Glucose metabolism and insulin sensitivity in patients on chronic hemodialysis

Glucose

Metabolism and Insulin Sensitivity on Chronic Hemodialysis

in Patients

H. Graf, R. Prager, J. Kovarik, A. Luger, G. Schernthaner, and W.F. Pinggera In order to evaluate the potential role of parathyroid hormone on glucose metabolism in patients on chronic hemodialysis hyperglycemic clamp studies were performed in 7 parathyroidectomired and 11 nonparathyroidectomized patients on chronic hemodialysis and in healthy controls. There were no significant differences in the peripheral glucose uptake of the 3 groups. The beta cell response to hyperglycemia during the early phase as well as during the steady state was almost identical in controls and in nonparathyroidectomized uremics, whereas in the parathyroidectomized group a markedly enhanced insulin secretion was found. Calculated tissue sensitivity to insulin therefore was equal in controls and in nonparathyroidectomized uremics, whereas patients after parathyroidectomy had peripheral insulin resistance. Our results demonstrate that patients on chronic hemodialysis apparently have normal peripheral glucose uptake. The subgroup of patients who have undergone parathyroidectomy, however, show an enhanced insulin response to hyperglycemia suggesting peripheral insulin resistance. We conclude that longstanding and severe secondary hyperparathyroidism-the usual cause for parathyroidectomy in these patients-results in irreversible insulin resistance with a compensatory increase of insulin secretion. o 1985 by Grune 81 Stratton. Inc.

G sequence of chronic renal failure.’ While the site and mechanism of this seem to be a decreased sensitivLUCOSE INTOLERANCE

is a well-known con-

ity of peripheral tissues to insulin,2V3the etiology of this lesion is still unknown. Several reports have suggested the accumulation in uremia of small molecular weight substances that effect carbohydrate metabolism.4*s In addition, intermittent hemodialysis may be capable of normalizing carbohydrate metabolism in uremia in a relatively short time. 4v5However, studies233employing the euglycemic and hyperglycemic clamp technique show that although intermittent hemodialysis (for a mean of 10 weeks) resulted in a significant improvement of impaired glucose metabolism of uremia, complete normalization did not occur. Whereas in primary hyperparathyroidism hypersecretion of insulin is known to occur,6 the data on insulin secretion patterns in patients with secondary hyperparathyroidism is still controversial. Earlier studies have demonstrated hypersecretion of insulin to a hyperglycemic stimulus in patients with severe secondary hyperparathyroidism which was reversible following parathyroidectomy suggesting that parathyroid hormone augments the beta-cell response to glucose.’ Others failed to show any significant relationship between secondary hyperparathyroidism and glucose tolerance, insulin secretion, and insulin sensitivity.* More recent data from From the second Depurtment of Internal Medicine, University of Vienna and Ludwig Boltzmann-Institute for Clinical Endocrinology, Vienna. Presented in part at the IXth International Congress of Nephrology, Los Angeles. June 1984. Address reprint requesis to Helmut Graf. MD, University of California Service, Veterans Administration Medical Center 4150 Clement Street (1 I I J). San Francisco, CA 94121. 0 1985 by Grune & Stratton, Inc. 0026~495/85/3400-0016$03.00/0 974

animal9 as well as human studies” suggest that excess parathyroid hormone does inhibit a compensatory increase of insulin secretion in an insulin-resistant state thus contributing to glucose intolerance of chronic renal failure. In order to investigate the consequences of longterm hemodialysis therapy on glucose metabolism we studied a larger series of patients on chronic hemodialysis by means of the hyperglycemic clamp technique. To further address the question of a potential role of parathyroid hormone patients who had undergone parathyroidectomy, and nonparathyroidectomized patients were compared. MATERIALS AND METHODS Eighteen patients on chronic intermittent hemodialysis were studied and the results compared to 10 healthy age- and sexmatched controls. Clinical and biochemical details of the patients and mean values of controls are summarized in Table 1. Seven patients had undergone subtotal parathyroidectomy 24 to 50 months previously. The effect of parathyroidectomy was demonstrated by a decrease of parathyroid hormone levels after parathyroid gland removal, normalization of clinical signs such as bone pain or x-ray findings of bone lesions and normal values of calcium for at least 12 months prior to the study. In all patients hemodialysis was performed for six hours 3 times per week using hollow fiber capillary kidneys (Cordis Dow, CDA-K), and a dialysate containing glucose (440 mg/dL) and acetate as buffer beneath the usual electrolyte content. Plasma phosphate levels were controlled to a level below 5.5 mg/dL by administration of aluminum hydroxide. None of the patients were on any kind of vitamin D therapy at the time of the study and none received any drugs known to affect carbohydrate metabolism. All patients were studied by the hyperglycemic clamp technique as originally described by DeFronzo.” For three days prior to the study all patients were placed on a standardized diet containing 40 Kcal/kg body weight with 40% carbohydrate. The glucose levels were acutely raised 120 mg/dL above baseline by a logarithmically decreasing glucose infusion and maintained at this level for 120 minutes by a variable glucose infusion, adjusted on the basis of glucose levels measured at five-minute intervals. The plasma insulin Metabolism, Vol34,

No 10 (October). 1985: pp 974-977

GLUCOSE METABOLISM AND INSULIN SENSITIVITY

975

Table 1. Summary of Clinical and Laboratory

Years Height Patient

Weight

obesity

on Dialysis

sex

cm

kg

Index

Data of Patients and Means f SD of Controls Fasting Blood GlWXXc hng/cu 80-105

Cr

BUN

(mg/dL) 0.4- 1.4

(mg/dL) 5-20

ca lmg/dI 8.5-10.5

P

PTH

(mgldL1 2.5-4.5

(mu/L) 1.3-3.8

NCVllMl Range

PTX (months since) 1

36

61

F

158

52

20.8

6.5

92

9.2

55

8.7

3.7

3.7

2

24

29

F

165

56

20.5

2

97

10.1

70

8.6

4.5

4.8

3

27

63

F

158

60

24.1

4.3

89

10.2

82

8.6

4.2

5.1

4

42

47

M

170

70

24.2

3.2

91

13.4

86

9.4

5.0

3.2

5

29

51

M

173

66

22.0

2

102

12.6

75

8.8

3.9

6.0

6

32

38

F

160

45

17.6

5.4

86

8.7

64

9.1

4.2

5.4

61

M

7 x +SD

181

65

19.8

7.5

88

12.5

73

9.4

4.7

4.9

34.2

50

166

59

21.3

4.4

92

10.9

72

8.9

4.3

4.7

9.1

13

9

9

2.3

2.1

5

1.8

10

0.3

0.4

0.9

50

N-PTX 1

43

M

180

74

22.8

2.5

99

13.4

79

9.4

4.2

12.3

2

40

M

188

77

21.8

1

87

9.7

80

8.3

5.2

7.9

3

32

M

167

83

29.8

3.8

89

12.8

75

8.3

5.2

8.8

4

22

M

179

75

23.4

2.2

101

13.5

82

9.5

5.0

10.1

5

19

F

170

51

17.6

9.1

103

9.2

60

9.8

4.7

9.4

6

45

M

178

102

29.2

3.1

92

15.4

78

8.7

3.9

11.5

7

29

M

176

70

22.6

4.2

90

14.0

79

9.2

4.5

9.7

8

58

F

169

59

20.7

4.5

95

8.1

68

9.3

4.0

6.8

9

56

M

186

78

22.6

6.0

104

12.8

77

8.7

4.1

8.9

10

62

M

174

72

23.8

5.1

88

13.1

80

9.0

3.8

9.5

M

12.1

11

38

176

70

22.6

0.6

90

11.0

79

9.5

3.8

x

40

177

74

23.3

3.8

94

12.0

76

9.1

4.2

9.7*

+SD

14

7

13

3.5

2.4

6

2.3

7

0.4

0.5

1.7

172

71

22.7

91

1.1

15

9.2

3.2

2.4*

14

4.1

5

0.2

6

0.5

0.6

1.1

Controls

(n = 10)

j;

42

*SD

13

7M/3F

Cr. plasma creatinine; BUN, blood urea nitrogen; Ca. plasma calcium; P, Plasma phosphate, PTH, serum parathyroid hormone. lP < 0.00 1 PTX Y N-PTX and PTX v controls. response was assessed at two-minute intervals during the first ten minutes of the study (early phase) and every 20 minutes thereafter (late phase). The study was performed after a 12 hours overnight fast 40 + 4 hours after the last hemodialysis. Blood samples were drawn from an indwelling catheter in the arteriovenous fistula. Glucose was administered via an antecubital vein. Glucose was measured by a rapid glucose analyzer using the glucose oxidase method (Beckman Instruments), BUN, creatinine, calcium, and phosphate were measured by routine methods (Technicon Auto Analyzer). Immunoreactive insulin’* and immunoreactive parathyroid hormone” were measured by radioimmunoassay. Computations of the amount of peripheral glucose uptake (M), plasma insulin response to glucose (I) as well as tissue sensitivity to insulin (M/I ratio) were done as originally described by DeFronzo.” Values are given as mean + SE unless otherwise stated. For statistical analysis the Student’s t test for unpaired data was used.

RESULTS

There were no significant differences between the parathyroidectomized and nonparathyroidectomized patients and the controls with respect to age, sex, height, weight, and obesity index (Table 1). The blood chemistry values of fasting blood glucose, calcium, and phosphate were also not significantly different in the three groups. The BUN and creatinine values were not significantly different in the parathyroidectomized

and nonparathyroidectomized patients (Table 1). Nonparathyroidectomized patients had significantly higher serum parathyroid hormone levels than parathyroidectomized patients (P < 0.001) who in turn had higher PTH values than the normal controls (P < 0.001; Table 1). None of the patients showed any signs of overt secondary hyperparathyroidism. Peripheral glucose uptake was found not to be significantly different in the three groups (Fig 1). Parathyroidectomized patients showed a significantly enhanced pancreatic beta-cell response when compared to nonparathyroidectomized patients (P -e0.05) and controls (P -c0.025), however, there was no significant difference between nonparathyroidectomized patients and controls (Fig 1). This enhanced beta-cell response of parathyroidectomized patients was observed in the initial phase (cl0 minutes) as well as during the steady state conditions of hyperglycemia (20-120 minutes; Fig 2). Calculating the tissue sensitivity to insulin by dividing the M-value by the corresponding plasma insulin level parathyroidectomized patients show a significantly decreased tissue sensitivity to insulin when compared to nonparathyroidectomized patients (P -c0.025) as well as to controls

GRAF ET AL

976

PLASMA

INSULIN

TISSUE SENSmvrrYTo

META8OLlSM (ho ktlg&mk~

8

Fig 1. Means and SEM of the physiologic variables quantified by the hyperglycemic clamp in parathyroidectomized (PTX) and nonparathyroidectomized (N-PTX) patients on chronic hemodialysis and in control subjects (CON). (P

<

O.OOS), but the nonparathyroidectomized patients were not significantly different from controls (Fig 1). DISCUSSION

Our results demonstrate that peripheral glucose uptake (M) in patients on chronic hemodialysis as assessed by the hyperglycemic clamp technique is in the normal range. Different results might have been obtained by performing the clamp procedures 24 hours rather than 40 hours after the last hemodialysis. In general, however, our results are in good agreement with studies from Hutchings et al4 and Hampers and coworkers’ which suggested that the glucose intolerance of chronic renal failure can be reversed by hemodialysis therapy, and are also consistent with the data of Lindal16 showing normal decay constants for glucose after an intravenous (IV) glucose load in a larger series of patients on chronic hemodialysis. However, our data are in some contrast to the recent findings of DeFronzo3 using the hyperglycemic clamp technique to show impaired glucose metabolism in patients undergoing dialysis therapy for a mean period of 10 weeks. This difference is most likely due to the prolonged course of dialysis therapy in our patients, who have been on therapy for at least 7 months. Our results therefore suggest that chronic hemodialysis therapy is capable of normalizing peripheral glucose utilization of chronic uremia at least when secondary hyperparathyroidism is effectively controlled. In both of the patient groups there were no signs of overt hyperparathyroidism at the time of the study. Although peripheral glucose uptake (M) was essentially normal in all patients parathyroidectomized uremics exhibited peripheral insulin resistance that was compensated for by an exaggerated early- and late-phase insulin secretion. Thus, these findings sug-

Fig 2. Time course of the endogenous insulin response during the hyperglycemic clamp in parathyroidectomized (ptx) and nonparathyroidectomized In-ptx) patients on chronic hemodblysis and in control subjects (con).

gest that secondary hyperparathyroidism might play a key role in the development of impaired carbohydrate metabolism in uremia. The role of the parathyroids in the pathogenesis of glucose intolerance of chronic renal failure was recently addressed by Akmal et al.’ In this study chronic uremia in dogs resulted in a peripheral insulin-resistant state which in previously parathyroidectomized dogs was compensated by hyperinsulinism. Nonparathyroidectomized dogs, however, were not able to overcome peripheral insulin resistance by increased insulin secretion and therefore became glucose intolerant. Similar results have been obtained by Mak et al.” In their study parathyroidectomy for secondary hyperparathyroidism resulted in an improvement of peripheral glucose uptake attributable to an increased insulin response, whereas peripheral insulin sensitivity remained unchanged. Both studiesg,lo therefore suggest that extremely high values of parathyroid hormone might have a suppressive effect on insulin secretion. These findings are in apparent discrepancy to results found in primary hyperparathyroidism where an augmented insulin response has been found.6,‘41’5 The most likely explanation for alterations in carbohydrate metabolism in primary hyperparathyroidism must be in factors others than parathyroid hormone, such as elevated calcium and lowered phosphate levels.lG2’ Since there were no significant differences in calcium and phosphate levels between parathyroidectomized and nonparathyroidectomized patients in our study, alterations in the concentrations of these ions can not account for the increased insulin secretion and peripheral insulin resistance in our para-

GLUCOSE METABOLISM AND INSULIN SENSITIVITY

977

thyroidectomized patients. In primary hyperparathyroidism complete normalization of glucose metabolism and insulin sensitivity can occur after the removal of the adenoma.” The results on glucose metabolism of effectively controlling secondary hyperparathyroidism are different. Animal’ and human” studies show persistent peripheral insulin resistance in the presence of a normal peripheral glucose uptake in uremia, even when secondary hyperparathyroidism is effectively controlled. In this regard our parathyroidectomized patients correspond exactly to the patients reported by Mak and coworkers.” Normal peripheral glucose uptake in these patients obviously is the result of a compensatory augmented insulin secretion that counterbalances peripheral insulin resistance. Concerning their glucose metabolism apparently there are at least two groups of patients with chronic uremia. In patients with sufficient control of secondary hyperparathyroidism chronic hemodialysis therapy is capable of normalizing glucose uptake and insulin sensitivity. On the other hand, patients with inadequately controlled sec-

ondary hyperparathyroidism might remain insulin resistant and glucose intolerant despite dialysis therapy and there is evidence that a compensatory insulin hypersecretion is blunted by high levels of parathyroid hormone.‘,” After parathyroidectomy these patients seem to be able to normalize their glucose tolerance by a compensatory insulin hypersecretion. In summary our results suggest that chronic hemodialysis therapy together with continuous effective control of secondary hyperparathyroidism can achieve normal peripheral glucose uptake and normal tissue sensitivity to insulin, and that severe secondary hyperparathyroidism, which is the usual indication for parathyroidectomy, might be at least one important cause of glucose intolerance and insulin resistance observed in patients with chronic uremia.

ACKNOWLEDGMENT We want to thank Evelyne Sedlack for her skillful technical assistance.

REFERENCES 1. DeFronzo RA, Andres R, Edgar P, et al: Carbohydrate metabolism in uremia: A review. Medicine 52:469-481, 1973 2. DeFronzo RA, Tobin DJ, Rowe JW, et al: Glucose intolerance in uremia. Quantification of pancreatic B-cell sensitivity to glucose and tissue sensitivity to insulin. J Clin Invest 62:425435, 1978 3. DeFronzo RA, Alvestrand A: Glucose intolerance in uremia: Site and mechanism. Am J Clin Nutr 33:1438-1445, 1980 4. Hutchings RH, Hegstrom RM, Scribner BH: Glucose intolerance in patients on long-term intermittent dialysis. Ann Intern Med 65:275-185.1966 5. Hampers DL, Soeldner JS, Doak PB, et al: Effect of chronic renal failure and hemodialysis on carbohydrate metabolism. J Clin Invest 45:1719-1731,1966 6. Prager R, Kovarik J, Schernthaner G, et al: Peripheral insulin resistance in primary hyperparathyroidism. Metabolism 32:80& 8051983 7. Linda11 A, Carmena R, Cohen S, et al: Insulin hypersecretion in patients on chronic hemodialysis. Role of parathyroids. J Clin Endocrinol32:653-658, 1971 8. Amend WJC, Steinberg SM, Lowrie EG, et al: The influence of serum calcium and parathyroid hormone upon glucose metabolism in uremia. J Lab Clin Med 86:435444, 1975 9. Akmal M, Massry SG, DeFronzo RA: Role of parathyroid hormone in insulin secretion and glucose metabolism in chronic renal failure. Kidney International 25:241, 1984 10. Mak RHK, Turner C, Haycock GB, et al: Secondary hyperparathyroidism and glucose intolerance in children with uremia. Kidney International 24:8128-S133, 1983 11. DeFronzo RA, Tobin DJ, Andres R: Glucose clamp technique: A method for quantifying insulin secretion and resistance. Am J Physiol237:E214-E223,1979

12. Yalow RS, Berson SA: Immunoassay of endogenous plasma insulin in man. J Clin Invest 39:1157-l 175, 1960 13. Berson SA, Yalow RS: Parathyroid hormone in plasma in adenomatous hyperparathyroidism, uremia and bronchogenic carcinoma. Science 154:907-9 10, 1966 14. Kim H, Kalkhoff RK, Costrini NV, et al: Plasma insulin disturbances in primary hyperparathyroidism. J Clin Invest 50:2596-2605,197l 15. Ljunghall S, Palmer M, Akerstrom G, et al: Diabetes mellitus, glucose tolerance and insulin response to glucose in patients with primary hyperparathyroidism before and after parathyroidectomy. Eur J Clin Invest 13:373-377, 1983 16. Harter HR. Santiago JV, Rutherford WE, et al: The relative roles of calcium, phosphorus and parathyroid hormone in glucoseand tolbutamid-mediated insulin release. J Clin Invest 58:359-367, 1976 17. Curry DC, Benett LL, G&sky GM: Requirement for calcium ion in insulin secretion by perfused rat pancreas. Am J Physiol 214:174-178, 1968 18. Devis G, Somers G, Malaisie W: Stimulation of insulin release by calcium. Biochem Biophys Res Comm 67:525-529, 1975 19. Hellman B: Stimulation of insulin release after intracellular calcium. FEBS Lett 63:125-128, 1976 20. DeFronzo RA, Lang R: Hypophosphatemia and glucose intolerance: Evidence for tissue insensitivity to insulin. N Engl J Med 303:1259-1263, 1980 21. Marshall WP, Banasiak MF, Kalkhoff RK: Effect of phosphate deprivation on carbohydrate metabolism. Horm Metabol Res l&369-373,1978