STRATEGIES TO REDUCE GLUCOSE EXPOSURE IN PERITONEAL DIALYSIS PATIENTS. Clifford J. Holmes and Ty R. Shockley

Transcription

1 VIIth International Course on Peritoneal Dialysis May 23 26, 2000, Vicenza, Italy Peritoneal Dialysis International, Vol. 20, Suppl /00 $ Copyright 2000 International Society for Peritoneal Dialysis Printed in Canada. All rights reserved. STRATEGIES TO REDUCE GLUCOSE EXPOSURE IN PERITONEAL DIALYSIS PATIENTS Clifford J. Holmes and Ty R. Shockley Renal Division, Baxter Healthcare, McGaw Park, Illinois, U.S.A. Glucose has been used successfully for more than two decades in peritoneal dialysis, and in this regard, must be considered a safe and effective osmotic agent. Recently, however, insight has been growing about the potential for metabolic and peritoneal effects arising from long-term exposure to high glucose concentrations for example, hyperlipidemia and loss of peritoneal ultrafiltration. Clinical concerns over exposure to excessive glucose and glucose degradation products (GDPs) during peritoneal dialysis can be significantly ameliorated by the use of non-glucose-based peritoneal dialysis (PD) solutions, in combination with more biocompatible glucosebased formulations. Peritoneal exposure to GDPs can be reduced by using low-gdp-containing glucose formulations and non glucose solutions such as amino acids and icodextrin. Peritoneal glucose exposure, hyperosmolar stress, and carbohydrate absorption can be reduced by using a combination of icodextrin and amino acids. KEY WORDS: Glucose exposure. The ideal osmotic agent for continuous forms of peritoneal dialysis (PD) must meet several requirements: It must produce a predictable ultrafiltration profile and a large ultrafiltration volume per unit mass absorbed. It must be easily metabolized while avoiding both local (that is, peritoneal) and systemic toxicity. It must be easily manufactured and made available at an acceptable cost. For more than twenty years, glucose has been employed as the sole osmotic agent that best meets these requirements. Many other osmotic agents have been evaluated for example, sorbitol, mannitol, xylitol, dextran, polyanionic and polycationic polymers, and milk-whey peptides but none have offered a profile superior to that of glucose in terms of safety and efficacy. This fact testifies to the appropriateness of glucose for use in long-term peritoneal dialysis. Correspondence to: C.J. Holmes, 1620 Waukegan Road, McGaw Park, Illinois U.S.A. Although the dialysis community generally accepts that complete substitution of glucose as the osmotic agent in PD therapy does not appear to be on the horizon, interest regarding the strategies that can be used by the clinician to minimize the amount of glucose required for adequate dialysis is being renewed (1). This review, as part of a series of viewpoints from industry, describes the clinical concerns putatively associated with the continuous use of glucose-based PD solutions and provides a series of alternative clinical strategies by which glucose exposure can be minimized using products either available today or potentially available in the next few years. THE CLINICAL CONCERNS The clinical concerns associated with the use of glucose-based PD solutions can be broadly grouped into two main categories: systemic metabolic effects and local biocompatibility effects on the peritoneum. Systemic Metabolic Effects: Studies performed to determine the absorption of glucose from the peritoneal cavity suggest that approximately 60% 80% of the glucose instilled into the peritoneal cavity is absorbed during a 6-hour dwell. From 3.86% solution, g of glucose is absorbed; from a 2.27% solution, g; and from a 1.36% solution, g (2). Peritoneal dialysis patients are estimated to absorb g of glucose per day through their dialysate. This glucose load contributes 12% 34% of daily energy intake and is of benefit to peritoneal dialysis patients, although concern exists that it may contribute to obesity in some patients. The glucose load is also believed to contribute to a number of metabolic abnormalities. Uremic patients have abnormal glucose and insulin metabolism with glucose intolerance, hyperinsulinemia, and reduced peripheral sensitivity to insulin (2). Although continuous ambulatory peritoneal dialysis (CAPD) exchanges with a 1.36% glucose solution have marginal effects on blood glucose and insulin levels, use of a 3.86% glucose solution yields biochemical changes similar to those observed after S37

2 HOLMES and SHOCKLEY MAY 2000 VOL. 20, SUPPL. 2 PDI an oral glucose load. Because sustained hyperinsulinemia may possibly increase atherogenesis, the elevated circulating insulin levels constitute a potential risk factor for patients during long-term treatment with PD solutions containing glucose. Several studies have demonstrated a hyperlipidemic effect with peritoneal dialysis. This effect has been attributed to the continuous absorption of glucose. However, the results vary considerably between reports owing to large inter-individual variations, varying energy intakes, and fluctuation of serum lipid levels over time in the individual patient (2). Peritoneal Membrane Effects: Glucose concentrations used in PD solutions ( mmol/l) are 15 times to 40 times physiological levels. Figure 1 summarizes the clinical concerns putatively associated with the long-term use of high glucose concentrations in PD. Changes in peritoneal structure such as vasculopathy, reduplication of basement membrane, and thickening of the peritoneal membrane are suspected to be causally related to functional changes such as loss of ultrafiltration. Impairment of peritoneal leukocyte function by elevated glucose concentration may also play an important role in peritoneal host defense. Mechanisms by which elevated glucose may contribute to structural and functional changes of the peritoneum are illustrated in Figure 2 and discussed in more detail below. Hyperosmolar Stress: Hyperosmolality has been shown to independently suppress both peritoneal leukocyte and mesothelial cell function. By controlling the osmolality of test solutions (using glucose or NaCl), Liberek et al were able to demonstrate that respiratory burst activation, phagocytosis, and cytokine release were osmolality-dependent phenomena (3). Recently, Sitter et al have shown that prostaglandin E2 (PGE 2 ) synthesis by human peritoneal mesothe- Figure 1 A diagrammatic view of the peritoneal membrane changes associated with excessive glucose use in peritoneal dialysis patients. S38 Figure 2 Mechanisms of peritoneal glucotoxicity. lial cells is mediated, in part, by osmolality (4). Effects on Peritoneal Cell Metabolism via the Polyol Pathway, Protein Kinase C Activation and Induction of Genes, such as Transforming Growth Factor β (TGFβ): In vitro evidence exists that exposure of mesothelial cells to high concentrations of glucose can lead to glucose metabolism via the polyol pathway, a pathway long suspected to be responsible for some diabetic complications such as neuropathy and retinopathy (5). Glucose-induced polyol pathway activation has also been shown to be further enhanced in a dose-dependent manner by lactate (5). Protein kinase C activation resulting from polyol pathway activity is one of the earliest signal-transduction mechanisms through which high glucose stimulates extracellular matrix (ECM) production. TGFβ is one of the cytokine growth factors that is known to play a central role in ECM regulation. Exposure of mesothelial cells to glucose is associated with increased fibronectin and TGFβ1 mrna expression and synthesis, effects that are both glucose and osmolality dependent (6). Glycation of Peritoneal Proteins and Subsequent Amadori Adduct and Advanced Glycosylation End- Product Formation: Non enzymatic glycosylation of proteins may occur in the peritoneal cavity with the subsequent formation of intermediate Amadori adducts and the eventual appearance of advanced glycosylated end-products (AGEs) in the peritoneal vasculature, interstitium, and mesothelial layer (7). Glycated proteins, Amadori products, and AGEs have individually been shown to be potent biological mediators of the structural changes, or functional changes, or both, thought to occur in the long-term

3 PDI MAY 2000 VOL. 20, SUPPL. 2 REDUCING GLUCOSE EXPOSURE IN PD glucose dialyzed peritoneal membrane. However, a causal relationship with the glucose content of PD solution remains unproven. Presence of Glucose Degradation Products: To meet the sterility requirements imposed by the various regulatory bodies around the world, PD solutions are terminally heat-sterilized. To minimize glucose degradation during heat sterilization, the ph of the dialysis solution must be kept low (approximately ph 5.3). GDPs are thought to affect biocompatibility via two mechanisms. The first mechanism is direct cytotoxicity. The second is an accelerated process of AGE formation. The aldehydes such as acetaldehyde and formaldehyde found in conventional heattreated glucose-containing PD solutions are suspected to be the GDPs that predominantly contribute to the cytotoxic characteristic seen in vitro as inhibition of fibroblast proliferation and leukocyte function. A second group of GDPs, also recognizable as carbonyl stress compounds (3-deoxyglucosone [3-DG], methylglyoxal, and glyoxal), have been suggested to play an important role in the AGE pathway. Although these compounds are present in much lower concentrations than is glucose, they are known to be significantly more reactive in terms of AGE formation (8). The relative importance of each GDP to AGE formation remains obscure in the PD setting. For instance, although the levels of 3-deoxyglucosone are approximately 100-fold higher than those of the glyoxal compounds, the latter apparently are significantly more reactive than 3-DG. Additionally, the formation of pentosidine, one of the better-characterized AGE compounds, has recently been shown to be inhibited in the presence of acetaldehyde, which is itself a GDP (9). Consequently, it can be speculated that reducing the level of all GDPs in current formulations might reduce the AGE formation associated with the presence of carbonyl stress compounds, while at the same time permitting glucose-associated AGE formation to occur at an accelerated rate owing to the lower levels of acetaldehyde. The net AGE formation remains unknown. STRATEGIES FOR TODAY The goal for the clinician is to identify how to reduce patient exposure to glucose, to minimize the total amount of glucose absorbed systemically by the patient, and to avoid the hyperosmolar stress, high glucose, and GDP exposure to the peritoneum that exists today. Fortunately, non-glucose-based PD solutions, such as amino acids and polyglucose-based formulations, are currently available in many European countries, and offer the opportunity to significantly reduce metabolic and local glucose-induced stress. Glucose-containing solutions with a lower GDP content, which can be used in combination with nonglucose-based solutions, are also becoming available. The benefits of each formulation are discussed below. Icodextrin: Icodextrin is a glucose polymer that employs colloidal, rather than crystalline, osmotic pressure to effect a sustained ultrafiltration profile beneficial for long dwells in CAPD and continuous cycling peritoneal dialysis (CCPD). Owing to its colloidal properties, a 7.5% icodextrin solution (Extraneal: Baxter Healthcare Corporation, Deerfield, Illinois, U.S.A.) exerts an osmotic pressure of only 282 mosm. Glucose PD solutions at 1.36%, 2.27%, and 3.86% exert osmotic pressures of 358 mosm, 401 mosm, and 511 mosm, respectively. Furthermore, the GDP content of Extraneal is very low compared to that of glucose-containing solutions, yielding a significant reduction in in vitro cell cytotoxicity, glycation of proteins, Amadori adduct formation, and AGE formation (10,11). Amore et al have focussed their research initially on the biological benefits of the reduced Amadori adduct formation. They demonstrated that, in marked contrast to results for glucose-based solutions, apoptosis and inducible nitric oxide synthetase (inos) activity in peritoneal mesothelial cells exposed to Extraneal were equivalent to control basal values (12). In a four-week rat study that included repetitive inflammatory episodes, Kim et al recently reported significantly less AGE formation and less peritoneal thickening in animals exposed to daily Extraneal versus 2.27% glucose PD solution (13). Clinically, several ex vivo studies have now been performed that have shown some improvement in both peritoneal and mesothelial cell function in patients using icodextrincontaining solutions, while using conventional solutions for all other exchanges (13 15). In terms of systemic metabolic benefits, an analysis by the MIDAS study group has shown a significant reduction in total cholesterol and low-density lipoprotein (LDL) cholesterol, especially in patients with baseline hypercholesterolemia (16). Mahiout et al have suggested that polyglucose degradation products exist in very low concentrations that may form a special group of AGEs and that, if internalized within cells, can inhibit the activity of the enzyme α-galactosidase, thereby leading to the accumulation of glycogen (17). However, no evidence exists that this process occurs in vivo, either from the chronic animal toxicology studies required for regulatory approval, or from thousands of years of cumulative patient exposure accrued to date. Amino acids: Amino acid based PD solutions have been developed to aid in the management of malnutrition and to replace amino acids lost during peritoneal dialysis. Amino-acid formulations are limited to one bag per day, as they generate nitrogenous waste; S39

4 HOLMES and SHOCKLEY MAY 2000 VOL. 20, SUPPL. 2 PDI but they inherently possess the characteristic of being non glucose based. In this latter respect, it has been reported that a 1.1% amino acid based solution (Nutrineal: Baxter Healthcare Corporation, Deerfield, Illinois, U.S.A.) has undetectable levels of the carbonyl stress compounds 3-deoxyglucosone, methylglyoxal, and glyoxal (10). As expected, the in vitro generation of the AGEs pentosidine and (carboxymethyl)lysine is also undetectable. The in vivo benefit of amino-acid solutions was recently highlighted by Garosi et al in a 60-day rabbit dialysis model, in which virtually all the changes in peritoneal morphology induced by glucose, such as cubic transformation and continuity of mesothelial cells, were avoided with amino acids (18). Glucose-Based PD Solutions: Non-glucose-based PD solutions most effectively address all aspects of both systemic and local glucotoxicity. However, if glucose-based solutions cannot be used for all exchanges in CAPD or automated peritoneal dialysis (APD) therapy, they are still required for at least part of the dialysis regimen. Today, it is possible to produce glucose-based solutions with an alternative GDP profile that may improve their peritoneal biocompatibility. The recent development of two-chamber bags permits glucose to be separated from other solution components, thereby allowing the glucose to be sterilized at a ph lower than is possible in singlechamber bags. Mixing the two compartments raises the ph, which, in the case of bicarbonate-containing formulations, can result in a neutral ph. For bicarbonate/lactate buffered two-chamber systems, sterilization of glucose at a ph lower than is currently performed today produces a reduction in most of the identified GDPs, including acetaldehyde, 3-deoxyglucosone, and methylglyoxal. Some GDPs appear to be unchanged (for example, glyoxal), while others are increased (for example, 5-hydroxymethylfurfural). The latter, however, appears to be innocuous in cytotoxicity assays. In vitro cytotoxicity and AGE formation are also significantly reduced in such formulations (19). Combinations of Solutions: Figure 3 shows the additive effect that can be achieved by using combinations of the new-generation solutions described above. A regimen of low-gdp-containing glucosebased solutions may have significant peritoneal effects, especially if bicarbonate-based neutral ph formulations are employed. Augmenting this regimen with an iso-osmolar icodextrin solution significantly reduces all glucotoxic aspects. The regimen can be further optimized with an exchange of non-glucosebased amino-acid solution. This latter combination significantly reduces carbohydrate absorption and peritoneal glucose and GDP exposure as compared to a standard CAPD regimen (3 1.36% %). S40 Figure 3 The additive benefits of using new-generation peritoneal dialysis solutions. A continuous ambulatory peritoneal dialysis (CAPD) regimen of three 5-hour dwells during the day and one 9-hour dwell overnight has been assumed. Peritoneal glucose exposure was calculated from the addition of area-under-the-curve values for each exchange over 24 hours. G = conventional glucose-based solution; B\L = bicarbonate/lactate glucose-based solution with a modified glucose degradation product profile; I = 7.5% icodextrin solution; AA = 1.1% amino-acid solution. STRATEGIES FOR TOMORROW? Research into other osmotic agents, such as glycerol, is ongoing and may offer alternatives to glucosebased therapy in the future when used in combination with amino acids or icodextrin. Continuous-flow techniques of APD may also be a technical possibility in the future, potentially providing a way to reduce glucose exposure both systemically and locally. For instance Raj et al reported adequate ultrafiltration by maintaining dialysate glucose concentration in the 0.2% 0.5% range during a continuous-flow PD experimental set-up (20). However, the safety, efficacy,

5 PDI MAY 2000 VOL. 20, SUPPL. 2 REDUCING GLUCOSE EXPOSURE IN PD and cost of such techniques are unknown and will require extensive evaluation. CONCLUSIONS Clinical concerns over excessive glucose exposure during peritoneal dialysis can be significantly ameliorated by the use of non-glucose-based PD solutions in combination with more biocompatible glucosebased formulations. Reduced peritoneal exposure to GDPs can be achieved by using low-gdp-containing glucose formulations and non-glucose solutions such as amino acids and icodextrin. Peritoneal glucose exposure, hyperosmolar stress, and carbohydrate absorption can be reduced by using a combination of icodextrin and amino acids. ACKNOWLEDGMENTS The authors thank Leo Martis for invaluable input. REFERENCES 1. Lameire N, Van Biesen W, Vanholder R. Consequences of using glucose in peritoneal dialysis fluid. Semin Dial 1998; 11: Martis L, Topley N, Holmes CJ. Conventional and newer peritoneal dialysis solutions. In: Owen WF, Pereira JG, Sayegh MH, eds. Dialysis and Transplantation: A companion to Brenner and Rector s The Kidney. Philadelphia: W.B. Saunders; 2000: Liberek T, Topley N, Jörres A, Coles GA, Gahl GM, Williams JD. Peritoneal dialysis fluid inhibition of phagocyte function: Effects of osmolality and glucose concentration. J Am Soc Nephrol 1993; 3: Sitter T, Haslinger B, Mandl S, Fricke H, Held E, Sellmayer A. High glucose increases prostaglandin E2 synthesis in human peritoneal mesothelial cells: Role of hyperosmolarity. J Am Soc Nephrol 1998; 9: Kaur D, Williams JD, Phillips AO, Topley NT. Improved cell function in pyruvate-buffered peritoneal dialysis fluids is related to specific antagonism of the polyol pathway [Abstract]. J Am Soc Nephrol 1997; 8:266A. 6. Witowski J, Williams JD, Topley N. D-glucose induces transforming growth factor-b1 (TGFβ1) mrna expression and secretion in human peritoneal mesothelial cells (HPMC): Effect of hyperosmolality [Abstract]. Perit Dial Int 1997; 17(Suppl 1):S Nakayama M, Kawaguchi Y, Yamada K, Hasegawa T, Takazoe K, Katoh N, et al. Immunohistochemical detection of advanced glycosylation end-products in the peritoneum and its possible pathophysiological role in CAPD. Kidney Int 1997; 51: Dawnay A, Millar DJ. The pathogenesis and consequences of AGE formation in uraemia and its treatment. Cell Mol Biol (Noisy-le-grand) 1998; 44: Al Abed Y, Mitsuhashi T, Li H, Lawson JA, FitzGerald GA, Founds H, et al. Inhibition of advanced glycation endproduct formation by acetaldehyde: Role in the cardioprotective effect of ethanol. Proc Natl Acad Sci U S A 1999; 96: Ueda Y, Miyata T, Izuhara Y, Inagi R, Nangaku M, Ishibashi Y, et al. Lower carbonyl stress in polyglucose or amino acid PD solutions than in conventional glucose based PD fluids [Abstract]. J Am Soc Nephrol 1999; 10:230A. 11. Dawnay A, Millar D. Glycation and advanced end-product formation with icodextrin and dextrose. Perit Dial Int 1997; 17: Amore A, Cirina P, Conti G, Peruzzi L, Gianoglio B, Ricotti E, et al. Icodextrin avoids the enhancement of inducible nitric oxide (inos) activity in peritoneal mesothelial cells given by glucose solutions [Abstract]. J Am Soc Nephrol 1998; 9:278A. 13. Kim YL, Kim JH, Kim CD, Cho DK, Kim IS, Kim YJ, et al. Effects of icodextrin on advanced glycation end-product formation and peritoneal morphology in rats [Abstract]. J Am Soc Nephrol 1999; 10:317A. 14. Posthuma N, ter Wee PM, Donker AJM, Dekker HAT, Oe PL, Verbrugh HA. Peritoneal defense using icodextrin or glucose for daytime dwell in CCPD patients. Perit Dial Int 1999; 19: Bajo MA, Selgas R, Castro MA, Castro MJ, Del Peso G, Aguilera A, et al. Effect of icodextrin on mesothelial cell growth in culture. In: Ritz E, ed. Abstracts. Proceedings of the XXXVI Congress of the European Renal Association, European Dialysis and Transplantation Association; 5 8 September 1999; Madrid, Spain. Gambro Renal Care; 1999: Gokal R, Moberly J, Ogrinc F, Gordon A, Peers E, and the MIDAS study group. Improvement of hyperlipidemia with icodextrin use in CAPD patients [Abstract]. J Am Soc Nephrol 1998; 9:283A. 17. Mahiout A, Rahni T, Brunkhorst U, Neuendorf S, Brunkhorst R. Cellular uptake of advanced glycated end-products of polyglucose and its potential impact on glycogen metabolism [Abstract]. Perit Dial Int 1999; 19(Suppl 1):S Garosi G, Gaggiotti E, Monaci G, Brardi S, Di Paolo N. Biocompatibility of a peritoneal dialysis solution with amino acids: Histological evaluation in the rabbit. Perit Dial Int 1998; 18: Cooker LA, Luneburg P, Faict D, Choo C, Holmes CJ. Reduced glucose degradation products in bicarbonate/ lactate buffered peritoneal dialysis solutions produced in two-chambered bags. Perit Dial Int 1997; 17: Raj DSC, Self M, George H, Work J. Hybrid dialysis [Abstract]. J Am Soc Nephrol 1998; 9:193A. S41