Long-Term Impact of Chronic Hemodialysis on Glycemic Control and Serum Lipids in Insulin-Treated Type 2-Diabetic Patients

Abstract

There are only a few data in the literature concerning metabolic control in insulin-treated diabetic patients with end stage renal disease (ESRD). The aim of the study was to find out the long-term impact of hemodialysis on glycemic control and lipid values in type 2 diabetic patients. Twenty insulin-treated type 2 diabetic patients (age 62 ± 9 years, f:m = 6:14) were evaluated. We compared HbA1c, fasting blood glucose (FBG), body weight, serum lipids, insulin requirement, and blood-pressure (BP) 12 and 6 months before dialysis, at the start of dialysis, and 6 as well as 12 months after the start. Results: The mean HbA1c- and FBG-values were not significantly different before and after the start of dialysis therapy. The average insulin requirement was 26 ± 10 IU/day in the predialysis period, 25 ± 12 IU/day at the start, and 24 ± 13 as well as 22 ± 13 IU/day after the start of dialysis. The mean cholesterol level fell significantly from 199 ± 63 and 190 ± 49 mg/dL in the predialysis phase to 167 ± 62 and 157 ± 38 mg/dL after dialysis began. The triglyceride concentrations decreased only slightly after the start of dialysis. The incidence of hypoglycemia (n/patient/month) was markedly lower in the predialysis phase (0.4 vs. 0.6, NS) than after start of dialysis. In patients with residual diuresis (< 500 mL urine/day) the needed insulin doses decreased significantly by 29% compared to patients with higher residual diuresis, whose insulin requirement remained unchanged. In summary, hemodialysis had no significant long-term effect on glycemic control in insulin-treated type 2 diabetic patients, but incidence of hypoglycemia tended to be higher under hemodialysis than in the predialysis period. Lipid levels tended to be lower after the initiation of dialysis therapy. Insulin requirement under hemodialysis decreased only in patients with loss of residual urine volume (below 500 mL urine/day).

• chronic hemodialysis
• glycemic control
• lipid levels

INTRODUCTION

Diabetes mellitus is the leading cause of end-stage renal disease (ESRD) worldwide. The survival rate is still poor in patients with ESRD caused by diabetes mellitus, especially in type 2 diabetic patients.[1] There are few data in the literature concerning the impact of hemodialysis on metabolic control in insulin-treated diabetic patients with ESRD, although good glycemic control during the first months of dialysis predicts a better survival for type 2 diabetic patients.[2] We compared the glycemic control, the insulin requirement, and the lipid levels 12 months before and 12 months after the start of hemodialysis in insulin-treated diabetic patients with diabetic nephropathy. The aim of the study was to find out the long-term impact of hemodialysis on metabolic control in type 2 diabetic patients with ESRD.

PATIENTS AND METHODS

Twenty insulin-treated type 2 diabetic patients (age 62 ± 9 years, f:m = 6:14, BMI 24 ± 2) were evaluated in this retrospective study. All patients received an intensified conventional insulin therapy with 2–3 insulin injections/day. We selected only those patients who were in control, who had been seen in our outpatient care unit for at least one year before the start of dialysis, and who had been on hemodialysis one year and more. The diagnosis ofdiabetic nephropathy was based on evidence of proteinuria (> 0.5 g protein/24 h urine), normal urine sediment, sonographically normal kidney size, and long-acting diabetes (> 10 years) with concomitant retinopathy at the start of dialysis.

We compared HbA1c, fasting blood glucose (FBG), body weight, cholesterol, triglycerides (Hitachi autoanalyzer), insulin requirement, and blood pressure (BP) 12 and 6 months before dialysis, at the start of dialysis, and 6 as well as 12 months after the start. The HbA1c and FBG were additionally measured 4 weeks after initiating dialysis (Figure 1).

Figure 1. HbA1c and fasting blood glucose (FBG) before and after the start of dialysis.

Moreover, we evaluated the frequency of severe hypoglycemia (with needing help, n/patient/month) using a patient questionnaire. All patients measured blood glucose at least 3 times daily by self monitoring. All patients were dialyzed 3 times a week, for a total of 12 to 15 hours per week. At the start of dialysis period we also measured serum albumin. In all patients hemodialysis was performed with a low-flux dialyzator using a dialysis fluid with 100 mg/dL glucose concentration. Baseline data is summarized in Table 1.

Table 1 Baseline data of the patients at the start of dialysis

Age (years)	62 ± 9
Sex (m:f)	14:6
BMI	24 ± 2
Diabetes (years)	16 (10–22)
C-peptid basal (ng/mL)	2.1 ± 1.1
HbA1c (%)	7.3 ± 1.2
Fasting blood glucose (mg/dL)	151 ± 47
Serum albumin (g/L)	34.46 ± 6.42
Insulin injections (n/day)	2 (1–3)
Insulin requirement (IU/day)	24 ± 12

Data are given as Mean ± SD or range.

In addition, the patients were divided into two groups: those with maintained diuresis during the first year after the start of dialysis (residual urine volume 0.5 L/day) and those with significantly decreased diuresis (urine volume < 0.5 L/day). In both patient groups the needed insulin doses and HbA1c-values were compared to evaluate the influence of residual diuresis on the insulin requirement under hemodialysis therapy.

For statistical calculations the SPSS for Windows statistical program was used. The Student's t-test was used for comparing continuous variables, and the Qui-test was performed for dichotomous variables. For comparison of differences between the groups with data not normally distributed, the Wilcoxon test was used. A p value of less than 0.05 was considered statistically significant.

RESULTS

The insulin requirement for our survey patients was 26 ± 10 and 26 ± 12 IU/day in the predialysis period, 25 ± 12 IU/day at the start, and 24 ± 13 as well as 22 ± 13 IU/day after the start of dialysis. The decrease of needed insulin doses from the start until one year after the start of dialysis was 8% (NS). Moreover, there was no statistically significant difference before and after the start of dialysis therapy with respect to HbA1c and FBG as well as insulin requirement. Total cholesterol fell significantly from 215 ± 59, 199 ± 63, and 190 ± 49 mg/dL in the predialysis phase to 167 ± 62 and 157 ± 38 mg/dL after initiating dialysis (p < 0.05). The triglyceride concentrations did notdecrease significantly after the start of dialysis. Twenty-five percent vs. 35% of patients received states therapy (NS) during both periods.

The incidence of hypoglycemia (n/patient/month) was markedly lower in the predialysis phase (0.4 vs. 0.6, NS) than it was after start of dialysis. The hypoglycemic episodes occurred at any time during the day. After the start of chronic dialysis therapy 33% of all hypoglycemic events occurred during or within 3 hours after hemodialysis. There was no difference between the frequency of hypoglycemia in patients with dialysis in the morning or later in the day. The patients with Body Mass Index (BMI) > 26 (n = 6) had a higher insulin requirement than did patients with BMI < 26 (n = 14) with 24 ± 14 vs. 19 ± 12 IU of insulin per day. There was no correlation between serum albumin concentration and needed insulin dose.

The mean morning BP values were similar before and after the start of hemodialysis. All patient data are shown in Table 2.

Table 2 HbA1c, fasting blood glucose (FBG), insulin requirement, cholesterol, triglycerides, and incidence of hypoglycemia before and after the start of hemodialysis

	− 12 months	− 6 months	Start of dialysis	+ 6 months	+ 12 months
HbA1c (%)	7.4 ± 1.8	7.3 ± 1.2	7.3 ± 1.3	7.4 ± 1.2	7.3 ± 1.1
FBG (mg/dL)	149 ± 51	147 ± 51	151 ± 46	152 ± 56	150 ± 52
Insulin requirement (IU/day)	26 ± 12	26 ± 10	25 ± 12	23 ± 13	22 ± 13
BMI	24 ± 4	24 ± 3	24 ± 3	23 ± 4	24 ± 3
Cholesterol (mg/dL)	215 ± 59	199 ± 63	190 ± 49	167 ± 62	157 ± 38
Triglycerides (mg/dL)	175 ± 75	169 ± 98	158 ± 83	137 ± 80	148 ± 98
Hypoglycemia (n/patient/month)	0.3	0.4	0.4	0.6	0.6

^*p < 0.05.

In the patients with residual diuresis of > 500 mL/day under hemodialysis therapy the insulin requirement did not change significantly, in the patients with residual diuresis below 0.5 L/day the needed insulin doses decreased by 29%. The data for both groups are presented in Table 3.

Table 3 HbA1c, fasting blood glucose (FBG), and insulin requirement at the start of hemodialysis and after 1 year in patients with high or low residual urine volume (< and > 500 mL urine/day)

	Urine volume
	> 500 mL/day		< 500 mL/day
	Start	After 1 year	Start	After 1 year
HbA1c (%)	7.2. ± 1.4	7.2 ± 1.2	7.3 ± 1.1	7.3 ± 0.8
FBG (mg/dL)	147 ± 81	151 ± 59	154 ± 56	145 ± 47
Insulin requirement (IU/day)	26 ± 12	27 ± 13	24 ± 11	17 ± 8

^*p < 0.05.

DISCUSSION

The impact of a kidney transplantation on the metabolic control of diabetic patients treated with insulin, was intensively investigated.[3] In contrast, there are few data in the literature concerning the impact of hemodialysis on glycemic control in insulin-treated diabetic patients.

Impairment of renal function leads to insulin resistance;[4] this uremia-associated insulin resistance may be improved after the start of dialysis therapy.[5] On the other side, insulin clearance is decreased in renal insufficiency, this is associated with a lower insulin requirement in insulin-dependent diabetic patients.[6] Therefore, many nephropathic insulin-treated patients have to reduce their insulin dose in relation to glomerular filtration rate (GFR).[7&8] After the start of dialysis therapy, the residual renal function often completely disappears within few months. It can be assumed that in this oligoanuric state the insulin clearance will further bedecreased and, consecutively, the insulin requirement has to be reduced. In our study of insulin-treated type 2 diabetic patients, the start of hemodialysis had no impact on glycemic control (mean HbA1c and FBG) and insulin requirement decreased not significantly by 8%. Ten of our study patients showed a high residual diuresis (> 500 mL urine/day) 12 months after the start of dialysis. Insulin requirement did not change in these patients. In the patients with decreased residual diuresis (< 500 mL/day) the needed insulin doses decreased by 29%, the difference in urine volume after one year dialysis between both patient groups was significant (p < 0.05). These results confirm the assumption that the residual diuresis can influence the insulin requirement in insulin-treated type 2 diabetic patients under hemodialysis. The data for both groups are presented in Table 3.

The frequency of hypoglycemia tended to be higher in hemodialyzed diabetic patients due to more frequent hypoglycemic events during hemodialysis, as reported in the literature:[9] this higher frequency can be explained by a higher incidence of hypoglycemic episodes during hemodialysis, although the glucose concentration in our dialysate was 100 mg/dL. The decrease in cholesterol levels after the start of dialysis may be explained by more compliance patient after initiating dialysis therapy. During the predialysis period 28% of the patients received statin therapy, during the dialysis period 36% The mean morning BP values were similar before and after the start of hemodialysis, although blood pressure can be elevated in dialysis patients due to hyperhydration.[10] Blood volume and blood temperature control were equal in both patient groups.

In conclusion, hemodialysis had no significant long-term effect on the glycemic control in insulin-treated type 2 diabetic patients, but incidence of hypoglycemia tended to be higher under hemodialysis than in the predialysis period. The lipid levels tended to be lower after initiating dialysis therapy. In patients with significant loss of residual diuresis within the first year after the start of hemodialysis the insulin requirement was significantly lower.