Much concern has been expressed regarding the diabetogenic potential of the immunosuppressive drugs cyclosporin A (CSA) and tacrolimus in liver transplant patients, but all prospective studies relative to this subject have used an oral glucose test to screen for glucose intolerance (1–3). Oral glucose tolerance tests (OGTT) (4) reveal a wide range of plasma glucose and insulin levels, for the most part due to individual differences in the capacity to suppress endogenous glucose production, and to different levels in splanchnic and peripheral glucose uptake (5, 6). Moreover, portosystemic shunting and abnormal systemic hemodynamics (7), which are also present after liver transplantation, may contribute to the wide variety of these results. These factors influencing glucose tolerance, as assessed by OGTT, may explain the different and partly contradictory results concerning the impact of CSA or tacrolimus treatment on glucose tolerance in patients after orthotopic liver transplantation (OLT).
Posttransplant diabetes characterized by insulin resistance and absent biphasic insulin secretion from pancreatic β-cells (8) is commonly diagnosed after OLT (9, 10) and seems to be related to long-term therapy with immunosuppressive drugs (10–12). An overwhelming number of clinical and experimental studies indicate that pancreatic β-cells are the primary targets for the diabetogenic effects of CSA and tacrolimus. Andersson et al. (13) reported that CSA impairs pancreatic islet DNA synthesis, proinsulin biosynthesis, and insulin release from cultured mouse islets. Studies in perfused pancreas (14) showed that CSA suppresses glucose-induced first- and second-phase insulin response. Moreover, the inhibitory effect of CSA on insulin release may be specific for this drug (15). Animal and clinical studies indicate that tacrolimus has similar diabetogenic potency, in that it causes β-cell dysfunction (9, 16–18). Both drugs appear to have similar diabetogenic activity (19) in the long term—however, tacrolimus may have a less detrimental effect on glucose metabolism, because its higher immunosuppressive potency allows for a lower total steroid requirement (17). Because all of these studies were performed in liver-grafted patients with concomitant steroid therapy, these results should be interpreted cautiously with regard to their diabetogenic potency.
The pathogenesis of posttransplant hyperglycemia is multifactorial, and factors other than immunosuppressive therapy may also contribute to the impairment of glucose tolerance after OLT, for example, family history of diabetes (8, 20), recurrence of virus-induced liver disease (21), and the number of rejection episodes with higher cumulative steroid medication (10, 22). To clearly define the diabetogenic potency of CSA and tacrolimus, we selected only those liver-grafted patients from our recently published study (1) in whom these factors that interfere with glucose metabolism could be excluded. The first question of this study was to investigate whether nondiabetic, stable, long-term liver-grafted patients with single immunosuppressive therapy, i.e. CSA or tacrolimus, have impaired regulation of glucose tolerance compared with healthy control subjects, and second, whether this therapy has any detrimental influence on β-cell function as assessed by measurements of proinsulin serum concentrations as a marker of β-cell dysfunction (23) in the fasting state and after intravenous glucose stimulus.
Glucose tolerance is normally maintained by the adjustment of insulin sensitivity, first- and second-phase insulin secretion (24). The frequently sampled intravenous glucose tolerance test (FSIGTT), interpreted with the minimal model for glucose (24, 25) and C-peptide (26) disappearance, is a powerful non-invasive tool for the investigation of glucose metabolism in healthy (24), diabetic, (27) and cirrhotic subjects (28, 29). The application of such techniques allows the simultaneous determination of insulin-mediated glucose uptake as well as first- and second-phase insulin responsiveness of β-cells to glucose of an individual from data from a single FSIGTT (24, 26). In healthy as well as in cirrhotic subjects, a compensatory relationship between these factors exists, thus permitting normal glucose tolerance (24, 28–31). Pathological changes of these parameters are predictive and may indicate the risk for development of diabetes mellitus (32).
PATIENTS AND METHODS
After OLT, 20 patients (10 of whom received CSA and the other 10, tacrolimus) who proved to have normal glucose tolerance after OGTT, according to the WHO criteria (1), underwent FSIGTT. All patients suffered from histologically proven liver cirrhosis and had portal hypertension before transplantation as assessed by ultrasound measurements of hepatic/portal flow and by clinical manifestations such as ascites and esophageal varices. Some of them had the characteristic signs of malnutrition and were in a severe catabolic stage (CSA n=6, tacrolimus n=3). None of the liver transplant patients had diabetes mellitus before transplantation as assessed by normal glycosylated hemoglobin levels. Exclusion criteria were any rejection episode after OLT, peri- or postoperative insulin treatment, cardiovascular, renal, and infectious diseases, alcoholism and drug abuse, estrogen therapy, and a positive family history of diabetes mellitus. Graft function and liver enzymes were in the normal range. Only patients with non-virus-induced liver disease (CSA: alcoholic liver disease n=4, primary biliary cirrhosis n=5, autoimmune liver disease n=1; and tacrolimus: alcoholic liver disease n=6, primary biliary cirrhosis n=3, primary sclerosing cholangitis n=1) underwent these tests. The mean interval after OLT was 58±4 months in the CSA-treated group and 53±4 months in the tacrolimus-treated group. The immunosuppressive peri- and postoperative regimens in both groups were standardized as previously described (1). Concomitant steroid therapy was normally withdrawn 1 year after OLT, and the mean interval of monoimmunosuppressive therapy with CSA was 39±3 months and with tacrolimus was 42±4 months. The serum concentrations of CSA and tacrolimus are shown in Table 2. Antihypertensive medication (α blockers) were administered equally in the CSA- (n=4) and tacrolimus- (n=4) treated groups. None of these patients were taking β-blockers or diuretics.
The patients who received transplants were compared with glucose-tolerant, body weight-, and age- and sex-matched control subjects from our metabolic research unit (Table 1). All subjects consented to participate in the study, which was approved by the local ethics committee.
Before FSIGTT, blood samples were used to measure the serum levels of CSA and tacrolimus. After overnight fasting, baseline samples for glucose, insulin, and C-peptide were obtained at −15 min, −10 min, −5 min, and 0 min. Glucose (300 mg/kg body weight, 50% solution) was administered intravenously within 2 min, and the samples were collected as reported (27). Proinsulin concentrations were measured baseline at 0 min before and 2, 6, 10, 20, and 60 min after intravenous glucose loading.
Plasma glucose concentration was measured in duplicate by the glucose oxidase method with a glucose analyzer (Beckmann Clinical System 700). Blood samples for plasma insulin [immunoreactive insulin (IRI)] were immediately centrifuged at 4°C and stored at −20°C until analysis. To avoid any reaction with proinsulin, insulin concentrations were measured by Microparticle-Enzyme Immunoassay (MEIA Insulin, IMX System, Abbott, Germany). The assay coefficient of variation (CV) was 5.3%; total assay variation was 6.2%. The C-peptide immunoradiometric assay was purchased from Diagnostic Products DPC, Bad Nauheim, Germany. Human proinsulin was determined by ELISA (DRG Proinsulin ELISA; interassay CV, 5.8%; total assay variation, 5.0%). The lower limit of detection of the proinsulin assay was 0.5 pmol/L. CSA and tacrolimus serum levels were analyzed as described (1). Before FSIGTT was started, glucagon, cortisol, serum growth hormone, cholesterol, and triglycerides were measured with commercially available kits.
Glucose disappearance rate and kinetic models.
Intravenous glucose tolerance as assessed by the glucose disappearance rate (KG) was calculated—for purposes of comparison of the model-derived parameters—as the slope of the regression of logarithm of blood glucose concentration against time between 10 and 50 min after intravenous glucose injection (27). FSIGTT data analysis was based on minimal model of glucose disappearance (25). Glucose and insulin profiles were analyzed with SAAM II software, version 1.0.2 (University of Washington, Seattle, WA). Weights for the weighted least-squares optimization were chosen equal to the inverse of the variance of glucose measurement error, determined to be Gaussian, zero mean and with a standard deviation equal to 2% of the measured glucose value. To mitigate the consequences of the single-compartment approximation of glucose kinetics on which the minimal model relies, glucose data points up to 8 min were neglected during the fitting process. Basal end-test glucose and insulin concentrations, necessary for the minimal model identification, were calculated as the mean of the last three data points. Precision of parameter estimates was expressed as fractional standard deviation (the ratio between the standard deviation of the estimate and the estimated value, expressed as a percentage).
Insulin sensitivity (SI, 10−4 min−1 per μU/ml) represents the increase in net fractional glucose clearance rate per unit change in plasma insulin concentration subsequent to intravenous glucose load, and measures the effect of insulin on enhancing the glucose disappearance rate and inhibiting endogenous glucose release [i.e. it cannot segregate the contribution of the liver and the peripheral tissue to plasma glucose kinetics (24, 33)]. We used the modeling technique for C-peptide (26), which estimates both C-peptide kinetic and secretion from FSIGTT data. In keeping with the biphasic pattern of insulin secretion, the model assesses the first-phase sensitivity (Φ1) and the second-phase sensitivity (Φ2) to glucose. Φ1 is the amount of X0 of C-peptide in the β-cells secreted during the first phase normalized to maximum increment ΔG (maximal glucose − minimal glucose) of plasma glucose concentration after the injection, Φ1=X0/ΔG (dimensionless). The second-phase Φ2 (min−1) equals β and measures the stimulatory effect of glucose on insulin provision of new C-peptide Y which is controlled by the glucose concentration above the threshold level h: Φ2=Y=β (G − h) (26).
Data are expressed as mean±SE. To evaluate differences between the controls and the CSA- and tacrolimus-treated patients, we analyzed the data by one-way analysis of variance, or on Ranks if normality was not achieved. Correlations were determined by linear regression; a P value of < 0.05 was considered significant.
The physical characteristics and laboratory data of the liver-grafted patients and controls are shown in Tables 1 and 2. The CSA and tacrolimus groups did not differ demographically. Counter-regulatory hormones, fasting triglycerides, and cholesterol were not significantly different between liver-grafted patients and control subjects (Table 1).
Fasting glucose and insulin concentrations were similar in control subjects and patients (Fig. 1). Fasting C-peptide levels, however, were significantly elevated (P <0.05;Fig. 1). Note that fasting proinsulin concentrations and proinsulin/IRI ratio were in the same range in both the control group and the transplant patients (Table 1). After glucose loading, 2-, 6-, 10-, 20-, and 60-min proinsulin concentrations did not differ between the control subjects and both transplant groups as did the proinsulin/IRI ratio (data not shown).
FSIGTT profiles of glucose, insulin, and C-peptide are given in Figure 1. The rise of glucose and insulin levels after intravenous glucose loading and the subsequent decrease were similar in all groups. In contrast to the insulin courses, the C-peptide profiles of liver-grafted patients were different: a second increase followed the first peak immediately after glucose stimulus, the decay of C-peptide concentrations was slower, and the levels remained higher after 30 min in both liver transplant groups than in the control group (P <0.05).
Intravenous glucose tolerance index was above the diabetic range (>1.0 min−1) (27, 34), and there was no significant difference between control subjects and liver-grafted patients. Modeling analysis of FSIGTT data (Table 2) revealed no significant differences in insulin sensitivity between liver-grafted patients (CV 2–16%) and the control group (CV 2–11%). CSA but not tacrolimus serum concentrations were negatively related to insulin sensitivity (CSA:r = −0.67, P <0.05). With regard to first-phase insulin secretion (CV for X0 4–19%), there was no difference between the patient and the control groups, whereas in both liver transplant groups, a higher (P >0.05) second-phase responsiveness to glucose (CV 4–12%) was measured (Table 2).
Insulin sensitivity of our control group was negatively related to fasting glucose (r = −0.62, P <0.05), insulin (r = −0.69, P <0.01), and body mass index (BMI) (r = −0.74, P <0.01). In the liver-grafted groups, we also detected a negative relationship between insulin sensitivity to fasting glucose (CSA, r = −0.79; tacrolimus, r = −0.63, P <0.05), to fasting insulin (CSA, r = −0.66; tacrolimus, r = −0.74, P <0.05 for both), and BMI (CSA, r = −0.81, P <0.01; tacrolimus r = −0.66, P <0.05). Intravenous glucose tolerance index of all groups was strongly influenced by first-phase insulin secretion (CSA, r = 0.66, P <0.05; tacrolimus, r = 0.72, P <0.03; controls, r = 0.76, P <0.01). Second-phase insulin secretion was positively related to BMI (r = 0.76, P <0.05) and fasting insulin (r = 0.66, P =0.02) in the control group. However, these relationships were lacking in the liver-grafted groups. No relationship of CSA and tacrolimus serum concentrations to first- and second-phase insulin secretion was found. Because the relationship between insulin sensitivity and first-phase insulin response is hyperbolic (31), i.e., SI × *1 = constant, we also performed a regression analysis between SI and Φ1 −1 and found a positive relationship (r = 0.63, P <0.05) in our controls subjects. However, this compensatory interplay between insulin sensitivity and first-phase insulin secretion was lacking in both CSA (r = 0.22) and tacrolimus (r = 0.43, P >0.05) -treated patients.
The present study offers an integrated view on the regulation of glucose tolerance in stable, long-term liver-grafted patients treated with a single immunosuppressive drug. We could show that neither CSA nor tacrolimus impairs insulin sensitivity in liver-grafted patients. CSA treatment, however, has a concentration-dependent influence on insulin sensitivity, as demonstrated in animal and human studies (8, 35). In contrast to CSA which is a low-clearance drug, tacrolimus is absorbed over many hours, with peak plasma concentrations that are blunted in comparison to CSA. The oral formulation of tacrolimus may have “slow release” qualities and presents fewer fluctuations (36). This pharmacokinetic property of tacrolimus may be the reason for the absence of a relationship between serum tacrolimus concentrations and insulin sensitivity, and therefore, we cannot exclude the possibility that this drug has a similar concentration-dependent effect on insulin sensitivity. The body mass is commonly considered the predominant factor that influences insulin sensitivity (37). Our study clearly shows that body mass is also strongly associated with changes of insulin sensitivity in liver-grafted patients. The negative relationship of insulin sensitivity with fasting glucose in our liver-grafted patients may reflect the intact regulation of hepatic glucose production in the fasting state in patients after liver transplantation, as reported recently (38). Diminished insulin sensitivity—probably caused by the combination of reduced hepatic insulin degradation and subsequent hyperinsulinemia (39)—is common in nearly all patients with liver cirrhosis (40). In this regard, we can assume that liver transplantation is capable of correcting this characteristic abnormality of glucose metabolism present in cirrhosis.
We investigated the influence of CSA and tacrolimus on pancreatic β-cell function by measuring serum proinsulin concentrations before and after glucose stimulus. Proinsulin is formed in the process of insulin synthesis and is secreted along with insulin (23). Epidemiological studies revealed that elevated proinsulin levels and proinsulin/insulin ratio points to an impaired insulin synthesis (23) and are markers for metabolic decompensation in prediabetic subjects (41). Neither the fasting proinsulin nor the proinsulin/insulin ratio in our liver-grafted patients was abnormal, compared with those of healthy controls. Therefore, our study clearly shows that long-term immunosuppressive therapy with CSA or tacrolimus does not impair insulin synthesis in liver-grafted patients.
However, some abnormalities exist when C-peptide concentrations in the fasting state and after intravenous glucose load are compared with insulin levels. Because C-peptide is secreted in equimolar concentrations with insulin and is not removed by the liver, its kinetic behavior can be used as an indirect monitor of prehepatic pancreas insulin production (42). Therefore, the elevated fasting C-peptide in contrast to normal insulin levels and the rapid hepatic destruction of insulin (post hepatic insulin) in contrast to the increased C-peptide concentrations (pre hepatic insulin) during FSIGTT may point to an accelerated hepatic insulin clearance rate in our liver-grafted patients. Subsequently, the analysis of C-peptide data from FSIGTT that evaluates the prehepatic insulin secretion resulted in a similar first-phase insulin response in both liver-grafted groups and in the control group. However, the significant enhancement of second-phase insulin secretion in the liver-grafted groups was independent of the drug employed. Perseghin et al. (43) showed that the suppression of β-cell secretion under euglycemia and hyperinsulinemia was similar in liver grafted patients and in patients with chronic uveitis under CSA therapy. Fasting C-peptide levels, however, were higher in liver grafted patients only whereas fasting insulin concentrations did not differ between both groups. This effect was not observed in patients with chronic uveitis under CSA therapy. Francavilla et al. (44) also measured elevated fasting C-peptide levels, in contrast to lower insulin levels in liver-grafted patients. The authors of both studies suggested that this difference may result from increased hepatic insulin clearance. Portal hypertension and portal systemic shunting characterizes the hemodynamic alterations in cirrhotics (7). Although most of the hemodynamic abnormalities appear to be reversed after liver transplantation (45), hepatic blood flow is still increased over years (46–48) with subsequently accelerated, higher first-pass clearance of insulin. The discrepancy between insulin and C-peptide levels appears to be present in patients with advanced chronic liver diseases only and not in patients with normal liver function before OLT (44). The persistence of such hemodynamic alterations in liver-grafted patients may have detrimental metabolic consequences, resulting in a prolonged insulin production and, finally, in β-cell exhaustion.
The calculated intravenous glucose tolerance (KG) is closely related to first-phase insulin secretion in our healthy subjects and in our liver-grafted patients. First-phase insulin response after intravenous glucose load is mainly involved in the early decline of glucose after FSIGTT (24, 49) by inhibiting hepatic glucose production (50), and it determines the range of intravenous glucose disposal rate of an individual (24). Therefore, the suppression of hepatic glucose production by first-phase insulin secretion (50) after glucose loading appears to be intact after OLT.
Second-phase insulin secretion normally adjusts the overall glucose tolerance during FSIGTT to individual parameters such as body mass (24) and is modulated only by fasting insulin concentrations and not by fasting glucose concentrations (51). The close relationship of fasting insulin levels and body weight with second-phase insulin secretion in our control subjects may reflect this interplay. However, these relations were absent in our transplant patients, probably because of the enhanced insulin secretion which compensates for increased hepatic insulin clearance. The pancreatic hypersecretion may cause higher variations of insulin concentrations in the fasting state and may dissolve the relationship between fasting insulin and second-phase insulin secretion in our liver-grafted patients. Moreover, the insulin sensitivity and first-phase insulin secretion of our control subjects were linked through a negative feedback loop [as described by other studies in normal and nondiabetic cirrhotic patients (24, 29)], so that as an individual becomes more insulin resistant, β-cell secretion is enhanced. The large variations of first-phase insulin secretion measured in our liver-grafted patients were independent of the drug used. The absence of a relationship between insulin sensitivity and first-phase insulin secretion in liver transplant patients probably is caused by the higher hepatic first-pass effect of insulin, which may induce the larger variations of first-phase insulin secretion.
Insulin sensitivity and first-phase insulin response, which are strong predictors for the development of diabetes mellitus (32), are not impaired by long-term therapy with CSA and tacrolimus at normal clinical dosages. The normal course of insulin in contrast to elevated C-peptide levels during FSIGTT, the normal insulin sensitivity (posthepatic insulin), and the higher second-phase insulin secretion (prehepatic insulin) point to higher hepatic insulin degradation, which is compensated by a prolonged pancreatic hypersecretion. Although we did not evaluate hepatic and portal flow in these patients after OLT, we suggest that the hemodynamic alterations after OLT cause these effects and may strongly influence pancreatic function in stable long-term liver-grafted patients. Therefore, the prolonged insulin secretion after liver transplantation may contribute to the early exhaustion of β-cells, finally causing posttransplant diabetes.
The authors thank A. Sewell, Ph.D., for the preparation of the manuscript and the critical comments.
1. Steinmüller TM, Gräf K-J, Schleicher J, et al. The effect of FK 506 versus cyclosporin on glucose and lipid metabolism: a randomized trial. Transplantation 1994; 58:669.
2. Krentz AJ, Dousset B, Mayer D, et al. Metabolic effects of cyclosporin A and FK 506 in liver transplant recipients. Diabetes 1993; 42:1753.
3. Krentz AJ, Dmitrewski J, Mayer D, et al. Postoperative glucose metabolism in liver transplant recipients. Transplantation 1994; 57:1666.
4. Reaven GM, Brand RJ, Chen I, Mathur AK, Goldfine I. Insulin resistance and insulin secretion are determinants of oral glucose tolerance in normal individuals. Diabetes 1993; 42:1324.
5. Ferrannini E, Bjorkman O, George AR, et al. The disposal of an oral glucose load in healthy subjects. Diabetes 1985; 34:580.
6. Kelley D, Mitrakou A, Marsh H, et al. Skeletal muscle glycolysis, oxidation, and storage of an oral glucose load. J Clin Invest 1988; 81:1563.
7. Henderson MJ. Abnormal splanchnic and systemic hemodynamics of end-stage-liver disease: what happens after liver transplantation? Hepatology 1993; 17:514.
8. Ekstrand AV, Erikson JG, Grönhagen-Riska C, Ahonen JP, Groop LC. Insulin resistance and insulin deficiency in the pathogenesis of posttransplantation diabetes in man. Transplantation 1992; 53:563.
9. Jindal RM. Posttransplant diabetes mellitus: a review. Transplantation 1994; 58:1289.
10. Navasa M, Bustamente J, Marroni C, et al. Diabetes mellitus after liver transplantation: prevalence and predictive factors. J Hepatol 1996; 25:64.
11. Roth D, Milgrom M, Esquenazi V, Fuller L, Burke G, Miller J. Posttransplant hyperglycemia. Transplantation 1989; 47:278.
12. Boudreaux JP, McHugh L, Canafax DM, et al. The impact of cyclosporine and combination immunosuppression on the incidence of posttransplant diabetes in renal allograft recipients. Transplantation 1987; 44:376.
13. Andersson A, Borg H, Hallberg A, Hellerström C, Sandeler S, Schnell A. Long-term effects of cyclosporin A on cultured mouse pancreatic islets. Diabetologia 1984; 27:66.
14. Gillison SL, Bartlett ST, Curry DL. Synthesis-secretion coupling of insulin. Diabetes 1989; 38:465.
15. Gillison SL, Bartlett ST, Curry DL. Inhibition by cyclosporin of insulin secretion: a beta-cell-specific alteration of islet tissue function. Transplantation 1991; 52:890.
16. Strumph P, Kirsch D, Gooding W, Carrol P. The effect of FK506 on glycemic response as assessed by the hyperglycemic clamp technique. Transplantation 1995; 60:147.
17. Tabasco-Minguillan J, Mieles L, Carrol P, Gavaler J, Van Thiel DH, Starzl TE. Insulin requirements after liver transplantation and FK506 immunosuppression. Transplantation 1993; 56:862.
18. Mor E, Sheiner PA, Schwartz ME, Emre S, Guy S, Miller CM. Reversal of severe FK 506 side effects by conversion to cyclosporin-based immunosuppression. Transplantation 1994; 58:381.
19. Jindal RM, Popescu I, Schwartz ME, Emre S, Boccagni P, Miller CM. Diabetogenicity of FK 506 versus cyclosporin in liver transplant recipients. Transplantation 1994; 58:370.
20. De Fronzo RA. Pathogenesis of type 2 (non-insulin dependent) diabetes mellitus: a balanced overview. Diabetologia 1992; 35:389.
21. Rayfield EJ, Curnow RT, George DT, Beisel WR. Impaired carbohydrate metabolism during mild viral illness. N Engl J Med 1973; 289:618.
22. Delaunay F, Khan A, Cintra A, et al. Pancreatic β-cells are important targets for the diabetogenic effects of glucocorticoids. J Clin Invest 1997; 100:2094.
23. Kahn SE, Leonetti DL, Progeon RL, et al. Relationship of proinsulin and insulin with non-insulin-dependent diabetes mellitus and coronary heart disease in Japanese-American men—impact of obesity: clinical research center study. J Clin Endocrinol Metab 1995; 80:1399.
24. Bergman RN, Phillips LS, Cobelli C. Physiologic evaluation of factors controlling glucose tolerance in man. J Clin Invest 1981; 68:1456.
25. Bergman RN, Ider ZY, Bowden CR, Cobelli C. Quantitative estimation of insulin sensitivity. Am J Physiol 1979; 236:E667.
26. Toffolo G, De Grandi F, Cobelli C. Estimation of β-cell sensitivity from intravenous glucose tolerance test C-peptide data. Diabetes 1995; 44:845.
27. Konrad T, Vicini P, Kusterer K, et al. α-Lipoic acid treatment decreases serum lactate and pyruvate concentrations and improves glucose effectiveness in lean and obese patients with type 2 diabetes. Diabetes Care 1999; 22:280.
28. Kruszynska YT, Harry DS, Bergman RN, McIntyre N. Insulin sensitivity, insulin secretion and glucose effectiveness in diabetic and non-diabetic cirrhotic patients. Diabetologia 1993; 36:121.
29. Marchesini G, Pacini G, Bianchi G, Patrono D, Cobelli C. Glucose disposal, β-cell secretion, and hepatic insulin extraction in cirrhosis: a minimal model assessment. Gastroenterology 1990; 99:1715.
30. Reaven GM, Hollenbeck CB, Chen Y-DI. Relationship between glucose tolerance, insulin secretion, and insulin action in non-obese individuals with varying degrees of glucose tolerance. Diabetologia 1989; 32:52.
31. Kahn SE, Prigeon RL, McCulloch DK, et al. Quantification of the relationship between insulin sensitivity and β-cell function in human subjects. Diabetes 1993; 42:1663.
32. Haffner SM, Miettinen H, Gaskill SP, Stern MP. Decreased insulin secretion and increased insulin resistance are independently related to the 7-year risk of NIDDM in Mexican-Americans. Diabetes 1995; 44:1386.
33. Bergman RN, Ider YZ, Bowden CR, Cobelli C. Quantitative estimation of insulin sensitivity. Am J Physiol 1979; 236:E667.
34. Cerasi E, Luft R. The plasma insulin response to glucose infusion in healthy subjects and in diabetes mellitus. Acta Endocrinol 1967; 55:278.
35. Wahlstrom HE, Akimoto R, Endres D, Kolterman O, Moossa AR. Recovery and hypersecretion of insulin and reversal of insulin resistance after withdrawal of short-term cyclosporin treatment. Transplantation 1992; 53:1190.
36. Abu-Elmagd K, Fung JJ, Alessiani M, et al. The effect of graft function on FK506 plasma levels, dosages, and renal function, with particular references to the liver. Transplantation 1991; 52:71.
37. Bergman RN. Toward physiological understanding of glucose tolerance. Diabetes 1989; 38:1512.
38. Yale JF, Roy RD, Grose M, Seemayer TA, Murphy GF, Marliss EB. Effects of cyclosporin on glucose tolerance in the rat. Diabetes 1985; 34:1309.
39. Petrides AS, Stanley T, Matthews DE, Vogt C, Bush AJ, Lambeth H. Insulin resistance in cirrhosis: prolonged reduction of hyperinsulinemia normalizes insulin sensitivity. Hepatology 1998; 28:141.
40. Petrides AS. Liver disease and diabetes mellitus. Diabetes Reviews 1994; 2:2.
41. Haffner SM, Mykkänen L, Stern MP, Valdez RA, Heissermann JA, Bowsher RR. Relationship of proinsulin and insulin to cardiovascular risk factors in nondiabetic subjects. Diabetes 1993; 42:1297.
42. Cobelli C, Pacini G. Insulin secretion and hepatic extraction in humans by minimal modeling of C-peptide and insulin kinetics. Diabetes 1988; 37:223.
43. Perseghin G, Regalia E, Battezzati A, et al. Regulation of glucose homeostasis in humans with denervated livers. J Clin Invest 1997; 100:931.
44. Francavilla A, Polimeno L, Van Thiel DH, et al. Pancreatic hormones and amino acid levels following liver transplantation. Hepatology 1987; 7:918.
45. Navasa M, Feu F, García-Pagán JC, et al. Hemodynamic and humoral changes after liver transplantation in patients with cirrhosis. Hepatology 1992; 17:355.
46. Hadengue A, Lebrec D, Moreau R, et al. Persistence of systemic and splanchnic hyperkinetic circulation in liver transplant patients. Hepatology 1992; 17:175.
47. Henderson MJ, Millikan WJ, Hooks M, Noe B, Kutner MH, Warren DW. Increased galactose clearance after liver transplantation: a measure of increased blood flow through the denerved liver? Hepatology 1989; 10:288.
48. Chezmar JL, Redvanly RD, Nelson RC, Henderson MJ. Persistence of portosystemic collaterals and splenomegaly on CT after orthotopic liver transplantation. AJR 1991; 159:317.
49. Cobelli C, Pacini G, Toffolo G, Saccà L. Estimation of insulin sensitivity and glucose clearance from minimal model: new insights from labeled IVGTT. Am J Physiol 1986; 250:E591.
50. Luzi L, DeFronzo RA. Effect of loss of first-phase insulin secretion on hepatic glucose production and tissue glucose disposal in humans. Am J Physiol 1989; 257:E241.
51. Piatti PM, Pontiroli AE, Caumo A, et al. Hyperinsulinemia decreases second-phase but not first-phase arginine-induced insulin release in humans. Diabetes 1994; 43:1157.