Statin therapy is associated with the development of new-onset diabetes after transplantation in liver recipients with high fasting plasma glucose levels

Authors

  • Yongin Cho,

    1. Division of Endocrinology and Metabolism, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Min Jung Lee,

    1. Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Eun Yeong Choe,

    1. Division of Endocrinology and Metabolism, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    Search for more papers by this author
  • Chang Hee Jung,

    1. Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
    Search for more papers by this author
  • Dong Jin Joo,

    1. Department of Transplantation, Yonsei University College of Medicine, Seoul, Korea
    Search for more papers by this author
  • Myoung Soo Kim,

    1. Department of Transplantation, Yonsei University College of Medicine, Seoul, Korea
    Search for more papers by this author
  • Bong Soo Cha,

    1. Division of Endocrinology and Metabolism, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    2. Brain Korea 21 Project for Medical Science, Yonsei University College of Medicine, Seoul, Korea
    3. Institute of Endocrine Research, Yonsei University College of Medicine, Seoul, Korea
    Search for more papers by this author
  • Joong-Yeol Park,

    Corresponding author
    1. Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
    • Address reprint requests to Joong-Yeol Park, M.D., Ph.D., Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-Ro, 43-Gil, Songpa-Gu, Seoul 138-736, Korea. Telephone: +82-2-3010-3240; FAX: +82-2-3010-6962; E-mail: jypark@amc.seoul.kr Address reprint requests to Eun Seok Kang, M.D., Ph.D., Division of Endocrinology and Metabolism, Department of Internal Medicine, Yonsei University College of Medicine, 50 Yonsei-Ro, Seodaemun-Ku, Seoul 120-752, Korea. Telephone: +82-2-2228-1968; FAX: +82-2-393-6884; E-mail: edgo@yuhs.ac

    Search for more papers by this author
  • Eun Seok Kang

    Corresponding author
    1. Division of Endocrinology and Metabolism, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    2. Brain Korea 21 Project for Medical Science, Yonsei University College of Medicine, Seoul, Korea
    3. Institute of Endocrine Research, Yonsei University College of Medicine, Seoul, Korea
    • Address reprint requests to Joong-Yeol Park, M.D., Ph.D., Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-Ro, 43-Gil, Songpa-Gu, Seoul 138-736, Korea. Telephone: +82-2-3010-3240; FAX: +82-2-3010-6962; E-mail: jypark@amc.seoul.kr Address reprint requests to Eun Seok Kang, M.D., Ph.D., Division of Endocrinology and Metabolism, Department of Internal Medicine, Yonsei University College of Medicine, 50 Yonsei-Ro, Seodaemun-Ku, Seoul 120-752, Korea. Telephone: +82-2-2228-1968; FAX: +82-2-393-6884; E-mail: edgo@yuhs.ac

    Search for more papers by this author

  • Yongin Cho and Min Jung Lee participated in the design and performance of the study, the data collection and analysis, the data interpretation, and the writing of the manuscript. Eun Yeong Choe participated in the design and performance of the study, the data analysis, and the data interpretation. Dong Jin Joo participated in the data collection. Myoung Soo Kim, Bong Soo Cha, and Joong-Yeol Park contributed to the discussion. Chang Hee Jung participated in the design and performance of the study and the data collection and analysis. Eun Seok Kang participated in the design and performance of the study, the data collection and analysis, the data interpretation, and the writing of the manuscript.

  • All the authors declare that there is no duality of interest associated with this article.

  • This work was financially supported by the Kiturami Faculty Research Assistance Program of the Yonsei University College of Medicine (6-2012-0148) and by a National Research Foundation grant funded by the Korean Ministry of Education, Science, and Technology through the Basic Research Promotion Fund (NRF-2010-013-E0008 and NRF-2012-000891).

Abstract

New-onset diabetes after transplantation (NODAT) and dyslipidemia are important metabolic complications after liver transplantation (LT) that can adversely affect both allograft and patient survival. Statins are used as first-line therapies for dyslipidemia because of their effectiveness and safety profile. However, it has recently been reported that statin therapy is associated with new-onset diabetes in the nontransplant population. The aim of this study was to investigate the association between statin therapy and the development of NODAT in LT recipients. Three hundred sixty-four LT recipients who underwent transplantation between the ages of 20 and 75 years without a previous history of diabetes were enrolled in this study. We evaluated the incidence of NODAT with respect to statin use as well as other risk factors. The incidence of NODAT was significantly higher in the statin group (31.7%) versus the control group (17.6%, P = 0.03). The mean follow-up period was 37.8 ± 19.0 months for the statin group and 42.7 ± 16.0 months for the control group (P = 0.07). Statin use was significantly associated with NODAT development after adjustments for other risk factors [hazard ratio (HR) = 2.32, 95% confidence interval (CI) = 1.23-4.39, P = 0.01]. Impaired fasting glucose before transplantation was also a risk factor for NODAT development (HR = 2.21, 95% CI = 1.36-3.62, P = 0.001). There were no significant differences in age, body mass index, cumulative corticosteroid dose, or fasting plasma glucose (FPG) levels between the groups. Patients with high FPG levels were more likely to develop NODAT when they were placed on statins after LT (P = 0.002). In conclusion, statin treatment could contribute to the development of NODAT in LT recipients, especially if they have high baseline FPG levels. Liver Transpl 20:557–563, 2014. © 2014 AASLD.

In recent years, liver transplantation (LT) has become a widely accepted treatment modality for patients with end-stage liver disease. With advances in surgical techniques, organ preservation, and immunosuppressant therapy, transplantation outcomes have continued to improve.[1] As life expectancy after LT has increased, metabolic complications such as dyslipidemia and new-onset diabetes after transplantation (NODAT) have also been on the rise.[2] Thus, it has become exceedingly important to prevent and/or manage these metabolic complications because they can negatively affect both graft and patient survival.

NODAT in patients with liver allografts is significantly associated with hepatic artery thrombosis as well as acute and chronic rejection.[3, 4] In addition, morbidity and mortality are greater in patients with NODAT.[5, 6] It has been shown that the diabetogenic effects of immunosuppressants play a major role in the development of NODAT.[7, 8] Current immunosuppressant regimens used in this population include steroids, calcineurin inhibitors (cyclosporine and tacrolimus), and sirolimus. Other variables, including genetic factors,[9, 10] older age,[11, 12] a high body mass index (BMI),[13] the presence of hepatitis C virus,[11] and impaired fasting glucose (IFG),[14] are also associated with NODAT development.[15] Notably, NODAT is a major independent risk factor for cardiovascular morbidity and mortality as well as graft loss.[16] Although dyslipidemia is uncommon in patients with liver cirrhosis, it is often observed in LT patients.[17, 18]

Statins (3-hydroxy-methylglutaryl coenzyme A reductase inhibitors) are widely used for the treatment of dyslipidemia in the general population and are considered to be first-line treatments for LT recipients to prevent cardiovascular disease by lowering serum low-density lipoprotein cholesterol levels.[19] However, several recent meta-analyses have revealed that statin therapy increases the risk of new-onset diabetes mellitus.[20] Few studies on the diabetogenic effects of statins in transplant patients have been reported.

The aim of this study was to investigate the association between statin therapy and NODAT development in LT recipients.

PATIENTS AND METHODS

Subjects

We conducted a retrospective, longitudinal, observational study of patients who underwent LT between January 1, 2005 and May 31, 2012 at Severance Hospital (Seoul, Korea) or Asan Medical Center (Seoul, Korea). Recipients between the ages of 20 and 75 years who survived for longer than 12 months after transplantation were included in this study. All patients with a history of diabetes mellitus or diabetes medication use before transplantation were excluded. All patients on statins at the time of LT were also excluded. Clinical data, including height, weight, and BMI, were measured, and the prevalence of hypertension and viral hepatitis was evaluated at the baseline. Fasting plasma glucose (FPG) levels and lipid profiles were also measured at the baseline (before transplantation) and 6, 12, 24, and 36 months after transplantation. BMI was also measured 12, 24, and 36 months after transplantation. From these data, we evaluated the incidence of NODAT according to statin use and other associated risk factors. Transient hyperglycemia after transplantation that was resolved within 6 months without treatment[21] was frequently observed, and these patients were not included in the NODAT group. The endpoint of this study was the development of NODAT 6 months or more after transplant surgery. NODAT was defined as an FPG level ≥ 126 mg/dL on 2 consecutive measurements, a hemoglobin A1c value ≥ 6.5%, or the use of glucose-lowering medications according to the 2003 international consensus guidelines.[22] Patients with FPG levels between 100 and 125 mg/dL were classified as having IFG. The study protocol was approved by the ethics committee of the Yonsei University College of Medicine.

Statistical Analysis

All statistical analyses were performed with the SPSS statistical analysis program (SPSS System for Windows 20.0, SPSS, Inc., Chicago, IL). All values are expressed as means and standard deviations or numbers and percentages. The normality of the distribution of continuous variables was examined with Shapiro-Wilk tests. Between-group differences of mean values were compared with unpaired t tests, and between-group differences of numbers and percentages were compared with χ[2] tests.

Any variable with a P value < 0.05 in the univariate analysis was included in the final multivariate model. Kaplan-Meier survival curves were used to compare disease-free survival. Multivariate Cox regression was performed with adjustments for significant predictors in the univariate analysis as well as other factors previously shown to be associated with the development of NODAT. For all statistical tests, a P value < 0.05 was considered to be statistically significant.

RESULTS

Clinical Characteristics of the Study Population and NODAT Development

Three hundred sixty-four LT recipients who underwent surgery between the ages of 20 and 75 years without a previous history of diabetes were enrolled in this study. One hundred fifty-four of these patients underwent LT at Severance Hospital, and 210 patients underwent LT at Asan Hospital. Three hundred fifty-four subjects were followed for at least 12 months after surgery, and 251 recipients were followed for at least 48 months. Forty-one recipients started on statins before the diagnosis of NODAT and used them for more than 6 months after LT. Twenty-one patients were treated with atorvastatin, 8 were treated with fluvastatin, 6 were treated with pravastatin, 4 were treated with pitavastatin, and 2 were treated with rosuvastatin. The primary reason for starting statin therapy was hypercholesterolemia, and the mean total cholesterol level before statin use was 251.4 ± 52.5 mg/dL. None of the patients in the statin group had been diagnosed with NODAT before they started statin therapy.

NODAT developed in 13 of the 41 statin users. The mean overall follow-up period was 42.1 ± 16.4 months. The control group was followed longer than the statin group, although this difference was not statistically significant (42.7 ± 16.0 versus 37.8 ± 19.0 months, P = 0.07; Table 1). The proportion of men was higher in the control group versus the statin group (70.9% versus 53.7%, P = 0.03; Table 1). There were no differences in the baseline BMI values or BMI changes (0, 1, 2, and 3 years after transplantation) between the 2 groups (Table 1). The prevalence of hypertension was higher in the statin group versus the control group (34.1% versus 17.3%, P = 0.01), as was the total cholesterol level before LT (128.1 ± 69.3 versus 114.6 ± 40.7 mg/dL, P = 0.07). There were no significant differences in age, BMI, prevalence of obesity, cumulative corticosteroid dose, prevalence of hepatitis C, donor type, incidence of acute rejection, or FPG levels between the groups (Table 1).

Table 1. Baseline Clinical Characteristics of the Patients
 Control Group (n = 323)Statin Group (n = 41)P Value
  1. NOTE: Values are presented as means and standard deviations unless otherwise noted. Differences between groups were analyzed with chi-square tests or unpaired t tests. Cumulative doses of steroids are equivalents of prednisone. Bolded values are significant.

Male sex [n (%)]229 (70.9)22 (53.7)0.03
Hepatitis B [n (%)]239 (74.0)30 (73.2)0.91
Hepatitis C [n (%)]13 (4.0)1 (2.4)0.62
IFG before LT [n (%)]109 (33.7)9 (22.0)0.13
Hypertension [n (%)]56 (17.3)14 (34.1)0.01
BMI (kg/m2)23.6 ± 3.323.8 ± 2.70.69
Changes in BMI (kg/m2)   
—From 0 to 1 year after LT−11.1 ± 0.1−10.8 ± 0.40.45
—From 1 to 2 years after LT+0.3 ± 0.1+0.1 ± 0.20.35
—From 2 to 3 years after LT+0.1 ± 0.1−10.0 ± 0.20.57
—From baseline to endpoint−10.8 ± 0.1−10.7 ± 0.40.98
Obesity: BMI ≥ 25 kg/m2 [n (%)]119 (36.8)14 (34.1)0.74
Age (years)49.8 ± 9.351.4 ± 8.10.31
Treatment with tacrolimus [n (%)]305 (94.4)41 (100.0)0.12
FPG before transplantation (mg/dL)93.8 ± 14.490.8 ± 16.40.22
Total cholesterol before LT (mg/dL)114.6 ± 40.7128.1 ± 69.30.07
Change in total cholesterol from baseline to endpoint (mg/dL)+55.5 ± 2.9+43.0 ± 15.40.20
Cumulative dose of steroids over course of 6 months (mg)2797 ± 9372542 ± 9990.11
Duration of follow-up (months)42.7 ± 16.037.8 ± 19.00.07
Donor type: living/cadaver (n/n)264/5929/120.09
Acute rejection [n (%)]46 (14.2)9 (22.0)0.39
NODAT development [n (%)]57 (17.6)13 (31.7)0.03

The mean duration of statin use before NODAT development was 10.4 months. There was no significant difference in the mean statin treatment duration between the NODAT group and the no-NODAT group (24.5 versus 21.5 months, P = 0.62). In addition, there was no difference in the dose of atorvastatin between the NODAT group and the no-NODAT group (Supporting Table 1).

Effects of Statin Treatment on Metabolic Parameters

The mean fasting blood glucose and total cholesterol levels were significantly higher in the statin group versus the control group 6 to 12 months after LT. Although total cholesterol levels were higher for the statin treatment group versus the control group, there were no significant differences in the total cholesterol levels between the 2 groups after statin treatment (Fig. 1). The incidence of NODAT was significantly higher in the statin group versus the control group (31.7% versus 17.6%, P = 0.03; Table 1).

Figure 1.

Changes in the mean total cholesterol level after LT. *P < 0.05 versus the control group.

Independent Predictors of NODAT Development in LT Patients

The use of statins [hazard ratio (HR) = 2.32, 95% confidence interval (CI) = 1.23-4.39, P = 0.01] and IFG before transplantation (HR = 2.21, 95% CI = 1.36-3.62, P = 0.001) showed significant associations with NODAT development after adjustments for other confounding factors (Table 2). Male sex, age, changes in BMI, hypertension, and hepatitis C did not significantly increase the risk of NODAT in the multivariate analysis (Table 2). According to the Cox regression analysis, statin use for longer than 6 months increased the risk of developing NODAT after LT after adjustments for age, sex, changes in BMI, hypertension, and IFG before transplantation (Fig. 2).

Table 2. Independent Predictors of NODAT Development in LT Patients
 Univariate AnalysisMultivariate Analysis
HR (95% CI)P ValueHR (95% CI)P Value
  1. NOTE: Bolded values are significant.

Use of statins1.83 (1.00-3.34)0.052.32 (1.23-4.39)0.01
Male sex1.31 (0.77-2.24)0.321.47 (0.85-2.55)0.17
Age (years)1.03 (0.99-1.06)0.031.03 (1.00-1.06)0.07
IFG before LT2.11 (1.32-3.37)<0.012.21 (1.36-3.62)<0.01
Hypertension0.96 (0.52-1.75)0.880.71 (0.38-1.35)0.30
BMI (kg/m2)1.01 (0.94-1.09)0.80  
Change in BMI from baseline to endpoint (kg/m2)1.01 (0.99-1.04)0.271.01 (0.99-1.03)0.39
Hepatitis B1.10 (0.64-1.90)0.73  
Hepatitis C1.68 (0.61-4.61)0.321.55 (0.55-4.35)0.40
FPG before LT (mg/dL)1.03 (1.02-1.05)<−0.01  
Total cholesterol before LT (mg/dL)1.00 (0.99-1.00)0.12  
Change in total cholesterol from baseline to endpoint (mg/dL)1.00 (0.99-1.01)0.13  
Acute rejection after LT1.33 (0.76-2.33)0.33  
Cumulative dose of steroids over course of 6 months (mg)1.00 (1.00-1.00)0.47  
Figure 2.

Analysis of NODAT development according to the use of statins. Kaplan-Meier curves for NODAT development are shown. Adjustments have been made for age, sex, BMI changes, the prevalence of hypertension, and hepatitis C.

Clinical Characteristics of NODAT Patients With Statin Treatment

Patients who developed NODAT showed higher FPG levels at the baseline (before transplantation) than those who did not develop NODAT (102.2 ± 15.8 versus 85.8 ± 14.4 mg/dL, P = 0.002; Table 3). In addition, hypertension was more common in subjects with NODAT versus patients without NODAT (61.5% versus 25.0%, P = 0.02; Table 3).

Table 3. Clinical Characteristics of Patients Undergoing Statin Treatment According to NODAT Development
 No NODAT (n = 28)NODAT (n = 13)P Value
  1. NOTE: Values are presented as means and standard deviations unless otherwise noted. Bolded values are significant.

Duration of statin treatment (months)21.5 ± 16.724.5 ± 19.60.62
Follow-up duration (months)39.3 ± 19.834.7 ± 17.50.48
Baseline BMI (kg/m2)23.9 ± 2.623.7 ± 3.10.85
BMI 1 year after LT (kg/m2)22.8 ± 2.423.3 ± 3.20.63
Change in BMI from 0 to 1 year after LT (kg/m2)−1.0 ± 2.3−0.4 ± 3.80.60
BMI 2 years after LT (kg/m2)22.9 ± 2.523.3 ± 2.80.69
Change in BMI from 1 to 2 years after LT (kg/m2)+0.1 ± 0.9+0.0 ± 1.80.85
BMI 3 years after LT (kg/m2)22.8 ± 2.723.7 ± 2.70.31
Change in BMI from 2 to 3 years after LT−0.2 ± 0.9+0.4 ± 1.20.08
Change in BMI from baseline to endpoint (kg/m2)−1.1 ± 2.0+0.0 ± 3.80.34
Age at LT (years)51.9 ± 7.650.5 ± 10.10.64
Male sex [n (%)]13 (46.4)9 (69.2)0.17
Cumulative dose of steroids over course of 6 months (mg)2455.8 ± 724.92624.8 ± 858.10.52
FPG before LT (mg/dL)85.8 ± 14.4102.2 ± 15.80.002
FPG at the start of statin treatment (mg/dL)104.5 ± 29.4101.8 ± 15.10.78
Total cholesterol before LT (mg/dL)143.1 ± 76.2106.5 ± 45.80.12
Total cholesterol at the start of statin treatment (mg/dL)255.3 ± 64.3242.6 ± 15.60.53
Total cholesterol 3 years after LT (mg/dL)168.0 ± 38.1188.3 ± 67.80.23
Change in total cholesterol from baseline to 3 years after LT (mg/dL)24.9 ± 99.981.8 ± 87.40.09
Donor type: living/cadaver (n/n)20/89/40.89
Acute rejection [n (%)]5 (17.9)4 (30.8)0.35
Hepatitis B [n (%)]22 (78.6)9 (69.2)0.52
Hepatitis C [n (%)]1 (3.6)0 (0)0.49
Hypertension [n (%)]7 (25.0)8 (61.5)0.02

DISCUSSION

Our study showed that the incidence of NODAT was significantly higher for patients who used statins after LT (31.7%) versus those who did not (17.6%, P = 0.03). Statin use for longer than 6 months was the principal factor related to the development of NODAT after LT (HR = 2.32, 95% CI = 1.23-4.39, P = 0.01).

A recent meta-analysis showed that statin therapy was associated with an increased risk of new-onset diabetes (HR = 1.10, 95% CI = 1.01-1.20),[20] whereas another study revealed that high-dose statin therapy could increase the risk of developing diabetes in comparison with moderate-dose statin therapy.[23] Additionally, it has been shown that some populations, including the elderly, women, and Asians, are more vulnerable to developing diabetes.[24] There are few studies, however, that have examined the effects of statin use on the development of diabetes in liver allograft recipients. The prevalence of diabetes in LT recipients has been reported to vary between 2.5% and 25% because of differences in the criteria for diagnosis.[25] Morbidity was found to be significantly higher for patients who developed NODAT versus an age- and sex-matched group of patients without NODAT.[4] Also, NODAT has been reported to be a significant risk factor for liver-related death after transplantation.[26]

Risk factors for NODAT, including pretransplant hyperglycemia (IFG), a higher BMI, an older age at the time of transplantation, higher steroid dose, genetic factors, and the type of immunosuppressant administered, have been widely studied.[9, 11-15, 19, 27-29] Our study has similarly confirmed that IFG is a strong risk factor for the development of NODAT in patients who have undergone LT (HR = 2.21, 95% CI = 1.36-3.62, P < 0.01). Other investigators also reported that IFG before LT was a risk factor for NODAT development (odds ratio = 3.8, 95% CI = 1.5-9.6, P < 0.01) in a cohort of 211 LT recipients.[13] Pretransplant IFG in renal transplant patients has been associated with a higher incidence of NODAT,[30] as has an older age at the time of transplantation. Although statistically insignificant in this study, an older age at the time of LT tended to increase the risk of NODAT (HR = 1.03, 95% CI = 1.00-1.06, P = 0.07). We did not observe a significant association between pretransplant BMI and NODAT. It is thought that weight gain after transplantation may be a more important risk factor for NODAT than the baseline BMI, and because of ascites and peripheral edema before LT, BMI may have to be calculated according to the dry weight.[31] Hepatitis C infection is another important risk factor for NODAT.[25] Because only 3.8% of our enrolled patients had evidence of a hepatitis C viral infection, we did not observe an association in this study. The cumulative steroid dose within the first year after LT has been reported to increase the risk of NODAT development (odds ratio = 1.01, 95% CI = 1.00-1.02, P = 0.05),[8] although this was not observed in this study. Low maintenance steroid dosages may have contributed to this observation.

To date, the precise pathophysiological mechanism underlying the diabetogenic effects of statins is poorly understood. Some studies have reported that impaired insulin secretion due to disrupted voltage-gated calcium channel function in pancreatic β cells,[32] the disruption of mitochondrial function in pancreatic β cells,[33] myocytes,[34] and adipocytes,[35] statin-induced sarcopenia,[36] and muscle destruction[37, 38] may play a role in the hyperglycemic effects of statins. Because transplant patients must take immunosuppressants for the duration of their life, drug interactions are an important concern. Cyclosporine is metabolized by cytochrome P450 3A4, and most statins (with the exception of pravastatin and fluvastatin) are also metabolized by cytochrome P450 3A4. Cyclosporine blood levels are slightly elevated with statin coadministration, whereas statin levels are increased several-fold.[39] Tacrolimus, on the other hand, does not increase statin levels to the same degree as cyclosporine.[40] It is possible that cyclosporine inhibits the metabolism of statins and thereby enhances statin toxicity. As a result, statin toxicity may be more obviously present in LT recipients.

The cardiovascular and mortality benefits of using statins have been reported to exceed the diabetes risk, and this includes subjects at high risk of developing diabetes.[41] This would also apply to LT recipients. For this reason, a restriction of statins could be inappropriate for LT patients with hypercholesterolemia. In our opinion, statins should be used in LT patients with caution; cessation is not necessary.

We performed additional analyses to assess differences in subject characteristics based on the presence of NODAT. Subjects who developed NODAT showed higher FPG levels at the baseline (P = 0.002), and a greater proportion of the patients had hypertension (P = 0.02). These findings indicate that patients with existing abnormalities of cardiometabolic components (eg, dysglycemia and hypertension) might be susceptible to the development of NODAT after statin treatment. In agreement with our results, a recent analysis from the Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER) trial[41] demonstrated that participants with 1 or more major diabetes risk factor had an increased risk of developing diabetes during the statin trial follow-up (incidence rate: 1.88 versus 0.18 per 100 person-years, HR = 10.5, 95% CI = 7.0-15.8, P = 0.001).

Our study has several limitations. First, this was not a randomized, blinded, placebo-controlled, prospective clinical trial. Therefore, we could not prove a causal relationship between statin use and NODAT development. Additionally, we could not perform oral glucose tolerance tests in all patients before transplantation. Therefore, preexisting diabetes or impaired glucose tolerance without IFG could have led to an overestimation of NODAT. The sample size was too small to draw a clear causal relationship and investigate other possible interactions. However, our study benefits from long-term observation (the mean follow-up period was 42.1 ± 16.4 months after transplantation). Lastly, we could not analyze which statins are more or less diabetogenic because of the relatively small number of statins used in our patient population. However, our study is the first to confirm statin use as a risk factor for NODAT development in liver allograft recipients.

In conclusion, chronic statin treatment for longer than 6 months could contribute to the development of NODAT in LT recipients, especially if they have higher baseline FPG levels. Patients with high baseline FPG levels are more likely to develop NODAT when they are placed on statin treatment after LT. Although statins are a cornerstone of modern cardiovascular medicine with benefits that clearly outweigh the risk of diabetes, physicians should monitor LT patients for diabetes when statins are used for the treatment of dyslipidemia.

Abbreviations
BMI

body mass index

CI

confidence interval

FPG

fasting plasma glucose

HR

hazard ratio

IFG

impaired fasting glycemia

LT

liver transplantation

NODAT

new-onset diabetes after transplantation

Ancillary