A 6-year prospective study on new onset diabetes mellitus, insulin release and insulin sensitivity in renal transplant recipients

Transcription

1 Nephrol Dial Transplant (2003) 18: DOI: /ndt/gfg338 Original Article A 6-year prospective study on new onset diabetes mellitus, insulin release and insulin sensitivity in renal transplant recipients Monica Hagen 1, Jøran Hjelmesæth 1, Trond Jenssen 1, Lars Mørkrid 2 and Anders Hartmann 1 1 Department of Medicine, Section of Nephrology and 2 Department of Clinical Chemistry, Rikshospitalet, Oslo, Norway Abstract Background. It is well known that both insulin resistance and insulin deficiency are involved in the pathogenesis of post-transplant diabetes mellitus (PTDM), but the relative importance of the two different mechanisms is still under debate. The present prospective longitudinal study was performed over 6 years to investigate the impact of impaired insulin secretion (ISec) and insulin sensitivity (IS) in the development of PTDM in renal transplant recipients. Methods. A total of 95 non-diabetic patients underwent a 75 g oral glucose tolerance test (OGTT) 10 weeks post-transplant. Six years later, 63 of these recipients were re-examined, the majority (n ¼ 58) with an OGTT. Fasting, 1- and 2-h insulin and glucose levels were measured and used to estimate the insulin secretory response and IS both at baseline and at follow-up. Results. The proportion of recipients with normal glucose tolerance (NGT) rose from 46% (baseline) to 65% (follow-up) (P ¼ 0.008), and median fasting and 2-h serum glucose were reduced by 0.7 mmol/l (P < 0.001) and 1.3 mmol/l (P ¼ 0.039), respectively. The recipients with PTDM at follow-up had a significant decline in the estimated median first and second phase ISec ( 58 and 47%, respectively, P ¼ for both). The patients who normalized their glucose tolerance from PTDM or IGT at baseline to NGT at follow-up increased their IS significantly (68%, P ¼ 0.002) without significant alterations in ISec. Conclusions. Impaired ISec seems to be the dominant mechanism in the development of PTDM after renal transplantation. In contrast, normalization of glucose intolerance is associated with improved IS. Correspondence and offprint requests to: Monica Hagen, Department of Medicine, Section of Nephrology, Laboratory for Renal Physiology, Rikshospitalet, 0027 Oslo, Norway. monica.hagen@rikshospitalet.no Keywords: insulin release; insulin sensitivity; posttransplant diabetes mellitus; renal transplantation Introduction Post-transplant diabetes mellitus (PTDM) and impaired glucose tolerance (IGT) are common complications after renal transplantation [1 3]. Both fasting and post-prandial hyperglycaemia may lead to microvascular complications as well as acceleration of macrovascular disease [4,5]. Impaired insulin secretion (ISec) and peripheral insulin resistance have both been identified as potential mechanisms in the pathogenesis of PTDM. It has been argued that the former may be more important than the latter [6]. We have shown recently that defects in insulin release as assessed 10 weeks after renal transplantation indicate a poor prognosis regarding normalization of glucose tolerance during the following year [7]. Moreover, in non-transplanted type 2 diabetics, like the Pima Indians, worsening of glucose tolerance from normal glucose tolerance (NGT) to IGT, and from IGT to diabetes mellitus, is accompanied by a progressive decline in the acute insulin secretory response and an increase in body weight [8]. In contrast, the subjects who remain normoglycaemic over time are characterized by maintained acute insulin response and glucose disposal. Long-term studies of renal transplant patients are scarce, still questioning whether PTDM may persist over several years in these patients, and whether lateonset diabetes may develop in the individuals who are normoglycaemic shortly after transplantation. The objectives of the present single-centre prospective observational study were first, to assess the glucose tolerance and the occurrence of PTDM in a group of renal transplant recipients over a period of 6 years after transplantation. Secondly, we tested the hypothesis of whether impaired insulin release is more important than insulin resistance in the development of PTDM. ß 2003 European Renal Association European Dialysis and Transplant Association

2 Diabetes mellitus after renal transplantation 2155 Subjects and methods Immunosuppressive therapy Subjects In 1995, 95 Caucasian renal transplant recipients underwent a 75 g oral glucose tolerance test (OGTT) 10 weeks after transplantation. None of these had diabetes prior to transplantation [2]. Six years later, the recipients in this cohort were evaluated for inclusion in the follow-up study. Eighteen patients had died, two had been retransplanted, two were in dialysis and 10 declined to participate, leaving a total of 63 patients to be included in the follow-up study (Figure 1). Each subject gave informed written consent before participating in the study, which was approved by the Regional Committee for Medical Research Ethics, Healthregion South, Norway. Out of 63 patients, 58 were re-examined with an OGTT, the majority (n ¼ 54) at our centre. The glucose challenge was not performed in five recipients, due to manifest PTDM treated with insulin (n ¼ 4) and lack of patient compliance (n ¼ 1). The OGTT was carried out after an overnight fast and included measurements of fasting, 1- and 2-h insulin and glucose concentrations. There were no statistically significant differences between the recipients included in the study and those who were lost to follow-up in terms of gender, age, blood pressure, use of antihypertensive and immunosuppressive medication, glucose tolerance, serum creatinine and number of rejection episodes. Stratification of the patients The recipients were divided into different categories of glucose tolerance according to the criteria given by the Expert Committee [4]: PTDM with fasting serum glucose 7.0 mmol/l or 2-h serum glucose 11.1 mmol/l; IGT with fasting glucose <7.0 mmol/l and 2-h glucose mmol/l; impaired fasting glucose (IFG) with fasting glucose mmol/l and 2-h glucose <7.8 mmol/l; and NGT with fasting serum glucose <6.1 mmol/l and 2-h serum glucose <7.8 mmol/l. Ten weeks after renal transplantation, all patients (n ¼ 63) were treated with prednisolone and cyclosporin A (CsA) (Sandimmun Neoral Õ ), and 81% with azathioprine. During the follow-up period, six recipients were switched to tacrolimus and two to mycophenolate mofetil, and three patients had their prednisolone treatment withdrawn. Excluding the subjects who changed immunosuppressive therapy from the statistical analyses did not change the overall findings. Analytical procedures The analysis of baseline serum glucose was performed using a glucose dehydrogenase method (Cobas Mira, Roche, Switzerland). At follow-up, whole blood glucose was measured with a HemocueAB Õ B-glucose Analyser (Angelholm, Sweden) in 54 recipients, whereas glucose was analysed in venous serum directly in nine recipients. Whole blood glucose values were recalculated to venous serum values [9]. Serum insulin was determined by a commercial radioimmunoassay (Coat-A-Count TM, Diagnostic Products Corporation, Los Angeles, CA) at baseline and by a fluoroimmunoassay (Auto DELFIA TM Insulin, Wallac Oy, Turku, Finland) at follow-up. Creatinine clearance was calculated according to the Cockcroft and Gault formula [10]. Insulin release and sensitivity indices Insulin release was estimated by the use of three OGTTderived indices. These equations have been validated in patients with varying degrees of glucose tolerance and correlate well with the results from hyperglycaemic clamp studies [11,12]. Using the trapezoid rule, the area under the curve (AUC) insulin and the AUC glucose during the OGTT were calculated and implemented in the insulin release index: Secr AUC ¼ AUC Ins /AUC Gluc [11,12]. The first and second phase insulin release were estimated according to the following equations: Secr 1.phase ¼ 1194 þ Ins Gluc 1 þ Ins 1 ;Secr 2.phase ¼ 295 þ Ins Gluc 1 þ Ins 0, where Ins 0 and Ins 1 are serum insulin levels at 0 and 1 h after an OGTT, respectively, and Gluc 1 represents serum glucose 1 h post-load [12]. The OGTT-derived insulin sensitivity (IS) index [ISI TX ¼ body mass index (BMI) Ins Gluc 2 ], which is a modification of the IS index proposed by Stumvoll et al. [12], was implemented as the surrogate estimate of IS [11,13]. The ISI TX has recently been validated in renal transplant recipients and correlated closely with the results from hyperinsulinaemic euglycaemic clamp (r ¼ 0.58, P < 0.001) [13]. Fig. 1. Flow chart of patient inclusion. Statistical analyses Results are given as median and range. For continuous data, the Mann Whitney or the Wilcoxon matched pairs signed rank sum test was used, whereas Pearson 2 or McNemar s test was implemented for categorical data as appropriate. Linear regression was implemented in the analysis of potential independent continuous variables, with 2-h serum glucose or IS as the dependent variables. Multiple linear regression was used to assess any independent predictor of change in glucose

3 2156 M. Hagen et al. tolerance or IS. All variables associated with 2-h serum glucose or IS in the univariate analyses with P-values <0.1 were included in the multiple regression model. Two tailed P-values are reported, and values <0.05 considered significant. The analyses were performed using the Statistical Package for the Social Sciences (SPSS version 10.0 for Windows, Chicago, IL). Results Changes in glucose tolerance from baseline to follow-up The prevalence of PTDM, IGT and NGT at baseline and follow-up is shown in Figure 2. The proportion of recipients with NGT rose from 46 to 65% during the 6 years (P ¼ 0.008). At baseline, 12 recipients (19%) had PTDM, IGT was found in 22 recipients (35%) whereas 29 patients (46%) had NGT. At follow-up, 14 recipients (22%) had PTDM, eight recipients (13%) had IGT and 41 patients (65%) had NGT. None of the patients had IFG, either at baseline or at follow-up. Five out of 12 patients with diabetes at baseline improved to NGT (n ¼ 4) or IGT (n ¼ 1) during the follow-up period. Among the 22 recipients with IGT at baseline, 11 patients improved to NGT, whereas six progressed to PTDM. One recipient with NGT at baseline deteriorated to PTDM during the follow-up period. The majority (26 out of 29) of the patients with NGT 10 weeks after renal transplantation were still euglycaemic 6 years later. In the total patient population, median fasting (P < 0.001) and 2-h (P ¼ 0.039) serum glucose were Fig. 2. Changes in glucose tolerance category from baseline (10 weeks) to follow-up 6 years after renal transplantation. PTDM, post-transplant diabetes mellitus; IGT, impaired glucose tolerance; NGT, normal glucose tolerance. significantly lower 6 years after renal transplantation as compared with baseline (Table 1). This coincided with a 50% decrease in median daily prednisolone dose and CsA whole blood trough concentration, a significant increase in median BMI and a decrement in median serum triglyceride concentration. The renal function improved slightly from baseline to follow-up. When the change in 2-h serum glucose was considered as being the dependent variable, univariate linear regression analysis showed a significant association between reduction in daily prednisolone dose and improvement in glucose tolerance ( ¼ 0.12, r 2 ¼ 0.069, P ¼ 0.046). On the other hand, neither the increase in BMI nor the decline in CsA whole blood trough concentration were associated with an alteration in glucose tolerance. Hypoglycaemic therapy Seven recipients had PTDM both at baseline and at follow-up. Of these, two patients were treated with insulin both at baseline and at follow-up, one recipient was switched from glipizide to insulin during the period and one had metformin withdrawn. The three remaining subjects were not treated with any hypoglycaemic drugs, either at baseline or at follow-up. None of the five recipients who improved their glucose tolerance from PTDM to either IGT (n ¼ 1) or NGT (n ¼ 4) were on glucose-lowering drugs at baseline. One recipient, who deteriorated from IGT to PTDM, was on a combination treatment with insulin and metformin at follow-up. None of the recipients who were using hypoglycaemic therapy at follow-up went through the OGTT. Insulin secretion (ISec) and insulin sensitivity (IS) Calculated indices of ISec and IS for all recipients are given in Table 2. All the indices of ISec were significantly reduced 6 years after renal transplantation (Secr 1.phase, P < 0.001; Secr 2.phase, P < 0.001; Secr AUC, P < 0.001), whereas median IS increased significantly (ISI TX, P ¼ 0.033). The changes in BMI and IS from baseline to followup were inversely correlated in the univariate model (linear regression: ¼ 0.006, r 2 ¼ 0.262, P < 0.001), whereas the reduction in median daily prednisolone dose and triglyceride level tended to be associated with improved insulin action ( ¼ 0.001, r 2 ¼ 0.067, P ¼ and ¼ 0.008, r 2 ¼ 0.057, P ¼ 0.098, respectively). When the change in prednisolone dose, triglyceride concentration and BMI were included in a multiple linear regression model, dose reduction of prednisolone was significantly associated with improved IS, whereas an increase in BMI was associated with a reduction in IS (P < for both, r 2 ¼ 0.34). Serum triglycerides were, on the other hand, not related to changes in IS in this model. None of these parameters were associated with changes in ISec.

4 Diabetes mellitus after renal transplantation 2157 Table 1. Changes in serum glucose, body weight, renal function, immunosuppressive and antihypertensive therapy during the first 6 years after renal transplantation (n ¼ 63) Baseline Follow-up Difference (%) P Serum glucose (mmol/l) Fasting 5.4 ( ) 4.7 ( ) 0.7 ( 13) < h 7.9 ( ) 6.6 ( ) 1.3 ( 16) Prednisolone dose (mg/day) 10 (10 30) 5 (0 10) 5 ( 50) <0.001 CsA trough concentration (mg/l) 255 ( ) 120 (35 215) 135 ( 53) <0.001 Creatinine (mmol/l) 136 (81 350) 130 (68 363) 6 ( 4) Creatinine clearance (ml/min) 55 (20 109) 60 (13 135) 5 (9) Body weight (kg) 69 (43 117) 73 (47 142) 4 (6) <0.001 BMI (kg/m 2 ) 22.8 ( ) 24.3 ( ) 1.5 (7) <0.001 Triglycerides (mmol/l) 1.9 ( ) 1.4 ( ) 0.5 ( 26) <0.001 Number of patients using Furosemide 20 (32%) 19 (30%) 1 ( 5) >0.9 -Blocker 25 (40%) 25 (40%) 0 (0) >0.9 Values are presented as median (range). Statistics: Wilcoxon matched pairs signed rank sum- or McNemar s test as appropriate. Table 2. Changes in estimated insulin secretion and insulin sensitivity during the first 6 years after renal transplantation (n ¼ 55) Association of changes in ISec and IS with changes in glucose tolerance 6 years after renal transplantation To assess the relative importance of ISec and IS on changes in glucose tolerance during the follow-up, the recipients were divided into two groups (Table 3). The group of patients who had PTDM at follow-up (n ¼ 14) was labelled diabetic. The group of recipients who normalized their glucose tolerance, from PTDM or IGT at baseline to NGT at follow-up (n ¼ 15), was labelled normoglycaemic. At baseline, these two groups were not significantly different in terms of IS, ISec, BMI, fasting and post-prandial serum glucose, kidney function and treatment with antihypertensive medication (P > 0.117). The diabetic group had a decline in estimated median insulin release of 58% (Secr 1.phase ) and 47% (Secr 2.phase ), whereas the ISI TX did not change significantly (Table 3). In contrast, the median ISI TX increased significantly (68%; P ¼ 0.002) whereas the median first and second phase insulin release were unchanged in the normoglycaemic group. The patients who had NGT both at baseline and at follow-up (n ¼ 26) tended to increase their IS [8.8 ( ) vs 9.3 ( ) 10 2 ; P ¼ 0.101]. All the ISec indices were significantly reduced during the period (P < 0.001). Baseline Follow-up Difference (%) P Insulin secretion indices Secr 1.phase 1293 ( 262 to 4537) 905 ( 448 to 3001) 388 ( 30) <0.001 Secr 2.phase 339 ( 7 to 1134) 245 ( 61 to 751) 94 ( 28) <0.001 Secr AUC 44 (10 159) 28 (7 124) 16 ( 36) <0.001 Insulin sensitivity index ISI TX ( 1.0 to 12.0) 9.1 ( 2.0 to 13.0) 1.8 (25) Values are presented as median (range). Statistics: Wilcoxon matched pairs signed rank sum test. Discussion Glucose tolerance In general, median fasting and 2-h serum glucose values declined significantly from baseline to follow-up. One half of the recipients with IGT and one-third of the recipients with PTDM improved to NGT. The recipients with PTDM who improved their glucose tolerance were not on glucose-lowering therapy at baseline. The majority of the recipients with NGT at baseline remained normoglycaemic after 6 years. Thus, NGT shortly after renal transplantation indicates a favourable prognosis for normoglycaemia in the long term [3]. On the other hand, about half (n ¼ 7) of the patients with PTDM at follow-up were diabetic already 10 weeks after transplantation. The other half (n ¼ 6) were mainly patients with IGT in the early course who progressed to PTDM, which supports the idea that IGT early after renal transplantation represents a risk factor for later development of PTDM [14,15]. Insulin sensitivity The IS improved significantly in the cohort in general over the 6 years. We previously have reported that tapering off prednisolone during the first year after renal transplantation is significantly associated with

5 2158 M. Hagen et al. Table 3. Changes in estimated insulin secretion and insulin sensitivity, BMI and daily prednisolone dose according to changes in glucose tolerance Diabetic a (n ¼ 14) Normoglycaemic b (n ¼ 15) Baseline Follow-up Difference (%) P Baseline Follow-up Difference (%) P Insulin secretion indices Secr 1.phase 858 ( ) 359 ( 448 to 914) 499 ( 58) ( ) 1000 ( ) 88 ( 8) Secr2.phase 239 (63 451) 127 ( 61 to 257) 112 ( 47) ( ) 266 (90 650) 31 ( 10) SecrAUC 30 (14 59) 25 (6 57) 5 ( 17) (19 159) 32 (15 117) 11 ( 26) Insulin sensitivity index ISI TX ( ) 5.8 ( ) 1.4 ( 19) ( 1.0 to 10.1) 9.6 ( 0.3 to 12.3) 3.9 (68) BMI (kg 2 /m) 22.9 ( ) 24.5 ( ) 1.6 (7) ( ) 23.8 ( ) 1.1 (5) Prednisolone (mg/day) 10 (10 30) 5 (0 10) 5 ( 50) (10 30) 5 (5 10) 10 ( 67) Triglycerides (mmol/l) 2.0 ( ) 1.4 ( ) 0.6 ( 30) ( ) 1.3 ( ) 0.9 ( 41) Values are presented as median (range). a Diabetic: Patients with PTDM at follow-up, who either stayed diabetic or progressed to PTDM from IGT or NGT. b Normoglycaemic: patients with PTDM or IGT at baseline who normalized their glucose tolerance (NGT) till follow-up. Statistics: Wilcoxon matched pairs signed rank sum test. improved glucose tolerance [3]. The present findings of a significant association between prednisolone tapering and improvement in IS may explain and support our previous data [3]. The group of recipients with PTDM or IGT at baseline who normalized their glucose tolerance was characterized by a significant increase in IS and preserved ISec. Interestingly, they also used higher daily doses of prednisolone than the progressors did in the early course. The subsequent greater dose reduction can be one plausible explanation for the improvement in IS [3]. Furthermore, hyperglycaemia itself predisposes for insulin resistance [16] and, as these patients gradually became normoglycaemic, they may have improved their IS by reducing the glucose exposure in insulin-sensitive tissue. However, the weight gain in both groups may have counteracted the improvement in IS. Insulin secretion The OGTT-derived first and second phase ISec decreased by half in the total patient population. Older age is known to be an important determinant of impaired -cell function [2,17]. However, we were not able to show a significant association between estimated ISec and age in the present study (data not shown). Since the glucose tolerance was improved during the period, the reduction in ISec seems mainly to be caused by increased IS. It is widely believed that diminution of first phase ISec is the earliest detectable defect of -cell function in the non-transplant human population [18]. Further, in the Pima Indians, the reduction in ISec is more pronounced than the fall in IS among subjects developing type 2 diabetes [9]. In the present study, the group of patients who developed PTDM during the follow-up suffered from a steep decline in ISec. Both the estimated acute and late phase insulin responses were diminished to a similar degree (Table 3). Thus, impaired ISec seems to be a prerequisite for the development of both PTDM and type 2 diabetes. A positive family history of diabetes is well known to be associated with increased risk of developing IGT [2,18,19]. The inability of the -cells to compensate for a decrease in insulin action distinguishes individuals who develop diabetes from those who retain NGT [18]. Five of the 14 diabetic recipients at follow-up reported diabetes mellitus in first-degree relatives, whereas this was the case in only one of the 15 non-diabetics. The immunosuppressive therapy, and particularly high doses of steroids in the early course, has probably revealed early glucose intolerance in predisposed pre-transplant non-diabetic individuals [2,6,13,20]. Although it has been suggested that the mechanism behind the diabetogenic effect of CsA is impaired insulin release, we were not able to find any association between the change in whole blood CsA trough levels and the change in estimated insulin secretion from baseline to follow-up. If any such an effect exists, it cannot be strictly dose dependent [3].

6 Diabetes mellitus after renal transplantation 2159 A limitation of the present study is that the surrogate estimates for ISec and IS may be inferior compared with insulin measures derived from hyperglycaemic and euglycaemic clamp studies. However, the ISec indices have been validated extensively in patients with different degrees of glucose tolerance and are strongly correlated with results from clamp studies [11,12]. In addition, the ISI TX has recently been validated in renal transplant subjects and found to correlate well with results from hyperinsulinaemic euglycaemic clamp [13]. Moreover, one should be aware that 32 patients were lost to follow-up, of whom 18 died. We do not know the glucose tolerance status of these recipients over the years. This may have biased our conclusions on the transitions between normoglycaemia and hyperglycaemia in the cohort. However, the fact that the persons lost to follow-up were not significantly different from the subjects included in the study at baseline makes it less likely that this is a major flaw of the study. In conclusion, a decline in insulin secretion seems to be the most important factor for the development of long-term PTDM. On the other hand, improvement in insulin sensitivity together with sufficient insulin release is associated with improved glucose tolerance during the first 6 years after renal transplantation. Acknowledgements. We appreciate the skilled technical assistance provided by Jean Stenstrøm, Kirsten Lund, Janicke Narverud and Els Breistein in the Laboratory for Renal Physiology. This study was financed by grants from the Norwegian Foundation for Health and Rehabilitation, the Norwegian Research Council and the Laboratory for Renal Physiology at Rikshospitalet. Conflict of interest statement. None declared. References 1. Jindal RM. Posttransplant diabetes mellitus a review. Transplantation 1994; 58: Hjelmesaeth J, Hartmann A, Kofstad J et al. Glucose intolerance after renal transplantation depends upon prednisolone dose and recipient age. Transplantation 1997; 64: Hjelmesaeth J, Hartmann A, Kofstad J, Egeland T, Stenstrom J, Fauchald P. Tapering off prednisolone and cyclosporine the first year after renal transplantation: the effect on glucose tolerance. Nephrol Dial Transplant 2001; 16: Expert Committee on Diagnosis and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 2000; 23: S4 S19 5. de Vegt F, Dekker JM, Ruhe HG et al. Hyperglycaemia is associated with all-cause and cardiovascular mortality in the Hoorn population: the Hoorn Study. Diabetologia 1999; 42: Nam JH, Mun JI, Kim SI et al. -Cell dysfunction rather than insulin resistance is the main contributing factor for the development of postrenal transplantation diabetes mellitus. Transplantation 2001; 71: Hjelmesaeth J, Hagen M, Hartmann A, Midtvedt K, Egeland T, Jenssen T. The impact of impaired insulin release and insulin resistance on glucose intolerance after renal transplantation. Clin Transplant 2002; 16: Weyer C, Bogardus C, Mott DM, Pratley RE. The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin Invest 1999; 104: Stahl M, Brandslund I, Jorgensen LG, Hyltoft Petersen P, Borch-Johnsen K, de Fine Olivarius N. Can capillary whole blood glucose and venous plasma glucose measurements be used interchangeably in diagnosis of diabetes mellitus? Scand J Clin Lab Invest 2002; 62: Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976; 16: Stumvoll M, Mitrakou A, Pimenta W et al. Use of the oral glucose tolerance test to assess insulin release and insulin sensitivity. Diabetes Care 2000; 23: Stumvoll M, Van Haeften T, Fritsche A, Gerich J. Oral glucose tolerance test indexes for insulin sensitivity and secretion based on various availabilities of sampling times. Diabetes Care 2001; 24: Hjelmesaeth J, Midtvedt K, Jenssen T, Hartmann A. Insulin resistance after renal transplantation: impact of immunosuppressive and antihypertensive therapy. Diabetes Care 2001; 24: Jarrett RJ, Keen H, McCartney P. The Whitehall Study: ten year follow-up report on men with impaired glucose tolerance with reference to worsening to diabetes and predictors of death. Diabet Med 1984; 1: Hjelmesaeth J, Hartmann A, Midtvedt K et al. Metabolic cardiovascular syndrome after renal transplantation. Nephrol Dial Transplant 2001; 16: Yki-Järvinen H. Glucose toxicity. Endocr Rev 1992; 13: Chiu KC, Lee NP, Cohan P, Chuang LM. Beta cell function declines with age in glucose tolerant Caucasians. Clin Endocrinol 2000; 53: Gerich JE. Is reduced first-phase insulin release the earliest detectable abnormality in individuals destined to develop type 2 diabetes? Diabetes 2002; 51: S117 S Aitman TJ, Todd JA. Molecular genetics of diabetes mellitus. Baillieres Clin Endocrinol Metab 1995; 9: Henriksen JE, Alford F, Ward GM, Beck-Nielsen H. Risk and mechanism of dexamethasone-induced deterioration of glucose tolerance in non-diabetic first-degree relatives of NIDDM patients. Diabetologia 1997; 40: Received for publication: Accepted in revised form: