TAKE CARE IN HOW YOU STORE YOUR PD FLUIDS ...

Peritoneal Dialysis International, Vol. 25, pp. 583–590 Printed in Canada. All rights reserved.

0896-8608/05 $3.00 + .00 Copyright © 2005 International Society for Peritoneal Dialysis

TAKE CARE IN HOW YOU STORE YOUR PD FLUIDS: ACTUAL TEMPERATURE DETERMINES THE BALANCE BETWEEN REACTIVE AND NON-REACTIVE GDPs

Martin Erixon,1,2 Anders Wieslander,1 Torbjörn Lindén,1 Ola Carlsson,1 Gunita Forsbäck,1 Eva Svensson,1 Jan Åke Jönsson,2 and Per Kjellstrand1 Corporate Research,1 Gambro AB; Analytical Chemistry,2 University of Lund, Lund, Sweden

Correspondence to: M. Erixon, Gambro AB, Box 10101, S-220 10 Lund, Sweden. [email protected] Received 28 October 2004; accepted 12 March 2005.

tremely careful with the temperatures conventional PDFs are exposed to. Perit Dial Int 2005; 25:583–590

www.PDIConnect.com

KEY WORDS: GDPs; storage; temperature; 3,4-DGE; glucose.

G

lucose is still used with good outcome by all patients on peritoneal dialysis (PD), simply because it is cheap and nontoxic. However, during sterilization and storage, glucose degrades to reactive carbonyl compounds referred to as glucose degradation products (GDPs) (1–6). More than 80% of all PD patients use fluids that are produced in the conventional way, with all ingredients mixed in one compartment at a pH around 5.5. Many side effects of PD, such as those on the peritoneal cell population, submesothelial thickening, and chemical peritonitis seem to be linked to the presence of GDPs in the PD fluid (7–9). In order to avoid the GDPs, one might replace glucose as osmotic agent. However, this has for various reasons proved to be difficult. Sophisticated manufacturing processes, such as optimizing pH and separating glucose from catalyzing substances, may be used to decrease GDP concentrations (10,11), but the overwhelming majority of PD bags used today are still manufactured in a conventional way. Consequently, the patients using them are exposed to unnecessarily high concentrations of GDPs. Many investigators have tried to identify the source of the biological effects among the specific GDPs. However, the lack of effects at the concentrations found in the fluids was a major problem until the identification of 3,4-dideoxyglucosone-3-ene (3,4-DGE). This molecule has been shown to have immunosuppressive effects at concentrations as low as 0.7 µmol/L, which is far below the 10 µmol/L normally found in a 1.5% PD fluid (12). Remesothelialization in a wound healing system was shown to be retarded at the concentrations present in 583

Downloaded from www.pdiconnect.com by on October 14, 2011

♦♦Objective: During heat sterilization and during prolonged storage, glucose in peritoneal dialysis fluids (PDF) degrades to carbonyl compounds commonly known as glucose degradation products (GDPs). Of these, 3,4-dideoxyglucosone3-ene (3,4-DGE) is the most cytotoxic. It is an intermediate in degradation between 3-deoxyglucosone (3-DG) and 5-hydroxymethyl-2-furaldehyde (5-HMF). We have earlier reported that there seems to be equilibrium between these GDPs in PDF. The aim of the present study was to investigate details of this equilibrium. ♦♦Methods: Aqueous solutions of pure 3-DG, 3,4-DGE, and 5-HMF were incubated at 40°C for 40 days. Conventional and low-GDP fluids were incubated at various temperatures for up to 3 weeks. Formaldehyde, acetaldehyde, glyoxal, methylglyoxal, 3-DG, 3,4-DGE, and 5-HMF were analyzed using high performance liquid chromatography. µµ ♦♦Results: Incubation of 100 µmol/L 3,4-DGE resulted in µµ µµ the production of 36 µmol/L 3-DG, 4 µmol/L 5-HMF, and µµ 40 µmol/L unidentified substances. With the same incuµµ µµ bation, 200 µmol/L 3-DG was converted to 9 µmol/L µµ µµ 3,4-DGE, 6 µmol/L 5-HMF, and 14 µmol/L unidentified subµµ stances. By contrast, 100 µmol/L 5-HMF was uninfluenced by incubation. In a conventional PDF incubated at 60°C for 1 day, the 3,4-DGE concentration increased from 14 to a µµ maximum of 49 µmol/L. When the fluids were returned to room temperature, the concentration decreased but did not reach original values until after 40 days. In a low GDP fluid, 3,4-DGE increased and decreased in the same manner as in the conventional fluid but reached a maximum of only µµ 0.8 µmol/L. ♦♦Conclusions: Considerable amounts of 3,4-DGE may be recruited by increases in temperature in conventional PDFs. Lowering the temperature will again reduce the concentration but much more time will be needed. Precursors for 3,4-DGE recruitment are most probably 3-DG and the enol 3-deoxyaldose-2-ene, but not 5-HMF. Considering the ease at which 3,4-DGE is recruited from its pool of precursors and the difficulty of getting rid of it again, one should be ex-

ERIXON et al.

MATERIAL AND METHODS

PDI

PROLIFERATION TESTS

Mouse L-929 fibroblast cells (CCL-1; ATCC, Rockville, Maryland, USA) were cultured to confluence in 75-cm2 tissue-culture flasks with cell culture media (MEM; Gibco, Paisley, Scotland) completed with 10% fetal calf serum (FCS; Hy Clone Laboratories, Logan, Utah, USA), 2 mmol/L L-glutamine, and 1% nonessential amino acids. The cells were seeded in a 96-well tissue-culture plate at a density of 2500 cells/well. After 24 hours, the medium was removed and the samples to be tested added. The samples consisted of freshly thawed PD fluid mixed 3:2 with growth medium containing 20% FCS, 2.5% nonessential amino acids, and 4 mmol/L L-glutamine. The plates were incubated in humidified air with 5% CO2 for 72 hours at 37°C before determining inhibition of cell growth (ICG) by neutral red assay (15). Percent ICG was calculated using a control containing only growth medium as reference. The ICG values for the differently stored samples are given as the difference between a heat-sterilized sample and a corresponding filtered sample.

FLUIDS EXPERIMENTAL PROCEDURES

Pure 3,4-DGE and 3-DG were dissolved in water at the concentrations previously found immediately after sterilization, that is, 100 and 200 µmol/L respectively (1). A high concentration of 5-HMF (100 µmol/L) was used to test the reversibility of the 3,4-DGE–5-HMF reaction. In all the fluids, the pH was adjusted to 5.5 with HCl. In the experiments with PD fluids, either Gambrosol 2.5% (as an example of a conventional manufactured fluid) or Gambrosol trio (as an example of a low GDP fluid) was used (both made by Gambro AB, Lund, Sweden). CHEMICAL ANALYSIS OF GDPs

Methylglyoxal, 5-HMF, glyoxal, acetaldehyde, and formaldehyde were purchased from Sigma-Aldrich, Stockholm, Sweden, and 3-DG from Toronto Research Chemicals, North York, Ontario, Canada. The standards were diluted in water before analysis. The 3,4-DGE standard was extracted from a heat-sterilized glucose-containing fluid according to a previous description (4). The GDPs were analyzed using a high performance liquid chromatography (HPLC) system (Agilent Technologies, Waldbronn, Germany) equipped with UV detection (3–5). The low-GDP fluids were diluted to a glucose concentration of 2.5% (140 mmol/L) before chemical analysis; each sample collected was analyzed on three separate occasions. Values are given as mean ± standard error of the mean (SEM). 584

Equilibrium between the different GDPs was studied through incubation of 100 µmol/L 3,4-DGE, 200 µmol/L 3-DG, and 100 µmol/L 5-HMF at 40°C for 30 days in 50-mL tubes. Samples were collected on days 0, 1, 3, 6, 8, 15, 24, and 30. The conventional PD fluid was partitioned into 50-mL tubes and the 50% glucose solution from the low-GDP fluid into 15-mL tubes. The tubes were then incubated at 25°C, 40°C, and 60°C. Samples were collected on days 0, 1, 2, 3, 4, 5, 6, 7, 14, and 21. In another experiment, the conventional PD fluid was first incubated in 15-mL tubes for 24 hours at 60°C and then at room temperature for 40 days. Samples were collected after 0, 1, 2, 4, 6, 8, 16, and 24 hours and on days 2, 5, 7, 11, 20, 30, and 40. All samples were frozen at –80°C until analyzed. RESULTS 3,4-DGE, 3-DG, AND 5-HMF INCUBATED AT 40°C FOR 30 DAYS

Incubation of 100 µmol/L 3,4-DGE resulted in the production of 36 µmol/L 3-DG and 4 µmol/L 5-HMF; 3,4-DGE decreased by as much as 79 µmol/L [Figure 1(a); Table 1]. Incubation of 200 µmol/L 3-DG resulted in the production of 9 µmol/L 3,4-DGE and 6 µmol/L 5-HMF; 3-DG decreased by as much as 29 µmol/L [Figure 1(b); Table 1].


PD fluids, and 3,4-DGE was recently demonstrated to be the most reactive GDP for accelerating human neutrophil apoptosis (13,14). Furthermore, we have reported a correlation between 3,4-DGE in different fluids and proliferation of both human peritoneal mesothelial cells and mouse fibroblasts (1). Thus, among all identified GDPs, 3,4-DGE is by far the one most strongly associated with biological effects (1,4). 3,4-DGE has been proposed as an intermediate between 3-deoxyglucosone (3-DG) and 5-hydroxymethyl2-furaldehyde (5-HMF). In a conventionally produced laboratory-made PD fluid, a very high concentration of 3,4-DGE was measured immediately after sterilization. At room temperature, the concentration decreased by over 85% during 2 months of storage. The only GDP that concomitantly increased was 3-DG (1). The aim of the present study was to investigate the influence of temperature on the presence of 3,4-DGE and to further explore details of the equilibrium between 3,4-DGE, 3-DG, and 5-HMF.

NOVEMBER 2005 – VOL. 25, NO. 6

PDI


GDPs IN PD FLUIDS

Concentration of 100 µmol/L 5-HMF was stable during the entire incubation (Table 1). The pH in all the fluids decreased during the incubation (Table 1). Small amounts of acetaldehyde and formaldehyde could be detected in some of the samples after incubation. There were no detectable amounts of methylglyoxal or glyoxal.

during the rest of the incubation period [Figure 2(a)]. At 60°C, 3-DG decreased by 35 µmol/L and 5-HMF increased by 25 µmol/L during 21 days of incubation [Figures 2(b) and 2(c)]. Similar changes were observed at A

PERITONEAL DIALYSIS FLUIDS INCUBATED AT DIFFERENT TEMPERATURES FOR 21 DAYS

During incubation of a conventional PD fluid at 60°C, 3,4-DGE increased by about a factor of four, from 13 to 51 µmol/L, within 1 day. Thereafter it decreased slightly B A Downloaded from www.pdiconnect.com by on October 14, 2011

C B

Figure 1 — Pure water solutions of 100 µmol/L 3,4-dideoxyglucosone-3-ene (3,4-DGE; open squares) (A) and 200 µmol/L 3-deoxyglucosone (3-DG; open circles) (B) stored at 40°C for 30 days. 5-HMF = 5-hydroxymethyl-2-furaldehyde (open triangles). pH 5.5. Mean ± SEM; n = 3.

Figure 2 — Levels of 3,4-dideoxyglucosone-3-ene (3,4-DGE) (A), 3-deoxyglucosone (3-DG) (B), and 5-hydroxymethyl2-furaldehyde (5-HMF) (C) in conventional peritoneal dialysis fluid (2.5%) incubated at 25°C (closed squares), 40°C (closed triangles), and 60°C (closed circles) for 3 weeks. Mean ± SEM; n = 3.

TABLE 1 Pure Solutions of 3,4-Dideoxyglucosone-3-ene (3,4-DGE), 3-Deoxyglucosone (3-DG), and 5-Hydroxymethyl2-furaldehyde (5-HMF) Incubated at 40°C for 30 Days. Mean ± SEM; n = 3. Fluid Incubation days 3-DG (µmol/L) 3,4-DGE (µmol/L) 5-HMF (µmol/L) pH

3,4-DGE (100 µmol/L) 0 30 2±1 91±5 6±0 5.4

38±2 12±0 10±0 4.7

3-DG (200 µmol/L) 0 30 199±10 0 1±0 5.4

170±8 9±0 7±0 4.3

5-HMF (100 µmol/L) 0 30 0 0 94±1 5.8

0 0 101±3 4.9 585

ERIXON et al.

lower temperatures but these were slower and less pronounced (Figure 2). At 40°C, the concentration of 3,4-DGE increased by 5 µmol/L (representing a 38% increase) after 24 hours. Within 1 week, it had doubled and reached a maximum of 25 µmol/L. Incubation of a low GDP fluid at 60°C doubled the 3,4-DGE concentration from 0.4 to 0.8 µmol/L within 1 day of incubation. During the following 20 days, the 3,4-DGE concentrations decreased and ended up below the point of origin [Figure 3(a)]. Simultaneously, 3-DG decreased by 19 µmol/L and 5-HMF increased by 17 µmol/L [Figures 3(b) and 3(c)]. Again, the same


PDI

changes were observed at lower temperatures but they were slower and less pronounced (Figure 3). Figure 4 illustrates the overall results for 3,4-DGE when comparing 1-day storage of different fluids at different temperatures. The low GDP fluid contains less 3,4-DGE compared to the conventional fluids, regardless of storage conditions. The figure also demonstrates that an increase in storage temperature from 25°C to 40°C has a minor impact on the recruitment of 3,4-DGE compared to 60°C. CONVENTIONAL PD FLUID INCUBATED AT 60°C FOR 24 HOURS AND THEN RETURNED TO ROOM TEMPERATURE FOR 40 DAYS

During incubation of a conventional PD fluid for 24 hours at 60°C, the concentration of 3,4-DGE increased rapidly, from 14 to 49 µmol/L. During the same period, 3-DG decreased by only 19 µmol/L. When the fluids were returned to room temperature, 3,4-DGE decreased slowly and 3-DG increased, ending up at their original values but not until after 40 days [Figure 5(a)]. Cytotoxicity and 3,4-DGE concentrations increased and decreased in parallel [Figure 5(b)].

B

DISCUSSION

C

Figure 3 — Levels of 3,4-dideoxyglucosone-3-ene (3,4-DGE) (A), 3-deoxyglucosone (3-DG) (B), and 5-hydroxymethyl2-furaldehyde (5-HMF) (C) in fluid with low glucose degradation product (50% glucose) incubated at 25°C (open squares), 40°C (open triangles), and 60°C (open circles) for 3 weeks. Before analysis, the samples were diluted to a glucose concentration of 2.5%. Mean ± SEM; n = 3. 586

When a conventional PD fluid is incubated at high temperatures, the concentration of 3,4-DGE increases rapidly and doubles within a few hours. The concentration reaches a maximum of about 50 µmol/L within 24 hours. Such a 3,4-DGE concentration renders the fluid highly cytotoxic [Figure 5(b)]. When this fluid is returned to room temperature, the cytotoxicity and concentration of 3,4-DGE decrease slowly to original values. They are not reached, however, until after 40 days. Thus, even a few hours of careless storage might completely change the GDP profile in the fluids.

Figure 4 — A conventional 2.5% peritoneal dialysis fluid and a fluid with low glucose degradation product (GDP) incubated at different temperatures for 1 day. The glucose compartment in the low GDP fluid was diluted to 2.5% before analysis. 3,4-DGE = 3,4-dideoxyglucosone-3-ene. Mean ± SEM; n = 3.


A

PDI


GDPs IN PD FLUIDS

A


B

Figure 5 — Concentrations of 3,4-dideoxyglucosone-3-ene (3,4-DGE; closed circles) and 3-deoxyglucosone (3-DG; open circles) in a conventional peritoneal dialysis fluid (PDF) stored at 60°C for 24 hours, then brought to room temperature for 40 days (A). Axis for 3,4-DGE values is located to the left and 3-DG values to the right. Mean ± SEM, n = 3. Concentration of 3,4-DGE (shaded circles) and inhibition of cell growth (ICG; closed squares) in a conventional PDF stored at 60°C for 24 hours, then brought to room temperature for 40 days (B). Axis for 3,4-DGE values is located to the left and ICG values to the right. Mean ± SEM; n = 3 for 3,4-DGE; n = 4 for ICG.

Our experiments were performed in tubes under strict laboratory conditions in order to elucidate the temperature-dependent equilibrium among the different GDPs. Most likely these results will apply to real-life products, but we have no information to date on actual time and

temperature profiles of products transported and stored during the summertime in hot countries. Short incubation at lower temperatures, such as warming the fluid to 37°C before use, seems to be of less importance for the formation of 3,4-DGE. However, we 587

ERIXON et al.

588

PDI

this pool of the unknown molecule; this is then converted to 3-DG. We speculate (Figure 6) that the unknown molecule is 3-deoxyaldose-2-ene (3-DA). The existence of this intermediate was suggested by Anet in 1962 (19). The most likely reaction path from glucose to 3,4-DGE is glucose undergoing reversible 1,2-enolization and forming equilibrium with fructose. When the 1,2-enediol dehydrates, water is removed and 3-DA is formed. 3-DA, which can exist in open structure or in ring structure, is an enol (the hydroxy group is attached to a doublebonded carbon), and enols spontaneously rearrange to the isomeric carbonyl compound, in this case 3-DG (this process is called tautomerism). After further dehydration and removal of water, 3-DA forms 3,4-DGE in a reversible way. 3,4-DGE (open structure) can rearrange to a ring structure or can be dehydrated, forming 5-HMF in an irreversible way. It has been proposed that 5-HMF may degrade further, but in our experiments we have not been able to confirm this, as the concentration of 5-HMF did not change during incubation (Table 1) (19–24). In the low GDP fluid, the same equilibrium is found as in a conventional PD fluid. However, the total amount of GDPs is much lower. When comparing the concentrations in the low GDP fluid with those found in the conventional fluid, the difference is striking (Figure 4). We conclude that if a conventional PD fluid is exposed to a high temperature it may take less than 1 day to quadruple the 3,4-DGE concentrations and to reach cytotoxic levels. Temperatures capable of this may easily be reached in containers or in private cars during transport. Once the fluids have been exposed to such temperatures, it may take many weeks until the 3,4-DGE levels are back to normal again. The results indicate that 3-DG constitutes about half of the pool that 3,4-DGE is in equilibrium with. One possible candidate for the other half of the pool is 3-DA, the enol of 3-DG. The formation of 5-HMF seems to be an irreversible reaction. Thus, by using a low pH in the fluids (as in Gambrosol trio), not only is the total production of GDPs decreased, but the balance is driven in the direction away from the highly reactive 3,4-DGE and its pool molecules, 3-DG and 3-DA, toward the nonreactive 5-HMF. This will also provide the advantage that the total amount of GDPs will be so low that not even extremely careless handling will result in cytotoxic concentrations of GDPs. ACKNOWLEDGMENTS Gambro AB, Lund, Sweden, financed this study. Part of this work was presented as a poster at the 1st Joint ISPD/EuroPD Congress on Peritoneal Dialysis, 2004, Amsterdam, The Netherlands.


advise against longer storage periods at 37°C, as the concentration of 3,4-DGE will be doubled within 1 week’s incubation at 40°C. The overall recommendation is that PD fluids should be warmed in as short a time as possible prior to use. Microwaving would most likely be a good way of warming the fluid, as the temperature increase is fast. In small isolated volumes, as in the tubes, the temperature could be very high, causing recruitment of 3,4-DGE. Because of the small volumes, this will most likely only marginally affect the final concentration in the PD fluid. Peritonitis is still one of the major causes of hospitalization and dropout for PD. A higher rate has been reported during summertime (16–18). This has been suggested to be due to bacterial growth promoted by the temperature. An alternative explanation based on our results could be that the fluids have been subjected to high temperatures. Such temperatures might lead to a moderate increase in the concentration of 3,4-DGE, which in turn might impair local host defense reactions in the peritoneal cavity and open up for the possibility of microbial growth. A more intense temperature increase may, as demonstrated, lead to a several-fold increase in 3,4-DGE. This might cause acute chemical peritonitis. It has been reported for instance that a severe outbreak of sterile peritonitis was due to a threefold increase in breakdown products. When the patients were treated with fluids containing less degradation products the problems disappeared (9). In order to clarify how 3,4-DGE is formed in PD fluids, solutions of pure 3-DG, 3,4-DGE, and 5-HMF were subjected to temperature treatments. As seen in Figure 1, about 80 µmol/L 3,4-DGE was lost during a 30-day incubation. Only about half (47 µmol/L) of this was accounted for by the increases in 3-DG (which accounted for the major increase), 5-HMF, and aldehydes. Thus, roughly 50% of the 3,4-DGE was converted to 3-DG and 50% to some other substance that was neither 5-HMF nor any known aldehyde. When the conventional PD fluid was incubated at 60°C, there was a rapid increase in 3,4DGE (Figures 2 and 5). This increase was not paralleled by any rapid decrease in 3-DG or 5-HMF. On the contrary, while 3,4-DGE increased by about 40 µmol/L, there was only about a 20-µmol/L decrease in 3-DG. Thus, 3,4-DGE seems to have been formed from yet another GDP than 3-DG. On the other hand, the decrease in 3,4-DGE when the fluid was returned to 25°C seems to have been paralleled by an almost equal increase in 3-DG (Figure 5). Thus, 3,4-DGE seems to be rapidly recruited from a pool of an unknown GDP when the temperature is raised. The pool of 3-DG is drained slowly. When the temperature is again lowered, 3,4-DGE is supposedly pushed back into


PDI


GDPs IN PD FLUIDS

REFERENCES 1. Erixon M, Linden T, Kjellstrand P, Carlsson O, Ernebrandt M, Forsbäck G, et al. PD fluids contain high concentrations of cytotoxic GDPs directly after sterilization. Perit Dial Int 2004; 24:392–8. 2. Kjellstrand P, Erixon M, Wieslander A, Linden T, Martinson E. Temperature: the single most important factor for degradation of glucose fluids during storage. Perit Dial Int 2004; 24:385–91. 3. Nilsson-Thorell CB, Muscalu N, Andrén AHG, Kjellstrand PTT, Wieslander AP. Heat sterilization of fluids for peritoneal dialysis gives rise to aldehydes. Perit Dial Int 1993; 13:208–13. 4. Lindén T, Cohen A, Deppisch R, Kjellstrand P, Wieslander A. 3,4-Dideoxyglucosone-3-ene (3,4-DGE): a cytotoxic glucose degradation product in fluids for peritoneal dialysis. Kidney Int 2002; 62:697–703. 5. Lindén T, Forsbäck G, Deppisch R, Henle T, Wieslander A. 3-Deoxyglucosone, a promoter of advanced glycation end products in fluids for peritoneal dialysis. Perit Dial Int

1998; 18:290–3. 6. Schalkwijk C, Posthuma N, Brink H, ter Wee P, Teerlink T. Induction of 1,2-dicarbonyl compounds, intermediates in the formation of advanced glycation end-products, during heat-sterilization of glucose-based peritoneal dialysis fluids. Perit Dial Int 1999; 19:325–33. 7. Rippe B, Simonsen O, Heimbürger O, Christensson A, Haraldsson B, Stelin G, et al. Long-term clinical effects of peritoneal dialysis fluid with less glucose degradation products. Kidney Int 2001; 59:348–57. 8. Jonasson P, Braide M. A commercially available PD fluid with high pH and low GDP induces different morphological changes of rat peritoneum in intermittent PD. Adv Perit Dial 1998; 14:48–53. 9. Tuncer M, Sarikaya M, Sezer T, Özcan S, Suleymanlar G, Yakupoglu G, et al. Chemical peritonitis associated with high dialysate acetaldehyde concentrations. Nephrol Dial Transplant 2000; 15:2037–40. 10. Kjellstrand P, Martinsson E. Degradation in peritoneal dialysis fluids may be avoided by using low pH and high glucose concentration. Perit Dial Int 2001; 21:338–44. 589


Figure 6 — Glucose degrades when heat is applied during sterilization and subsequently during storage. In the degradation pathway, 3-deoxyaldose-2-ene (3-DA) is most probably formed, which is the enol of 3-deoxyglucosone (3-DG). After dehydration of 3-DA, 3,4-dideoxyglucosone-3-ene (3,4-DGE) is formed. This means that 3,4-DGE is in reversible equilibrium with 3-DA, which in turn produces 3-DG. After further dehydration of 3,4-DGE, 5-hydroxymethyl-2-furaldehyde (5-HMF) is formed. This reaction is irreversible and is favored at a low pH. 5-HMF may then be further degraded to other substances.

ERIXON et al.

590

PDI

17. Szeto CC, Chow KM, Wong FKM, Leung CB, Li PKT. Influence of climate on the incidence of peritoneal dialysisrelated peritonitis. Perit Dial Int 2003; 23:580–6. 18. Chan MK, Chan CY, Cheng IK, Ng WS. Climatic factors and peritonitis in CAPD patients. Int J Artif Organs 1989; 12: 366–8. 19. Anet EFLJ. Formation of furan compounds from sugars. Chem Ind 1962; 6:262. 20. Anet EFLJ. 3-Deoxyglycosuloses (3-deoxyglycosones) and the degradation of carbohydrates. Adv Carbohydr Chem Biochem 1964; 19:181–218. 21. Theander O, Nelson DA. Aqueous, high-temperature transformation of carbohydrates relative to utilization of biomass. Adv Carbohydr Chem Biochem 1978; 46:273–325. 22. van Dam HE, Kieboom APG, van Bekkum H. The conversion of fructose and glucose in acidic media: formation of hydroxymethylfurfural. Starch/Stärke 1986; 38(3):95–101. 23. Feather MS, Harris JF. Dehydration reactions of carbohydrates. Adv Carbohydr Chem Biochem 1973; 28:161–237. 24. Pigman W, Anet EFLJ. Mutarotations and actions of acids and bases. Biochem J 1972; 96:787–92.


11. Zimmeck T, Tauer A, Fuenfrocken M, Pischetsrieder M. How to reduce 3-deoxyglucosone and acetaldehyde in peritoneal dialysis fluids. Perit Dial Int 2002; 22:350–6. 12. Kato F, Mizukoshi S, Aoyama Y, Matsuoka H, Tanaka H, Nakamura K, et al. Immunosuppressive effects of 3,4dideoxyglucosone-3-ene, an intermediate in the Maillard reaction. J Agric Food Chem 1994; 42:2068–73. 13. Morgan LW, Wieslander A, Davies M, Horiuchi T, Ohta Y, Beavis J, et al. Glucose degradation products (GDP) retard remesothelialization independently of D-glucose concentration. Kidney Int 2003; 64:1854–66. 14. Penélope Catalan M, Santamaría B, Reyero A, Ortiz A, Egido J, Ortiz A. 3,4-Di-deoxyglucosone-3-ene promotes leukocyte apoptosis. Kidney Int 2005; 68:1303–11. 15. Wieslander AP, Andrén A, Martinson E, Kjellstrand P, Hultqvist M. Toxicity of effluent peritoneal dialysis fluid. Adv Perit Dial 1993; 9:31–5. 16. Kim MJ, Song JH, Park YJ, Kim GA, Lee SW. The influence of seasonal factors on the incidence of peritonitis in continuous ambulatory peritoneal dialysis in the temperate zone. Adv Perit Dial 2000; 16:243–7.
