Effect of Peritoneal Dialysis Solution Type on Serum Lipid Levels in End-Stage Renal Disease

Abstract

Among the different cardiovascular risk factors, lipid abnormalities dominate the high mortality in chronic ambulatory peritoneal dialysis patients. So far, no data comparing the effect of standard glucose-containing, amino acid-containing, and icodextrin-containing peritoneal dialysis solutions on serum lipid concentrations in a chronic ambulatory peritoneal dialysis population are available. To determine the effect of peritoneal dialysis solutions on parameters of lipid metabolism, 67 subjects who had continued their usual dialysis for the last six months were enrolled in the study. Group A consisted of 18 patients who were receiving only glucose-based peritoneal dialysis solutions, group B consisted of 18 patients who were receiving glucose and amino acid-based peritoneal dialysis solutions, and group C consisted of 31 patients who were receiving glucose and icodextrin-based peritoneal dialysis solutions. Serum lipid parameters including total cholesterol, low-density lipoprotein, high-density lipoprotein, triglyceride, and lipoprotein (a) were determined in all groups. No significant difference in serum lipid parameters was found between groups A, B, and C. These results demonstrate the lack of the effect of amino acid or icodextrin-based peritoneal solutions on dyslipidemia of CAPD patients.

Keywords:

• continuous ambulatory peritoneal dialysis
• high-density lipoprotein
• lipoprotein (a)
• low-density lipoprotein
• triglyceride
• total cholesterol

INTRODUCTION

Continuous ambulatory peritoneal dialysis (CAPD) has become an alternative therapy to hemodialysis in a growing proportion of the dialysis population.[1] Davies et al. have reported a five-year survival of 50–70% for CAPD, irrespective of the underlying disease.[2] Because cardiovascular complications are the major cause of morbidity and mortality in patients on dialysis,[3–6] risk factors for these events have become increasingly important as the possibility of longer survival on renal replacement therapy increases. Among the different cardiovascular risk factors, lipid abnormalities dominate the high mortality in CAPD patients.[1],[7] There are several studies demonstrating elevated cholesterol levels, hypertriglyceridemia, and low high-density lipoprotein (HDL) in CAPD patients compared to the general population. It has been suggested that CAPD patients may be at a greater risk from vascular disease in view of a combination of hypercholesterolemia and a low HDL.[8]

The mechanisms underlying altered lipid metabolism in CAPD patients are not completely understood. As hyperlipidemia is found in nephrotic syndrome and CAPD, it may relate to the massive protein loss that accompanies these conditions, leading to hepatic oversynthesis of lipoproteins. Also the excessive peritoneal glucose absorption from highly concentrated glucose-containing CAPD solutions may enhance metabolic disturbances.[9],[10] A reduction in cholesterol, triglyceride (TG), and low-density lipoprotein (LDL) levels has been reported based upon treatment with icodextrin.[11] Amino acid-based dialysate is also thought to reduce glucose-related elevations of lipids.[12]

So far, no data comparing the effect of standard glucose-containing, amino acid-containing, and icodextrin-containing peritoneal dialysis (PD) solutions on serum lipid concentrations in a CAPD population is available. This study was designed with the objective of determining the effects of peritoneal dialysis solutions on parameters of lipid metabolism.

MATERIALS AND METHODS

Sixty-seven patients—33 males and 34 females, aged 20–81 years (mean 44 years)—were enrolled to the study after giving informed consent. The primary renal diagnoses were: diabetic nephropathy (13 patients), hypertension (19 patients), chronic glomerulonephritis (17 patients), amyloidosis (6 patients), nephritic syndrome (1 patient), polycystic kidney disease (1 patient), and urolithiasis (1 patient), while the diagnosis was unknown in 9 patients. The following exclusion criteria were used: significant hepatic dysfunction (defined as aspartate aminotransferase or gamma glutamyl transferase greater than twice the upper limit of normal), other causes of hypoalbuminemia (i.e., malignancy, malabsorption, and serious infection), abnormal thyroid function tests, unstable clinical condition, lipid-lowering therapy during the preceding three months, and beta blocker therapy.

A cross-sectional study design was used. Patients who had continued their usual dialysis with standard glucose-containing (1.36%, 2.27%, and 3.86%, Dianeal, Eczacibasi-Baxter, Istanbul), amino acid-containing (Nutrineal, Eczacibasi-Baxter, Istanbul), or icodextrin-containing (Extraneal, Eczacibasi-Baxter, Istanbul) PD solutions for the last six months were included in the study and divided into three groups. Group A consisted of 18 patients who were receiving only glucose-based PD solutions, group B consisted of 18 patients who were receiving glucose- and amino acid-based PD solutions, and group C consisted of 31 patients who were receiving glucose- and icodextrin-based PD solutions.

Venous blood samples were collected for serum lipid and lipoprotein measurements after an overnight fast, before the first exchange in the morning. Biochemical parameters including urea, creatinine, calcium, phosphor, albumin, total cholesterol, triglyceride, and HDL concentrations were determined by colorimetric method on a Hitachi 912 analyzer. Low-density lipoprotein cholesterol was calculated using the Friedewald formula. Apoprotein A1, apolipoprotein B, and lipoprotein(a) [(Lp(a)] were measured by immunochemical assay using a Beckman Coulter Nephelometer (Beckman Coulter Inc, Immage Biochemistry Systems, Fullerton, California, USA).

The study was approved by the Fatih University School of Medicine Ethics Committee and was conducted in accordance with the ethical principles described by the Declaration of Helsinki.

Statistical Analysis

All statistical analyses were performed using the SPSS program, version 11.5 (SPSS Inc, Chicago, Illinois, USA). Parameters are given as mean±standard deviation. Kruskal Wallis test was used, with the level of significance set at p<0.05.

RESULTS

Sixty-seven patients on CAPD were entered into the study with a mean age of 44±15 years (20–81 years). There were 34 women (50.7%) and 33 men (49.3%). Demographic and outcome data are presented in Table 1. Group A consisted of 18 patients (6 women, 12 men) who were receiving only glucose-based PD solutions with a mean age of 40.8±16.2 years, group B consisted of 18 patients (14 women, 4 men) who were receiving glucose- and amino acid-based PD solutions with a mean age of 39±12.4 years, and group C consisted of 31 patients (14 women, 17 men) who were receiving glucose- and icodextrin-based PD solutions with a mean age of 49±16.1 years. There was no significant difference between the ages of the groups (p > 0.05).

Table 1 Demographic and outcome data of groups A, B, and C

Group (n)	A (18)	B (18)	C (31)	p
Age (years)	40.8 ± 16.2	39 ± 12.4	49 ± 16.1	>0.05
Height (m)	1.6 ± 0.1	1.6 ± 0.04	1.6 ± 0.1	>0.05
Weight (kg)	59.9 ± 9.6	58.6 ± 9.6	69.3 ± 14.2	>0.05
Gender (female/male)	6/12	14/4	14/17	>0.05
CAPD time (months)	18.4 ± 13.5	17.4 ± 12.3	18.5 ± 14	>0.05
CAPD glucose concentration (gr/day)	157 ± 38.5	95.1 ± 20.4	98.5 ± 24.6	<0.05
Urea [SI unit (mg/dL)]	40.8 ± 12.4 (114.4 ± 34.8)	35.9 ± 5.9 (100.5 ± 16.6)	37.8 ± 12.2 (106 ± 34.1)	>0.05
Creatinine [SI unit (mg/dL)]	686 ± 290 (9.0 ± 3.8)	602 ± 137 (7.9 ± 1.8)	572 ± 214 (7.5 ± 2.8)	>0.05
Calcium [SI unit (mg/dL)]	2.42 ± 0.2 (9.7 ± 0.8)	2.2 ± 0.12 (8.8 ± 0.5)	2.35 ± 0.25 (9.4 ± 1.0)	>0.05
Phosphor [SI unit (mg/dL)]	1.61 ± 0.84 (5.0 ± 2.6)	1.39 ± 0.29 (4.3 ± 0.9)	1.42 ± 0.36 (4.4 ± 1.1)	>0.05
Albumin [SI unit (g/dL)]	36 ± 3 (3.9 ± 0.3)	31 ± 7 (3.1 ± 0.7)	36 ± 5 (3.6 ± 0.5)	>0.05
Parathormone [SI unit (pg/mL)]	443 ± 360 (443.4 ± 360)	367 ± 198 (367 ± 197.8)	318 ± 239 (317.6 ± 238.6)	>0.05
Ferritin [SI unit (ng/mL)]	339 ± 185 (339.3 ± 185.2)	490 ± 347 (490.1 ± 347.2)	395 ± 347 (395.1 ± 347.2)	>0.05

Abbreviation: SI = systeme international.

CAPD glucose concentration per day was higher in group A than groups B and C (157±38.5 gr/day, 95.1±20.4 gr/day and 98.5±24.6 gr/day, respectively, p < 0.05). Serum albumin levels in groups A, B, and C were 36±3 SI unit (3.6±0.3 g/dL), 31±7 SI unit (3.1±0.7 g/dL), and 36±5 SI unit (3.6±0.5 g/dL), respectively (p > 0.05). No significant difference was observed in serum levels of urea, creatinine, calcium, and phosphor between the groups (p > 0.05). Serum total cholesterol (TC), LDL cholesterol, HDL cholesterol, TG, and Lp(a) levels are shown in Table 2. No significant difference was found in serum lipid parameters between groups A, B, and C (p > 0.05).

Table 2 Parameters of lipid metabolism in groups A, B, and C

Group	A	B	C	p
TC [SI unit (mg/dL)]	4.97 ± 1.62 (192.2 ± 62.8)	4.09 ± 1.31 (158.2 ± 50.6)	4.93 ± 1.17 (190.5 ± 45.4)	>0.05
LDL-C [SI unit (mg/dL)]	2.38 ± 0.95 (92.1 ± 36.9)	2.62 ± 0.89 (101.4 ± 34.3)	2.77 ± 0.81 (107.7 ± 31.3)	>0.05
HDL-C [SI unit (mg/dL)]	1 ± 0 (38.1 ± 13.6)	1 ± 0 (33.5 ± 5.4)	1 ± 0 (36.9 ± 7.7)	>0.05
TG [SI unit (mg/dL)]	2.14 ± 1.43 (189.3 ± 126.4)	1.48 ± 0.52 (131.4 ± 46.1)	2.16 ± 1.23 (191.6 ± 108.8)	>0.05
APO-A [SI unit (mg/dL)]	0 (1.1 ± 0.3)	0 (0.9 ± 0.2)	0 (1.0 ± 0.2)	>0.05
APO-B [SI unit (mg/dL)]	0 (1.0 ± 0.4)	0 (1.1 ± 0.4)	0 (1.1 ± 0.3)	>0.05
Lp(a) [SI unit (mg/dL)]	1.98 ± 1.65 (55.4 ± 46.2)	1.66 ± 1.4 (46.6 ± 39.1)	1.17 ± 1.1 (32.9 ± 30.0)	>0.05
NONHDL [SI unit (mg/dL)]	3.65 ± 1.53 (141.2 ± 59)	3.23 ± 1.18 (125.0 ± 45.8)	3.88 ± 1.16 (150.0 ± 44.9)	>0.05

Abbreviations: SI = systeme international, TC = total cholesterol, LDL-C = low-density lipoprotein cholesterol, HDL-C = high-density lipoprotein cholesterol, TG = triglyceride, APO-A = apolipoprotein A, APO-B, apolipoprotein B, Lp(a) = lipoprotein (a).

DISCUSSION

Because of reported changes in glucose and lipid metabolism involving the high cardiovascular morbidity and mortality in CAPD, the effect of glucose, icodextrin, and amino acid solutions on cardiovascular risk factors was investigated. The etiology of dyslipidemia in patients on chronic CAPD is complex.[13],[14] Apart from the general abnormalities of lipid metabolism associated with uremia, glucose absorption from the dialysis fluid may theoretically contribute to dyslipidemia, particularly to hypertriglyceridemia, with lipogenesis induced through a hyperinsulinemic response. In contrast to intraperitoneal glucose solutions, maltose and related metabolites of icodextrin are reported not to affect glucose metabolism or lead to hyperinsulinemia.[15]

The MIDAS study on efficacy and safety of icodextrin in CAPD demonstrated an overnight ultrafiltration achieved by icodextrin equal to a 3.86% glucose solution.[11] Also, long-term use of icodextrin had no detrimental influence on peritoneal ultrafiltration capacity,[11] and the routine use of icodextrin did not deteriorate the peritoneal defense mechanism.[16] Because of its demonstrated efficacy and safety, icodextrin has been increasingly used for objectives other than improved ultrafiltration.

A reduction in cholesterol, triglyceride, and LDL levels has been reported based upon six months of treatment with icodextrin.[11] Sebastian et al. also reported a beneficial effect of icodextrin on lipid profiles in a six-week study.[17] On the other hand, serum lipid levels of our patients on icodextrin-based PD were not different from amino acid- or glucose-based PD.

Amino acid-based dialysate is also thought to reduce glucose-related elevations of lipids. Oreopoulos et al.[12] first evaluated the use of amino acid dialysate as an alternative to glucose to reduce the hyperlipidemia seen in patients on CAPD. Since then, variable effects on serum lipids have been reported with amino acid-based dialysis. Two studies reported a decrease in plasma cholesterol.[18],[19] In one of these studies, there was no change in serum triglycerides over 12 weeks,[18] whereas a rise in serum triglycerides has been reported by another study.[20] In the present study, although serum triglyceride and cholesterol levels of patients on amino acid-based PD were lower than glucose-based PD, this was not significant. These observations are in partial agreement with Bruno et al.,[19] who noted significant lower serum triglyceride and total cholesterol levels. Misra et al. have reported that the use of an amino acid-based PD had no effect on dyslipidemia in a group of CAPD patients.[21] Dibble et al. reported a decrease in LDL cholesterol and Apo B (>85% of which is the sole lipoprotein component of LDL particles) during the use of amino acid-based PD. On the other hand, HDL cholesterol, ApoA, triglyceride, and VLDL levels remained unchanged.[22] In the current study, no significant change was observed in LDL cholesterol, HDL cholesterol, non-HDL cholesterol, ApoA, and ApoB levels between the three groups. The reasons for variance in these studies are hard to define. Young et al.[18] analyzed eight patients with a serum albumin of 35 gr/L, whereas Bruno et al. studied six patients who were overweight and Misra et al. selected their patients (n = 18) on the basis of their serum cholesterol rather than their nutritional status.[21] Patients in the present study were selected according to their PD type, and no difference was observed according to their nutritional status. Also, the amino acid-based PD fluid used in this study differed in chemical composition from that of the fluid used by Young et al.[18] and Bruno et al.[19] in that it had a higher (1.1%) amino acid content, a higher osmolality (365 mosm/L), and was more biocompatible (lactate content 40 mmol/L, pH 6.7). These findings are in agreement with four other smaller studies.[23–26] In this study, if the patient number of the groups was higher, at least the triglyceride levels of group A and group B would be significantly different from each other.

Lipoprotein(a) is a genetically determined risk factor for atherosclerotic vascular disease. Several studies have described a correlation between high Lp(a) plasma levels and coronary heart disease, stroke, and peripheral atherosclerosis.[27] Shoji et al. have found the median Lp(a) level to be significantly higher in a group of CAPD patients with ischemic heart disease.[28] Markedly elevated Lp(a) levels in CAPD patients have been reported.[2],[29],[30] A correlation between Lp(a) and peritoneal glucose absorption was reported, indicating that the increased plasma Lp(a) levels in CAPD may be related to the dialysis procedure, particularly peritoneal transport of proteins and glucose. In this study, no significant difference in Lp(a) levels between the three groups was observed. Misra et al. have also reported that serum Lp(a) was not correlated to PD solution.[21]

In conclusion, this study demonstrates the lack of effect of amino acid or icodextrin dialysate on the dyslipidemia of CAPD patients. It also suggests that hyperlipidemia, which is so common in CAPD patients, may not be solely attributed to the dialysate type alone.