PERITONEAL DIALYSIS Glucotoxicity in Peritoneal Dialysis—Solutions for the Solution! Clifford J. Holmes Glucose has served well as the prototypical osmotic agent in peritoneal dialysis for more than 2 decades, because it affords many of the characteristics required of a safe and effective osmotic agent. The disadvantages of glucose include its rapid dissipation from the peritoneum and its resulting limited UF efficiency capacity in high and high-average transporters, the associated metabolic response to absorbed glucose in all patients, and the local effects of glucose, glucose degradation products, and hyperosmolality on peritoneal membrane structure and function. This paper briefly reviews the salient elements of glucotoxicity associated with conventional glucose-based peritoneal dialysis (PD) solution use, and then discusses emerging clinical benefits of newer nonglucose PD solutions. Potential future strategies designed to abrogate glucose-associated toxicity are then reviewed. These approaches include bimodal long-dwell solutions, nonglucose crystalloid osmotic agent mixtures, and administration of pharmacologically active agents. © 2007 by the National Kidney Foundation, Inc. Index Words: Glucose; solution; peritoneal dialysis; osmotic agents; additives T he requirements for an effective osmotic agent for peritoneal dialysis have been described by Martis et al 1 and include a pre- dictable ultrafiltration profile, a high osmotic driving force at relatively low concentrations, and a large ultrafiltration volume per unit mass absorbed. In addition, it should be easily metabolized and have an absence of local and systemic toxicity. Few molecules will ever com- ply with all of these requirements, especially in terms of metabolic rate, potential for accumula- tion and toxicity of both parent molecules and their metabolites. Thus, the selection of glucose as the original prototypic osmotic agent for peri- toneal dialysis can be appreciated; it offers good osmotic driving force and rapid ultrafiltration, especially in the first few hours of the intraperi- toneal dwell. Glucose—The Prototypical Osmotic Agent for Peritoneal Dialysis For nearly 3 decades, the success of peritoneal dialysis as a viable choice of renal replacement therapy has relied upon the success of glucose as an osmotic agent that can provide adequate ultrafiltration with a clinically acceptable safety profile. The metabolic fate of glucose is well understood, with no acute toxicity or serious long-term adverse events. It can be heat sterilized, thereby providing a large de- gree of sterility assurance, and is stable for long periods of time in dialysis-solution for- mulations at ambient temperature ranges. Glucose provides a source of energy and, when used in conjunction with amino acids, can promote protein synthesis. 2 Despite the described attractive attributes of glucose for peritoneal dialysis, in recent years, awareness has developed of the poten- tial clinical disadvantages of excessive glucose exposure in dialysis patients. 3 Peritoneal-dial- ysis (PD) patients absorb from 100 to 300 g of glucose per day, or about 30 to 100 kg per year, depending upon their prescription and transporter type. 3,4 Likewise, the peritoneal membrane is continuously exposed to supra- physiologic concentrations of glucose for sev- eral years. Consequently, concerns over exces- sive glucose exposure in PD encompass both local and systemic effects. Putative Pathogenetic Mechanisms of Excessive Glucose Exposure in PD Patients Systemic Effects The potential systemic metabolic effects of excessive intraperitoneal glucose absorption From the Renal Division, Baxter Healthcare, McGaw Park, IL. Address correspondance to C.J. Holmes, PhD, Baxter Healthcare, 1620 Waukegan Road, William Graham Building, Mail code MPGR-A2N, McGaw Park, IL 60085. E-mail: [email protected] © 2007 by the National Kidney Foundation, Inc. 1548-5595/07/1403-0011$32.00/0 doi:10.1053/j.ackd.2007.03.009 Advances in Chronic Kidney Disease, Vol 14, No 3 (July), 2007: pp 269-278 269
Glucotoxicity in Peritoneal Dialysis—Solutions for the Solution!
Feb 24, 2023
Glucose has served well as the prototypical osmotic agent in peritoneal dialysis for more than 2
decades, because it affords many of the characteristics required of a safe and effective osmotic agent.
The disadvantages of glucose include its rapid dissipation from the peritoneum and its resulting
limited UF efficiency capacity in high and high-average transporters, the associated metabolic
response to absorbed glucose in all patients, and the local effects of glucose, glucose degradation
products, and hyperosmolality on peritoneal membrane structure and function.
Welcome message from author
This paper briefly reviews the salient elements of glucotoxicity associated with conventional glucose-based peritoneal dialysis (PD) solution use, and then discusses emerging clinical benefits of newer nonglucose PD solutions. Potential future strategies designed to abrogate glucose-associated toxicity are then reviewed. These approaches include bimodal long-dwell solutions, nonglucose crystalloid osmotic agent mixtures, and administration of pharmacologically active agents.
Transcript
doi:10.1053/j.ackd.2007.03.009G D C
T d d d a m m s p t t t a t o e t
G A
F d t a u s w s h g l m
ERITONEAL DIALYSIS
lucotoxicity in Peritoneal ialysis—Solutions for the Solution!
lifford J. Holmes Glucose has served well as the prototypical osmotic agent in peritoneal dialysis for more than 2
decades, because it affords many of the characteristics required of a safe and effective osmotic agent.
The disadvantages of glucose include its rapid dissipation from the peritoneum and its resulting
limited UF efficiency capacity in high and high-average transporters, the associated metabolic
response to absorbed glucose in all patients, and the local effects of glucose, glucose degradation
products, and hyperosmolality on peritoneal membrane structure and function. This paper briefly
reviews the salient elements of glucotoxicity associated with conventional glucose-based peritoneal
dialysis (PD) solution use, and then discusses emerging clinical benefits of newer nonglucose PD
solutions. Potential future strategies designed to abrogate glucose-associated toxicity are then
reviewed. These approaches include bimodal long-dwell solutions, nonglucose crystalloid osmotic
agent mixtures, and administration of pharmacologically active agents.
© 2007 by the National Kidney Foundation, Inc.
Index Words: Glucose; solution; peritoneal dialysis; osmotic agents; additives
G w c
o y t e y g y t m p e s l
P E P
H M c
he requirements for an effective osmotic agent for peritoneal dialysis have been
escribed by Martis et al1 and include a pre- ictable ultrafiltration profile, a high osmotic riving force at relatively low concentrations, nd a large ultrafiltration volume per unit ass absorbed. In addition, it should be easily etabolized and have an absence of local and
ystemic toxicity. Few molecules will ever com- ly with all of these requirements, especially in
erms of metabolic rate, potential for accumula- ion and toxicity of both parent molecules and heir metabolites. Thus, the selection of glucose s the original prototypic osmotic agent for peri- oneal dialysis can be appreciated; it offers good smotic driving force and rapid ultrafiltration, specially in the first few hours of the intraperi- oneal dwell.
lucose—The Prototypical Osmotic gent for Peritoneal Dialysis
or nearly 3 decades, the success of peritoneal ialysis as a viable choice of renal replacement
herapy has relied upon the success of glucose s an osmotic agent that can provide adequate ltrafiltration with a clinically acceptable afety profile. The metabolic fate of glucose is ell understood, with no acute toxicity or
erious long-term adverse events. It can be eat sterilized, thereby providing a large de- ree of sterility assurance, and is stable for
ong periods of time in dialysis-solution for-
ulations at ambient temperature ranges.
Advances in Chronic Kidney Disease, Vol
lucose provides a source of energy and, hen used in conjunction with amino acids,
an promote protein synthesis.2
Despite the described attractive attributes f glucose for peritoneal dialysis, in recent ears, awareness has developed of the poten- ial clinical disadvantages of excessive glucose xposure in dialysis patients.3 Peritoneal-dial- sis (PD) patients absorb from 100 to 300 g of lucose per day, or about 30 to 100 kg per ear, depending upon their prescription and ransporter type.3,4 Likewise, the peritoneal
embrane is continuously exposed to supra- hysiologic concentrations of glucose for sev- ral years. Consequently, concerns over exces- ive glucose exposure in PD encompass both ocal and systemic effects.
utative Pathogenetic Mechanisms of xcessive Glucose Exposure in PD atients
ystemic Effects
From the Renal Division, Baxter Healthcare, McGaw Park, IL. Address correspondance to C.J. Holmes, PhD, Baxter
ealthcare, 1620 Waukegan Road, William Graham Building, ail code MPGR-A2N, McGaw Park, IL 60085. E-mail:
[email protected] © 2007 by the National Kidney Foundation, Inc. 1548-5595/07/1403-0011$32.00/0
doi:10.1053/j.ackd.2007.03.009
a t u
P p d e e T t g S s t
W i i h t b w t a l o a i 1 t b a h d
M s d D s f a D m c t m m i fl
D d t a h t a d p t p i m t t p c V t
270 Clifford J. Holmes
re described in Figure 1 and can be attributed o a combination of the effects of carbohydrate ptake, caloric uptake, and hyperglycemia.
Carbohydrate Load and Dyslipidemia
D patients are dyslipidemic, and their lipid rofile is more atherogenic than that of hemo- ialysis (HD) patients.5 The greatest differ- nce between the modalities is seen in the levated levels of apo B and LDL cholesterol. he cause for the overproduction of LDL par-
icles remains unknown, although excessive lucose absorption has been implicated. tronger evidence suggests that glucose ab- orption plays an important role in the hyper- riglyceridemia observed in PD patients.5
Caloric Uptake and Increases in Fat Mass
eight gain is frequently observed in patients nitiating PD because of the extra caloric daily ntake from the PD solution.6 Nordfors et al7
ave demonstrated a genetic determinant to his weight gain in PD patients, characterized y an increase in truncal fat mass associated ith a polymorphism in the uncoupling pro-
ein 2 gene. Weight gain in PD patients can lso be a distressing event, affecting quality of ife. The role that high BMI plays in PD patient utcomes, however, remains controversial. In retrospective analysis of Medicare patients
n the United States initiating dialysis between 995 and 2000, Synder and coworkers8 found hat overweight and obese PD patients had a etter survival that those with lower BMI, lthough those PD patients with a high BMI ad an increasing risk of death with time on
ialysis. In 2004, data from the USRDS and y
ortality Wave II registries reported that obe- ity in PD patients, in contrast to HD patients, id not confer any protection.9 In fact, Mc- onald10 reported in 2003 that obesity, at the
tart of renal replacement therapy, was a risk actor for poor outcome in PD patients in an nalysis of the Australian and New Zealand ialysis and Transplant Registry. Clearly, ore research is needed in this field, espe-
ially given the vast amount of information hat is accumulating on the role of truncal fat
ass and the associated release of proinflam- atory cytokines and adipokines that may be
mportant mediators of insulin resistance, in- ammation, and atherosclerosis in ESRD.11
Hyperglycemia
elarue and colleagues12 have succinctly emonstrated that absorption of glucose from
he peritoneum, in contrast to that seen in oral bsorption, results in an extended period of yperglycemia (Fig 2), with an associated ex-
ended period of hyperinsulinemia. The met- bolic consequences of hyperglycemia in non- iabetic and poor glycemic control in diabetic atients has been well described in the litera-
ure and are summarized in Figure 1.13,14 Hy- erglycemia increases mitochondrial superox-
de, which, in turn, leads to activation of the 4 ajor pathways of hyperglycemic damage:
he polyol pathway, the hexosamine pathway, he protein kinase C pathway, and the AGE athway. Downstream events include in- reased gene expression of endothelin-1, EGF, TGF-, fibronectin, collagen, and lep-
in; down regulation of epithelial nitrogen ox-
Figure 1. The potential sys- temic metabolic effects of excessive intraperitoneal glu- cose absorption.
gen synthase (eNOS); and activation of the
p I r p p i c
S w w s e c r u h m v
o s b t p a h t s g E f p
v o p s g s t o c
P
T s a g p c a h w s e a n b d m f m ( s p i b t e o p w
p P s h m r s t s
F p l
roinflammatory transcription factor NF-B. n conjunction with increased production of eactive oxygen species (ROS), such altered atterns of gene expression are believed to be ivotal in the pathobiology of inflammation,
nsulin resistance, diabetic tissue damage, and ardiovascular disease.13
Hemodynamic Effects
everal studies by Selby and coworkers,15
ho used continuous noninvasive pulse- ave analysis at the digital artery, have
hown that hypertonic glucose has an acute ffect of increasing blood pressure. This in- rease occurs during the dwell period as a esult of increased heart rate and stroke vol- me. The authors speculate that such adverse emodynamics seen with hypertonic glucose ay play an important role in overall cardio-
ascular health. As the above arguments indicate, a spectrum
f confidence exists in the evidence that exces- ive glucose absorption is a factor in the mor- idity and mortality of dialysis patients. Never- heless, the combination of an atherogenic lipid rofile, hyperglycemia and insulin resistance, ccumulation of truncal fat mass, and adverse emodynamics provides a compelling basis to
est the hypothesis that these physiologic re- ponses are driven, at least in part, by excessive lucose absorption. Upon careful analysis of all SRD mortality studies to date, as recently per-
ormed by Vonesh et al,16 the cohort of PD
igure 2. The kinetics of glycemia after oral or eritoneal glucose 50 g dosing. Adapted from De- arue et al.12
atients that have lower survival versus con- p
entional HD patients are shown to be diabetics lder than 45 years. Survival of nondiabetic PD atients is currently equivalent or superior to urvival of HD patients. Thus, the promise of lucose-sparing regimens is one of improved urvival for both diabetic and nondiabetic pa- ients, which could result in an overall superi- rity in survival for PD patients compared with onventional HD patients.
eritoneal Membrane Function
he scientific literature is replete with a con- tellation of cell-based in vitro assays and nimal models that suggest both glucose and lucose degradation products (GDP) may lay an important role in the longitudinal hanges of peritoneal-membrane structure nd function seen in some patients. This topic as been the subject of recent reviews and so ill not be dealt with here.17,18 However, de-
pite the plethora of suggestive preclinical vidence that glucose degradation products re important mediators of membrane change, o clinical benefit with low-GDP, glucose- ased solutions has been demonstrated. In- eed, perhaps the best evidence to date of a embrane-protective effect with any new PD
ormulation comes from the European Auto- ated Peritoneal Dialysis Outcome Study
EAPOS). In a secondary analysis of this ob- ervational cohort of functionally anuric atients treated in 28 centers across Europe,
codextrin use was associated with a mem- rane-protective effect.19 In this study, pa- ients treated with icodextrin did not experi- nce a decline in UF capacity over the 2-year bservation period, unlike the full-glucose rescription group, despite starting therapy ith worse membrane function. The realization that the dogma that all
atients’membranes deteriorate over time on D is not supported by the literature is in- tructive. At least 3 large, longitudinal studies ave now been published that delineate only odest changes in membrane functional pa-
ameters in incident patients over the first everal years of therapy that uses conven- ional glucose solutions.20-22 Scrutiny of these tudies portend that randomized clinical trials
owered on such modest changes in perito-
n n l l
C S
T a i n s m a w s fl c p t a w t g f
i U c t s g c t e o s o 3 d g r fl c a H o i c o
T P
272 Clifford J. Holmes
eal functional parameters will require large umbers of patients to be enrolled and fol-
owed for several years and, thus, may be ogistically impractical.
ontemporary Strategies for Glucose paring
wo complementary strategies can be envis- ged for the implementation of glucose spar- ng in a clinical setting: first, reduction of the eed for peritoneal ultrafiltration (UF) and econd, optimization of peritoneal UF with inimal glucose use. The first strategy is best
chieved with reduction in dietary salt and ater intake and the use of diuretics. Dietary-
alt restriction can significantly reduce the uid burden and has been shown to be suc- essful in the absence of any change in dialysis rescription.23 Diuretic use in nonoliguric pa-
ients is also successful but requires use of ppropriate doses of loop diuretics with or ithout thiazides or thiazide congeners.24 In-
erventions under this first strategy require no lucose absorption and, therefore, are highly avorable physiologically.
able 1. A Comparison of Icodextrin with Dextros eritoneal Dialysis (CAPD) and APD Patients
Dwell Time (hour
odified from Holmes et al.26
The second strategy relies on the use of phys- ologic principles to optimize the conditions for F by matching patients’ peritoneal-transport
haracteristics with the design of the prescrip- ion. One concept that is crucial to glucose- paring approaches, and which is starting to ain appreciation, is that of UF efficiency. The oncept of UF efficiency was first introduced in he pediatric nephrology literature by Fischbach t al25 and is defined as the amount of net UF btained for every gram of carbohydrate ab- orbed. A simple illustration of the UF efficiency f glucose and icodextrin is provided in Figure . The ultrafiltration efficiency of glucose is high uring the short dwell, with 15 mL of UF per ram of carbohydrate absorbed, which deterio- ates over the long-dwell period as glucose and uid are lost from the peritoneum. The UF effi- iency index will vary between patients and is ffected by the conditions of the dwell. Recently, olmes and Mujais26 have reported clinical data
n the ultrafiltration efficiency of glucose and codextrin solutions during the long dwell that learly demonstrate a benefit of the colloidal smotic agent icodextrin (Table 1).
Figure 3. An illustration of the ultrafiltration efficiency of 4.25% glucose versus 7.5% icodextrin.
ing Long Dwells in Continuous Ambulatory
UF Efficiency (mL/g absorbed)
e Dur
s)
d T m U b a s a i d t i
E G
I o c t r t a t m t f
a t d s a s l b t t a s w a p i r h t p fi g d e d f t
T P
273Glucotoxicity in Peritoneal Dialysis
Additionally, use of icodextrin in the long well will readily lead to better UF efficiency. he impact of icodextrin in glucose sparing ay go beyond the long dwell, as enhanced F during the long dwell may lessen the urden of required UF during the short dwell, nd, consequently, lower glucose use in the hort dwells is then possible.27 Use of amino cid–based solutions during the short dwell n either continuous ambulatory peritoneal ialysis (CAPD) or automated dialysis pa-
ients (APD) will also maximize glucose spar- ng.
merging Clinical Benefits of lucose-Sparing Regimens
codextrin and amino acids are nonglucose smotic agents that have been used in the linical management of PD patients for more han a decade, at least in some geographic egions. The benefits of improved UF during he long dwell with 7.5% icodextrin solution nd the effect of amino acid solution on nu- ritional parameters have been the subject of
uch research. Only of late, however, have he benefit of glucose sparing offered by these ormulations been explored. Table 2 provides
able 2. A Summary of Recently Demonstrated an rescriptions
Patient Population Prescription
Nondiabetic CAPD Icodextrin vs glucose Decrease Diabetics Icodextrin vs glucose HbA1c d
7.90. CAPD with
only All PD patients Icodextrin vs glucose Gastric em
icodex All PD patients Icodextrin vs glucose No increa
group Diabetic CAPD Icodextrin, amino
acids and glucose vs all glucose
Significan
Improved Increas oxidati
Increased cardiac pressu icodex
Nondiabetic patients Icodextrin vs glucose Decrease in icod HDL a
icodextrin gro
summary of more recent clinical observa- ions that suggest that both diabetic and non- iabetic patients may benefit from a glucose- paring approach. By employing icodextrin to chieve a reduction in total carbohydrate ab- orption while maintaining adequate UF, a ess atherogenic lipid profile can apparently e attained, with avoidance of the weight gain hat is often observed in glucose-using pa- ients.28-30 At least 2 independent studies have lso identified the potential for improving in- ulin sensitivity.31,32 In diabetic patients, a ell-designed, albeit small, study by Marshall
nd coworkers33 demonstrated much im- roved glycemic control with a glucose-spar-
ng regimen, and Johnson and colleagues34
eported preliminary observations of reduced emoglobin (Hb) A1c. Recent research illus-
rates the numerous avenues for further ex- loration of glucose-sparing regimens: bene- cial changes in plasma adipokine levels,32
lucose and lipid oxidation,35 systemic hemo- ynamic effects,15,36 and improved gastric mptying.37 The accumulated experience to ate is promising, but clearly a need exists for
urther well-designed studies to confirm and o extend these observations.
erging Benefits of Glucose-Sparing Solution
Observations Author Year
insulin Improved insulin Amid 2001
a total cholesterol decreased LDL Bredie 2001 with icodextrin use 8.00.7% to
.05 Johnson 2001
time significantly shorter with up
Van 2002
Davies 2003
e and lipid metabolism: ose oxidation, decreased lipid
Martikainen 2005
ate, stroke volume and thus leading to increased blood g dwell with glucose versus
Selby 2005
a leptin, insulin and triglycerides roup. Increased adiponectin, roved insulin sensitivity in
Furuya 2005
d Em
ptying trin gro se in n unlike g tly imp
glucos ed gluc on heart r output
re durin trin d plasm extrin g nd imp
up also
M S
T i s t t t b e c p t d f a t o d i c m a t c r r p i s a n a t s t t o a t s s t w
M s r
A p e v c c r h s l L c d l A w c a n l c
T d o a o s t b a T i h t p b t b u o b r s n a I w p
274 Clifford J. Holmes
uture Solutions to Glucotoxicity in PD
aximized Use of Available Nonglucose olutions
he introduction of icodextrin and amino ac- ds as nonglucose osmotic agents has enabled tudies into the glucose-sparing benefits of hese solutions, either alone or in combina- ion, as described above. Randomized con- rolled trials are currently being planned in oth diabetic and nondiabetic PD patients to xplore the local and systemic benefits of glu- ose-sparing prescriptions in appropriately owered study populations. Clearly, icodex-
rin, used to replace glucose for the long well, has a significant glucose-sparing effect
or a 24-hour therapy prescription, and the ddition of amino acids can further enhance his impact. However, these regimens can nly approach replacement of 30% to 50% of aily glucose absorption. A further reduction
n glucose absorption using a second ex- hange of amino acid solution is not recom- ended because of a potential risk of acidosis
nd azotemia-related side effects.38 In con- rast, 2 preliminary reports that describe the linical use of 2 exchanges of icodextrin ecently appeared with the goal of further educing glucose exposure.39,40 As an exam- le, Fontan and colleagues40 used a random-
zed crossover design to compare a glucose- paring regimen in APD patients consisting of n icodextrin daytime dwell combined with a ighttime tidal prescription using amino acid nd glucose exchanges to a similar regimen hat was further glucose reduced by the sub- titution of a glucose bag for a second icodex- rin bag during the night. They observed that he already low peritoneal-glucose exposure f 72 g/day was further reduced to 46 g/day, nd the systemic glucose uptake was reduced o only 14 g/day from 35 g/day. Long-term afety studies will be needed to confirm the afety of these approaches, as icodextrin me- abolite levels will be higher than that seen
ith a single exchange.
Nonglucose Osmotic Agents
any alternatives to glucose have been ought over the past 3 decades and recently
eviewed extensively by Van Biesen et al.41 t…
T d d d a m m s p t t t a t o e t
G A
F d t a u s w s h g l m
ERITONEAL DIALYSIS
lucotoxicity in Peritoneal ialysis—Solutions for the Solution!
lifford J. Holmes Glucose has served well as the prototypical osmotic agent in peritoneal dialysis for more than 2
decades, because it affords many of the characteristics required of a safe and effective osmotic agent.
The disadvantages of glucose include its rapid dissipation from the peritoneum and its resulting
limited UF efficiency capacity in high and high-average transporters, the associated metabolic
response to absorbed glucose in all patients, and the local effects of glucose, glucose degradation
products, and hyperosmolality on peritoneal membrane structure and function. This paper briefly
reviews the salient elements of glucotoxicity associated with conventional glucose-based peritoneal
dialysis (PD) solution use, and then discusses emerging clinical benefits of newer nonglucose PD
solutions. Potential future strategies designed to abrogate glucose-associated toxicity are then
reviewed. These approaches include bimodal long-dwell solutions, nonglucose crystalloid osmotic
agent mixtures, and administration of pharmacologically active agents.
© 2007 by the National Kidney Foundation, Inc.
Index Words: Glucose; solution; peritoneal dialysis; osmotic agents; additives
G w c
o y t e y g y t m p e s l
P E P
H M c
he requirements for an effective osmotic agent for peritoneal dialysis have been
escribed by Martis et al1 and include a pre- ictable ultrafiltration profile, a high osmotic riving force at relatively low concentrations, nd a large ultrafiltration volume per unit ass absorbed. In addition, it should be easily etabolized and have an absence of local and
ystemic toxicity. Few molecules will ever com- ly with all of these requirements, especially in
erms of metabolic rate, potential for accumula- ion and toxicity of both parent molecules and heir metabolites. Thus, the selection of glucose s the original prototypic osmotic agent for peri- oneal dialysis can be appreciated; it offers good smotic driving force and rapid ultrafiltration, specially in the first few hours of the intraperi- oneal dwell.
lucose—The Prototypical Osmotic gent for Peritoneal Dialysis
or nearly 3 decades, the success of peritoneal ialysis as a viable choice of renal replacement
herapy has relied upon the success of glucose s an osmotic agent that can provide adequate ltrafiltration with a clinically acceptable afety profile. The metabolic fate of glucose is ell understood, with no acute toxicity or
erious long-term adverse events. It can be eat sterilized, thereby providing a large de- ree of sterility assurance, and is stable for
ong periods of time in dialysis-solution for-
ulations at ambient temperature ranges.
Advances in Chronic Kidney Disease, Vol
lucose provides a source of energy and, hen used in conjunction with amino acids,
an promote protein synthesis.2
Despite the described attractive attributes f glucose for peritoneal dialysis, in recent ears, awareness has developed of the poten- ial clinical disadvantages of excessive glucose xposure in dialysis patients.3 Peritoneal-dial- sis (PD) patients absorb from 100 to 300 g of lucose per day, or about 30 to 100 kg per ear, depending upon their prescription and ransporter type.3,4 Likewise, the peritoneal
embrane is continuously exposed to supra- hysiologic concentrations of glucose for sev- ral years. Consequently, concerns over exces- ive glucose exposure in PD encompass both ocal and systemic effects.
utative Pathogenetic Mechanisms of xcessive Glucose Exposure in PD atients
ystemic Effects
From the Renal Division, Baxter Healthcare, McGaw Park, IL. Address correspondance to C.J. Holmes, PhD, Baxter
ealthcare, 1620 Waukegan Road, William Graham Building, ail code MPGR-A2N, McGaw Park, IL 60085. E-mail:
[email protected] © 2007 by the National Kidney Foundation, Inc. 1548-5595/07/1403-0011$32.00/0
doi:10.1053/j.ackd.2007.03.009
a t u
P p d e e T t g S s t
W i i h t b w t a l o a i 1 t b a h d
M s d D s f a D m c t m m i fl
D d t a h t a d p t p i m t t p c V t
270 Clifford J. Holmes
re described in Figure 1 and can be attributed o a combination of the effects of carbohydrate ptake, caloric uptake, and hyperglycemia.
Carbohydrate Load and Dyslipidemia
D patients are dyslipidemic, and their lipid rofile is more atherogenic than that of hemo- ialysis (HD) patients.5 The greatest differ- nce between the modalities is seen in the levated levels of apo B and LDL cholesterol. he cause for the overproduction of LDL par-
icles remains unknown, although excessive lucose absorption has been implicated. tronger evidence suggests that glucose ab- orption plays an important role in the hyper- riglyceridemia observed in PD patients.5
Caloric Uptake and Increases in Fat Mass
eight gain is frequently observed in patients nitiating PD because of the extra caloric daily ntake from the PD solution.6 Nordfors et al7
ave demonstrated a genetic determinant to his weight gain in PD patients, characterized y an increase in truncal fat mass associated ith a polymorphism in the uncoupling pro-
ein 2 gene. Weight gain in PD patients can lso be a distressing event, affecting quality of ife. The role that high BMI plays in PD patient utcomes, however, remains controversial. In retrospective analysis of Medicare patients
n the United States initiating dialysis between 995 and 2000, Synder and coworkers8 found hat overweight and obese PD patients had a etter survival that those with lower BMI, lthough those PD patients with a high BMI ad an increasing risk of death with time on
ialysis. In 2004, data from the USRDS and y
ortality Wave II registries reported that obe- ity in PD patients, in contrast to HD patients, id not confer any protection.9 In fact, Mc- onald10 reported in 2003 that obesity, at the
tart of renal replacement therapy, was a risk actor for poor outcome in PD patients in an nalysis of the Australian and New Zealand ialysis and Transplant Registry. Clearly, ore research is needed in this field, espe-
ially given the vast amount of information hat is accumulating on the role of truncal fat
ass and the associated release of proinflam- atory cytokines and adipokines that may be
mportant mediators of insulin resistance, in- ammation, and atherosclerosis in ESRD.11
Hyperglycemia
elarue and colleagues12 have succinctly emonstrated that absorption of glucose from
he peritoneum, in contrast to that seen in oral bsorption, results in an extended period of yperglycemia (Fig 2), with an associated ex-
ended period of hyperinsulinemia. The met- bolic consequences of hyperglycemia in non- iabetic and poor glycemic control in diabetic atients has been well described in the litera-
ure and are summarized in Figure 1.13,14 Hy- erglycemia increases mitochondrial superox-
de, which, in turn, leads to activation of the 4 ajor pathways of hyperglycemic damage:
he polyol pathway, the hexosamine pathway, he protein kinase C pathway, and the AGE athway. Downstream events include in- reased gene expression of endothelin-1, EGF, TGF-, fibronectin, collagen, and lep-
in; down regulation of epithelial nitrogen ox-
Figure 1. The potential sys- temic metabolic effects of excessive intraperitoneal glu- cose absorption.
gen synthase (eNOS); and activation of the
p I r p p i c
S w w s e c r u h m v
o s b t p a h t s g E f p
v o p s g s t o c
P
T s a g p c a h w s e a n b d m f m ( s p i b t e o p w
p P s h m r s t s
F p l
roinflammatory transcription factor NF-B. n conjunction with increased production of eactive oxygen species (ROS), such altered atterns of gene expression are believed to be ivotal in the pathobiology of inflammation,
nsulin resistance, diabetic tissue damage, and ardiovascular disease.13
Hemodynamic Effects
everal studies by Selby and coworkers,15
ho used continuous noninvasive pulse- ave analysis at the digital artery, have
hown that hypertonic glucose has an acute ffect of increasing blood pressure. This in- rease occurs during the dwell period as a esult of increased heart rate and stroke vol- me. The authors speculate that such adverse emodynamics seen with hypertonic glucose ay play an important role in overall cardio-
ascular health. As the above arguments indicate, a spectrum
f confidence exists in the evidence that exces- ive glucose absorption is a factor in the mor- idity and mortality of dialysis patients. Never- heless, the combination of an atherogenic lipid rofile, hyperglycemia and insulin resistance, ccumulation of truncal fat mass, and adverse emodynamics provides a compelling basis to
est the hypothesis that these physiologic re- ponses are driven, at least in part, by excessive lucose absorption. Upon careful analysis of all SRD mortality studies to date, as recently per-
ormed by Vonesh et al,16 the cohort of PD
igure 2. The kinetics of glycemia after oral or eritoneal glucose 50 g dosing. Adapted from De- arue et al.12
atients that have lower survival versus con- p
entional HD patients are shown to be diabetics lder than 45 years. Survival of nondiabetic PD atients is currently equivalent or superior to urvival of HD patients. Thus, the promise of lucose-sparing regimens is one of improved urvival for both diabetic and nondiabetic pa- ients, which could result in an overall superi- rity in survival for PD patients compared with onventional HD patients.
eritoneal Membrane Function
he scientific literature is replete with a con- tellation of cell-based in vitro assays and nimal models that suggest both glucose and lucose degradation products (GDP) may lay an important role in the longitudinal hanges of peritoneal-membrane structure nd function seen in some patients. This topic as been the subject of recent reviews and so ill not be dealt with here.17,18 However, de-
pite the plethora of suggestive preclinical vidence that glucose degradation products re important mediators of membrane change, o clinical benefit with low-GDP, glucose- ased solutions has been demonstrated. In- eed, perhaps the best evidence to date of a embrane-protective effect with any new PD
ormulation comes from the European Auto- ated Peritoneal Dialysis Outcome Study
EAPOS). In a secondary analysis of this ob- ervational cohort of functionally anuric atients treated in 28 centers across Europe,
codextrin use was associated with a mem- rane-protective effect.19 In this study, pa- ients treated with icodextrin did not experi- nce a decline in UF capacity over the 2-year bservation period, unlike the full-glucose rescription group, despite starting therapy ith worse membrane function. The realization that the dogma that all
atients’membranes deteriorate over time on D is not supported by the literature is in- tructive. At least 3 large, longitudinal studies ave now been published that delineate only odest changes in membrane functional pa-
ameters in incident patients over the first everal years of therapy that uses conven- ional glucose solutions.20-22 Scrutiny of these tudies portend that randomized clinical trials
owered on such modest changes in perito-
n n l l
C S
T a i n s m a w s fl c p t a w t g f
i U c t s g c t e o s o 3 d g r fl c a H o i c o
T P
272 Clifford J. Holmes
eal functional parameters will require large umbers of patients to be enrolled and fol-
owed for several years and, thus, may be ogistically impractical.
ontemporary Strategies for Glucose paring
wo complementary strategies can be envis- ged for the implementation of glucose spar- ng in a clinical setting: first, reduction of the eed for peritoneal ultrafiltration (UF) and econd, optimization of peritoneal UF with inimal glucose use. The first strategy is best
chieved with reduction in dietary salt and ater intake and the use of diuretics. Dietary-
alt restriction can significantly reduce the uid burden and has been shown to be suc- essful in the absence of any change in dialysis rescription.23 Diuretic use in nonoliguric pa-
ients is also successful but requires use of ppropriate doses of loop diuretics with or ithout thiazides or thiazide congeners.24 In-
erventions under this first strategy require no lucose absorption and, therefore, are highly avorable physiologically.
able 1. A Comparison of Icodextrin with Dextros eritoneal Dialysis (CAPD) and APD Patients
Dwell Time (hour
odified from Holmes et al.26
The second strategy relies on the use of phys- ologic principles to optimize the conditions for F by matching patients’ peritoneal-transport
haracteristics with the design of the prescrip- ion. One concept that is crucial to glucose- paring approaches, and which is starting to ain appreciation, is that of UF efficiency. The oncept of UF efficiency was first introduced in he pediatric nephrology literature by Fischbach t al25 and is defined as the amount of net UF btained for every gram of carbohydrate ab- orbed. A simple illustration of the UF efficiency f glucose and icodextrin is provided in Figure . The ultrafiltration efficiency of glucose is high uring the short dwell, with 15 mL of UF per ram of carbohydrate absorbed, which deterio- ates over the long-dwell period as glucose and uid are lost from the peritoneum. The UF effi- iency index will vary between patients and is ffected by the conditions of the dwell. Recently, olmes and Mujais26 have reported clinical data
n the ultrafiltration efficiency of glucose and codextrin solutions during the long dwell that learly demonstrate a benefit of the colloidal smotic agent icodextrin (Table 1).
Figure 3. An illustration of the ultrafiltration efficiency of 4.25% glucose versus 7.5% icodextrin.
ing Long Dwells in Continuous Ambulatory
UF Efficiency (mL/g absorbed)
e Dur
s)
d T m U b a s a i d t i
E G
I o c t r t a t m t f
a t d s a s l b t t a s w a p i r h t p fi g d e d f t
T P
273Glucotoxicity in Peritoneal Dialysis
Additionally, use of icodextrin in the long well will readily lead to better UF efficiency. he impact of icodextrin in glucose sparing ay go beyond the long dwell, as enhanced F during the long dwell may lessen the urden of required UF during the short dwell, nd, consequently, lower glucose use in the hort dwells is then possible.27 Use of amino cid–based solutions during the short dwell n either continuous ambulatory peritoneal ialysis (CAPD) or automated dialysis pa-
ients (APD) will also maximize glucose spar- ng.
merging Clinical Benefits of lucose-Sparing Regimens
codextrin and amino acids are nonglucose smotic agents that have been used in the linical management of PD patients for more han a decade, at least in some geographic egions. The benefits of improved UF during he long dwell with 7.5% icodextrin solution nd the effect of amino acid solution on nu- ritional parameters have been the subject of
uch research. Only of late, however, have he benefit of glucose sparing offered by these ormulations been explored. Table 2 provides
able 2. A Summary of Recently Demonstrated an rescriptions
Patient Population Prescription
Nondiabetic CAPD Icodextrin vs glucose Decrease Diabetics Icodextrin vs glucose HbA1c d
7.90. CAPD with
only All PD patients Icodextrin vs glucose Gastric em
icodex All PD patients Icodextrin vs glucose No increa
group Diabetic CAPD Icodextrin, amino
acids and glucose vs all glucose
Significan
Improved Increas oxidati
Increased cardiac pressu icodex
Nondiabetic patients Icodextrin vs glucose Decrease in icod HDL a
icodextrin gro
summary of more recent clinical observa- ions that suggest that both diabetic and non- iabetic patients may benefit from a glucose- paring approach. By employing icodextrin to chieve a reduction in total carbohydrate ab- orption while maintaining adequate UF, a ess atherogenic lipid profile can apparently e attained, with avoidance of the weight gain hat is often observed in glucose-using pa- ients.28-30 At least 2 independent studies have lso identified the potential for improving in- ulin sensitivity.31,32 In diabetic patients, a ell-designed, albeit small, study by Marshall
nd coworkers33 demonstrated much im- roved glycemic control with a glucose-spar-
ng regimen, and Johnson and colleagues34
eported preliminary observations of reduced emoglobin (Hb) A1c. Recent research illus-
rates the numerous avenues for further ex- loration of glucose-sparing regimens: bene- cial changes in plasma adipokine levels,32
lucose and lipid oxidation,35 systemic hemo- ynamic effects,15,36 and improved gastric mptying.37 The accumulated experience to ate is promising, but clearly a need exists for
urther well-designed studies to confirm and o extend these observations.
erging Benefits of Glucose-Sparing Solution
Observations Author Year
insulin Improved insulin Amid 2001
a total cholesterol decreased LDL Bredie 2001 with icodextrin use 8.00.7% to
.05 Johnson 2001
time significantly shorter with up
Van 2002
Davies 2003
e and lipid metabolism: ose oxidation, decreased lipid
Martikainen 2005
ate, stroke volume and thus leading to increased blood g dwell with glucose versus
Selby 2005
a leptin, insulin and triglycerides roup. Increased adiponectin, roved insulin sensitivity in
Furuya 2005
d Em
ptying trin gro se in n unlike g tly imp
glucos ed gluc on heart r output
re durin trin d plasm extrin g nd imp
up also
M S
T i s t t t b e c p t d f a t o d i c m a t c r r p i s a n a t s t t o a t s s t w
M s r
A p e v c c r h s l L c d l A w c a n l c
T d o a o s t b a T i h t p b t b u o b r s n a I w p
274 Clifford J. Holmes
uture Solutions to Glucotoxicity in PD
aximized Use of Available Nonglucose olutions
he introduction of icodextrin and amino ac- ds as nonglucose osmotic agents has enabled tudies into the glucose-sparing benefits of hese solutions, either alone or in combina- ion, as described above. Randomized con- rolled trials are currently being planned in oth diabetic and nondiabetic PD patients to xplore the local and systemic benefits of glu- ose-sparing prescriptions in appropriately owered study populations. Clearly, icodex-
rin, used to replace glucose for the long well, has a significant glucose-sparing effect
or a 24-hour therapy prescription, and the ddition of amino acids can further enhance his impact. However, these regimens can nly approach replacement of 30% to 50% of aily glucose absorption. A further reduction
n glucose absorption using a second ex- hange of amino acid solution is not recom- ended because of a potential risk of acidosis
nd azotemia-related side effects.38 In con- rast, 2 preliminary reports that describe the linical use of 2 exchanges of icodextrin ecently appeared with the goal of further educing glucose exposure.39,40 As an exam- le, Fontan and colleagues40 used a random-
zed crossover design to compare a glucose- paring regimen in APD patients consisting of n icodextrin daytime dwell combined with a ighttime tidal prescription using amino acid nd glucose exchanges to a similar regimen hat was further glucose reduced by the sub- titution of a glucose bag for a second icodex- rin bag during the night. They observed that he already low peritoneal-glucose exposure f 72 g/day was further reduced to 46 g/day, nd the systemic glucose uptake was reduced o only 14 g/day from 35 g/day. Long-term afety studies will be needed to confirm the afety of these approaches, as icodextrin me- abolite levels will be higher than that seen
ith a single exchange.
Nonglucose Osmotic Agents
any alternatives to glucose have been ought over the past 3 decades and recently
eviewed extensively by Van Biesen et al.41 t…
Related Documents
