- Email: [email protected]
Increased resistin blood levels are not associated with insulin resistance in patients with renal disease
Increased Resistin Blood Levels Are Not Associated With Insulin Resistance in Patients With Renal Disease Jan T. Kielstein, MD, Bjo¨rn Becker, MD, Susanne Graf, Georg Brabant, MD, Hermann Haller, MD, and Danilo Fliser, MD ● Background: Resistin is a newly discovered insulin inhibitor secreted by adipocytes. We explored the potential role of resistin in the pathophysiological process of insulin resistance encountered in patients with renal disease. Methods: Resistin blood concentrations, insulin sensitivity index (by intravenous glucose tolerance test), and glomerular filtration rate (GFR; by inulin clearance) were assessed in 30 male patients with immunoglobulin A glomerulonephritis in different stages of renal disease. Results: Patients with increased resistin blood concentrations had more advanced renal failure and were significantly older. Plasma resistin levels correlated significantly with GFR (r ⴝ ⴚ0.82; P < 0.0001), plasma homocysteine concentration (r ⴝ 0.68; P < 0.001), and age (r ⴝ 0.42; P ⴝ 0.05), but not with fasted plasma insulin (r ⴝ ⴚ0.34; P ⴝ 0.12), glucose (r ⴝ 0.25; P ⴝ 0.19), and leptin (r ⴝ ⴚ0.24; P ⴝ 0.21) concentrations; body mass index (r ⴝ ⴚ0.06; P ⴝ 0.75), waist-hip ratio (r ⴝ 0.09; P ⴝ 0.63), or insulin sensitivity (r ⴝ ⴚ0.05; P ⴝ 0.79). In multiple regression analysis, GFR was the only independent predictor of plasma resistin concentrations in renal patients (r ⴝ ⴚ0.812; P < 0.0001). Conclusion: Resistin blood concentrations increase with progressive impairment of renal function. Thus, the kidney seems to be an important site of resistin elimination. However, the greater than 5-fold increase in resistin blood levels apparently is not associated with deterioration in insulin sensitivity in patients with renal disease. Am J Kidney Dis 42:62-66. © 2003 by the National Kidney Foundation, Inc. INDEX WORDS: Insulin resistance; glomerular filtration rate (GFR); renal disease; resistin; uremia.
P
ATIENTS WITH renal diseases are characterized by resistance to the action of insulin, which is accompanied by hyperinsulinemia and glucose intolerance.1-3 We and others have documented that insulin resistance is present very early in the course of renal disease, ie, in patients with only minor impairment of renal function and even in patients with apparently normal renal function.4-7 The cause(s) of insulin resistance in renal patients has not been elucidated in detail. However, experimental studies and studies of humans have shown that the kidney is an important organ for glucose metabolism.8-11 Thus, subclinical injury to renal metabolic function may predispose to derangement of glucose metabolism in patients with renal disease. Resistin is a recently discovered cysteine-rich
From the Department of Internal Medicine, Division of Nephrology and the Division of Endocrinology, Medical School Hannover, Germany. Received November 18, 2002; accepted in revised form March 19, 2003. Address reprint requests to Danilo Fliser, MD, Associate Professor of Medicine, Division of Nephrology, Department of Internal Medicine, Hannover Medical School, CarlNeuberg-Strasse 1, 30625, Hannover, Germany. E-mail: [email protected] © 2003 by the National Kidney Foundation, Inc. 0272-6386/03/4201-0007$30.00/0 doi:10.1016/S072-6386(03)00409-8 62
protein secreted by adipocytes that inhibits the action of insulin.12 Administration of recombinant resistin to laboratory animals impairs glucose tolerance and the action of insulin, whereas an increase in insulin sensitivity is noted when animals are treated with resistin-neutralizing antibodies.12 Preliminary evidence from genetic studies of humans suggests that a variant at the resistin gene locus might be associated with risk for type 2 diabetes13; however, the role of resistin in human syndromes of insulin resistance, such as uremia, remains unclear. To explore the role of resistin in the insulin resistance syndrome present in patients with renal diseases, we performed a cross-sectional study in patients with immunoglobulin A (IgA) glomerulonephritis (GN) at different degrees of renal function. We measured resistin blood levels, as well as insulin sensitivity and glomerular filtration rate (GFR). PARTICIPANTS AND METHODS
Study Protocol The study protocol was approved by the local ethics committee. All participants gave their written informed consent. Thirty nonsmoking Caucasian men with biopsyconfirmed IgA GN were examined. Patients with known diabetes mellitus of any type were excluded from the study. All patients studied had stable renal function for at least 3 months before the study and none of them was treated with vitamin D, erythropoietin, fish oil, or immunosuppressive agents. Antihypertensive drugs with the potential of con-
American Journal of Kidney Diseases, Vol 42, No 1 (July), 2003: pp 62-66
RESISTIN AND RENAL DISEASE
founding the assessment of insulin sensitivity and GFR were washed out for periods depending on their half-life of action; ie, short-acting drugs were withdrawn for at least 3 days before examination, whereas long-acting drugs were washed out for at least 1 week before examination. All renal patients were examined using a frequently sampled intravenous glucose tolerance test with plasma glucose and insulin measurements. For 3 days before the test, they adhered to a diet with normal carbohydrate content. They were examined while in a recumbent position in a quiet environment after an overnight fast of at least 12 hours. After blood samples for the measurement of basal (fasted) plasma glucose and insulin concentrations were withdrawn, an intravenous bolus injection of a 40% glucose solution was administered over 3 minutes (0.3 g/kg body weight).14 Blood samples for measurement of plasma glucose and insulin concentrations were withdrawn at regular times thereafter. Blood samples for measurement of routine chemistry and resistin, leptin, intact parathyroid hormone (PTH), and total homocysteine (Hcy) plasma concentrations were obtained from all patients. In addition to this protocol, GFR was assessed in patients on a separate day by using the steady-state inulin infusion clearance technique as described elsewhere.15 In parallel, mean arterial blood pressure was recorded oscillometrically using Dinamap (Criticon Inc, Tampa, FL). For comparison, we assessed resistin blood concentrations and GFR in 13 nonsmoking healthy male Caucasian subjects and 15 male Caucasian patients with type 2 diabetes mellitus and normal plasma creatinine concentrations. Antihypertensive drugs with the potential of confounding GFR measurement were washed out in patients with diabetes for periods depending on their half-life of action, as described previously.
Measurements and Calculations Plasma glucose concentrations were measured using the Glucoanalyzer II (Beckmann Instruments, Munich, Germany), and plasma insulin levels, using an enzyme-linked immunosorbent assay (ELISA) with monoclonal insulin antibodies (Enzymun; Boehringer-Mannheim, Mannheim, Germany). Intra-assay and interassay coefficients of variation were 5.0% and 7.5%, respectively. Data for plasma insulin and glucose levels obtained during the frequently sampled intravenous glucose tolerance test were analyzed using a computer program (MINMOD), described in detail elsewhere.6,14,16 Insulin sensitivity is expressed as the insulin sensitivity index (SI). Plasma resistin concentrations were measured using an ELISA (Human Resistin ELISA; BioVendor Laboratory Medicine Inc, Brnø, Czech Republic). Intraassay and interassay coefficients of variations were 5.2% and 7.2%, respectively. Leptin blood concentrations were measured using a radioimmunoassay (Linco Research Inc, St Charles, MT). PTH was measured using an immunoradiometric assay; Hcy, with a fluorescence-polarization immunoassay; and inulin, enzymatically using inulinase. All other measurements were performed using certified assay methods.
63 Table 1.
Baseline Clinical Data for 30 Male Patients With IgA GN
No. of subjects Age (y) BMI (kg/m2) WHR GFR (mL/min/1.73 m2) Plasma creatinine (mg/dL) Mean arterial blood pressure (mm Hg) Plasma resistin (ng/mL) Insulin SI (min⫺1 [U/mL]) Plasma insulin (U/mL) Plasma glucose (mg/dL) Plasma leptin (ng/mL) Serum cholesterol (mg/dL) Serum triglycerides (mg/dL) Plasma Hc (mol/L) Intact PTH (pmol/L)
30 43 ⫾ 2 25.6 ⫾ 0.7 0.87 ⫾ 0.02 78 ⫾ 8 1.95 ⫾ 0.24 106 ⫾ 2 50.6 ⫾ 6.7 5.0 ⫾ 0.4 11.7 ⫾ 1.2 77 ⫾ 2 6.1 ⫾ 0.6 197 ⫾ 7 140 ⫾ 10 15.6 ⫾ 1.3 15.4 ⫾ 3.8
NOTE. To convert plasma creatinine concentration in mg/dL to mol/L, multiply by 88.4; plasma insulin concentration in U/mL to pmol/L, multiply by 7.175; plasma glucose concentration in mg/dL to mmol/L, multiply by 0.0555; serum cholesterol concentration in mg/dL to mmol/L, multiply by 0.0259; serum triglyceride concentration in mg/dL to mmol/L, multiply by 0.0113; plasma Hcy, concentration in mg/dL to mol/L, multiply by 7.397.
Statistical Analysis Clinical data for patients with IgA GN (n ⫽ 30) were arranged in tertiles of resistin blood concentrations and analyzed by means of a 2-tailed analysis of variance (ANOVA) using the SPSS package (SSPS Inc, Chicago, IL). If ANOVA showed significant differences, a t-test for random data was used to compare patient data in tertiles. Bonferroni correction for multiple comparisons was applied. In addition, resistin plasma concentrations in renal patients, healthy subjects, and patients with type 2 diabetes were compared using ANOVA, and post hoc analysis with a t-test for random data was performed. Differences were considered significant at P of 0.05. All data are presented as mean ⫾ SEM. Pearson⬘s correlation analyses between plasma resistin levels and age, body mass index (BMI), waist-hip ratio (WHR), GFR, SI, and fasted blood insulin, leptin, glucose, Hcy, PTH, cholesterol, triglyceride, and creatinine concentrations were performed using data from patients with IgA GN only. Bonferroni-Holmes correction for multiple correlation analysis was applied. A multiple stepwise regression analysis was performed between resistin plasma level as the dependent variable and age, BMI, WHR, GFR, SI, and fasted blood insulin, leptin, glucose, PTH, cholesterol, triglyceride, and Hcy concentrations as independent variables.
RESULTS
Table 1 lists baseline clinical data for our 30 patients with IgA GN, and rearranged data using tertiles of plasma resistin concentrations are listed
64 Table 2.
KIELSTEIN ET AL Clinical Data for 30 Male Patients With IgA GN Arranged by Tertiles of Plasma Resistin Concentrations Tertile
No. of subjects Plasma resistin concentration (ng/mL) Age (y) BMI (kg/m2) WHR GFR (mL/min/1.73 m2) Plasma creatinine (mg/dL) Mean arterial blood pressure (mm Hg) Insulin SI (min⫺1 [U/mL]) Plasma insulin (U/mL) Plasma glucose (mg/dL) Plasma leptin (ng/mL) Serum cholesterol (mg/dL) Serum triglycerides (mg/dL) Plasma Hcy (mol/L) Intact PTH (pmol/L)
Lower
Middle
Upper
10 12.6 ⫾ 1.2 38 ⫾ 2 24.9 ⫾ 0.7 0.87 ⫾ 0.02 114 ⫾ 3 1.1 ⫾ 0.1 103 ⫾ 2 5.6 ⫾ 0.4 12.6 ⫾ 1.2 76 ⫾ 2 4.8 ⫾ 0.5 181 ⫾ 7 118 ⫾ 10 10.4 ⫾ 0.6 4.5 ⫾ 0.2
10 43.8 ⫾ 2.4 42 ⫾ 1 25.8 ⫾ 0.6 0.89 ⫾ 0.02 76 ⫾ 8† 2.0 ⫾ 0.2 104 ⫾ 2 4.7 ⫾ 0.3 11.4 ⫾ 1.1 75 ⫾ 2 7.5 ⫾ 0.6 216 ⫾ 5 161 ⫾ 10 16.2 ⫾ 0.8 17.8 ⫾ 4.2
10 95.5 ⫾ 2.8 49 ⫾ 2* 24.9 ⫾ 0.6 0.87 ⫾ 0.02 43 ⫾ 3‡ 3.0 ⫾ 0.3‡ 109 ⫾ 2 5.2 ⫾ 0.4 10.8 ⫾ 1.2 79 ⫾ 2 6.8 ⫾ 0.6 187 ⫾ 4 140 ⫾ 5 19.9 ⫾ 1.4‡ 23.1 ⫾ 3.3*
NOTE. To convert plasma creatinine concentration in mg/dL to mol/L, multiply by 88.4; plasma insulin concentration in U/mL to pmol/L, multiply by 7.175; plasma glucose concentration in mg/dL to mmol/L, multiply by 0.0555; serum cholesterol concentration in mg/dL to mmol/L, multiply by 0.0259; serum triglyceride concentration in mg/dL to mmol/L, multiply by 0.0113; plasma Hcy concentration in mg/dL to mol/L, multiply by 7.397. *P ⬍ 0.05, comparison between lower and upper tertile. †P ⬍ 0.05, comparison between lower and middle tertile. ‡P ⬍ 0.01, comparison between lower and upper tertile.
in Table 2. Renal patients with the greatest plasma resistin blood levels were older and had more advanced impairment of renal function, documented by significantly greater mean serum creatinine concentrations and significantly lower mean GFRs. In addition, they had significantly greater Hcy and PTH levels as a manifestation of their advanced renal failure. There were no significant differences between patients in the lower, middle, and upper tertiles of resistin plasma concentrations with respect to fasted blood insulin, glucose, leptin, cholesterol, and triglyceride levels. In addition, mean SIs were not significantly different among patients in the 3 tertiles of resistin plasma concentration (Table 2). Mean resistin plasma concentration in the group of renal patients (47.6 ⫾ 6.5 ng/mL) was significantly greater (P ⬍ 0.001) than in healthy men (age, 26 ⫾ 1 years; BMI, 22.5 ⫾ 0.5 kg/m2; resistin level, 10.9 ⫾ 1.5 ng/mL; GFR, 121 ⫾ 3 mL/min/1.73 m2) and male patients with type 2 diabetes mellitus (age, 61 ⫾ 1 years; BMI, 29.8 ⫾ 1.1 kg/m2; resistin level, 13.6 ⫾ 1.6 ng/mL; GFR, 119 ⫾ 3 mL/min/1.73 m2). However, mean resistin plasma concentrations in healthy sub-
jects and patients with diabetes were similar to that of patients with IgA GN in the lower tertile of resistin plasma levels who had a mean GFR within the normal range (Table 2). Results of univariate correlation analysis with data from patients with IgA GN are listed in Table 3. Resistin blood levels correlated significantly with GFR (Fig 1), plasma creatinine, Hcy, and PTH concentrations, and age, but not with fasted blood insulin, glucose, leptin, cholesterol, and triglyceride concentrations, BMI, WHR, and insulin sensitivity. Results of multiple stepwise regression analysis for determinants of plasma resistin levels are listed in Table 4. After adjustment for potential confounding by age, BMI, WHR, SI, and fasted blood insulin, leptin, glucose, cholesterol, triglyceride, Hcy, and PTH concentrations, only GFR was independently related to plasma resistin concentration. DISCUSSION
One novel finding in the present study is that in patients with primary renal disease (IgA GN), blood concentrations of resistin, a recently discov-
RESISTIN AND RENAL DISEASE
65
Table 3. Univariate Correlation Analysis Between Plasma Resistin Concentrations and Other Clinical and Laboratory Variables in 30 Male Patients With IgA GN
GFR Plasma creatinine Plasma Hcy Intact PTH Age Plasma insulin Plasma glucose Plasma leptin Serum triglycerides WHR BMI Insulin SI Serum cholesterol
Coefficient
P
⫺0.82 0.71 0.68 0.45 0.42 ⫺0.34 0.25 ⫺0.24 0.16 0.09 ⫺0.06 ⫺0.05 ⫺0.02
0.0001 0.0005 0.001 0.05 0.05 0.12 0.19 0.21 0.42 0.63 0.75 0.79 0.93
ered naturally occurring putative inhibitor of insulin action, increase with progressive impairment of renal function. This observation is reminiscent of the increase in blood levels of other substances in patients with renal failure, such as the classic uremic toxins Hcy and PTH. The mechanism of the increase in resistin plasma concentrations with declining renal function is unknown, but the close relationship between GFR and plasma resistin level (r ⫽ ⫺0.81) favors the possibility of reduced filtration with declining GFR. This assumption is corroborated because mean resistin blood levels were similar in renal patients with a GFR within the normal range, healthy subjects, and patients with type 2 diabetes mellitus who had normal renal function. Furthermore, resistin plasma levels in middle-
Fig 1. Correlation between resistin plasma concentrations and GFR in 30 male patients with IgA GN. There was a highly significant correlation between both variables (r ⴝ ⴚ0.82; P < 0.0001).
Table 4. Stepwise Multiple Regression Analysis for Independent Determinants of Resistin Plasma Concentrations in 30 Male Patients With IgA GN
GFR Age Insulin SI Plasma Hcy Intact PTH BMI Plasma glucose Serum triglycerides Serum cholesterol Plasma insulin Plasma leptin WHR
Coefficient
P
⫺0.812 0.214 0.189 0.182 ⫺0.134 ⫺0.129 0.122 ⫺0.115 ⫺0.113 ⫺0.109 ⫺0.065 ⫺0.054
0.0001 0.086 0.146 0.331 0.420 0.291 0.322 0.368 0.355 0.390 0.609 0.661
aged healthy subjects with normal renal function were in this range.17 Resistin is a 12.5-kd protein and should be free filterable, at least theoretically. A close relationship between GFR and Hcy concentration also exists although Hcy is not filtered by the kidney and its renal metabolism is still not unfolded in detail.18 Thus, further studies are needed to elucidate the kidney’s role in resistin elimination and mechanisms of resistin accumulation in renal failure. The second important finding of the present study is the observation that a substantial increase in plasma resistin levels in our patients with renal insufficiency apparently is not accompanied by deterioration in insulin sensitivity. Resistin blood levels in our patients did not correlate with insulin sensitivity across a wide range of plasma resistin concentrations, ie, between 5 and 100 ng/L, and only GFR was a predictor of resistin blood levels in multiple regression analysis. Furthermore, BMI and WHR also were not independently related to resistin blood levels. This observation is of interest because in experimental studies, resistin was found to be secreted mainly by adipocytes.12 We did not assess the amount and distribution of fat tissue in our patients, for example, by using computed tomographic scanning, but similar to insulin sensitivity, there was no correlation between resistin blood level and BMI or WHR in our patients. These indirect observations at least question the role of body fat as a major secretory organ for resistin and the development of insulin resistance in renal patients. Together, our findings do not support the no-
66
KIELSTEIN ET AL
tion that resistin may be of importance in the pathophysiological process of insulin resistance syndrome present in patients with renal disease. Our results are in line with observations from 2 recently published studies in which findings obtained in rodents were not fully confirmed in human tissue.19,20 However, the present study has some limitations. We examined only male patients with IgA GN because no data are available on the relationship between menstrual cycle (an important confounder of insulin sensitivity) and resistin blood level. However, in a recent publication reporting data in control subjects, no sex difference with respect to resistin blood levels was observed.17 The number of renal patients examined in the present study was limited, but in the face of the poor correlation between resistin blood levels and insulin sensitivity (r ⫽ ⫺0.05), it seems unlikely that examining even a larger cohort would show a closer relationship between both variables. In conclusion, in patients with primary renal disease, plasma resistin levels increase with progressive impairment of renal function, but this increase apparently is not associated with deterioration in insulin sensitivity. Because the role of the kidney in glucose metabolism is increasingly appreciated, additional studies are needed to elucidate the metabolism of resistin and mechanisms of its accumulation with renal failure. REFERENCES 1. DeFronzo RA, Alverstrand A, Smith D, Hendler R, Hendler E, Wahren J: Insulin resistance in uremia. J Clin Invest 67:563-572, 1981 2. Kautzky-Willer A, Pacini G, Barnas U, et al: Intravenous calcitriol normalizes insulin sensitivity in uremic patients. Kidney Int 47:200-206, 1995 3. Alvestrand A: Carbohydrate and insulin metabolism in renal failure. Kidney Int Suppl 52:S48-S52, 1997 4. Eidemak I, Feldt-Rasmussen B, Kanstrup IL, Nielsen SL, Schmitz O, Strandgaard S: Insulin resistance and hyperinsulinemia in mild to moderate progressive chronic renal failure and its association with aerobic work capacity. Diabetologia 38:565-572, 1995 5. Vareesangthip K, Tong P, Wilkinson R, Thomas TH:
Insulin resistance in adult polycystic kidney disease. Kidney Int 52:503-508, 1997 6. Fliser D, Pacini G, Engelleiter R, et al: Insulin resistance and hyperinsulinaemia are present already in patients with incipient renal disease. Kidney Int 53:1243-1247, 1998 7. Kato Y, Hayashi M, Ohno Y, Suzawa T, Sasaki T, Saruta T: Mild renal dysfunction is associated with insulin resistance in chronic glomerulonephritis. Clin Nephrol 54: 366-373, 2000 8. Wirthensohn G, Guder WG: Renal substrate metabolism. Physiol Rev 66:469-497, 1986 9. Stumvoll M, Chintalapudi U, Perriello G, Welle S, Gutierrez O, Gerich J: Uptake and release of glucose by the human kidney. Postabsorptive rates and responses to epinephrine. J Clin Invest 96:2528-2533, 1995 10. Cersosimo E, Garlick P, Ferretti J: Insulin regulation of renal glucose metabolism in humans. Am J Physiol 276:E78-E84, 1999 11. Meyer C, Stumvoll M, Dostou J, Welle S, Haymond M, Gerich J: Renal substrate exchange and gluconeogenesis in normal postabsorptive humans. Am J Physiol 282:E428E434, 2002 12. Steppan CM, Bailey ST, Bhat S, et al: The hormone resistin links obesity to diabetes. Nature 409:307-312, 2001 13. Ma X, Warram JH, Trischitta V, Doria A: Genetic variants at the resistin locus and risk of type 2 diabetes in Caucasians. J Clin Endocrinol Metab 87:4407-4410, 2002 14. Bergman RN: Toward physiological understanding of glucose tolerance: Minimal-model approach. Diabetes 38: 1512-1527, 1989 15. Fliser D, Zeier M, Nowack R, Ritz E: Renal functional reserve in healthy elderly people. J Am Soc Nephrol 3:1371-1377, 1993 16. Pacini G, Bergman RN: MINMOD: A computer program to calculate insulin sensitivity and pancreatic responsitivity from the frequently sampled intravenous glucose tolerance test. Comp Methods Progr Biomed 23:113-122, 1986 17. Hegele RA, Kraw ME, Ban MR, Miskie BA, Huff MW, Cao H: Elevated serum CRP and free fatty acids among nondiabetic carriers of missense mutations in the gene encoding lamin A/C (LMNA) with partial dystrophy. Arterioscler Thromb Vasc Biol 23:111-116, 2003 18. Bostom AG, Culleton BF: Hyperhomocysteinemia in chronic renal disease. J Am Soc Nephrol 10:891-900, 1999 19. Nagaev I, Smith U: Insulin resistance and type 2 diabetes are not related to resistin expression in human fat cells or skeletal muscle. Biochem Biophys Res Commun 285:561-564, 2001 20. Savage DB, Sewter CP, Klenk ES, et al: Resistin/ Fizz3 expression in relation to obesity and peroxisome proliferator-activated receptor-gamma action in humans. Diabetes 50:2199-2202, 2001
P
ATIENTS WITH renal diseases are characterized by resistance to the action of insulin, which is accompanied by hyperinsulinemia and glucose intolerance.1-3 We and others have documented that insulin resistance is present very early in the course of renal disease, ie, in patients with only minor impairment of renal function and even in patients with apparently normal renal function.4-7 The cause(s) of insulin resistance in renal patients has not been elucidated in detail. However, experimental studies and studies of humans have shown that the kidney is an important organ for glucose metabolism.8-11 Thus, subclinical injury to renal metabolic function may predispose to derangement of glucose metabolism in patients with renal disease. Resistin is a recently discovered cysteine-rich
From the Department of Internal Medicine, Division of Nephrology and the Division of Endocrinology, Medical School Hannover, Germany. Received November 18, 2002; accepted in revised form March 19, 2003. Address reprint requests to Danilo Fliser, MD, Associate Professor of Medicine, Division of Nephrology, Department of Internal Medicine, Hannover Medical School, CarlNeuberg-Strasse 1, 30625, Hannover, Germany. E-mail: [email protected] © 2003 by the National Kidney Foundation, Inc. 0272-6386/03/4201-0007$30.00/0 doi:10.1016/S072-6386(03)00409-8 62
protein secreted by adipocytes that inhibits the action of insulin.12 Administration of recombinant resistin to laboratory animals impairs glucose tolerance and the action of insulin, whereas an increase in insulin sensitivity is noted when animals are treated with resistin-neutralizing antibodies.12 Preliminary evidence from genetic studies of humans suggests that a variant at the resistin gene locus might be associated with risk for type 2 diabetes13; however, the role of resistin in human syndromes of insulin resistance, such as uremia, remains unclear. To explore the role of resistin in the insulin resistance syndrome present in patients with renal diseases, we performed a cross-sectional study in patients with immunoglobulin A (IgA) glomerulonephritis (GN) at different degrees of renal function. We measured resistin blood levels, as well as insulin sensitivity and glomerular filtration rate (GFR). PARTICIPANTS AND METHODS
Study Protocol The study protocol was approved by the local ethics committee. All participants gave their written informed consent. Thirty nonsmoking Caucasian men with biopsyconfirmed IgA GN were examined. Patients with known diabetes mellitus of any type were excluded from the study. All patients studied had stable renal function for at least 3 months before the study and none of them was treated with vitamin D, erythropoietin, fish oil, or immunosuppressive agents. Antihypertensive drugs with the potential of con-
American Journal of Kidney Diseases, Vol 42, No 1 (July), 2003: pp 62-66
RESISTIN AND RENAL DISEASE
founding the assessment of insulin sensitivity and GFR were washed out for periods depending on their half-life of action; ie, short-acting drugs were withdrawn for at least 3 days before examination, whereas long-acting drugs were washed out for at least 1 week before examination. All renal patients were examined using a frequently sampled intravenous glucose tolerance test with plasma glucose and insulin measurements. For 3 days before the test, they adhered to a diet with normal carbohydrate content. They were examined while in a recumbent position in a quiet environment after an overnight fast of at least 12 hours. After blood samples for the measurement of basal (fasted) plasma glucose and insulin concentrations were withdrawn, an intravenous bolus injection of a 40% glucose solution was administered over 3 minutes (0.3 g/kg body weight).14 Blood samples for measurement of plasma glucose and insulin concentrations were withdrawn at regular times thereafter. Blood samples for measurement of routine chemistry and resistin, leptin, intact parathyroid hormone (PTH), and total homocysteine (Hcy) plasma concentrations were obtained from all patients. In addition to this protocol, GFR was assessed in patients on a separate day by using the steady-state inulin infusion clearance technique as described elsewhere.15 In parallel, mean arterial blood pressure was recorded oscillometrically using Dinamap (Criticon Inc, Tampa, FL). For comparison, we assessed resistin blood concentrations and GFR in 13 nonsmoking healthy male Caucasian subjects and 15 male Caucasian patients with type 2 diabetes mellitus and normal plasma creatinine concentrations. Antihypertensive drugs with the potential of confounding GFR measurement were washed out in patients with diabetes for periods depending on their half-life of action, as described previously.
Measurements and Calculations Plasma glucose concentrations were measured using the Glucoanalyzer II (Beckmann Instruments, Munich, Germany), and plasma insulin levels, using an enzyme-linked immunosorbent assay (ELISA) with monoclonal insulin antibodies (Enzymun; Boehringer-Mannheim, Mannheim, Germany). Intra-assay and interassay coefficients of variation were 5.0% and 7.5%, respectively. Data for plasma insulin and glucose levels obtained during the frequently sampled intravenous glucose tolerance test were analyzed using a computer program (MINMOD), described in detail elsewhere.6,14,16 Insulin sensitivity is expressed as the insulin sensitivity index (SI). Plasma resistin concentrations were measured using an ELISA (Human Resistin ELISA; BioVendor Laboratory Medicine Inc, Brnø, Czech Republic). Intraassay and interassay coefficients of variations were 5.2% and 7.2%, respectively. Leptin blood concentrations were measured using a radioimmunoassay (Linco Research Inc, St Charles, MT). PTH was measured using an immunoradiometric assay; Hcy, with a fluorescence-polarization immunoassay; and inulin, enzymatically using inulinase. All other measurements were performed using certified assay methods.
63 Table 1.
Baseline Clinical Data for 30 Male Patients With IgA GN
No. of subjects Age (y) BMI (kg/m2) WHR GFR (mL/min/1.73 m2) Plasma creatinine (mg/dL) Mean arterial blood pressure (mm Hg) Plasma resistin (ng/mL) Insulin SI (min⫺1 [U/mL]) Plasma insulin (U/mL) Plasma glucose (mg/dL) Plasma leptin (ng/mL) Serum cholesterol (mg/dL) Serum triglycerides (mg/dL) Plasma Hc (mol/L) Intact PTH (pmol/L)
30 43 ⫾ 2 25.6 ⫾ 0.7 0.87 ⫾ 0.02 78 ⫾ 8 1.95 ⫾ 0.24 106 ⫾ 2 50.6 ⫾ 6.7 5.0 ⫾ 0.4 11.7 ⫾ 1.2 77 ⫾ 2 6.1 ⫾ 0.6 197 ⫾ 7 140 ⫾ 10 15.6 ⫾ 1.3 15.4 ⫾ 3.8
NOTE. To convert plasma creatinine concentration in mg/dL to mol/L, multiply by 88.4; plasma insulin concentration in U/mL to pmol/L, multiply by 7.175; plasma glucose concentration in mg/dL to mmol/L, multiply by 0.0555; serum cholesterol concentration in mg/dL to mmol/L, multiply by 0.0259; serum triglyceride concentration in mg/dL to mmol/L, multiply by 0.0113; plasma Hcy, concentration in mg/dL to mol/L, multiply by 7.397.
Statistical Analysis Clinical data for patients with IgA GN (n ⫽ 30) were arranged in tertiles of resistin blood concentrations and analyzed by means of a 2-tailed analysis of variance (ANOVA) using the SPSS package (SSPS Inc, Chicago, IL). If ANOVA showed significant differences, a t-test for random data was used to compare patient data in tertiles. Bonferroni correction for multiple comparisons was applied. In addition, resistin plasma concentrations in renal patients, healthy subjects, and patients with type 2 diabetes were compared using ANOVA, and post hoc analysis with a t-test for random data was performed. Differences were considered significant at P of 0.05. All data are presented as mean ⫾ SEM. Pearson⬘s correlation analyses between plasma resistin levels and age, body mass index (BMI), waist-hip ratio (WHR), GFR, SI, and fasted blood insulin, leptin, glucose, Hcy, PTH, cholesterol, triglyceride, and creatinine concentrations were performed using data from patients with IgA GN only. Bonferroni-Holmes correction for multiple correlation analysis was applied. A multiple stepwise regression analysis was performed between resistin plasma level as the dependent variable and age, BMI, WHR, GFR, SI, and fasted blood insulin, leptin, glucose, PTH, cholesterol, triglyceride, and Hcy concentrations as independent variables.
RESULTS
Table 1 lists baseline clinical data for our 30 patients with IgA GN, and rearranged data using tertiles of plasma resistin concentrations are listed
64 Table 2.
KIELSTEIN ET AL Clinical Data for 30 Male Patients With IgA GN Arranged by Tertiles of Plasma Resistin Concentrations Tertile
No. of subjects Plasma resistin concentration (ng/mL) Age (y) BMI (kg/m2) WHR GFR (mL/min/1.73 m2) Plasma creatinine (mg/dL) Mean arterial blood pressure (mm Hg) Insulin SI (min⫺1 [U/mL]) Plasma insulin (U/mL) Plasma glucose (mg/dL) Plasma leptin (ng/mL) Serum cholesterol (mg/dL) Serum triglycerides (mg/dL) Plasma Hcy (mol/L) Intact PTH (pmol/L)
Lower
Middle
Upper
10 12.6 ⫾ 1.2 38 ⫾ 2 24.9 ⫾ 0.7 0.87 ⫾ 0.02 114 ⫾ 3 1.1 ⫾ 0.1 103 ⫾ 2 5.6 ⫾ 0.4 12.6 ⫾ 1.2 76 ⫾ 2 4.8 ⫾ 0.5 181 ⫾ 7 118 ⫾ 10 10.4 ⫾ 0.6 4.5 ⫾ 0.2
10 43.8 ⫾ 2.4 42 ⫾ 1 25.8 ⫾ 0.6 0.89 ⫾ 0.02 76 ⫾ 8† 2.0 ⫾ 0.2 104 ⫾ 2 4.7 ⫾ 0.3 11.4 ⫾ 1.1 75 ⫾ 2 7.5 ⫾ 0.6 216 ⫾ 5 161 ⫾ 10 16.2 ⫾ 0.8 17.8 ⫾ 4.2
10 95.5 ⫾ 2.8 49 ⫾ 2* 24.9 ⫾ 0.6 0.87 ⫾ 0.02 43 ⫾ 3‡ 3.0 ⫾ 0.3‡ 109 ⫾ 2 5.2 ⫾ 0.4 10.8 ⫾ 1.2 79 ⫾ 2 6.8 ⫾ 0.6 187 ⫾ 4 140 ⫾ 5 19.9 ⫾ 1.4‡ 23.1 ⫾ 3.3*
NOTE. To convert plasma creatinine concentration in mg/dL to mol/L, multiply by 88.4; plasma insulin concentration in U/mL to pmol/L, multiply by 7.175; plasma glucose concentration in mg/dL to mmol/L, multiply by 0.0555; serum cholesterol concentration in mg/dL to mmol/L, multiply by 0.0259; serum triglyceride concentration in mg/dL to mmol/L, multiply by 0.0113; plasma Hcy concentration in mg/dL to mol/L, multiply by 7.397. *P ⬍ 0.05, comparison between lower and upper tertile. †P ⬍ 0.05, comparison between lower and middle tertile. ‡P ⬍ 0.01, comparison between lower and upper tertile.
in Table 2. Renal patients with the greatest plasma resistin blood levels were older and had more advanced impairment of renal function, documented by significantly greater mean serum creatinine concentrations and significantly lower mean GFRs. In addition, they had significantly greater Hcy and PTH levels as a manifestation of their advanced renal failure. There were no significant differences between patients in the lower, middle, and upper tertiles of resistin plasma concentrations with respect to fasted blood insulin, glucose, leptin, cholesterol, and triglyceride levels. In addition, mean SIs were not significantly different among patients in the 3 tertiles of resistin plasma concentration (Table 2). Mean resistin plasma concentration in the group of renal patients (47.6 ⫾ 6.5 ng/mL) was significantly greater (P ⬍ 0.001) than in healthy men (age, 26 ⫾ 1 years; BMI, 22.5 ⫾ 0.5 kg/m2; resistin level, 10.9 ⫾ 1.5 ng/mL; GFR, 121 ⫾ 3 mL/min/1.73 m2) and male patients with type 2 diabetes mellitus (age, 61 ⫾ 1 years; BMI, 29.8 ⫾ 1.1 kg/m2; resistin level, 13.6 ⫾ 1.6 ng/mL; GFR, 119 ⫾ 3 mL/min/1.73 m2). However, mean resistin plasma concentrations in healthy sub-
jects and patients with diabetes were similar to that of patients with IgA GN in the lower tertile of resistin plasma levels who had a mean GFR within the normal range (Table 2). Results of univariate correlation analysis with data from patients with IgA GN are listed in Table 3. Resistin blood levels correlated significantly with GFR (Fig 1), plasma creatinine, Hcy, and PTH concentrations, and age, but not with fasted blood insulin, glucose, leptin, cholesterol, and triglyceride concentrations, BMI, WHR, and insulin sensitivity. Results of multiple stepwise regression analysis for determinants of plasma resistin levels are listed in Table 4. After adjustment for potential confounding by age, BMI, WHR, SI, and fasted blood insulin, leptin, glucose, cholesterol, triglyceride, Hcy, and PTH concentrations, only GFR was independently related to plasma resistin concentration. DISCUSSION
One novel finding in the present study is that in patients with primary renal disease (IgA GN), blood concentrations of resistin, a recently discov-
RESISTIN AND RENAL DISEASE
65
Table 3. Univariate Correlation Analysis Between Plasma Resistin Concentrations and Other Clinical and Laboratory Variables in 30 Male Patients With IgA GN
GFR Plasma creatinine Plasma Hcy Intact PTH Age Plasma insulin Plasma glucose Plasma leptin Serum triglycerides WHR BMI Insulin SI Serum cholesterol
Coefficient
P
⫺0.82 0.71 0.68 0.45 0.42 ⫺0.34 0.25 ⫺0.24 0.16 0.09 ⫺0.06 ⫺0.05 ⫺0.02
0.0001 0.0005 0.001 0.05 0.05 0.12 0.19 0.21 0.42 0.63 0.75 0.79 0.93
ered naturally occurring putative inhibitor of insulin action, increase with progressive impairment of renal function. This observation is reminiscent of the increase in blood levels of other substances in patients with renal failure, such as the classic uremic toxins Hcy and PTH. The mechanism of the increase in resistin plasma concentrations with declining renal function is unknown, but the close relationship between GFR and plasma resistin level (r ⫽ ⫺0.81) favors the possibility of reduced filtration with declining GFR. This assumption is corroborated because mean resistin blood levels were similar in renal patients with a GFR within the normal range, healthy subjects, and patients with type 2 diabetes mellitus who had normal renal function. Furthermore, resistin plasma levels in middle-
Fig 1. Correlation between resistin plasma concentrations and GFR in 30 male patients with IgA GN. There was a highly significant correlation between both variables (r ⴝ ⴚ0.82; P < 0.0001).
Table 4. Stepwise Multiple Regression Analysis for Independent Determinants of Resistin Plasma Concentrations in 30 Male Patients With IgA GN
GFR Age Insulin SI Plasma Hcy Intact PTH BMI Plasma glucose Serum triglycerides Serum cholesterol Plasma insulin Plasma leptin WHR
Coefficient
P
⫺0.812 0.214 0.189 0.182 ⫺0.134 ⫺0.129 0.122 ⫺0.115 ⫺0.113 ⫺0.109 ⫺0.065 ⫺0.054
0.0001 0.086 0.146 0.331 0.420 0.291 0.322 0.368 0.355 0.390 0.609 0.661
aged healthy subjects with normal renal function were in this range.17 Resistin is a 12.5-kd protein and should be free filterable, at least theoretically. A close relationship between GFR and Hcy concentration also exists although Hcy is not filtered by the kidney and its renal metabolism is still not unfolded in detail.18 Thus, further studies are needed to elucidate the kidney’s role in resistin elimination and mechanisms of resistin accumulation in renal failure. The second important finding of the present study is the observation that a substantial increase in plasma resistin levels in our patients with renal insufficiency apparently is not accompanied by deterioration in insulin sensitivity. Resistin blood levels in our patients did not correlate with insulin sensitivity across a wide range of plasma resistin concentrations, ie, between 5 and 100 ng/L, and only GFR was a predictor of resistin blood levels in multiple regression analysis. Furthermore, BMI and WHR also were not independently related to resistin blood levels. This observation is of interest because in experimental studies, resistin was found to be secreted mainly by adipocytes.12 We did not assess the amount and distribution of fat tissue in our patients, for example, by using computed tomographic scanning, but similar to insulin sensitivity, there was no correlation between resistin blood level and BMI or WHR in our patients. These indirect observations at least question the role of body fat as a major secretory organ for resistin and the development of insulin resistance in renal patients. Together, our findings do not support the no-
66
KIELSTEIN ET AL
tion that resistin may be of importance in the pathophysiological process of insulin resistance syndrome present in patients with renal disease. Our results are in line with observations from 2 recently published studies in which findings obtained in rodents were not fully confirmed in human tissue.19,20 However, the present study has some limitations. We examined only male patients with IgA GN because no data are available on the relationship between menstrual cycle (an important confounder of insulin sensitivity) and resistin blood level. However, in a recent publication reporting data in control subjects, no sex difference with respect to resistin blood levels was observed.17 The number of renal patients examined in the present study was limited, but in the face of the poor correlation between resistin blood levels and insulin sensitivity (r ⫽ ⫺0.05), it seems unlikely that examining even a larger cohort would show a closer relationship between both variables. In conclusion, in patients with primary renal disease, plasma resistin levels increase with progressive impairment of renal function, but this increase apparently is not associated with deterioration in insulin sensitivity. Because the role of the kidney in glucose metabolism is increasingly appreciated, additional studies are needed to elucidate the metabolism of resistin and mechanisms of its accumulation with renal failure. REFERENCES 1. DeFronzo RA, Alverstrand A, Smith D, Hendler R, Hendler E, Wahren J: Insulin resistance in uremia. J Clin Invest 67:563-572, 1981 2. Kautzky-Willer A, Pacini G, Barnas U, et al: Intravenous calcitriol normalizes insulin sensitivity in uremic patients. Kidney Int 47:200-206, 1995 3. Alvestrand A: Carbohydrate and insulin metabolism in renal failure. Kidney Int Suppl 52:S48-S52, 1997 4. Eidemak I, Feldt-Rasmussen B, Kanstrup IL, Nielsen SL, Schmitz O, Strandgaard S: Insulin resistance and hyperinsulinemia in mild to moderate progressive chronic renal failure and its association with aerobic work capacity. Diabetologia 38:565-572, 1995 5. Vareesangthip K, Tong P, Wilkinson R, Thomas TH:
Insulin resistance in adult polycystic kidney disease. Kidney Int 52:503-508, 1997 6. Fliser D, Pacini G, Engelleiter R, et al: Insulin resistance and hyperinsulinaemia are present already in patients with incipient renal disease. Kidney Int 53:1243-1247, 1998 7. Kato Y, Hayashi M, Ohno Y, Suzawa T, Sasaki T, Saruta T: Mild renal dysfunction is associated with insulin resistance in chronic glomerulonephritis. Clin Nephrol 54: 366-373, 2000 8. Wirthensohn G, Guder WG: Renal substrate metabolism. Physiol Rev 66:469-497, 1986 9. Stumvoll M, Chintalapudi U, Perriello G, Welle S, Gutierrez O, Gerich J: Uptake and release of glucose by the human kidney. Postabsorptive rates and responses to epinephrine. J Clin Invest 96:2528-2533, 1995 10. Cersosimo E, Garlick P, Ferretti J: Insulin regulation of renal glucose metabolism in humans. Am J Physiol 276:E78-E84, 1999 11. Meyer C, Stumvoll M, Dostou J, Welle S, Haymond M, Gerich J: Renal substrate exchange and gluconeogenesis in normal postabsorptive humans. Am J Physiol 282:E428E434, 2002 12. Steppan CM, Bailey ST, Bhat S, et al: The hormone resistin links obesity to diabetes. Nature 409:307-312, 2001 13. Ma X, Warram JH, Trischitta V, Doria A: Genetic variants at the resistin locus and risk of type 2 diabetes in Caucasians. J Clin Endocrinol Metab 87:4407-4410, 2002 14. Bergman RN: Toward physiological understanding of glucose tolerance: Minimal-model approach. Diabetes 38: 1512-1527, 1989 15. Fliser D, Zeier M, Nowack R, Ritz E: Renal functional reserve in healthy elderly people. J Am Soc Nephrol 3:1371-1377, 1993 16. Pacini G, Bergman RN: MINMOD: A computer program to calculate insulin sensitivity and pancreatic responsitivity from the frequently sampled intravenous glucose tolerance test. Comp Methods Progr Biomed 23:113-122, 1986 17. Hegele RA, Kraw ME, Ban MR, Miskie BA, Huff MW, Cao H: Elevated serum CRP and free fatty acids among nondiabetic carriers of missense mutations in the gene encoding lamin A/C (LMNA) with partial dystrophy. Arterioscler Thromb Vasc Biol 23:111-116, 2003 18. Bostom AG, Culleton BF: Hyperhomocysteinemia in chronic renal disease. J Am Soc Nephrol 10:891-900, 1999 19. Nagaev I, Smith U: Insulin resistance and type 2 diabetes are not related to resistin expression in human fat cells or skeletal muscle. Biochem Biophys Res Commun 285:561-564, 2001 20. Savage DB, Sewter CP, Klenk ES, et al: Resistin/ Fizz3 expression in relation to obesity and peroxisome proliferator-activated receptor-gamma action in humans. Diabetes 50:2199-2202, 2001