Nephrol Dial Transplant (1998) 13 [Suppl 6 ]: 158163

Nephrology
Dialysis
Transplantation

Implications of the CanadaUSA (CANUSA) study of the adequacy of

dialysis on peritoneal dialysis schedule
D. N. Churchill
Department of Medicine, Department of Clinical Epidemiology & Biostatistics, McMaster University, Hamilton, Ontario,
Canada

Introduction
The CanadaUSA (CANUSA) study of peritoneal
dialysis [1] has increased awareness of the association
between increased removal of small molecular weight
solutes and improved patient survival. The data also
suggest that patients who start continuous ambulatory
peritoneal dialysis (CAPD) with little residual renal
function are less well nourished and that a poor
nutritional status is associated with decreased patient
survival [2]. There is also a suggestion that patients
with higher peritoneal transport, as defined by the
peritoneal equilibrium test (PET ) [3], have a decreased
probability of survival [4]. These CANUSA-derived
data suggest that dialysis treatment should be initiated
earlier, that the total weekly clearance of small molecular weight solutes should be considerably higher than
is current practice and that higher peritoneal transport
is not a desirable feature for CAPD treatment. These
suggestions are controversial [5,6 ]. There are concerns,
in an incident dialysis population, about the relative
value of clearance due to residual renal function and
that due to peritoneal dialysis. Others are concerned
that earlier initiation of dialysis may not be accepted
by funding agencies and that the increased economic
burden is not justifiable. Higher adequacy targets are
not considered feasible for heavier patients with little
residual function, and the increased cost of higher
dialysate volumes may discourage some providers.
The objective of this review is to present the conclusions reached in the peer-reviewed publications from
the CANUSA peritoneal dialysis study group [1,2,4]
and to consider the controversy regarding the clinical
inferences derived from those conclusions.

Methods

The conclusions, as stated in the publications, and the clinical

inferences or predictions from these conclusions were listed.
The arguments for and against these inferences were
reviewed. Based on this evidence, a summary of the clinical
inferences was produced.

Adequacy of dialysis and patient survival

The first peer-reviewed publication ( Table 1) addressed
the association of adequacy and nutritional status with
clinical outcomes [1].
Methodology
The CANUSA study group included 14 centres in
North America; 10 in Canada and four in the US.
There were 680 new CAPD or continuous cyclic peritoneal dialysis (CCPD) patients enrolled between
September 1, 1990 and December 31, 1992 with followup recorded until December 31, 1993.
Adequacy of dialysis was estimated by measurement
of total weekly Kt/V for urea and total weekly creatinine clearance (CCr). Daily peritoneal Kt was estimated
from a 24 h dialysate urea excretion and the serum
urea concentration at the conclusion of the collection
period. Renal Kt was estimated from the concurrent
24 h urine urea excretion. Total body water was estimated from the formula of Watson et al. [7]. Peritoneal
CCr was estimated from a 24 h dialysate excretion and
the serum creatinine concentration at the end of the
collection period. Dialysate creatinine concentration
was corrected for glucose interference using a validated technique-specific correction factor. The renal
contribution was an estimate of glomerular filtration
Table 1. CANUSA study publications

The methodology used in each of the three publications

(Table 1) was reviewed and the results were summarized.

Subject

Publications

Correspondence and offprint requests to: David N. Churchill, MD,

Department of Medicine, Department of Clinical Epidemiology &
Biostatistics, Faculty of Health Sciences, McMaster University,
Hamilton, Ontario, Canada.

1. Adequacy of dialysis
2. Initiation of dialysis
3. Membrane transport

J Am Soc Nephrol 1996 [1]

Kidney Int 1996 [2]
Peritoneal Dial Int 1996 [4]

1998 European Renal AssociationEuropean Dialysis and Transplant Association

Implications of the CANUSA study on peritoneal dialysis schedule

rate (GFR) defined as the average of urea and creatinine clearance. Daily clearances were converted to
weekly by multiplying the daily values by seven. These
estimates of adequacy of dialysis were determined
when CAPD or CCPD training was completed and at
6 monthly intervals thereafter.
Estimates of nutritional status were obtained at the
same intervals as for adequacy. These included the
serum albumin concentration, the subjective global
assessment (SGA) [8], protein catabolic rate (PCR)
according to Randerson et al. [9] normalized to standard body weight (0.58/V ) and percentage lean body
mass (%LBM ) from creatinine kinetics [10]. The statistical analyses are described in detail elsewhere [1].
Results
The relative risk (RR) of death was increased with
increased age, dialysis in the USA, insulin-dependent
diabetes mellitus, history of cardiovascular disease,
lower serum albumin concentration, and worse nutritional status according to SGA and %LBM. The RR
of death was decreased by 6% for a 0.1 greater weekly
Kt/V and by 7% for a 5 l/1.73 m2 greater weekly CCr.
Conclusions
Controlling for other independent variables, there is
an association of greater weekly Kt/V and greater
weekly CCr with a decreased RR of death. There was
also an association of better nutritional status with a
decreased RR of death.
Prediction
Data from the multivariate statistical analysis was used
to create a mathematical model to predict survival at
different levels of weekly Kt/V and weekly CCr. This
model assumes that peritoneal and renal clearances are
equivalent, and predicts the probability of survival, in
this population, if total weekly clearances could be
maintained at stable levels by increasing peritoneal
clearance as renal clearance is lost.
The predicted 2 year survival probabilities are shown
in Table 2. Over the range of adequacy studied, there
was a progressive increase in the probability of survival
with increased Kt/V and CCr.
Clinical inference
The theoretical constructs addressing adequacy of dialysis for CAPD [1113] were reviewed recently [14].

159

These suggest that a weekly Kt/V of 2.02.2 would

provide adequate dialysis. For a weekly Kt/V of 2.1,
the predicted 2 year survival probability was 78% [1].
There are no theoretical constructs for CCr in CAPD.
However, a 2 year survival probability of 78% was
predicted by a CCr of 70 l/1.73 m2. These were suggested as reasonable targets for CAPD [1]. The data
also predict increased survival probability with
increasing Kt/V and CCr.
Controversy
The target Kt/V selected as being reasonable is consistent with theoretical constructs [1113]. Gotch has
proposed [15] that the reference Kt/V for normal renal
function be 12.96 weekly while the normal CCr is
~1000 l weekly. Expressed as a percentage of normal,
a weekly Kt/V of 2.1 is 16% of normal and a weekly
CCr of 70 l is only 7% of normal. It would appear
logical to consider these to be reasonable targets but,
when possible, to achieve greater Kt/V and CCr.
On the other hand, increased peritoneal dialysis dose
for CAPD extracts a quality of life price in the time
taken to perform exchanges and discomfort from
larger dialysis volumes. Attainment of these targets
with CCPD is associated with increased cost. Gotch
argues, from the haemodialysis literature, that the RR
of death does not decrease further with single-pool
Kt/V >1.2 per dialysis for patients with thrice-weekly
dialysis [15,16 ]. The equivalent Kt/V for CAPD is 2.0
per week. The choice of a CCr target of 70 l weekly in
the CANUSA study reflects the greater relative contribution of residual renal function to CCr than to Kt/V.
For patients with little residual renal function, Nolph
et al. have reported equivalence between a Kt/V of 2.0
and a CCr of 60 l weekly [17]. Despite the convention
of considering renal and peritoneal clearance equivalent, there are few data to support this concept.
Summary
A weekly Kt/V of at least 2.0 and/or a weekly CCr of
at least 60 l are reasonable adequacy targets. Clearance
from residual renal function and peritoneal clearance
should be considered equivalent until there is evidence
to reject this assumption. If higher clearances can be
achieved without impairing patient quality of life or
significantly increasing cost, this is desirable. Research
addressing the equivalence of residual renal clearance
with peritoneal clearance and clinical outcome research
addressing the effect of higher adequacy targets should
be given high priority.

Table 2. Predicted 2 year survival probabilities according to weekly

Kt/V and CCr ( l/1.73 m2)

Renal function and nutrition at initiation of dialysis

Kt/V

Survival %

CCr

Survival %

2.3
2.1
1.9
1.7
1.5

81
78
74
71
66

95
80
70
55
40

86
81
78
72
65

The second peer-reviewed publication from the

CANUSA study group ( Table 1) addressed the relationship between residual renal function and nutritional status at baseline [2]. It further evaluated the
association between baseline nutritional status and
patient survival.

160

D. N. Churchill
Table 5. Two year survival probabilities according to nutritional
status at initiation of dialysis

Methods
The residual renal function at initiation of CCPD was
determined for the 680 patients enrolled in the
CANUSA study [1] and expressed as renal Kt/V and
as an estimate of GFR determined from the mean of
renal urea and creatinine clearance. For each of renal
Kt/V and GFR, three cohorts were created representing the upper, middle and lower thirds of residual
renal function. For each tertile of residual renal
function, nutritional status was expressed as the mean
of serum albumin concentration, SGA, %LBM and
nPCR. For several levels of each of these baseline
nutritional estimates, the association with the actual
survival probability was determined using the
KaplanMeier method [18] and compared by the
log-rank test.

Albumin
g/l
<30
3035
>35
Log-rank

LBM
%
64%
75%
85%
P<0.001

<63
6373
>73
<0.001

nPCR
g/kg
65%
81%
88%
<0.001

<0.09
0.91.02
>1.02

71%
80%
83%

values >1.02 g/kg, 0.91.02 g/kg and <0.9 g/kg, the

2 year survival probabilities were 83, 80 and 71%
respectively (P<0.001; log-rank test). These data are
summarized in Table 5.
Conclusions

Results
For weekly renal Kt/V, the mean value at initiation of
continuous peritoneal dialysis was 0.71 [1]; one-third
had values >0.89, one-third were 0.440.89 and onethird were <0.44. For weekly renal GFR, the mean
value at initiation was 39 l [1]; one-third had values
>50, one-third were 2550 and one-third were
<25 l/1.73 m2. The data for Kt/V and GFR defined
tertiles are shown in Tables 3 and 4 respectively. For
each of the estimates of nutritional status, higher initial
residual renal function is associated with better nutritional status. For patients with initial serum albumin
concentrations >35 g/l, 3035 g/l and <30 g/l, the 2
year survival probabilities were 85, 75 and 64% respectively (P<0.001; log-rank test). For those with initial
%LBM >73%, 6373% and <63%, the 2 year survival
probabilities were 88, 81 and 65% respectively
(P<0.001; log-rank test). For those with initial nPCR
Table 3. Mean values of nutritional status at initiation of dialysis
according to residual renal function estimated by weekly Kt/V
tertiles (n=680)

Kt/V
>0.89
0.440.89
<0.44
ANOVA P

Albumin
g/l

SGA
units

LBM
%

nPCR
g/kg

35.4
35.0
34.2
<0.05

5.4
5.1
5.1
<0.05

71.6
68.0
64.8
<0.001

1.12
1.03
0.98
<0.001

Table 4. Mean values of nutritional status at initiation of dialysis

according to residual renal function estimated by weekly
GFR/1.73 m2 tertiles (n=680)
GFR

Albumin
g/l

SGA
units

LBM
%

nPCR
g/kg

>50
2550
<25
ANOVA P

35.8
34.8
34.1
<0.01

5.4
5.2
5.0
<0.05

73.8
68.1
58.0
<0.001

1.10
1.04
0.99
<0.001

Lower residual renal function at the initiation of

dialysis was associated with a worse nutritional status.
In turn, worse initial nutritional status was associated
with a decreased probability of survival.
Inference
Earlier initiation of dialysis would be associated with
a better initial nutritional state and this would result
in improved patient survival. If the adequacy target
were a weekly total Kt/V of 2.0 or a weekly GFR of
60 l (6 ml/min GFR or ~12 ml/min endogenous
creatinine clearance), it would be illogical to allow the
residual renal Kt/V or CCr to fall significantly below
these values before initiating dialysis.
Controversy
Evidence supporting an earlier initiation of dialysis has
been provided by Ikizler et al. [19] who reported a
decrease in protein intake when CCr fell below
50 ml/min or 500 l/week. For non-dialysed patients
with a CCr <10 ml/min, the dietary protein intake
was 0.54 g/kg body weight. Poor nutritional status is
associated with worse outcome in both haemodialysis
and peritoneal dialysis patients [20,21]. Bonomini has
demonstrated superior survival among patients in
whom haemodialysis was initiated early, and attributed
the improvement to better nutritional status [22,23].
They reported an 88% 10 year survival among
patients commencing haemodialysis therapy with CCr
>10 ml/min compared with 55% for a group starting
with a mean CCr of 4 ml/min [23]. Recently, Tattersall
et al. reported increased morbidity among patients
commencing dialysis with a weekly renal Kt/V <1.05
[24].
Arguments against an earlier start include the possibility of later initiation being a confounder which is
linked to decreased access to medical care, poor patient
compliance or suboptimal medical care. There may
also be a starting time bias which gives an apparent
rather than an actual survival advantage to those
commencing dialysis earlier. There are no controlled

Implications of the CANUSA study on peritoneal dialysis schedule

161

clinical trials demonstrating that earlier initiation does

more good than harm, with peritoneal dialysis complications and increased glucose loads being two potentially harmful effects. In some countries, funding
agencies may not reimburse for early initiation and
thus there will be a significant increase in cost.

transporters. This is clinically important and statistically significant. There is a trend to an increased RR of
technique failure among high transporters. Possible
mechanisms include fluid overload from ultrafiltration
failure, malnutrition from loss of proteins in the dialysate and adverse effects of increased glucose absorption.

Summary

Inference

A reasonable solution would be to offer peritoneal

dialysis to patients when the weekly renal Kt/V is
<2.0 or the weekly estimate of GFR is <60 l (CCr
<12 ml/min) if there is concurrent evidence of malnutrition. This could be defined as a serum albumin
<35 g/l or an nPCR <0.9 g/kg. A study comparing
an earlier initiation compared with conventional
initiation should be considered high priority for a
randomized clinical trial.

Patients with higher than average transport should be

observed for evidence of fluid overload and malnutrition. If present, the patient should be treated with
cycling peritoneal dialysis or with haemodialysis. The
former is preferable for those with high average
transport in that shorter dwell times will decrease
glucose absorption and improve ultrafiltration. Those
with high transport may have better outcomes on
haemodialysis.
Evidence

Peritoneal membrane transport and patient survival

The CANUSA study group have also addressed the
association of peritoneal membrane transport with
clinical outcomes and have reported these data in an
abstract ( Table 1) [4].
Methods
Patients enrolled in the CANUSA study of the
adequacy of peritoneal dialysis [1] underwent a PET
[3] at the completion of dialysis training and every 6
months thereafter. Patients with a 4 h dialysate:plasma
(D/P) creatinine >0.65 were considered high peritoneal transporters, while those with values <0.65 were
considered low transporters. This variable was added
to the Cox proportional hazards model. The probability of technique and patient survival according to
peritoneal transport status was determined using the
KaplanMeier method [18]. The RR of technique and
patient survival according to transport status was
evaluated with the Cox proportional hazards method.
Results
The probability of 2 year technique survival was 79%
among those with low transport compared with 71%
among those with high transport. The RR of technique
failure was 1.44 among those with high compared with
those with low transport (95% confidence interval;
0.942.20).
The probability of 2 year patient survival was 82%
among the low transporters and 72% among the high
transporters (P=0.04; log rank test). The RR of death
was 2.18 for high as compared with low transporters
(95% confidence interval; 1.33.6). The RR associated
with previously reported variables was unchanged [1].
Conclusions
Patients with higher than average peritoneal transport
have an RR of 2.18 for death compared with low

Other investigators have suggested worse outcomes

among patients with high transport. Nolph et al. [25]
reported that high transporters (as defined by
Twardowski et al. [3]) had lower drain volumes,
decreased nPCR, greater dialysate protein losses and
lower serum albumin values than did those with low,
low average and high average peritoneal transport.
Heaf has demonstrated a detrimental association of
high peritoneal transport with clinical morbidity [26 ].
This includes increased fatigue, nausea, pain, oedema,
a symptom index and hospitalization. Recently, Davies
et al. reported that a low Kt/V and increased peritoneal
membrane transport were both associated with an
increased risk of death among CAPD patients [27].
On the other hand, the biological rationale for the
increased morbidity and mortality associated with
higher peritoneal membrane transport is speculative.
In addition, the finding of worse outcomes is counterintuitive for those with high average transport in that
these patients are predicted to have better outcomes
by virtue of increased clearance of urea and creatinine.
Summary
Patients with high and high average peritoneal membrane transport should be monitored for evidence of
fluid overload and malnutrition, the latter as previously
defined. Transfer to cycling peritoneal dialysis should
be considered. This should decrease glucose absorption
and be associated with improved ultrafiltration. The
decreased glucose uptake should also increase appetite
and improve nutritional status. However, if fluid
overload or malnutrition fail to improve, transfer to
haemodialysis should be considered. Development of
alternative osmotic agents should be a research
priority.

Discussion
The three CANUSA studies reviewed [1,2,4] have
addressed adequacy of dialysis [1], the association

162

between residual renal function at initiation of dialysis

and nutritional status [2], the association between
nutritional status at initiation of dialysis and patient
survival [2] and the association between peritoneal
membrane transport and both technique and patient
survival [4].
These data suggest that increased clearance of urea
and creatinine are associated with a decreased RR of
death over the range of adequacy studied ( Kt/V
1.52.3 per week; GFR 4095 l/1.73 m2 per week) [1].
They also suggest that earlier initiation of dialysis
should be associated with better initial nutritional
status and with improved patient survival [2]. The
data also raise concerns regarding worse outcomes for
patients with high and high average transport treated
with CAPD, and suggest that cycling peritoneal dialysis
with shorter dwell times might be preferable for these
patients [4].
The suggestion, from the CANUSA study [1], that
there is an increased survival advantage with urea
clearance over the Kt/V range of 1.52.3 is consistent
with the suggestion that longer and/or more frequent
haemodialysis [2830] might be associated with better
clinical outcomes than that associated with conventional thrice-weekly haemodialysis. Protein intake,
estimated from nPCR, has been suggested as an
outcome for estimation of adequacy of dialysis [31].
As the Kt/V is increased, both in haemodialysis and
peritoneal dialysis, the nPCR increases to an asymptote
or plateau. For CAPD, this asymptote is reached at a
weekly Kt/V of 3.0; for conventional thrice-weekly
haemodialysis, the asymptote is at a weekly Kt/V of
6.0 [31,32]. Optimum dialysis is defined as the dose of
dialysis (added to residual renal function) above which
the incremental clinical gains are not worth the additional patient time or the additional cost. For CAPD,
this dose probably lies between the current recommendation of a weekly Kt/V of 2.0 and the level of
3.0 suggested by the asymptote for nPCR.
The suggestion that dialysis be initiated earlier is
not original. Bonomini and colleagues recommended
this strategy in 1978 [22] but, despite demonstrating
superior survival with an early initiation of dialysis,
there has been little support for this approach. The
CANUSA study data [2] have contributed to the
renewed interest in earlier initiation of dialysis.
The CANUSA study has identified higher than
average transport as associated with an increased RR
of death [4]. This extends the observation of worse
nutrition among high transporters made by Nolph and
colleagues [25] and the increased morbidity observed
by Heaf [26 ]. Recently, Davies and colleagues have
reported increased mortality with increased peritoneal
membrane transport [27]. This should stimulate clinical research to define the best treatment schedule
based on transport status and to consider alternative
osmotic agents.
However, the CANUSA study [1,2,4] used an observational, albeit prospective, rather than a randomized
interventional research design. The data can be used
to generate hypotheses which can be tested with a

D. N. Churchill

randomized clinical study design. Studies with patient

survival as the clinical outcome will take several years
to complete. Pending the results of these hypothesistesting studies, it would be reasonable to implement
the modified clinical strategies suggested by the
CANUSA study results.
These strategies are: (i) initiation of dialysis when
the CCr is <12 ml/min if there is any evidence
for malnutrition (e.g. serum albumin concentration
<30 g/l or nPCR <0.9 g/kg body weight); (ii) during
maintenance peritoneal dialysis, the total (renal and
peritoneal ) Kt/V should be >2.0 and the weekly
CCr (renal GFR and peritoneal CCr) should be
>60 l/1.73 m2; (iii) patients with high average and
high transport according to PET should be treated
with cycling peritoneal dialysis using shorter dwell
times if there is evidence for fluid overload or
malnutrition.
Acknowledgements. I would like to acknowledge the members of the
CANUSA study group for making this review possible. I would also
like to thank Ms D. Hobeck for secretarial assistance.

References
1. Churchill DN, Keshaviah PR, Taylor DW et al. for the
CANUSA Peritoneal Dialysis Study Group. Adequacy of dialysis and nutrition in continuous peritoneal dialysis; association
with clinical outcomes. J Am Soc Nephrol 1996; 7: 198207
2. McCusker FX, Teehan BP, Thorpe KE, Keshaviah PR,
Churchill DN for the CANUSA Peritoneal Dialysis Study
Group. Kidney Int 1996; 50 [Suppl 56 ]: S56S61
3. Twardowski ZJ, Nolph KD, Khanna R et al. Peritoneal
equilibrium test. Peritoneal Dial Bull 1987; 7: 138147
4. Churchill DN, Thorpe KE, Nolph KD et al. for the CANUSA
Peritoneal Dialysis Study Group. CAPD patient and technique
survivals are worse with increased membrane permeability.
Peritoneal Dial Int 1996; 16 [Suppl 2]: S21 (abstract)
5. Blake PG. A critique of the CanadaUSA (CANUSA) peritoneal
dialysis study. Peritoneal Dial Int 16: 243245
6. Jindal KK. Commentary on the findings of the CANUSA study.
Peritoneal Dial Int 1996; 16: 246247
7. Watson PE, Watson ID, Batt RD. Total body water estimates
for adult males and females estimated from simple anthropometric measurements. Am J Clin Nutr 1980; 33: 2739
8. Detsky AS, Mclaughlin JR, Baker JP et al. What is subjective
global assessment? J Parenter Enteral Nutr 1987; 11: 813
9. Randerson DH, Chapman GV, Farrell PC. Amino acid and
dietary status in CAPD patients. In: Atkins RC, Farrell PC,
Thompson N, eds. Peritoneal Dialysis. Churchill-Livingstone,
Edinburgh: 1981: 180191
10. Keshaviah PR, Nolph KD, Moore HL et al. Lean body mass
estimation by creatinine kinetics. J Am Soc Nephrol 1994; 4:
14751485
11. Popovich RP, Moncrief JW. Kinetic modeling of peritoneal
transport. Contrib Nephrol 1979; 17: 5972
12. Teehan BP, Scheifler CR, Sigler MH, Gilgore GS. A quantitative
approach to the CAPD prescription. Peritoneal Dial Bull 1985;
5: 152156
13. Keshaviah PR. Adequacy of CAPD: a quantitative approach.
Kidney Int 1992; 42 [Suppl 38]: S160S164
14. Churchill DN. Adequacy of peritoneal dialysis; how much do
we need? Kidney Int 1994; 46 [Suppl 48]: S2S6
15. Gotch F. The current place of urea kinetic modeling with respect
to different dialysis modalities. Nephrol Dial Transplant 13
[Suppl 6 ]: 1014
16. Gotch F, Gentile D, Keen M et al. The incident patient cohort
study design with uncontrolled dose: substantial overestimation

Implications of the CANUSA study on peritoneal dialysis schedule

17.
18.
19.
20.
21.
22.
23.

24.

of mortality as a function of peritoneal dialysis dose. Trans Am

Soc Artif Intern Org 1996; 42: M514M518
Nolph KD, Twardowski ZB, Keshaviah PR. Weekly clearances
of urea and creatinine on CAPD and NIPD. Peritoneal Dial Int
1992; 12: 298302
Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Statist Soc 1958; 53: 457481
Ikizler TA, Greene JH, Wingard RL, Parker RA, Hakim RM.
Spontaneous dietary protein intake during progression of
chronic renal failure. J Am Soc Nephrol 1995; 6: 13861391
Kopple JD. Effect of nutrition on morbidity and mortality in
maintenance dialysis patients. Am J Kidney Dis 1994; 24:
10021009
Bergstrom J. Nutrition and mortality in hemodialysis. J Am Soc
Nephrol 1995; 6: 13291341
Bonamini V, Vangelista A, Stefoni S. Early dialysis in renal
substitutive programs. Kidney Int 1978; 13 [Suppl 8]: 112116
Bonamini V, Baldreti I, Feletti C, Stefoni S, Vangelista A. Longterm early dialysis. In Giordano C, Friedman EA, eds. Uremia
Pathobiology of Patients Treated for 10 Years or More. Wichtig,
Milan: 1981: 133137
Tattersall JE, Greenwood RN, Farrington K. Urea kinetics and
when to commence dialysis. Am J Nephrol 1995; 15: 283289

163
25. Nolph KD, Moore HL, Prowant B et al. Continuous ambulatory
peritoneal dialysis with a high flux membrane. J Am Soc Artif
Intern Org 1993; 39: 904909
26. Heaf J. CAPD adequacy and dialysis morbidity: detrimental
effect of a high peritoneal equilibrium rate. Renal Failure 1995;
17: 575587
27. Davies SJ, Philips L, Russell GI. Peritoneal solute transfer is an
independent predictor of survival on CAPD. J Am Soc Nephrol
1996; 9: 1443 (abstract)
28. Charra B, Calemard E, Ruffet M et al. Survival as an index of
adequacy of dialysis. Kidney Int 1992; 41: 12861291
29. Buoncristiani U, Quintaliani G, Cozzari, Giombini L, Ragaiolo
M. Daily dialysis: long-term clinical metabolic results. Kidney
Int 1988; 33 [Suppl 24]: S137S140
30. Uldall R, Francoeur R, Ouwendyk M et al. Simplified nocturnal
home hemodialysis (SNHHD): a new approach to renal replacement therapy. J Am Soc Nephrol 1994; 5: 428 (abstract)
31. Nolph KD, Keshaviah PR, Emerson P et al. A new approach
to optimizing urea clearances in hemodialysis and CAPD. J Am
Soc Artif Intern Org 1995; 41: 446451
32. Kenley RS. Tearing down the barriers to daily home hemodialysis and achieving the highest value renal therapy through
holistic product design. Adv Renal Replace Ther 1996; 3: 137146