Repaglinide in the Management of New-Onset Diabetes Mellitus After Renal Transplantation
The purpose of this study was to investigate the use of the short-acting insulin secretion drug repaglinide in new-onset diabetes mellitus (NODM) after renal transplantation. Twenty-three Caucasian patients with NODM after renal transplantation were selected to receive repaglinide therapy and were followed for at least 6 months. A control group treated with rosiglitazone was chosen for comparison. Successful repaglinide treatment was defined as a significant improvement of blood glucose concentrations and HbA1c <7% in the absence of glucosuria and without the need for the addition of further anti-diabetic agents. After 6 months of treatment with repaglinide, 14 of the 23 patients were successfully treated. Mean HbA1c decreased from 7.6 ± 0.6% to 5.8 ± 0.6% in 14 patients treated successfully. In nine patients, hyperglycemia persisted, and they were switched to insulin treatment (HbA1c 8.5 ± 2.9% at the beginning to 7.4 ± 2.2%). Mean serum creatinine levels, cyclosporine A and tacrolimus blood levels did not change significantly following institution of repaglinide therapy. The rate of successful treatment and the degree of HbA1c decrease were similar compared to rosiglitazone-treated control patients. The data from our observational study indicate that repaglinide can be an effective treatment option in Caucasian patients with NODM after renal transplantation.
New-onset diabetes mellitus (NODM) in renal transplant recipients is a serious complication associated with increased long-term morbidity and mortality (1–3). The use of immunosuppressive drugs, such as corticosteroids, cyclosporine and tacrolimus, has been recognized as a risk factor for the development of NODM with an incidence ranging from 2 to 53% among renal transplant recipients (4,5). The risk for the development of NODM is highest during the first months after transplantation or during the course of anti-rejection treatment. Insulin resistance and a defect in insulin secretion have been identified as the key mechanisms of immunosuppression-associated diabetes mellitus (6,7).
Treatment guidelines for NODM have recently been published. These involve modifications in the immunosuppressive therapy and non-pharmacological treatment options and in case of persistent NODM, therapy with oral agents as the next step (8). However, only a few studies have been conducted with oral agents in patients with NODM. These studies involve treatment with the insulin-sensitizers rosiglitazone and pioglitazone (9–11). However, treatment with glitazones is contraindicated for patients with heart failure, which is common in patients with renal disease (12,13), and occasional development of pronounced edema has been observed during treatment with glitazones after renal transplantation (10).
The glinides, a new class of short-acting insulin-secretagogues, have not been formally studied in transplanted patients. The glinides are structurally dissimilar to sulfonylureas and act directly on the pancreatic beta cell to stimulate rapid insulin secretion dependent on ambient glucose. Repaglinide is one of the few oral agents that can be used in chronic renal failure and is therefore a potential treatment option for NODM after renal transplantation (14).
The glinides, repaglinide and nateglinide, which can be used as a monotherapy or in combination therapy, reduce HbA1c levels by 1–1.5% and are mainly indicated to cover post-prandial hyperglycemic peaks by enhancing the early-phase insulin secretion. The targeting of this early-phase insulin release is thought to be important in reducing the long-term cardiovascular complications of diabetes as post-prandial blood glucose is more strongly correlated to HbA1c formation than fasting blood glucose (FBG) (15). Furthermore, the glinides allow a more flexible lifestyle with the possibility to skip a meal or to incorporate an additional meal preceded by an extra dose (16). In addition, the glinides have been associated with a reduced risk of hypoglycemic episodes especially in comparison to sulfonylureas (17). The purpose of this study was to investigate the use of repaglinide in patients with NODM after renal transplantation.
Patients and Methods
Twenty-three Caucasian patients (10 males and 13 females) with NODM after renal transplantation were selected to receive repaglinide therapy from 2000 to 2005. The mean age at diagnosis of NODM was 49 ± 11 years, and the median time at diagnosis of NODM was 4 months after transplantation. The body weight was 68 ± 11 kg with a mean BMI of 25 ± 4 kg/m2. Four of the patients had received a living donor transplant, and 19 a deceased donor transplant. Data were collected from our transplant database of about 1500 kidney transplant recipients. Data for a control group with NODM that were treated with rosiglitazone within the same time frame were also collected for comparison as rosiglitazone and repaglinide were regarded the oral drugs of the first choice for NODM in our center. Patients had to be followed for at least 6 months after the initiation of repaglinide or rosiglitazone therapy. The diagnostic criteria for diabetes mellitus as recommended by the 2003 international consensus guidelines for new-onset diabetes after transplantation were used (8). These criteria were symptoms of diabetes plus casual plasma glucose concentrations >200 mg/dL or fasting plasma glucose >126 mg/dL or 2-h plasma glucose >200 mg/dL during an oral glucose tolerance test (8). These criteria had to be confirmed on another day to result in the diagnosis of NODM.
Fifteen of the 23 patients had developed diabetes mellitus within the first year following transplantation. None of the patients was treated for diabetes mellitus prior to renal transplantation. Three patients had a family history of diabetes mellitus. Successful repaglinide treatment was defined as a significant improvement of blood glucose concentrations and a HbA1c less than 7% in the absence of glucosuria and addition of further glycemic control agents. Treatment failure was defined as the need for insulin therapy or for the addition of other oral agents due to persistence of hyperglycemia and glucosuria (10).
To assess safety of repaglinide treatment, the medical files of the patients were searched for side effects and recorded hypoglycemic episodes, and all patients were retrospectively questioned if they had hypoglycemia periods under treatment. Liver function was assessed by monitoring serum transaminases and parameters of cholestasis (alkaline phosphatase, bilirubin, gamma-glutamyl-transferase).
The underlying disorder for terminal renal failure was glomerulonephritis in seven patients, cystic kidney disease in three patients, nephrosclerosis in two patients, interstitial renal disease or congenital kidney disease in eight patients, hemolytic-uremic syndrome in one patient, HELLP-syndrome in one patient and kidney disease of unknown origin in one patient. Five patients had chronic hepatitis C and two had chronic hepatitis B as a concomitant disorder.
Thirteen patients received a dual immunosuppressive drug regimen (prednisone and tacrolimus: 7; prednisone and cyclosporine A: 5; prednisone and mycophenolate mofetil: 1). Ten patients received a triple drug regimen (prednisone, tacrolimus and mycophenolate mofetil: 5; prednisone, cyclosporine A and sirolimus: 2; prednisone, sirolimus and mycophenolate mofetil: 1; and prednisone, sirolimus and tacrolimus: 2). Fourteen patients were treated with tacrolimus (mean trough level 11.0 ± 3.0 ng/mL), seven with cyclosporine A (mean trough level 173 ± 65 ng/mL) and five with sirolimus (mean level 7.2 ± 0.9 ng/dL). All patients received prednisone (mean dose 9.4 ± 3.5 mg/day), and seven received mycophenolate mofetil (mean dose 500–1000 mg b.i.d.). Cyclosporine A levels were measured using the mFPIA assay (TDxFLx analyzer/Abbott Diagnostics®) and tacrolimus levels using the MEIA assay (IMx analyzer/Abbott Diagnostics®).
As recommended in the treatment guidelines for NODM (8), modifications of the immunosuppressive therapy with decreases in steroid dosage or reduction in calcineurin inhibitor trough levels were tried as appropriate.
Comparison with a control group receiving rosiglitazone
A group of 21 Caucasian patients with NODM after renal transplantation (10 males and 11 females) receiving rosiglitazone treatment was used for comparison with the repaglinide-treated patients. The mean age at diagnosis of NODM was 59 ± 9 years, and the median time at diagnosis of NODM was 4 months after transplantation. The mean body weight was 74 ± 11 kg with a mean BMI of 26 ± 3 kg/m2. Preliminary data of this patient group were previously published (10). Eighteen of these 21 patients had developed diabetes mellitus within the first year of transplantation. Rosiglitazone treatment was started after a median of 2 months (range 0–58 months) after diagnosis of NODM with a dose of 4 mg/day and was increased within the next 3 weeks to 8 mg/day. Diagnostic criteria for diabetes mellitus as well as criteria for successful treatment were the same as in the repaglinide group. The kidney disease was glomerulonephritis in nine patients, cystic kidney disease in seven patients, interstitial renal disease or congenital kidney disease in three patients, nephrosclerosis in one patient and bilateral nephrectomy due to kidney cancer in one patient. There were nine patients on a triple immunosuppressive regimen, and 12 patients received a double immunosuppressive regimen. Sixteen patients were treated with tacrolimus (mean trough level 9.8 ± 4.5 ng/mL), four with cyclosporine (mean trough level 165 ± 37.3 ng/mL), and two with sirolimus (mean trough level 7.1 ± 4.5 ng/mL). All patients received prednisone (mean dose 9.0 ± 4.2 ng/day) and eight patients mycophenolate mofetil (mean dose 500–1000 mg b.i.d.).
In comparison of the two groups, the only parameter that attained statistical significance was the age of the patients: Patients in the control group were significantly older (p < 0.05). There was no significant difference between the two groups in the other baseline data: weight, BMI, serum creatinine, and time of diagnosis after transplantation.
All data are expressed as mean ± SD. The Fisher Exact Probability Test was used for the analysis of frequency data. Means of ordinal data were compared using the Student's t-test or a Mann-Whitney U-test when the variable was not normally distributed. A p-value <0.05 was considered to be significant. SPSS for Windows® Release 11.0.1 (SPSS Inc.®) was used for statistical analysis.
Repaglinide therapy was started after a median of 4 weeks (range 0–377 weeks) following the diagnosis of NODM in a daily dose of 3 mg in four patients, 2.5 mg in one patient, 2 mg in one patient, 1.5 mg in 15 patients and 1 mg in two patients. Repaglinide was usually given three times per day with meals. In two patients started on 1.5-mg repaglinide, the dosage was increased to 3 mg during follow-up; in one patient started on 2.5 mg, the dosage was increased to 6 mg and in another one starting with 1 mg the daily dose was increased up to 15 mg.
Ten of the 23 patients were pre-treated with other oral agents and experienced side effects or inefficiency of this therapy, and repaglinide was substituted (n = 5) or added (n = 5). Four of these patients had been on metformin treatment before, three of them had been switched to repaglinide due to inefficiency or side effects and one continued metformin treatment during the study period after addition of repaglinide. Six patients had been on rosiglitazone treatment before repaglinide was instituted. Two of these had been switched to repaglinide due to persistent edema and four of them continued rosiglitazone as a comedication. No patient had been on sulfonylurea or insulin treatment before repaglinide was instituted. Altogether, 18 patients received repaglinide monotherapy and 5 patients a combination therapy with repaglinide (metformin and repaglinide (n = 1), rosiglitazone and repaglinide (n = 4)). Data were obtained in these patients after the institution of repaglinide.
After 6 months of treatment with repaglinide, 14 of the 23 patients were successfully treated. Mean HbA1c decreased from 7.6 ± 0.6% to 5.8 ± 0.6% in 14 patients treated successfully (p < 0.05). In nine patients hyperglycemia persisted, and they were switched to insulin treatment (HbA1c 8.5 ± 2.9% at the beginning to 7.4 ± 2.2%, five of these nine patients showed a decrease in HbA1c >1% but this was not sufficient to normalize hyperglycemia and HbA1c). FBG improved from 155 ± 58 mg/dL before repaglinide treatment to 118 ± 32 mg/dL after 6 months in the whole treatment cohort (p < 0.05). FBG of 14 patients successfully treated with repaglinide (n = 14) decreased from 151 ± 65 to 104 ± 13 mg/dL (p < 0.01). FBG remained high in nine patients: 161 ± 43 mg/dL before repaglinide treatment decreasing only to 141 ± 41 mg/dL. Factors predictive of success of repaglinide therapy were a higher FBG and higher HbA1c levels in patients with treatment failure, although these did not achieve statistical significance (Table 1).
Table 1. Baseline parameters of the patients treated with repaglinide
|Age at diagnosis of NODM (years)||48.5 ± 12.5||48.4 ± 7.6|
|Time since transplantation (weeks)||32.9 ± 58.7||23.1 ± 35.5|
|Weight (kg)||70.7 ± 11.1||64.7 ± 10.8|
|BMI (kg/m2)||25.9 ± 3.1||23.9 ± 4.3|
|Serum creatinine (mg/dL)||1.9 ± 0.7||1.8 ± 0.8|
|Fasting blood glucose (mg/dL)||151 ± 65||161 ± 43|
|HbA1c (%)||7.6 ± 0.6||8.5 ± 2.9|
|Prednisone dose (mg/day)||9.1 ± 2.9||9.7 ± 4.2|
|Mean tacrolimus level (ng/mL)||11.0 ± 2.7||11.0 ± 3.4|
|On cyclosporine A||6||1|
|Mean cyclosporine A level (12 h; ng/mL)||182 ± 67||126|
|Sirolimus levels (ng/mL)||6.6 ± 0||7.5 ± 0.9|
In the five patients on combination therapy including repaglinide, the patient on combination therapy with metformin and repaglinide had to be switched to insulin therapy after 17 weeks because of persistent hyperglycemia. The four patients on combination therapy with rosiglitazone and repaglinide were all successfully treated. Mean HbA1c decreased in the patients on a combination therapy after the addition of repaglinide from 8.2 ± 3.6% to 7.1 ± 1.8% and in the patients receiving repaglinide monotherapy to a similar degree from 7.5 ± 1.3% to 6.2 ± 1.5%.
In two patients with successful treatment, repaglinide was stopped by the patients after 3 and 7 months because of persistent nausea (n = 1) and light diarrhea (n = 1). Both patients reported improvement of their symptoms after stopping repaglinide, and anti-diabetic treatment was continued with rosiglitazone.
During repaglinide treatment, body weight and BMI did not change significantly (before repaglinide 68.4 ± 11.4 kg vs 69.8 ± 13.2 kg (p = 0.71) and 25.1 ± 3.7 kg/m2 vs 25.6 ± 4.5 kg/m2 (p = 0.67)). Mean serum creatinine levels did not change significantly during treatment (1.83 ± 0.72 to 1.95 ± 0.90 mg/dL (p = 0.64)). Cyclosporine A, tacrolimus and sirolimus blood levels did not change significantly following institution of repaglinide therapy, and relevant dose adjustments were not necessary (tacrolimus: 9.2 ± 2.1 ng/mL vs 11.7 ± 3.8 ng/mL (p = 0.25), cyclosporine A: 131 ± 24.5 ng/mL vs 134 ± 31.9 ng/mL (p = 0.90), sirolimus: 7.5 ± 0.9 ng/mL vs 6.9 ± 0.2 ng/mL (p = 0.58), blood levels at the day of starting repaglinide and next available blood level 3–7 days later; patients with modifications of the immunosuppressive dose at the day repaglinide was instituted were excluded from this analysis).
In the patients with chronic viral hepatitis, transaminase levels did not change significantly during treatment (ALT: 79.7 ± 148.7 U/L vs 44.8 ± 27.6 U/L, p = 0.27, AST: 77.3 ± 153.3 U/L vs 33.3 ± 19.9 U/L, p = 0.22). The high levels in the beginning were caused by one patient having transaminase levels of nearly 400 U/L. There was also no clinical or other biochemical evidence of worsening of liver disease in these patients.
Three patients reported symptoms of a hypoglycemic episode during treatment with repaglinide. In one case, this was accompanied by a recorded blood glucose below 50 mg/dL.
Comparison with a control group treated with rosiglitazone
Two of 21 patients in the rosiglitazone group had to stop the medication within the first weeks due to edema formation and were excluded from this analysis. After 6 months of treatment, the rate of successful treatment was similar in 23 repaglinide and the remaining 19 rosiglitazone-treated patients (14 of 23 (61%) vs 14 of 19 (74%)). The decrease in HbA1c and FBG was almost identical in the complete cohorts of both treatments (mean delta HbA1c 1.28 vs 1.20%; mean delta FBG 37 and 36 mg/dL). Patients were then separated according to their mean FBG before initiation of treatment. For both treatment cohorts results were inferior for patients starting with FBG >160 mg/dL vs patients with FBG <160 mg/dL (success rates: repaglinide: 33% and 79%; rosiglitazone 54% and 100%) (Table 2).
Table 2. Comparison of 23 patients treated with repaglinide with a control group of 19 patients treated with rosiglitazone after 6 months of follow-up (2 patients that stopped rosiglitazone within a few weeks due to side effects were excluded from this analysis)
|Age at diagnosis of NODM (years)*||48.5 ± 10.8||58.6 ± 9.3|
|Median time since transplantation (months)||4||4|
|Weight (kg)||68.4 ± 11.4||74.2 ± 11.2|
|BMI (kg/m2)||25.1 ± 3.7||26.4 ± 2.9|
|Success rate||14 of 23 (61%)||14 of 19 (74%)|
|Mean delta HbA1c (%)||−1.3||−1.2|
|Mean delta FBG (mg/dL)||−37||−36|
|Success rate with starting FBG >160 mg/dL||3 of 9 (33%)||6 of 11 (54%)|
|Success rate with starting FBG <160 mg/dL||11 of 14 (79%)||8 of 8 (100%)|
|Success rate of patients on tacrolimus||8 of 14 (57%)||10 of 14 (71%)|
|Success rate of patients on cyclosporine A||6 of 7 (86%)||3 of 4 (75%)|
|Success rate of patients on sirolimus||3 of 5 (60%)||0 of 2 (0%)|
|Success rate of patients on tacrolimus + sirolimus||2 of 2||0 of 1|
|Success rate of patients on cyclosporine A + sirolimus||1 of 2||0 of 1|
Repaglinide treatment was successful in three of five patients with sirolimus therapy, whereas rosiglitazone was unsuccessful in two of two patients with sirolimus therapy. Success of treatment was independent of patients' age in both groups. Four patients with initially unsuccessful treatment with rosiglitazone monotherapy had repaglinide added in the later course with subsequent improvement in glycemic control.
The data from our observational study investigating the use of glinides for the first time in a transplant patient group indicate that repaglinide is a new treatment option in Caucasian patients with NODM after renal transplantation. Deleterious effects on the transplant or significant interactions with the immunosuppressive medications were not observed, and efficacy was similar to a control group treated with rosiglitazone. Side effects observed in two patients were mild and resolved completely after cessation of therapy, and only three mild hypoglycemic episodes were reported.
Although the recently published guidelines for NODM suggest oral monotherapy as the first step of medical anti-diabetic therapy (8), it is not clear which agents can be safely used in a transplant population as there are very few studies in these patients. As we and others have shown (9,10), the insulin sensitizer rosiglitazone can be safely administered to patients after solid organ transplantation, potentially improving glycemic control caused by insulin resistance. Metformin, which also improves insulin resistance, is rarely used after renal transplantation as it is absolutely contraindicated in patients with renal failure. The data presented in this study indicate that insulin secretion agents can also be useful in patients with NODM. The use of repaglinide adds another principle for oral therapy of NODM. Repaglinide seems to be preferable to the other insulin secretion group of drugs, the sulfonylureas, as it is not accumulating in renal failure (13). Furthermore, the sulfonylureas as well as metformin have not formally been studied in a transplant population so far.
Like other oral agents, repaglinide is not fully effective in all patients, although HbA1c levels decreased to a certain degree in all patients. This indicates that insulin resistance may have contributed to the development of diabetes mellitus in patients with treatment failure. Therapy with repaglinide had low success rates in patients with higher starting HbA1c and FBG >160 mg/dL. Therefore, as indicated by our data, a combination therapy of repaglinide with insulin sensitizers may be very useful in certain patients as insulin deficiency and insulin resistance can both contribute to the development of NODM. The marked improvement in fasting glucose observed after repaglinide therapy is explained by a general improvement of the metabolic situation. As repaglinide primarily reduces glucose concentrations after meals, this should also improve subsequent fasting concentrations.
In addition, it is of interest that repaglinide is metabolized by the human cytochrome P450 isoenzymes CYP3A4 and CYP2C8 (14). The immunosuppressive drugs cyclosporine A and tacrolimus are also metabolized via CYP3A4 (18). Our data did not show evidence for a clinically important interaction of repaglinide metabolism on cyclosporine A and tacrolimus blood levels, although we did not formally investigate this pharmacological interaction. Furthermore, it is known that the dual CYP biotransformation of repaglinide via CYP3A4 and CYP2C8 has consequences for drug–drug interactions as one of these CYP pathways has sufficient capacity to compensate if the other is inhibited (14). Therefore, relevant changes in repaglinide drug concentrations due to interactions with calcineurin inhibitors are unlikely to occur.
In conclusion, this is the first report showing that repaglinide can be an effective treatment option in Caucasian patients with NODM after renal transplantation.