Objectives: Diabetes mellitus is one of the metabolic consequences of solid-organ transplant. Most reports on this condition are from cross-sectional or retrospective studies. In this prospective study, we evaluated the incidence, risk factors, and short-term follow-up of diabetes mellitus in recipients of liver transplant at the Shiraz Liver Transplant Center (Shiraz, Iran).
Materials and Methods: Recipients of liver transplant who were ≥ 16 year old and were seen from February 2017 until February 2018 were included. Anthropologic measurements and diabetes history were taken between 2 and 4 weeks after transplant. Fasting blood sugar and 75-g oral glucose tolerance test were measured. We diagnosed patients with diabetes mellitus and patients with impaired fasting glucose or impaired glucose tolerance test based on American Diabetes Association criteria. These patients were promptly followed for at least 6 months.
Results: Of the 397 recipients who were included in this study, 35.5% were female and 64.5% were male. Overall, the most common reason for transplant was primary sclerosing cholangitis (22.5%). We had 42 living donors and 355 deceased donors, with none being unrelated donors. At first visit (3.8 ± 1.6 wk posttransplant), 20.4% of recipients did not have diabetes, 24.2% were diagnosed with preexisting diabetes mellitus, 31.2% had impaired fasting blood sugar or oral glucose tolerance test, and 24.2% were determined to have posttransplant diabetes mellitus. At last visit (13.6 ± 4.9 mo posttransplant), prevalence for posttransplant diabetes mellitus was 10.8%. Multivariate regression analysis showed that age correlated with development of impaired glucose tolerance test or posttransplant diabetes mellitus (odds ratio 1.060; 95% confidence interval, 1.026-1.095; P < .001).
Conclusions: In this prospective cohort study, we followed recipients posttransplant and reevaluated the prevalence of posttransplant diabetes mellitus. We found significant recovery for this type of diabetes mellitus. Further larger and multicenter studies are necessary to monitor and manage diabetes mellitus posttransplant.
Key words : Glucose tolerance test, Hyperglycemia, Impaired fasting blood sugar
Diabetes mellitus (DM) and impaired glucose tolerance are among the most important metabolic consequences of solid-organ transplant.1 Although posttransplant diabetes mellitus (PTDM) is not thought to be associated with microvascular and macrovascular complications, the risk of cardiovascular morbidity and mortality and graft loss increase and graft and patient survival decrease in patients with PTDM.2
Over the years, due to lack of specified criteria, the exact incidence of new-onset diabetes mellitus in transplant recipients has been difficult to determine.3 In 2003, an international expert panel set forth the International Consensus Guidelines for the diagnosis of new-onset diabetes after transplant (NODAT) according to fasting blood sugar (FBS), oral glucose tolerance test (OGGT), and symptomatic hyperglycemia, which was based on the American Diabetes Association (ADA) and World Health Organization criteria for nontransplant patients.4 An update on these criteria was published in 2014.5 The condition was renamed from NODAT to PTDM. This renaming was based on the fact that many patients who were considered to be new cases of PTDM were not truly new. In most centers, candidate screening for DM has been done with tests for FBS and HbA1c, which are less sensitive measurements than the OGTT. The consensus recommends OGTT as a screening test.6 The consensus also recommends delaying screening and diagnosing of PTDM until the patient is discharged from the hospital and when the patient is on a stable dose of immunosuppressant drugs.7 By this definition, the diagnosis of PTDM has no “end date” and thus may be shown even many years after organ transplant.5,8
Although the relative importance of increased insulin resistance or decreased insulin secretion on development of PTDM is controversial, current data highlight the importance of beta-cell failure compared with insulin resistance.9
There are some known risk factors for PTDM. These include underlying DM risk factors, age, obesity (defined by body mass index [BMI] or waist circumference), race, and type of immunosuppressive regimen. There is evidence of the importance of candidate gene single nucleotide polymorphisms and a higher association with some conditions, including hepatitis C for liver transplant recipients and polycystic kidney disease for renal transplant recipients.5 The use of statins,10 magnesium deficiency,11 and vitamin D deficiency12 have been also reported to be risk factors.
Based on the 2003 International Consensus Guidelines, rates of PTDM have been reported to be 10% to 74% after kidney transplant, 11% to 38% after heart transplant, 7% to 30% after liver transplant, and 32% after lung transplant.4 Reports from Iran have indicated an incidence of 18.8% for PTDM in renal transplant recipients13 and 18.81% in liver transplant recipients at the Shiraz Transplant Center.14 In general, most reports on this condition are from cross-sectional or retrospective studies and some have used outdated criteria.
In this prospective study, we evaluated the incidence, risk factors, and short-term follow-up of patients with PTDM after liver transplant at the Shiraz Liver Transplant Center (Shiraz, Iran).
Materials and Methods
In this prospective cohort study, all patients who received liver transplant at the Liver Transplant Center in Shiraz, Iran, from February 2017 to February 2018 and who were ≥ 16 years old were included.
Fasting blood sugar and HbA1c were measured before transplant with the high-performance liquid chromatography method (Variant II Turbo analyzer, Bio-Rad, Hercules, CA, USA).
All patients had follow-up visits 2 to 4 weeks after transplant, and anthropologic measurements (including weight, height, and waist circumference) were recorded. Weight was measured with a standard scale to the nearest 0.1 kg (Seca, Hamburg, Germany); height was measured to the nearest 0.5 cm with a wall-mounted meter (both with the participant wearing light clothing and no shoes). Body mass index was calculated by dividing weight (in kg) by height square (m2). We used nonstretchable tape to measure waist circumference, with patients standing with feet at shoulder width apart (25-30 cm) and with back straight, midpoint between lower rib margin and iliac crest.
Clinical assessments, including past history of DM, use of oral antidiabetics or insulin, family history of DM, and other risk factors for DM,8 were gathered for analyses. Patients with documented past history of DM and those on treatment with antidiabetic agents were considered to have preexisting DM. The cause of end-stage liver disease or the underlying disease for which transplant was done was reconfirmed based on reevaluation of pretransplant data and result of histopathology of liver explant when necessary. Blood samples for measuring FBS, total cholesterol, triglycerides, low-density lipoprotein cholesterol, and high-density lipoprotein cholesterol were taken by experienced laboratory technicians at the Shiraz Motahhari Clinic after 12 hours of fasting. Fasting blood sugar was rechecked in all patients after 1 week.
All patients without diabetes underwent a 75-g OGTT with measurement of 2-hour plasma glucose value. We diagnosed PTDM patients and patients with impaired FBS or impaired OGTT according to ADA criteria. After diagnoses, standard management for blood sugar control was started. These patients were followed for at least 6 months.
This study was approved by the Ethics Committee of the Shiraz University of Medical Science (number 94-01-01-10770). The study protocol conformed to the ethical guidelines of the 1975 Helsinki Declaration. Written, informed consent was obtained from patients.
Quantitative and qualitative data are shown as means and standard deviation (SD) and percent and frequencies. Parameter comparisons were performed with a t test and analyses of variance for quantitative parameters and chi-square analyses for qualitative parameters. Multivariate logistic regression analysis was performed to identify independent risk factors. Patient survival was analyzed by Cox regression. Statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 22.0, IBM Corporation, Armonk, NY, USA). P values < .05 were considered statistically significant.
From February 2017 to February 2018, 589 patients received liver transplant procedures at the Shiraz Liver Transplant Center; 135 recipients were < 16 years old and thus were excluded from our analyses. Of the 454 patients ≥ 16 years old, 397 recipients agreed to participate in this study (35.5% were female and 64.5% were male). We had 42 living donors and 355 deceased donors, with no donors being unrelated. Clinical information of participants is shown in Table 1.
Table 2 shows the causes of underlying liver disease in study patients. The average time until first outpatient visit was 3.8 ± 1.6 weeks after transplant. At this time, patients were on immunosuppressive medications, including prednisolone with daily mean dose of 17.4 ± 2.9 mg, mycophenolic acid with daily mean dose of 1703.6 ± 334.2 mg, and tacrolimus with daily mean dose of 3.6 ± 1.3 mg. Of total patients, 43 were on everolimus, with daily mean dose of 1.5 ± 0.5 mg, with or without tacrolimus.
Of total patients, 96 (24.2%) had been diagnosed as having preexisting DM. After these patients were excluded, we found at least 1 risk factor for PTDM in 42.3% of the other recipients.
Mean HbA1c in patients without DM was 4.8 ± 0.6% before transplant. At the first visit, based on past history of recipients, ADA criteria,8 and posttransplant laboratory tests (FBS or OGTT), patients were categorized as normal, having preexisting DM, having impaired FBS or OGTT after transplant, and having PTDM. We found that 20.4% of recipients did not have DM, 24.2% had preexisting DM, 31.2% had impaired FBS or OGTT, and 24.2% had PTDM. In the impaired group, we found impaired OGTT in 69.4%, impaired FBS in 11.3%, and impaired FBS and OGTT in 19.4% of recipients.
Figure 1 shows mean age, mean BMI, and distribution of sex, cause of cirrhosis, and DM risk factors in the 4 transplant recipient groups. We found no significant differences in male versus female recipients among these 4 groups (P = .56).
Patients with DM were older than those without DM (51.8 ± 10.1 y for those with preexisting DM, 47.9 ± 11.4 years for those with PTDM, 35.7 ± 11.9 y for patients without DM, and 41.6 ± 12.3 for patients with impaired FBS or OGTT; P < .001). Patients with preexisting DM were significantly older than PTDM patients (P = .014).
Body mass index was not significantly different between the 4 groups (P = .128). Of total study patients, 69.4% had BMI < 25 kg/m2 and 30.6% had BMI ≥ 25 kg/m2. The proportion of overweight to normal patients in the 4 groups was not significantly different (P = .98).
The most common reason for transplant in PTDM patients was cirrhosis due to hepatitis B virus (24.2%). In patients with preexisting DM, it was cryptogenic cirrhosis (33.7%). In patients without DM and in those in the impaired group, primary sclerosing cholangitis was the most common reason for transplant (43.2% and 25.0%; P < .001). Table 3 shows the distribution of different reasons for transplant in each group.
In our study patients, 196 (49.4%) had a positive family history of DM; from these, 14.3% showed no DM after transplant, 36.2% had preexisting DM, 26.0% had impaired FBS or OGTT posttransplant, and 23.5% had PTDM (P < .001).
We found risk factors for DM in a greater number of patients with DM than in those without DM (63.0% vs 38.7%; P < .001); we also observed risk factors for DM in a greater number of patients with preexisting DM than with PTDM (75.8% vs 50.0%; P < .001).
When we categorized recipients into subgroups based on reason for liver transplant (viral, metabolic, autoimmune, and other cause), we found significant differences among patients categorized as normal, having preexisting DM, having impaired tests, and having PTDM (P < .001) (Figure 2).
We followed 384 patients (96.7%) for 13.6 ± 4.9 months after transplant. Of these patients, 18 recipients (4.6%) had died (8 patients with preexisting DM, 6 with PTDM, 2 with impaired tests, and 2 with normal blood glucose results). The leading cause of death was infection (10 patients), intracerebral hemorrhage (1 patient), and rejection.
At the last visit, 206 recipients (51.9%) had normal FBS, 31 (7.8%) had impaired FBS, and 43 (10.8%) were diagnosed as having PTDM (with some showing PTDM at the first visit and some being cases of PTDM that presented during follow-up). Patients with PTDM showed no significant differences in age, risk factors for DM, and causes of end-stage liver disease versus patients without DM and patients with impaired tests (all P > .05).
A multivariate Cox proportional hazards analysis revealed that steroid dosage was correlated with PTDM (hazard ratio = 1.13; P = .009). Table 4 demonstrates this correlation. Multivariate logistic regression analysis of risk factors (age, BMI, risk factor of DM, cause of cirrhosis, and dose of immunosuppressive drugs) showed that only age correlated with development of impaired tests or PTDM after liver transplant (odds ratio 1.060; 95% confidence interval, 1.026-1.095; P < .001).
During the follow-up period, 14 transplant recipients with DM died compared with 4 transplant recipients without DM. At 6 and 12 months posttransplant, overall patient survival rate was 98% for recipients without DM versus and 92% for recipients with DM.
Hyperglycemia is common in the early posttransplant setting,15 with most cases of stress- or steroid-induced hyperglycemia resolving quickly with time.16 A diagnosis of NODAT excludes patients with undiagnosed and preexisting DM and those with transient hyperglycemia posttransplant.5 The term PTDM (“posttransplant diabetes mellitus”) better explains the presence of DM in the posttransplant setting regardless of the timing of its onset.9 The criterion standard test for diagnosis of PTDM is the OGTT,5,17,18 which is optimally made once the patient is stable on maintenance immunosuppressive drugs and in the absence of acute infection.16,19
In this prospective study, we evaluated 397 recipients of liver transplant who had been assessed for blood glucose and prevalence of PTDM according to ADA recommendation based on OGTT. Patients in our study were followed for about 1 year after transplant, allowing us to observe changes in prevalence of PTDM. At the first follow-up visit, the rate of PTDM in the study group was 24.2%. This rate agreed with a cross-sectional, multicenter study involving Spanish centers with liver transplant programs from Alvarez-Sotomayor and associates.20
Regardless of the definite criteria for diagnosis of PTDM,5 discrepancies persist in its prevalence (10.8% to 33%) in different studies due to the length of follow-up (1-5 years) and time of diagnosis.14,21-25 We suggest that PTDM patients in some studies had undiagnosed and/or untreated DM before transplant.4,26 However, in our study, we precisely excluded patients with preexisting DM. This difference could also be explained by the self-resolving and transient nature of PTDM in some individuals, as we showed in our study.
Some risk factors are well known for PTDM, including both general risks for DM (such as age) and transplant-specific factors (such as immunosuppressive use).19 Some immunosuppressant agents can have significant effects on glucose metabolism; therefore, these drugs can be important in the development of PTDM. Corticosteroids are known to induce insulin resistance and hepatic gluconeogenesis and can reduce insulin-induced glucose utilization in adipose tissue and muscle.27,28 In our study, our multivariate Cox proportional hazards analysis revealed that steroid dosage was correlated with PTDM.
In follow-up of patients, we found that hyperglycemia in some PTDM patients resolved, with prevalence decreasing to 10.8% at the last visit. In a retrospective study, Zayed and colleagues found that 25.8% of PTDM patients were transient, with recovery within 6 months to 1 year,29 which allowed them to divide NODAT patients into those with transient and permanent DM. In our study, we found 13.4% of the recipients had transient PTDM by this definition. Some other previous studies have also described this definition for transient hyperglycemia, which ended within 1 year after transplant.30
Diabetes mellitus and its consequences are common problems after solid-organ transplant. In this prospective study, we provided a preliminary overview of the presence of DM in patients who received transplants at the Liver Transplant Center in Shiraz, Iran. To the best of our knowledge, this study is advantageous to others due to its prospective design, as many previous studies had a retrospective or cross-sectional design. In this cohort study, we followed patients and reevaluated the prevalence of PTDM and found significant recovery for this type of DM. However, our study has some limitations. Although we followed most patients and measured FBS, OGTT was not performed in all patients. Second, we did not evaluate patients for other transplant complications and did not assess the impact of high blood sugar on morbidity and mortality in these individuals. For a more complete assessment, further larger and multicenter studies are needed.
Volume : 19
Issue : 9
Pages : 928 - 934
DOI : 10.6002/ect.2019.0158
From the 1Endocrinology and Metabolism Research Center, Shiraz University of Medical Sciences, the 2Department of Radiology, Shiraz School of Medicine, Shiraz University of Medical Sciences, the 3Shiraz Transplant Research Center, Shiraz University of Medical Sciences, and the 4Health Policy Research Center, Shiraz University of Medical Sciences, Shiraz, Iran
Acknowledgements: The authors are thankful to the Nemazee Hospital Organ Transplant Center, Shiraz, Iran, and especially thank Mr. Siavash Gholami and Miss Goli Mehrdad for help in gathering these data. The authors have no sources of funding for this study and have no conflicts of interest to declare.
Corresponding author: Kamran Bagheri Lankarani, Shiraz School of Medicine, Zand Blvd., Shiraz, Islamic Republic of Iran 7134845794
Phone: +98 7132309615
Table 1. Anthropometric Data of Liver Transplant Recipients
Table 2. Frequency of Underlying Liver Disease in Study Cohort
Table 3. Frequency of Underlying Liver Disease in the 4 Recipient Groups
Table 4. Risk Factors for Posttransplant Diabetes Mellitus
Figure 1. Characteristics and Risk Factors in the Patient Groups
Figure 2. Distribution of Viral, Autoimmune, and Metabolic Reasons for Liver Transplant in the 4 Groups of Recipients