Advances in Peritoneal Dialysis, Vol. 28, 2012
Dixie-Ann Sawin, Rainer Himmele, Jose A. Diaz–Buxo
Although dialytic removal of phosphate significantly
contributes to the management of phosphate levels in
end-stage renal disease, many patients on peritoneal
dialysis (PD) still do not reach optimal phosphate
control. The present review discusses the impact of
PD modality—continuous ambulatory (CAPD) or
automated (APD)—on phosphate removal. Relevant
factors are the diffusive properties of the phosphate
anion and the kinetics of phosphate distribution in
various body compartments. Confounders that potentially affect comparisons of phosphate clearances
in CAPD and APD are differences in residual renal
function, membrane transport status, and prescribed
dialysis dose. The evidence reviewed here is not strong
enough to clearly determine if one modality has a clear
advantage with respect to phosphate removal. In the
absence of final proof, the data suggest that, given the
same residual renal function and dialysis dose, CAPD
might be slightly more effective than APD at peritoneal
phosphate clearance, especially in low transporters.
Key words
APD, CAPD, phosphate clearance
Introduction
Dialytic removal of phosphate is a key aspect of
managing phosphate levels in end-stage renal disease.
However, many patients (about 40%) on peritoneal
dialysis (PD) still have higher-than-normal phosphate
levels (1), suggesting that phosphate management in
those patients is not optimal. One factor that may affect
phosphate clearance is the prescribed PD modality.
From: Fresenius Medical Care North America, Waltham,
Massachusetts, U.S.A.
Phosphate Clearance in
Peritoneal Dialysis:
Automated PD Compared
with Continuous
Ambulatory PD
Unfortunately, that issue has not been thoroughly
studied and is not well understood. Nevertheless, it is
of great interest, because it may allow for better management of phosphate levels in PD patients through
more appropriate prescriptions.
In the present review, we highlight important factors that dictate phosphate clearance in PD, such as
the properties of the phosphate anion and the kinetics
of phosphate in dialysis. We also review the available literature on phosphate clearance in continuous
ambulatory (CAPD) and automated PD (APD), and
we discuss various confounders that affect phosphate
clearance in both PD modalities, such as residual renal
function, membrane status, and total dialysis dose.
Discussion
Phosphate clearance properties
It is important to understand the properties of
phosphate that govern its clearance in dialysis.
Dialytic removal of phosphate depends on diffusive
and convective transport of phosphate across the
peritoneal membrane (2,3). Phosphate clearance is
also time-dependent and affected by the kinetics,
molecular size, and compartmental distribution of
the phosphate anion.
One peculiarity of phosphate is that it seems to
exhibit elimination characteristics different from
those of other small-molecular-weight toxins (2–5).
For instance, peritoneal clearance of phosphate [molecular weight (MW) 95] can be up to 50% lower than
that of urea (MW 60) and almost 20% lower than that
of creatinine (MW 113). One factor that contributes
to this dissimilarity of phosphate from other smallMW molecules is the aqueous shell that surrounds
the phosphate anion (Figure 1). The shell tends to
Sawin et al.
1 Schematic representation of the aqueous shell surrounding the hydrated phosphate anion [adapted from Pribil et
al. 2008 (6)].
figure
increase the effective MW and molecular radius of
the phosphate ion (6). In the shell, at least one water
molecule can form weak hydrogen bonds with each
of the four oxygen molecules in the phosphate anion.
Pribil and colleagues (6) showed that approximately
13 water molecules can participate in making up the
shell closest to the phosphate ion. In their study, those
authors demonstrated that the mean distance between
adjacent water molecules was, on average, 3.9 Å. In
addition, in contrast to urea, phosphate does not readily
diffuse across cell membranes (2,4,5), possibly because
approximately 5% of the circulating phosphate is a
component of sodium, calcium, or magnesium salts.
Phosphate kinetics
The kinetics of phosphate depend on phosphate movement from the various pools within the body. To assess
phosphate movement from the various phosphate
pools, Spalding and colleagues conducted an extensive modeling study based on patient data (7). Their
findings suggest that phosphate behavior depends on
serum phosphate levels.
Under healthy conditions, phosphate kinetics were
shown to follow a 4-compartment model.The model
comprises 4 distinct phosphate pools, of which mainly
121
the first 2 pools are the most relevant for hyperphosphatemic dialysis patients. In such patients, phosphate
levels rarely decline below the physiologic range;
therefore, release of phosphate from the 3rd and 4th
pools is usually not activated. Therefore, in dialysis
patients, phosphate kinetics tend to follow a 2-compartment or a pseudo 1-compartment model (8). The
1st compartment is the inaccessible compartment (for
example, bone). The 2nd compartment is the accessible compartment or the miscible pool. The authors
propose that the accessible pool is in equilibrium with
the plasma during the intradialytic and post-dialytic
periods. Phosphorus mobilization into that compartment comes from the larger inaccessible pool at a
rate proportional to dialytic phosphate clearance.
Because the inaccessible compartment is relatively
large, its phosphorus concentration is assumed to be
constant over time and equal to the pre-dialytic plasma
phosphorus concentration, making the 2-compartment
model mathematically identical to the pseudo 1-compartment model. Thus, during dialysis, phosphate is
removed primarily from the miscible pool.
Phosphate mass removal and clearance in CAPD
and APD
Very few data comparing the removal of phosphate
across PD modalities are available. A review of
the existing literature revealed only six studies that
investigated the total mass removal of phosphate in
PD (3,9–13). In CAPD, the overall mass removal of
phosphate by PD was determined to range from 2 g
to 3 g per week (Table I).
In an early study performed in 9 patients (14),
Twardowski and colleagues demonstrated that, compared with APD (n = 3), CAPD (n = 6) provided better
clearance of small solutes, including phosphate. In
that study, the CAPD regimen consisted of four 2-L
exchanges, and the APD regimen involved exchanges
of 13 – 24 L in-center over 8 hours during the daytime.
The study was limited by its small number of patients
and the absence of appropriate controls for transport
status. Further, clearance in APD was affected by
very short cycles involving 10-minute dwells and
27-minute fill/drain times.
A later study by Sedlacek et al. (15) showed that
the average weekly phosphate clearance did not significantly differ between CAPD (56.4 ± 16 L, n = 20)
and APD (51 ± 21 L, n = 36). Notably, 18 of the patients
in that study had residual renal function (RRF). The
122
table i
Phosphate Clearance in PD
Mass phosphate removal in continuous ambulatory peritoneal dialysis (CAPD) and automated peritoneal dialysis (APD)
Phosphate clearance (mg/wk)
Reference
APD
CAPD
Messa et al. (3)
Evenepoel et al. (9)
2028±1257 (n=35; 4×2 L)
~2941 (n=35, 4×3 L)
2739±1042 (n=34)
2790±1022 (n=16)
Delmez et al. (10)
2170 (4×2 L)
Winchester et al. (11)
3250 (4×3 L)
Ketteler et al. (12)
2100–2800
Botelho et al. (13)
3024 (renal plus peritoneal), 1862 (peritoneal alone)
authors speculated that larger dialysate volumes in
APD might have been offset by the shorter dwell times.
Badve et al. (16) conducted a 2-year cross-sectional retrospective study of 129 prevalent PD patients, of whom 62 were on CAPD (52 with RRF)
and 67 were on continuous cycling PD (37 with
RRF). Their daily dialysate volumes were 8.2 ± 1.6 L
for CAPD and 14.4 ± 3.5 L for APD (p < 0.001). In
their analysis across modalities, the authors did not
find a significant difference in weekly peritoneal
phosphate clearance between CAPD and continuous
cycling PD (38.3 ± 12 L/1.73 m2 vs. 40.9 ± 10.4 L/1.73 m2,
p = 0.199).
Data from Bernardo and colleagues (17) and
Botelho et al. (13) support the study by Badve and
coworkers, in that the overall phosphate clearance did
not differ between PD modalities (p = 0.928 and 0.926
respectively). In contrast, a study by Gallar et al. (18)
showed that peritoneal phosphate clearance was higher
in APD than in CAPD. Those authors discussed how
plasma phosphate levels, daily dialysate volume, and
peritoneal membrane characteristics were factors that
affected peritoneal phosphate clearance, but the effect
of membrane status was not assessed.
Evenepoel and colleagues (9) compared phosphate
mass removal and phosphate clearance between 16
CAPD patients (14 with RRF) and 34 APD patients
(27 with RRF). Although weekly total phosphate removal was not significantly different between CAPD
and APD (2.7 g vs. 2.8 g, p = 0.4), weekly peritoneal
phosphate clearance was significantly better with
CAPD than with APD (40.2 ± 13.0 L/1.73 m2 vs.
32.7 ± 14.8 L/1.73 m2, p < 0.0001), despite the larger
dialysate volume in APD (11.7 L vs. 9 L on average).
However, the study did not adjust for the group differences in transport status and RRF.
Based on the quality of the available studies comparing phosphate clearance between APD and CAPD,
it is quite difficult to assess whether modality choice
actually affects phosphate clearance. The available
studies are limited by their small sample size, their
design, and their lack of adjustment for confounders.
What the studies bring to light, however, is that many
factors play a role in determining phosphate clearance.
It appears that when choosing a PD modality with the
aim of optimizing or improving phosphate clearance,
transport status and the presence of RRF are crucial
factors that should be considered.
Confounders of phosphate clearance in PD
Assessing phosphate removal in CAPD compared with
APD is challenging because of a substantial bias by
indication attributable to differences in RRF, transport
status, and required dialysis dose. In the presence of
RRF, adequate fluid balance is maintained, overall
systemic inflammation is lower, and renal endocrine
functions and elimination mechanisms are protected
to varying degrees (19–25). Importantly, RRF also
contributes significantly to adequate Kt/V and phosphate removal. At the start of dialysis, RRF may
account for up to 65% of total phosphate clearance,
adding up to 40 – 50 mmol of clearance daily. As RRF
declines, there is a greater need for an increased dialysis dose. That situation often leads to an intensified
PD prescription and use of APD, resulting in higher
ultrafiltration (UF), increased peritoneal Kt/V, and
consequently, better phosphate clearance. A prospective study by Granja and colleagues (1) supports that
theory. Of phosphate removed in APD patients, 11%
was because of net UF, with a significant correlation
between the UF rate and phosphate removal (r = 0.81).
However, Botelho and colleagues did not find a
Sawin et al.
correlation between net UF and phosphate clearance.
They showed that UF accounted for only 5.9% of the
phosphate removed by the peritoneum (13). Interestingly, those investigators also determined that total
daily phosphate removal correlated negatively with
peritoneal Kt/V urea.
In another study, 22 of 24 patients who started on
CAPD were switched to APD with an increased dialysis
dose as RRF declined (5). The authors demonstrated
that peritoneal phosphate clearance increased and appeared to compensate for the loss of renal function during the study period. Thus, patients on APD might have
lower renal function but higher peritoneal phosphate
clearance because of three synergistic mechanisms:
the increase in dialysis dose, structural changes in the
peritoneal membrane toward high transport characteristics over time, and convective transport because
of increased UF. Although the follow-up period in the
study was short (median: 7.2 months), the investigators
concluded that implementation of APD could increase
clearance of water-soluble toxins such as phosphate.
However, whether the effect was independent of dose
could not be determined from the data.
To understand the effect of an increased dialysis
dose on phosphate removal in APD, Juergensen
et al. (26) studied 10 patients who were high or highaverage transporters. They showed a 19% increase
in phosphate clearance when the dialysate volume
was increased by 71% (p < 0.005). That increase in
phosphate clearance translates into less than 50 mg of
net phosphate removal in 9 hours, assuming a serum
phosphate of 6 mg/%.
In a pediatric study (n = 87), Schmitt and colleagues (27) also showed that dialytic phosphate
clearance correlated positively with total and nighttime fluid turnover, the number of cycles, total cycler
time, and net UF. In fact, PD fluid turnover was found
to be the major determinant of phosphate clearance.
The study determined that, for every liter of PD fluid
replaced per square meter, the daily peritoneal clearance of phosphate increased by 0.7 L/1.73 m2. Those
results support similar previous reports in adults,
which showed that 8 – 8.8 mmol phosphate could
be removed for every liter of fluid replaced (3,11).
Although the foregoing observations are interesting,
and clearly, loss of RRF could be counterbalanced by
APD and an increased dialysis dose, the efficiency of
increasing phosphate removal is low and comes with
the burden of significantly increased costs.
123
Transport status is another important confounder
of phosphate clearance in PD. High (“fast”) transporters tend to show better phosphate clearance than
do low (“slow”) transporters. They are also more
likely to be on APD because of their need for short
cycles. Schmitt and colleagues showed that high
transporters had better clearance of phosphate and
creatinine (27). In their study, dialysate-to-plasma
(D/P) creatinine and creatinine clearance correlated
well with D/P phosphate (r = 0.27) and phosphate
clearance (r = 0.52), although compared with D/P
creatinine, D/P phosphate was clearly a better predictor of phosphate clearance. The authors further
suggested that patients with a D/P phosphate of less
than 0.27 after 2 hours (or less than 0.41 after 4 hours)
could be considered low phosphate transporters.
Sedlacek et al. (15) concluded that the main
determinant of peritoneal phosphate clearance was
transport status, as determined by a peritoneal equilibration test (PET). Compared with slow transporters,
fast transporters had significantly better phosphate
clearance (p = 0.0003). Although phosphate clearance
in slow transporters was slightly better with CAPD,
it did not reach significance. When Badve et al. (16)
adjusted for transport status, modality choice was not
an important factor for high transporters (p = 0.72).
In contrast, high-average and low to low-average
transporters had significantly better phosphate clearance with CAPD (p = 0.01 and 0.034 respectively).
In general, compared with the other transport groups,
high transporters had better phosphate clearance.
With respect to transport status, Bernardo et al. (17)
concluded that, for fast and fast-average transporters,
there was no difference in the phosphate cleared by
APD or CAPD. For slow and slow-average transporters, however, their results indicated that CAPD might
be a more favorable option for improving phosphate
clearance (p = 0.006 for slow-average, p = 0.049 for
slow transporters).
Summary
The quality of the available studies is insufficient
to ultimately prove an independent effect of CAPD
or APD on peritoneal phosphate clearance. Several
confounders—including RRF, dialysis dose, transport
status, and UF—limit interpretation of the study outcomes. Higher transport status, dialysis dose, and UF
are associated with higher phosphate removal regardless of modality. Weighing the foregoing evidence,
124
and giving special consideration to the confounders in
each study, CAPD seems to be slightly favored over
APD with respect to peritoneal phosphate clearance,
especially in low transporters.
Disclosures
All authors are employees of Fresenius Medical Care
North America.
References
1 Granja CA, Juergensen P, Finkelstein FO. Phosphate
balance in peritoneal dialysis patients: role of ultrafiltration. Contrib Nephrol 2009;163:198–205.
2 Kuhlmann MK. Phosphate elimination in modalities
of hemodialysis and peritoneal dialysis. Blood Purif
2010;29:137–44.
3 Messa P, Gropuzzo M, Cleva M, et al. Behaviour of
phosphate removal with different dialysis schedules.
Nephrol Dial Transplant 1998;13(suppl 6):43–8.
4 Badve SV, Zimmerman DL, Knoll GA, Burns KD,
McCormick BB. Peritoneal phosphate clearance is influenced by peritoneal dialysis modality, independent
of peritoneal transport characteristics. Clin J Am Soc
Nephrol 2008;3:1711–17.
5 Bammens B, Evenepoel P, Verbeke K, Vanrenterghem
Y. Time profiles of peritoneal and renal clearances of
different uremic solutes in incident peritoneal dialysis
patients. Am J Kidney Dis 2005;46:512–19.
6 Pribil AB, Hofer TS, Randolf BR, Rode BM. Structure and dynamics of phosphate ion in aqueous solution: an ab initio QMCF MD study. J Comput Chem
2008;29:2330–4.
7 Spalding EM, Chamney PW, Farrington K. Phosphate
kinetics during hemodialysis: evidence for biphasic
regulation. Kidney Int 2002;61:655–67.
8 Agar BU, Akonur A, Lo YC, Cheung AK, Leypoldt
JK. Kinetic model of phosphorus mobilization during
and after short and conventional hemodialysis. Clin J
Am Soc Nephrol 2011;6:2854–60.
9 Evenepoel P, Bammens B, Verbeke K, Vanrenterghem
Y. Superior dialytic clearance of β2-microglobulin
and p-cresol by high-flux hemodialysis as compared
to peritoneal dialysis. Kidney Int 2006;70:794–9.
10 Delmez JA, Slatopolsky E, Martin KJ, Gearing BN,
Harter HR. Minerals, vitamin D, and parathyroid hormone in continuous ambulatory peritoneal dialysis.
Kidney Int 1982;2:862–7.
11 Winchester JF, Quijada VE, Rotellar C. Phosphate
control in peritoneal dialysis patients. In: LaGreca
G, Ronco C, Feriani M, Chiaramonte S, Conz P, eds.
Peritoneal dialysis. Milan, Italy: Wichtig Editore;
1991: 327–31.
Phosphate Clearance in PD
12 Ketteler M, Gross ML, Ritz E. Calcification and
cardiovascular problems in renal failure. Kidney Int
Suppl 2005;(94):S120–7.
13 Botelho C, Rodrigues A, Oliveira JC, Cabrita A.
Peritoneal phosphate removal varies by peritoneal dialysis regimen: an underestimated parameter of phosphate control. J Nephrol 2012;:[Epub
ahead of print].
14 Twardowski ZJ, Nolph KD, Khanna R, Gluck Z,
Prowant BF, Ryan LP. Daily clearances with continuous ambulatory peritoneal dialysis and nightly peritoneal dialysis. ASAIO Trans 1986;32:575–80.
15 Sedlacek M, Dimaano F, Uribarri J. Relationship between phosphorus and creatinine clearance in peritoneal dialysis: clinical implications. Am J Kidney Dis
2000;36:1020–4.
16 Badve SV, Zimmerman DL, Knoll GA, Burns KD,
McCormick BB. Peritoneal phosphate clearance is influenced by peritoneal dialysis modality, independent
of peritoneal transport characteristics. Clin J Am Soc
Nephrol 2008;3:1711–17.
17 Bernardo AP, Contesse SA, Bajo MA, et al. Peritoneal membrane phosphate transport status: a cornerstone in phosphate handling in peritoneal dialysis.
Clin J Am Soc Nephrol 2011;6:591–7.
18 Gallar P, Ortega O, Gutiérrez M, et al. Influencing
factors in the control of phosphorus in peritoneal
dialysis. Therapeutic options [Spanish]. Nefrologia
2000;20:355–61.
19 Chung SH, Heimbürger O, Stenvinkel P, Qureshi
AR, Lindholm B. Association between residual renal
function, inflammation and patient survival in new
peritoneal dialysis patients. Nephrol Dial Transplant
2003;18:590–7.
20 Pecoits–Filho R, Heimbürger O, Bárány P, et al. Associations between circulating inflammatory markers
and residual renal function in CRF patients. Am J
Kidney Dis 2003;41:1212–18.
21 Wang AY, Wang M, Woo J, et al. Inflammation, residual kidney function, and cardiac hypertrophy are
interrelated and combine adversely to enhance mortality and cardiovascular death risk of peritoneal dialysis patients. J Am Soc Nephrol 2004;15:2186–94.
22 Zappacosta AR, Caro J, Erslev A. Normalization of
hematocrit in patients with end-stage renal disease on
continuous ambulatory peritoneal dialysis: the role of
erythropoietin. Am J Med 1982;72:53–7.
23 Besarab A, Caro J, Jarrell BE, Francos G, Erslev AJ.
Dynamics of erythropoiesis following renal transplantation. Kidney Int 1987;32:526–36.
24 Erkan E, Moritz M, Kaskel F. Impact of residual renal
function in children on hemodialysis. Pediatr Nephrol
2001;16:858–61.
Sawin et al.
25 Lund B, Clausen E, Friedberg M, et al. Serum
1,25-dihydroxycholecalciferol in anephric, haemodialyzed and kidney-transplanted patients. Effect of
vitamin D3 supplement. Nephron 1980;25:30–3.
26 Juergensen P, Eras J, McClure B, Kliger AS, Finkelstein FO. The impact of various cycling regimens on
phosphorus removal in chronic peritoneal dialysis
patients. Int J Artif Organs 2005;28:1219–23.
27 Schmitt CP, Borzych D, Nau B, Wühl E, Zurowska A,
Schaefer F. Dialytic phosphate removal: a modifiable
125
measure of dialysis efficacy in automated peritoneal
dialysis. Perit Dial Int 2009;29:465–71.
Corresponding author:
Jose A. Diaz–Buxo, md facp, 309 East Morehead
Street, Suite 285, Charlotte, North Carolina 28202
U.S.A.
E-mail:
[email protected]