Prospective Analysis of Metabolic Parameters in the Detection of Diabetes and Metabolic Syndrome in Liver Transplant Recipients

Abstract

Background:

Liver transplant recipients are at increased risk of metabolic complications, including new-onset diabetes mellitus after transplantation (NODAT) and post-transplant metabolic syndrome (PTMS), both of which are associated with decreased patient survival. We prospectively monitored traditional and novel metabolic parameters in nondiabetic liver transplantation (LT) candidates to determine their role in detecting these conditions.

Methods:

Nondiabetic adults undergoing initial LT were prospectively identified. NODAT and PTMS were defined according to WHO and ATP III criteria. Metabolic measures were collected at pre-LT, 4, and 12 months post-LT.

Results:

Of 49 subjects enrolled, 24.5% were found to be diabetic pre-LT by 2-hr oral glucose tolerance test (OGTT) despite fasting glucose below the diabetic range. Two patients developed NODAT post-LT. A single patient was found to have MS at baseline, while PTMS developed in 26% and 31.3% of patients at 4 and 12 months. Novel metabolic markers did not detect these conditions.

Conclusions:

Screening OGTT detected pre-LT diabetes in patients with normal fasting glucose. Serial measurement of metabolic parameters allowed earlier detection of PTMS. Novel metabolic parameters did not correspond to post-LT outcomes, but provided a baseline for future study. More frequent and intensive metabolic monitoring appears reasonable, but larger studies are needed to clarify its efficacy.

Introduction

Liver transplantation (LT) is a life-saving procedure for over 5000 patients in the United States each year, with 5-year survival exceeding 75%.¹ Although advances in immunosuppression and post-transplant care have resulted in improved graft survival and patient longevity, cardiovascular disease remains the leading cause of nonhepatic mortality.^2,3 This heightened cardiovascular risk in LT recipients is attributable to their risk for a number of metabolic complications, including obesity, dyslipidemia, hypertension, and diabetes.⁴ New-onset diabetes mellitus after liver transplantation (NODAT) and post-transplant metabolic syndrome (PTMS) are common complications of LT resulting in inferior allograft survival, accelerated atherosclerosis, and cardiovascular mortality.^4

–8

NODAT and PTMS remain poorly understood phenomena, as the existing literature is largely confined to retrospective studies and limited by a lack of methodologic standardization. In addition to traditional risk factors for diabetes mellitus (DM), liver transplant recipients are exposed to known diabetogenic immunosuppression, including corticosteroids and calcineurin inhibitors.⁹ These agents have also been associated with the development of MS.¹⁰ Other risk factors specifically implicated in NODAT include older age,¹¹ higher body mass index (BMI),^12,13 known family history of DM,^14,15 presence of active hepatitis C virus infection,^13,16,17 male gender,^18,19 African American race,¹³ and corticosteroid therapy for acute cellular rejection.^19,20 PTMS shares many of these risk factors.²¹

Recent studies in renal transplant recipients have suggested the utility of additional markers for metabolic risk. Oral glucose tolerance testing (OGTT) is more sensitive for the detection of DM²² and has been shown to better predict NODAT in renal transplantation.²³ While prior guidelines for the detection and monitoring of NODAT in LT relied more on traditional risk factors and fasting plasma glucose (FPG),²⁴ recent updates have recommended the routine use of OGTT in pre-LT screening.²⁵ Its use in post-LT monitoring, however, is suggested only when FPG is abnormally elevated. Other indices of insulin sensitivity and response (e.g., HOMA-IR, corrected insulin response) have been correlated with NODAT in renal transplant recipients,²⁶ but have not been assessed in LT.

The measurement of a fasting lipid profile is currently recommended in pre-LT screening and annual post-LT monitoring. A prospective evaluation of more frequent monitoring for the interval detection of PTMS has not been undertaken. Other novel metabolic markers, such as the adipocytokine adiponectin and the inflammatory marker c-reactive protein (CRP), have been correlated with heightened metabolic risk in renal transplant recipients²⁷ and the general population,^28,29 but have not been assessed in LT.

With the scarcity of suitable donor organs³⁰ and the significant impact of NODAT and PTMS on patient and allograft survival, early identification and risk factor modification, both pre- and post-LT, is of paramount importance. We aimed to prospectively analyze a cohort of nondiabetic liver transplant candidates to assess the utility of serial metabolic monitoring and determine the role of novel metabolic markers in detection of these conditions.

Methods

Patient selection

All adult nondiabetic patients undergoing liver transplant evaluation between April 2010 and January 2013 were prospectively screened for inclusion. Routine diabetic screening was performed in the course of liver transplant evaluation. Nondiabetic was defined as no personal history of DM diagnosis, no current or past antihyperglycemic medication use, and a FPG and hemoglobin A1c in the normal reference range. Only patients approved for listing were included. Pregnant patients and those receiving a multiorgan transplant or using corticosteroids within 6 months of enrollment were excluded. Informed consent was obtained from all patients. The study was approved by the Mayo Clinic Institutional Review Board.

Definitions

DM, impaired glucose tolerance (IGT), and impaired fasting glucose (IFG) were defined according to WHO diagnostic criteria²²:

NODAT:

FPG ≥126 mg/dL or

2-hr OGTT ≥200 mg/dL

IGT:

2-hr OGTT between 140–200 mg/dL in the setting of FPG <126 mg/dL

IFG:

FPG ≥110 mg/dL but <126 mg/dL in the setting of an OGTT <140 mg/dL

PTMS was defined according to the National Cholesterol Education Program–Adult Treatment Plan III guidelines,³¹ in which patients must meet three or more of the following criteria:

(1) FPG ≥100 mg/dL

(2) Waist circumference >102 cm in men or >88 cm in women

(3) Fasting triglyceride level ≥150 mg/dL or drug treatment for high triglyceride levels

(4) Fasting HDL level <40 mg/dL in men, <50 mg/dL in women, or drug treatment for low HDL levels

(5) Blood pressure ≥130/85 or drug treatment for hypertension.

Calculations

After collection of serologic data, several composite metabolic indices were calculated. The homeostasis model assessment of insulin sensitivity (HOMA-IS) and insulin resistance (HOMA-IR) were utilized to determine insulin sensitivity and resistance, respectively. Both the original and revised quantitative insulin sensitivity check index (QUICK-I) also served as measures of insulin sensitivity. Assessment of insulin response was performed using the 2-hr corrected insulin response (CIR). These calculations were performed as follows³²:

(1) HOMA-IR = (fasting insulin [mIU/mL] × fasting glucose [mM])/22.5

(2) HOMA-IS = 1/HOMA-IR

(3) QUICK-I = 1/[log(fasting insulin) + log(fasting glucose)]

(4) Revised QUICK-I = 1/[log(fasting insulin) + log(fasting glucose) + log(free fatty acids)]

(5) 2-hr CIR = 2-hr insulin ×100/[2-hr glucose × (2-hr glucose–70)]

Transplant immunosuppression

All patients received a standard immunosuppression protocol after LT, which included the following: (1) methylprednisolone 1000 mg intraoperatively, followed by oral prednisone 25 mg on the first postoperative day and a subsequent 4-month taper; (2) tacrolimus adjusted to a trough level of 8–12 ng/mL; (3) mycophenolate mofetil 1000 mg twice daily. In the event of gastrointestinal adverse effects or myelosuppression, mycophenolate mofetil was discontinued.

Data collection

Demographic, clinical, and anthropometric characteristics were recorded for all study participants, including age, sex, race, ethnicity, BMI, waist and hip circumference, etiology of liver disease, and family history of DM. Patients then underwent pretransplant laboratory testing, including standard measurement of FPG, fasting lipid panel, and hemoglobin A1c. Each FPG measurement was repeated, in accordance with established guidelines. Several novel variables were also measured, including plasma adiponectin (Quest Diagnostics Nichols Institute, San Juan Capistrano, CA; ELISA; test code 15060), CRP, and insulin. The adiponectin assay measures total adiponectin, as multimers are denatured by heating. A 2-hr OGTT was performed after an overnight fast by measurement of plasma glucose 2 hrs after ingestion of a beverage containing 75 g of glucose, in accordance with World Health Organization (WHO) guidelines. All parameters were collected again during clinic visits at 4 and 12-month intervals after undergoing LT. All laboratory sample collections were scheduled for a single morning time point to minimize the effect of varying hormonal control by time of day.

Statistical analysis

Statistical analyses were performed utilizing JMP software (SAS Institute, Inc.). Differences in dichotomous and categorical variables were analyzed by Fisher exact test. The difference in means for continuous variables was assessed by independent t-test. All analyses were two-tailed and a value of P < 0.05 was utilized to denote statistical significance.

Results

Forty-nine LT candidates were prospectively enrolled in the study. Baseline 2-hr OGTT revealed the presence of diabetes in 12 (24.5%) patients despite no prior history of DM or hyperglycemia. Baseline data for all enrolled patients is displayed in Table 1, stratified by the presence or absence of DM. Only the OGTT (P < 0.001) and FPG (P = 0.046) demonstrated a statistically significant difference between groups. The patients found to be diabetic were excluded from further analysis. Of the remaining 37 patients, 15 (40.5%) did not return for subsequent testing due to voluntary (n = 2) or involuntary (n = 3) delisting, transfer of care to other liver transplant programs (n = 2), refusal to participate after consent (n = 1), intraoperative death (n = 1), death due to intracranial hemorrhage (n = 1), extended delay between consent and transplant (n = 2), and general loss to follow-up (n = 3).

Table 1.

Comparison of Baseline Metabolic Parameters Between Nondiabetic Enrollees and Those with Oral Glucose Tolerance Test-Detected Diabetes

Parameter	DM (n = 12)	Non-DM (n = 37)	P-value
Age (y)	53.3	54	0.82
Family history of DM	7/12 (58.3%)	14/38 (36.8%)	0.31
Male gender	7/12 (58.3%	28/38 (73.7%)	0.47
HCV infection	7/12 (58.3%)	18/38 (47.4%)	0.74
Fasting plasma glucose (mg/dL)	99.2	94.6	0.046
OGTT (mg/dL)	234.5	134.9	<0.001
BMI (kg/m²)	28.9	28.2	0.73
Hip circumference (cm)	103.9	108.4	0.43
Waist circumference (cm)	102.3	103.5	0.87
Total cholesterol (mg/dL)	123.6	134.3	0.56
HDL (mg/dL)	41.6	41.7	>0.99
LDL (mg/dL)	64.7	76.8	0.45
Triglycerides (mg/dL)	85.9	79.8	0.62
Hemoglobin A1C (%)	5.2	5	0.31
2-hr CIR	0.614	1.065	0.44
HOMA-IR	6.91	4.84	0.12
HOMA-IS	0.258	0.529	0.32
QUICK-I	0.302	0.325	0.18
Revised QUICK-I	0.347	0.365	0.58
Adiponectin (μg/mL)	41.2	77.1	0.68
CRP (mg/dL)	12.3	11.6	0.88

Data are presented as means except as otherwise noted.

Bold values indicate statistical significance.

BMI, body mass index; CIR, corrected insulin response; CRP, C-reactive protein; DM, diabetes mellitus; OGTT, oral glucose tolerance test; HOMA-IR, homeostasis model assessment of insulin resistance; HOMA-IS, homeostasis model assessment of insulin sensitivity; QUICK-I, quantitative insulin sensitivity check index.

The baseline demographic and clinical characteristics for the remaining 22 patients are presented in Table 2. Patients were predominantly Caucasian males receiving transplants from deceased donors, although living donor recipients made up 36.4% of the study population. Compared to living donor recipients, those receiving deceased donor organs had higher MELD scores at both enrollment (P = 0.02) and transplant (P < 0.01) and a longer time to transplant (P = 0.01). Hepatitis C infection was the most common etiology of liver disease. Six of the 22 patients did not complete 1 year of follow-up, having returned for 1-month (n = 1) and 4-month (n = 5) follow-up appointments. In one case, this was due to death before the 1-year mark. Adherence to the study protocol was not uniform and, despite patients returning for follow-up, several isolated values were omitted from data collection. Additionally, despite an intention to collect laboratory data at a consistent time of day across subjects, two patients deviated from this protocol and underwent afternoon lab draws rather than morning. Subsequent collections for these patients were performed in the afternoon to minimize variability.

Table 2.

Baseline Demographic and Clinical Data for Included Study Participants

Characteristic	Total Patients (n = 22)
Age (y)	55.1
Male gender	17 (77.3%)
Race
Caucasian	20 (90.9%)
Hispanic	2 (9.1%)
Etiology of liver disease
Hepatitis C infection	10 (45.5%)
Alcoholic liver disease	1 (4.5%)
NASH/cryptogenic	1 (4.5%)
Primary sclerosing cholangitis	3 (13.7%)
Other	7 (31.8%)
MELD score (mean)
Deceased donor, enrollment	21.8
Deceased donor, transplant	27.9
Living donor, enrollment	16.2
Living donor, transplant	18.1
Months, enrollment to transplant (mean)
All patients	3.7
Deceased donor	5.8
Living donor	0.8
Donor type
Deceased	14 (63.6%)
Living	8 (36.4%)
Family history of DM	5 (22.7%)

Data are presented as frequency of occurrence and percent of patients, except as otherwise noted.

The serially monitored metabolic parameters are presented in Table 3. NODAT developed in two instances. In the first, an FPG of 139 mg/dL was detected at 4 weeks in a patient with IGT at baseline. In the second, a patient with IGT at baseline developed an FPG of 129 mg/dL at 4 months with an OGTT of 194 mg/dL and was subsequently noted at 1 year to have an abnormal OGTT of 202 mg/dL. Metabolic syndrome was present in a single patient (4.5%) at baseline and developed in 4/21 (19%) at 4 months and 5/16 (31.3%) at 1 year. Only a single patient experienced acute cellular rejection during the study period. This occurred at 10 days post-transplant and was treated successfully with an increase in baseline immunosuppression without the need for intravenous corticosteroids.

Table 3.

Serial Measurement of Metabolic Parameters

Parameter	Baseline	4 months	1 year
Fasting plasma glucose (mg/dL)	94.6	98.6	100
OGTT (mg/dL)	134.2	118.5	116.7
IGT	12/22 (54.5%)	4/19^a (21.1%)	3/16 (18.8%)
IFG	1/22 (4.5%)	2/19^a (10.5%)	2/16 (12.5%)
BMI (kg/m²)	26.8	26.3	27.9
Hip circumference (cm)	104.3	103.7	109
Waist circumference (cm)	100.8	99.4	105
Total cholesterol (mg/dL)	147.5	159.8	160.9
HDL (mg/dL)	44.7	48.2	46.8
LDL (mg/dL)	86.5	83.5	82.4
Triglycerides (mg/dL)	83.5	121.4	158.5
Hemoglobin A1C (%)	5	5.3	5.3
2-hr CIR	0.9	1.09	1.06
HOMA-IR	4.77	3.39	2.84
HOMA-IS	0.449	0.493	0.486
QUICK-I	0.324	0.335	0.337
Revised QUICK-I	0.364	0.411	0.423
Adiponectin (μg/mL)	103.1	18	16
CRP (mg/dL)	11.1	8.6	9

Data are presented as means except as otherwise noted.

As noted in methods, data collection at follow-up was variable. Despite 21 patients returning for 4-month follow-up, only 19 underwent collection of FPG and OGTT.

FPG, fasting plasma glucose; IFG, impaired fasting glucose; IGT, impaired glucose tolerance.

Discussion

In this study, we prospectively analyzed various metabolic markers to elucidate their role in the detection and of NODAT and PTMS. We demonstrated that monitoring of these parameters allowed for earlier detection of pre- and post-LT DM and MS, a key factor in the development of adequate interventions that can avert fatal cardiovascular outcomes. Our serial measurements of several novel metabolic parameters were the first to be conducted in LT.

Baseline OGTT identified DM in 12 (24.5%) patients whose FPG would not have placed them in a diabetic range. Only three patients in this group had FPG levels within the IFG range. Their remaining metabolic and glycemic profile did not differ from the nondiabetic cohort. Consensus guidelines for the diagnosis and management of NODAT were updated in 2004 to incorporate OGTT pre-LT for patients with normal FPG,²⁵ in line with the WHO's observation that approximately 30% of DM is undetected by FPG alone.²² Current estimates suggest that DM is present in 25% of liver transplant recipients before transplantation.¹² We note that patients with established DM were excluded from our analysis, which yielded a diagnosis of diabetes in an additional 22.4% of patients. Thus, our results demonstrate the enhanced sensitivity of OGTT for detection of DM and suggest a far higher pre-LT prevalence of DM than previously reported.

In our patients, IGT was detected in 21.1% and 18.8% at 4 months and 1 year post-LT, respectively. The use of OGTT in the post-LT setting is currently recommended only if an intermediate FPG level (i.e., IFG) is detected at prescribed testing intervals.²⁵ Each of our cases of IGT was accompanied by FPG within the normal range. The presence of IGT carries a higher risk of progression to frank diabetes and predicts cardiovascular disease in both diabetic and nondiabetic patients.³³ Furthermore, patients in the initial stages of IGT and NODAT are asymptomatic, but susceptible to the adverse cardiovascular effects of abnormal glucose homeostasis.³⁴ Widespread adoption of OGTT as a component of routine post-transplant screening has not occurred, primarily due to increased cost, complexity, and poor reproducibility of the test.^35,36 The use of hemoglobin A1c has been more readily implemented due to its comparative simplicity and familiarity and is now recommended to be performed along with FPG starting 3 months post-LT.²⁵ However, it should be noted that, while OGTT is well-validated as predictive of DM and IGT in transplant recipients, A1c has not.³⁶

Using strict pre- and post-LT definitions, the incidence of NODAT in our patient cohort was 9%. Existing estimates of the incidence of NODAT after LT vary widely, with a recent meta-analysis of 16 prior studies demonstrating rates ranging from 0% to 40%.⁹ This variation is likely multifactorial, owing to heterogeneity in time of follow-up, patient population, immunosuppressive regimen, duration of NODAT, and diagnostic criteria for DM. A previous retrospective analysis at our institution demonstrated an incidence of 17% at 1 year,¹⁶ with prevalence increasing over time and reaching a rate of 28% by 5 years.³⁷ The smaller proportion of NODAT in our study is somewhat surprising, especially given the use of OGTT, which has been shown to uncover cases of NODAT undetected by FPG alone.³⁸ However, we note that provocative testing by OGTT resulted in the identification of DM in 24.5% of our study population despite no prior history. These patients would have been classified as nondiabetic at entry in existing studies of prevalence. The discordance is also likely partially attributable to the inclusion of living-donor LT, which has been demonstrated as a protective factor for NODAT.¹⁹

PTMS is estimated to occur in 44%–58% of LT recipients^21,39 and occurred in 31.3% of our patients by 1 year. Bianchi et al. detected PTMS at a mean of 43–61.4 months post-LT in a retrospective analysis, depending on liver disease etiology.²¹ In contrast, MS was detected in 4/21 (19%) of our patients by 4 months. Perhaps more notably, 1-year mean values for FPG, LDL, and waist circumference in our study population satisfied MS diagnostic criteria. Current guidelines recommend the pre-LT measurement of fasting lipid levels, with post-LT measurements occurring annually thereafter.²⁵ Studies in the general population have demonstrated that the presence of any single criterion of MS is associated with increased cardiovascular morbidity and patients meeting 3 or more criteria carry a 50% higher risk of a major coronary event.⁴ Our findings suggest that routine monitoring of anthropometric and metabolic criteria of MS identifies more at-risk patients and does so earlier in their post-LT course.

Several novel metabolic markers were measured serially in our study population. Adiponectin is synthesized in adipose tissue and acts to increase insulin sensitivity and decrease hepatic gluconeogenesis.⁴⁰ It is also believed to exert anti-inflammatory effects, protecting against atherosclerosis. Serial measurement of adiponectin pre- and post-LT demonstrated a pronounced downward trend, consistent with diminishing insulin sensitivity. In contrast, calculated measures of insulin resistance (HOMA-IR), sensitivity (HOMA-IS, original and revised QUICK-I), and response (CIR) collectively suggest a pattern of improving glycemic control post-LT.

Because systematic measurement of these novel markers in LT recipients has not been previously reported, interpreting the trends and values is difficult, particularly in such a small sample size. The incidence of PTMS was too low to allow meaningful statistical analysis. Further, measures and indices of insulin sensitivity and glycemic control demonstrated conflicting findings and the applicability of these measures to the detection of NODAT and PTMS remains speculative. We suggest that these findings should serve as a foundation for future studies examining the role of these measures in detection and, if validated, prediction of NODAT and PTMS.

The strengths of this study include its prospective design and the novel utilization of several metabolic parameters in the setting of LT. However, we acknowledge several important limitations. Foremost among these is the small sample size, which precluded more robust statistical analysis or the development of a predictive model for NODAT based on our measured variables. Follow-up and adherence to protocol was also not uniform and introduced an element of possible selection bias into the displayed measures, as patients lost to follow-up may have been more likely to be diagnosed with NODAT or MS. Living-donor LT has been implicated as a protective factor for NODAT, however, despite the prospective design, we did not control for donor type. The differential MELD and time to transplant for living versus deceased donor recipients introduces an additional element of possible bias, which could not be limited further due to the small sample size. Finally, both the high proportion of living donor recipients and single-center design may limit generalizability of these data to the general LT population.

In summary, this prospective analysis of a small patient cohort serves as a compelling preliminary demonstration that an expanded panel of metabolic testing in both pre-LT screening and post-LT monitoring enhances detection of conditions that predispose to a higher risk of cardiovascular mortality. Whether the more novel markers of metabolic function (e.g., adiponectin, CRP) have a role in the detection or prediction of MS or DM in LT recipients remains to be assessed by larger prospective studies that should be aided by these initial measurements. Ultimately, demonstration of the cost and outcome benefits of any enhanced diagnostic paradigm will require concerted research effort and improved medical and behavioral interventions.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Organ Procurement and Transplantation Network (OPTN) and Scientific Registry of Transplant Recipients (SRTR). OPTN/SRTR 2010 Annual Data Report. Department of Health and Human Services, Health Resources and Services Administration, Healthcare Systems Bureau, Division of Transplantation. Am J Transplant, 2012; 12 Suppl 1:1–156.

Pfitzmann

, Nussler

, Hippler-Benscheidt

, et al. Long-term results after liver transplantation. Transpl Int, 2008; 21:234–246.

Watt

, Pedersen

, Kremers

, et al. Evolution of causes and risk factors for mortality post-liver transplant: Results of the NIDDK long-term follow-up study. Am J Transplant, 2010; 10:1420–1427.

Watt

KDS

. Obesity and Metabolic Complications of Liver Transplantation. Liver Transplant, 2010; 16:S65–S71.

Baid

, Cosimi

, Farrell

, et al. Posttransplant diabetes mellitus in liver transplant recipients: Risk factors, temporal relationship with hepatitis C virus allograft hepatitis, and impact on mortality. Transplantation, 2001; 72:1066–1072.

Ducloux

, Kazory

, Simula-Faivre

, et al. One-year post-transplant weight gain is a risk factor for graft loss. Am J Transplant, 2005; 5:2922–2928.

Moon

, Barbeito

, Faradji

, et al. Negative impact of new-onset diabetes mellitus on patient and graft survival after liver transplantation: Long-term follow up. Transplantation, 2006; 82:1625–1628.

Navasa

, Bustamante

, Marroni

, et al. Diabetes mellitus after liver transplantation: Prevalence and predictive factors. J Hepatol, 1996; 25:64–71.

Heisel

, Heisel

, Balshaw

, Keown

. New onset diabetes mellitus in patients receiving calcineurin inhibitors: A systematic review and meta-analysis. Am J Transplant, 2004; 4:583–595.

10.

Marchetti

, Navalesi

. The metabolic effects of cyclosporin and tacrolimus. J Endocrinol Invest, 2000; 23:482–490.

11.

Kim

, Kim

, Lee

, et al. Effect of liver transplantation on glucose levels in patients with prediabetes or type 2 diabetes. Transplant Proc, 2014; 46:225–229.

12.

Carey

, Douglas

, Vargas

, et al. Incidence and risk factors for new onset diabetes mellitus after transplantation. The 59th Annual Meeting of the American Association for the Study of Liver Diseases. San Francisco, CA, 2008.

13.

Kuo

, Sampaio

, Ye

, et al. Risk factors for new-onset diabetes mellitus in adult liver transplant recipients, an analysis of the Organ Procurement and Transplant Network/United Network for Organ Sharing database. Transplantation, 2010; 89:1134–1140.

14.

Arner

, Gunnarsson

, Blomdahl

, et al. Some characteristics of steroid diabetes: A study in renal-transplant recipients receiving high-dose corticosteroid therapy. Diabetes Care, 1983; 6:23–25.

15.

Sumrani

, Delaney

, Ding

, et al. Diabetes mellitus after renal transplantation in the cyclosporine era—an analysis of risk factors. Transplantation, 1991; 51:343–347.

16.

Carey

, Aqel

, Byrne

, et al. Pretransplant fasting glucose predicts new-onset diabetes after liver transplantation. J Transplant, 2012; 2012:614781.

17.

Saliba

, Lakehal

, Pageaux

, et al. Risk factors for new-onset diabetes mellitus following liver transplantation and impact of hepatitis C infection: An observational multicenter study. Liver Transpl, 2007; 13:136–144.

18.

Saab

, Shpaner

, Zhao

, et al. Prevalence and risk factors for diabetes mellitus in moderate term survivors of liver transplantation. Am J Transplant, 2006; 6:1890–1895.

19.

Yadav

, Chang

, Aqel

, et al. New Onset Diabetes Mellitus in Living Donor versus Deceased Donor Liver Transplant Recipients: Analysis of the UNOS/OPTN Database. J Transplant, 2013; 2013:269096.

20.

Honda

, Asonuma

, Hayashida

, et al. Incidence and risk factors for new-onset diabetes in living-donor liver transplant recipients. Clin Transplant, 2013; 27:426–435.

21.

Bianchi

, Marchesini

, Marzocchi

, et al. Metabolic syndrome in liver transplantation: Relation to etiology and immunosuppression. Liver Transpl, 2008; 14:1648–1654.

22.

World Health Organization. Definition and diagnosis of diabetes mellitus and intermediate hyperglycemia. Geneva, Switzerland: World Health Organization; 2006.

23.

Tokodai

, Amada

, Haga

, et al. The 5-time point oral glucose tolerance test as a predictor of new-onset diabetes after kidney transplantation. Diabetes Res Clin Pract, 2014; 103:298–303.

24.

Davidson

, Wilkinson

, Dantal

, et al. New-onset diabetes after transplantation: 2003 International consensus guidelines. Proceedings of an international expert panel meeting. Barcelona, Spain, 19 February 2003. Transplantation, 2003; 75(10 Suppl):SS3–24.

25.

Wilkinson

, Davidson

, Dotta

, et al. Guidelines for the treatment and management of new-onset diabetes after transplantation. Clin Transplant, 2005; 19:291–298.

26.

Tokodai

, Amada

, Haga

, et al. Insulin resistance as a risk factor for new-onset diabetes after kidney transplantation. Transplant Proc, 2014; 46:537–539.

27.

Bayes

, Granada

, Pastor

, et al. Obesity, adiponectin and inflammation as predictors of new-onset diabetes mellitus after kidney transplantation. Am J Transplant, 2007; 7:416–422.

28.

Chen

, Gona

, Sutherland

, et al. Long-term C-reactive protein variability and prediction of metabolic risk. Am J Med, 2009; 122:53–61.

29.

Dehghan

, Kardys

, de Maat

, et al. Genetic variation, C-reactive protein levels, and incidence of diabetes. Diabetes, 2007; 56:872–878.

30.

Dutkowski

, Linecker

, DeOliveira

, et al. Challenges to Liver Transplantation and Strategies to Improve Outcomes. Gastroenterology, 2015; 148:307–323.

31.

Third Report of the National Cholesterol Education Program (NCEP). Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation, 2002; 106:3143–3421.

32.

Singh

, Saxena

. Surrogate markers of insulin resistance: A review. World J Diabetes, 2010; 1:36–47.

33.

Barr

, Zimmet

, Welborn

, et al. Risk of cardiovascular and all-cause mortality in individuals with diabetes mellitus, impaired fasting glucose, and impaired glucose tolerance: The Australian Diabetes, Obesity, and Lifestyle Study (AusDiab). Circulation, 2007; 116:151–157.

34.

Benhamou

, Penfornis

. Natural history, prognosis, and management of transplantation-induced diabetes mellitus. Diabetes Metab, 2002; 28:166–175.

35.

Inzucchi

. Clinical practice. Diagnosis of diabetes. N Engl J Med, 2012; 367:542–550.

36.

Pham

, Edling

, Chakkera

, et al. Screening strategies and predictive diagnostic tools for the development of new-onset diabetes mellitus after transplantation: An overview. Diabetes Metab Syndr Obes, 2012; 5:379–387.

37.

Carey

EJA

, Byrne

, Vargas

, Rakela

, Douglas

, Chakkera

. Incidence and risk factors for new onset diabetes mellitus after transplantation. The 60th Annual Meeting of the American Association for the Study of Liver Diseases, Boston: MA, 2009.

38.

Mirabella

, Brunati

, Ricchiuti

, et al. New-onset diabetes after liver transplantation. Transplant Proc, 2005; 37:2636–2637.

39.

Laryea

, Watt

, Molinari

, et al. Metabolic syndrome in liver transplant recipients: Prevalence and association with major vascular events. Liver Transpl, 2007; 13:1109–1114.

40.

Meier

, Gressner

. Endocrine regulation of energy metabolism: Review of pathobiochemical and clinical chemical aspects of leptin, ghrelin, adiponectin, and resistin. Clin Chem, 2004; 50:1511–1525.

41.

Rohner-Jeanrenaud

, Jeanrenaud

. Obesity, leptin, and the brain. N Engl J Med, 1996; 334:324–325.