuppl
1 l:99A.
56. Chen
H-C, Guh J-Y, Laf Y-H, Tsai J-H: Serial changes
in
serum
f3-2-microglobulin
in CAPD.
Pent
Dial mt 1993;
13:238-239.
57. Lysaght
MJ, Pollock
CA, Moran
JE, Ibels LS, Farrell
PC:
Beta-2
microglobulin
removal
during
continuous
ambulatory
penitoneal
dialysis
(CAPD).
Pent
Dial mt 1989:9:
29-35.
58. Karlson
FA, Wibell
L, Ervin PE: Beta-2
microglobulmn
in
clinical
medicine.
Scand
J Lab
Clin
Invest
1980;
4OESuppll:293-299.
59. US Renal
Data
System.
USRDS
1994
Annual
Report.
Bethesda,
MD: The National
Institute
of Health,
National
Institute
of Diabetes
and Digestive
and Kidney
Diseases;
1994.
60. Keshaviah
P, Emerson
PF, Vonesh
EF, Brandes
JC:
Relationship
between
body
size, fill volume,
and mass
transfer
area coefficient
in penitoneal
dialysis.
J Am Soc
Nephrol
1994:4:1820-1826.

207