Hyperinsulinism Reduction Associated with Icodextrin Treatment in Continuous Ambulatory Peritoneal Dialysis Patients

Gianpaolo Amici, Massimo Orrasch,1 Giorgio Da Rin,2 Carlo Bocci
From: Nephrology, 1Diabetology, and 2Laboratory, Regional Hospital S. Maria dei Battuti, Treviso, Italy.

Glucose absorption from peritoneal dialysis solutions causes a chronic stimulation of insulin secretion, which leads to hyperinsulinism. The use of solutions without glucose should correct this metabolic derangement together with the associated cardiovascular risk.
To verify this hypothesis, we studied the entire non diabetic continuous ambulatory peritoneal dialysis (CAPD) population of our center: 27 patients with a mean age of 62 ± 15 years, and a median 17 months on treatment. Morning fasting serum insulin was 32.8 ± 9.3 mU/mL; glucose, 104.4 ± 21.8 mg/dL; triglycerides, 162.4 ± 125.7 mg/dL; cholesterol, 221.9 ± 54.7 mg/dL; intact parathyroid hormone (iPTH), 212 ± 189 pg/mL; fibrinogen, 519 ± 112 mg/dL; body mass index, 24.1 ± 4.1; and daily erythropoietin subcutaneous therapy dose, 17 ± 6 U/kg. Insulin sensitivity, measured as ISI-HOMA (insulin sensitivity index, derived from the homeostasis model assessment) was 2.4 ± 0.7. Daily glucose load, calculated from dialytic schedules, was 135 ± 38 g.
Of the 27 patients, 12 were treated with standard glucose solutions during the day and with one icodextrin dwell during the night for a median of 9 months (range: 1 – 28). The remaining 15 patients were treated with standard glucose solutions.
The icodextrin group showed significantly lower serum insulin levels (28.6 ± 6.0 mU/mL vs 36.1 ± 10.2 mU/mL, p = 0.021) and significantly higher ISI-HOMA values (2.7 ± 0.5 vs 2.2 ± 0.7, p = 0.041) than the control group. The two groups showed no significant differences for glucose, triglycerides, cholesterol, iPTH, fibrinogen, body mass index, or erythropoietin therapy dose. Daily glucose load was lower in the icodextrin group, but without reaching statistical significance (128 ± 31 g vs 142 ± 43 g).
This study shows, in a preliminary way, that the chronic use of icodextrin in the long nighttime dwell can reduce serum insulin levels and increase insulin sensitivity in CAPD patients.

Key words

Introduction

Chronic renal failure is commonly associated with an increased insulin resistance that dialytic therapy reduces, but does not completely normalize (1–4). In continuous ambulatory peritoneal dialysis (CAPD), the baseline plasma insulin level is increased, and, at every exchange with glucose solution, a marked insulin response occurs (5,6). The repeated absorption of pure glucose from standard peritoneal dialysis solutions causes a chronic stimulation of insulin secretion in non diabetic patients (1,4). In these patients, this condition can lead to stable elevated plasma insulin levels and, consequently, to reduced tissue insulin sensitivity owing to downregulation of tissue receptor expression (7). This condition is generally known as hyperinsulinism, or insulin resistance. It probably represents a pre-diabetic state (7).
Elevated plasma insulin levels—together with impaired glucose tolerance, central obesity, hyperlipidemia, and hypertension—describe a metabolic derangement strongly linked to atherosclerosis (8). The introduction of new solutions with alternative osmotic agents now offers new opportunities for optimal prescription in CAPD, with a relevant reduction in glucose load (4,9–11). In particular, the use of icodextrin-based PD solutions should correct the metabolic derangement and the associated cardiovascular risk (12). To verify this hypothesis, we studied the glucose–insulin system of the entire non diabetic CAPD population of our center, comparing in particular the group of patients chronically treated with icodextrin solution during the nightly dwell with the group of patients treated with standard glucose solutions alone.

Patients and Methods

This cross-sectional study included the entire non diabetic CAPD population of our center: 27 patients, aged 62 ± 15 years, on dialytic treatment for a median 17 months (range: 1 – 107 months), body weight 67.6 ± 14.6 kg, height 1.67 ± 0.10 m, and body mass index 24.1 ± 4.1. Diabetes was excluded in these subjects by the absence of stable blood glucose levels above 140 mg/dL or by the absence of insulin or hypoglycemic drugs in their therapy regimens, or both. Patients with acute pathologic conditions such as peritonitis and fluid overload were excluded.
In all patients, a blood sample was taken in baseline conditions: that is, the morning after 12 hours of fasting, 2 hours after the first (standard glucose) exchange of the CAPD day. Serum concentrations of insulin, glucose, triglycerides, total cholesterol, intact parathyroid hormone (iPTH), and fibrinogen were measured. Insulin was determined by the solid-phase two-site chemiluminescent enzyme-labeled immunometric assay (Immulite: Diagnostic Products, Los Angeles, CA, U.S.A.). By this method, the 95% range for serum insulin in normal subjects is 6 – 27 mU/mL (interassay variability: 5.4%; assay range: 2 – 400 mU/mL). The insulin sensitivity index (ISI) derived from the homeostasis model assessment (HOMA) was then calculated using the baseline insulin and glucose serum concentrations and the simplified formula

ISI-HOMA = k / ( FPG × FPI )

where k = 22.5 × 18 (derived from HOMA); FPG = fasting plasma glucose (mmol/L); and FPI = fasting plasma insulin (mU/mL) (13–15). Approximate peritoneal glucose load was calculated from dialytic schedules, ignoring the percentage of effective absorption.
Across the entire group, 25 patients were being treated with subcutaneous recombinant human erythropoietin at a daily dose of 17 ± 6 U/kg. Twelve patients (44%) were treated with standard glucose solutions during the daytime and with one icodextrin dwell nightly for a median of 9 months (range: 1 – 28). The remaining 15 patients (56%; control group) were treated with standard glucose solutions for all exchanges.
Data are expressed as mean ± standard deviation, median and range, or percentage depending on variables and distributions. For data analysis, the unpaired Student t-test, the Wilcoxon rank sum test, and the Fisher exact test have been applied where appropriate. The null hypothesis was refused for all tests for two-tailed alpha values lower than 0.05. The software package JMP 3.2.2 (SAS Institute, Inc., Cary, NC, U.S.A.) was used for statistical analysis.

Results

Across the entire group, serum morning fasting insulin was 32.8 ± 9.3 mU/mL; glucose, 104.4 ± 21.8 mg/dL; triglycerides, 162.4 ± 125.7 mg/dL; cholesterol, 221.9 ± 54.7 mg/dL; iPTH, 212 ± 189 pg/mL; fibrinogen, 519 ± 112 mg/dL. Insulin sensitivity, measured as ISI-HOMA, was 2.4 ± 0.7, and daily glucose load, calculated from dialytic schedules, was 135 ± 38 g, corresponding to 556 ± 156 kcal daily. These overall data showed a hyperinsulinemic condition [20 subjects (74.1%) with serum insulin above 27 mU/mL], without evident hyperglycemia and dyslipidemia. Moreover, the ISI-HOMA index showed reduced tissue (liver and muscle) insulin sensitivity. Only 6 patients (22.2%) were overweight (BMI > 27). Twelve patients (44.4%) were hypertensive without significant association with serum insulin levels. All patients showed elevated plasma fibrinogen values without any correlation with hyperinsulinemia, hypertension, and overweight.
The icodextrin-treated group showed significantly lower serum insulin levels and significantly higher ISI-HOMA values than the control group (Figure 1). The difference in the approximate glucose load between the two groups was statistically nonsignificant, but the icodextrin-treated group showed lower values, as expected (Table I). The two groups showed similar values for serum glucose, triglycerides, cholesterol, iPTH, fibrinogen, body mass index, hypertension, and erythropoietin therapy dose (Table I).

Discussion

Glucose solutions have been successfully used for many years in CAPD, but the well-known drawbacks of glucose as osmotic agent are the systemic metabolic effects and low peritoneal biocompatibility (11). The absorption of glucose from the peritoneal cavity is generally between 60% and 80% in 6 hours, resulting in glucose absorption of approximately 100 – 300 g daily (4). This fast and considerable carbohydrate absorption may be linked to metabolic complications such as hyperglycemia, hyperinsulinemia, hyperlipidemia, and obesity (12). The hyperinsulinemic condition in CAPD patients reported by many researchers (1,4–6) and confirmed by the present study constitutes an independent cardiovascular risk factor (16).
The introduction of alternative osmotic agents is changing the therapeutic prospects for this type of dialysis, by reducing glucose load, improving ultrafiltration, and extending the durability of the peritoneal membrane (4,9,10,12). In particular, an initial finding of low insulin secretion stimulation by glucose polymer solutions (6) has also been reported in large studies (12), corroborating the difference in serum insulin level and insulin sensitivity seen between the icodextrin and non icodextrin chronically treated patients in our study (Figure 1).
Glucose polymers, like most substances, are subject to peritoneal absorption by lymphatic and non lymphatic pathways. The absorption of the polymer into the bloodstream, expressed in terms of carbohydrate load, is low and distributed progressively over a long dwell time (6,12). But the explanation for the absent insulin secretion stimulation with icodextrin solutions (6,12) is the complex carbohydrate structure and the consequent slow metabolism of the absorbed molecules.

figure 1 Mean baseline serum insulin and ISI-HOMA values (insulin sensitivity index derived from the homeostasis model assessment) in the icodextrin-treated and control groups. Significantly lower serum insulin and higher insulin sensitivity were seen in the icodextrin-treated group.

Conclusion

The results of our cross-sectional study show, in a preliminary way, the clear metabolic advantages of using chronic icodextrin in the long nighttime dwell in CAPD. The reduction or elimination of free glucose load through the use of alternative solutions is a promising strategy for reducing hyperinsulinism and the associated cardiovascular risk in CAPD patients.

Acknowledgment

Many thanks to Marta Tenan for data collection and to Elizabeth Tomlin for manuscript translation.

References

1. Castellino P, Luzi L, Giordano M, DeFronzo RA. Effect of insulin and amino acids on glucose and leucine metabolism in CAPD patients. J Am Soc Nephrol 1999; 10:1050–8.
2. Kobayashi S, Maejima S, Ikeda T, Nagase M. Impact of dialysis therapy on insulin resistance in end-stage renal disease: comparison of haemodialysis and continuous ambulatory peritoneal dialysis. Nephrol Dial Transplant 2000; 15:65–70.
3. DeFronzo RA, Alvestrand A, Smith D, Hendler R, Hendler E, Wahren J. Insulin resistance in uremia. J Clin Invest 1981; 67:563–8.
4. Holmes CJ, Shockley TR. Strategies to reduce glucose exposure in peritoneal dialysis patients. Perit Dial Int 2000; 20(Suppl 2):S37–41.
5. De Santo NG, Capodicasa G, Senatore R, et al. Glucose utilization from dialysate in patients on CAPD. Int J Artif Organs 1979; 2:119–25.
6. MD Thesis. Mistry CD. Glucose polymer as an osmotic agent in continuous ambulatory peritoneal dialysis. London: University of London; 1988: 1–300.
7. American Diabetes Association. Consensus Development Conference on Insulin Resistance. 5–6 November 1997. Diabetes Care 1998; 21:310–314.
8. Reaven GM. Banting lecture 1988. Role of insulin resistance in human disease. Diabetes 1988; 37:1595–607.
9. Ho-dac-Pannekeet MM, Weiss MF, de Waart DR, Erhard P, Hiralall JK, Krediet RT. Analysis of non enzymatic glycosylation in vivo: impact of different dialysis solutions. Perit Dial Int 1999; 19(Suppl 2): S68–74.
10. Krediet RT. Prevention and treatment of peritoneal dialysis membrane failure. Adv Ren Replace Ther 1998; 5:212–17.
11. Gokal R. Peritoneal dialysis solutions: nutritional aspects. Perit Dial Int 1997; 17(Suppl 3):S69–72.
12. Mistry CD, Gokal R, Peers E, and the Midas Study Group. A randomized multicenter clinical trial comparing isoosmolar icodextrin with hyperosmolar glucose solutions in CAPD. Kidney Int 1994; 46:496–503.
13. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic clamp. Diabetes Care 1999; 22:1462–70.
14. Turner RC, Holman RR, Matthews D, Hockaday TDR, Peto J. Insulin deficiency and insulin resistance interaction in diabetes: estimation of their relative contribution by feedback analysis from basal plasma insulin and glucose concentration. Metabolism 1979; 28:1086–96.
15. Matthews D, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28:412–19.
16. Prichard S. Major and minor risk factors for cardiovascular disease in peritoneal dialysis. Perit Dial Int 2000; 20(Suppl 2):S154–9.

Corresponding author:

Gianpaolo Amici, md, Nephrology and Dialysis Division, Regional Hospital S. Maria dei Battuti, 1 Piazzale Ca’Foncello, Treviso I-31100 Italy.