Metabolic derangements are almost universal in patients after liver transplantation. Reported rates are 40%-85% for hypertension,1, 2 13%-61% for diabetes mellitus,2-4 40%-66% for dyslipidemia (mainly hypertriglyceridemia),2, 4, 5 and 24%-40% for obesity.2, 4, 6, 7 Immunosuppressive medications play a significant role in their pathogenesis.8 Calcineurin inhibitors, such as cyclosporine and tacrolimus, and steroids as well, are associated with hypertension, hyperglycemia, and dyslipidemia.8 Sirolimus, a non-nephrotoxic drug, can also contribute independently to dyslipidemia.9 Other potential factors are lifestyle modifications, namely, the return to normal daily life and free food intake, which contribute to weight gain and the resultant insulin resistance (IR),6 and in some cases, the underlying liver disease itself; hemochromatosis, alcohol abuse, and autoimmune hepatitis have all been reported to be associated with diabetes. The relationship between hepatitis C virus (HCV) infection and diabetes or IR is well-known in the nontransplant population and probably involves the direct effect of the virus on insulin-signaling pathways.10 Cumulative data suggest that this is true in the transplant population as well.10, 11 Metabolic derangements might also be due to the persistence of factors that led to development of the primary nonalcoholic fatty liver disease (NAFLD) that was the indication for transplantation or their recurrence after transplantation in patients with nonalcoholic steatohepatitis (NASH)-related cirrhosis.12-14
Obesity, hypertension, hyperglycemia, and dyslipidemia comprise the metabolic syndrome, which is evidently associated with cardiovascular morbidity and mortality.15 In renal transplant recipients, cardiovascular disease is the main cause of graft loss and death in the long term.16 However, in liver transplant recipients, the leading causes of death are graft failure due to recurrence of the liver disease and chronic rejection.17 Nevertheless, with the increasing survival of liver transplant recipients, cardiovascular disease is emerging as the main cause of non–graft-related mortality,17, 18 especially in the older population.
Although the individual components of the metabolic syndrome have been well described in liver transplant recipients, data regarding full-blown posttransplantation metabolic syndrome (PTMS) and its effects in this population are still lacking. The aim of this study was to determine the prevalence of PTMS and to assess risk factors for its development and cardiovascular sequelae.
The Liver Transplant Unit at Rabin Medical Center is the largest such unit in Israel. For the present study, we reviewed the files of 248 consecutive patients who received a liver transplant from a cadaver or living donor from January 1991 to December 2007 and were followed for at least 6 months. Four patients who underwent liver transplantation abroad between 1985 and 1991 were added to the study group as well. All study patients survived at least 3 months after the procedure. An additional 12 patients lost to follow-up were excluded because of missing posttransplantation data. Follow-up visits were conducted on an outpatient basis, and all patients received the same intraoperative and postoperative care. Nearly all patients received standard immunosuppression with a calcineurin inhibitor (90 cyclosporine, 159 tacrolimus) and mycophenolate mofetil associated with tapered corticosteroids within 6 months; only 4 patients (a statistically negligible number) were switched from cyclosporine to tacrolimus. Five patients who underwent transplantation because of liver disease of autoimmune origin were still receiving low-dose methylprednisolone (2.5 to 5 mg/day) at the time of the study. The small number of subjects on sirolimus (n = 12) were not considered. The indications for liver transplantation are summarized in Table 1. Three patients in whom HCV infection was associated with hepatitis B virus (HBV) infection were included in the HCV group (35%) because the HCV infection had the higher metabolic impact. The HCV group also included 3 patients with alcoholism who had overcome their addiction before transplantation.
Table 1. Baseline Characteristics of Liver Transplant Recipients
Value (N = 252)
Mean ± SD. All other values are n (%).
“Other” indications included alcoholism (15 cases), idiopathic Budd-Chiari syndrome (3 cases), secondary biliary cirrhosis (2 cases), congenital biliary atresia (4 cases), glycogen-storage disease type III (2 cases), Wilson disease (7 cases), hyperoxalluria (3 cases), primary amyloidosis (1 case), fulminant hepatic failure (most were drug-induced, the rest idiopathic; 6 cases), anti-trypsin deficiency (2 cases), porphyria (1 case), cystic fibrosis (2 cases), congenital hepatic fibrosis (1 case), and sarcoidosis (1 case).
Follow-up monitoring included assessment of immunosuppressive therapy as well as clinical evaluation (vital signs and weight and questioning about current medications, adverse reactions, and new major complications) and laboratory tests (blood count, full chemistry including lipid profile, and blood level of immunosuppressive medications, as needed).
The following data were extracted from the medical records: patient sex and age, cause of liver disease, date of transplantation, length of follow-up (at least 6 months), pretransplant and posttransplantation weight and height, posttransplantation waist circumference, presence of pretransplant and posttransplantation diabetes mellitus, hypertension, or hyperlipidemia, and prescribed medications (immunosuppressive, antihypertensive, hypoglycemic, and lipid-lowering drugs). For patients who died before the study, we collected the most recent laboratory assay findings from the files. For patients who were alive, we collected fasting venous blood samples for same-day assay of the lipid profile (total cholesterol, low-density lipoprotein, high-density lipoprotein [HDL], and triglycerides), fasting glucose, and levels of creatinine and hemoglobin at their last follow-up visit. In all cases, when feasible, we recorded the pretransplantation serum glucose level and lipid profile. In addition, in a consecutive group of 90 patients, posttransplantation serum samples for measurement of insulin were drawn in the last follow-up visit over a period of 4-5 months in 2008. The serum was frozen at −20°C, and insulin was measured by using an enzyme-labeled chemiluminescent immunometric assay (Immulite 2000; Siemens). Insulin measurements were performed once monthly (total of 5 measurements) for several samples, to maintain standardization. Fasting glucose levels were determined in the same serum from which the frozen samples were derived. Diabetic patients receiving insulin treatment were excluded from this procedure because the possible formation of anti-insulin antibodies could interfere with the assay.
Assessment of Insulin Resistance
IR was estimated with the homeostasis model assessment (HOMA) according to the following formula: HOMA-IR = fasting insulin (mU/mL) × fasting glucose (mmol/L), divided by 22.5.19 This method has proved to be accurate at predicting IR in the normoglycemic population over a wide range of glucose blood levels, before glucose tolerance or overt diabetes develops.20 The cutoff for the diagnosis of IR was set at 2.5, which is an accepted significant value.
Definition of Metabolic Syndrome
Metabolic syndrome was defined according to the 2001 guidelines of the National Cholesterol Education Program Adult Treatment Panel III (ATP III)21 and their revision in 2004 by the National Heart, Lung and Blood Institute and the American Heart Association.22 Although body mass index (BMI) ≥30 kg/m2 has not been adopted by the ATP III in the definition of metabolic syndrome, being overweight clearly correlated with the metabolic risk factors. Therefore, in the present study, the metabolic syndrome was diagnosed when at least 3 of the following 5 criteria were met:
1BMI ≥30 kg/m2 or waist circumference >102 cm for men and 88 cm for women.
2Fasting plasma glucose ≥100 mg/dL (5.6 mmol/L).
3Blood pressure ≥130/85 mm Hg.
4Triglycerides ≥150 mg/mL (1.7 mmol/L).
5HDL <40 mg/dL (1.0 mmol/L) in men and <50 mg/dL (1.3 mmol/L) in women.
For purposes of the study, patients were also considered to have abnormal blood glucose, blood pressure, or triglyceride level if they were receiving hypoglycemic agents, antihypertensive agents, or fibrates, respectively.
Definition of Major Vascular Events
Major vascular events were defined as transient ischemic attack, cerebrovascular accident, acute coronary syndrome, and myocardial infarction. Coronary events were identified by coronary angiography or coronary revascularization. Most of the vascular events occurred in our institution; data on events that occurred in other institutions (minority) were obtained from medical records. Patients were not systematically screened for asymptomatic coronary events or cerebrovascular disease after transplantation.
Pearson correlation coefficient (r) and the significance for it (P) were calculated between the variables. To analyze statistically significant differences between patients with and without metabolic syndrome and between the pretransplant and posttransplant settings, we used a Student t test for continuous variables and chi-square test for categorical variables. The data were fitted to a multivariate stepwise logistic regression model, and odds ratios (OR) and 95% confidence intervals were estimated. P values ≤ 0.05 were considered statistically significant. To analyze a possible correlation between cardiovascular morbidity and IR to pretransplant and posttransplant metabolic derangements, we used a Student t test for continuous variables and chi-square test for categorical variables. P values ≤ 0.05 were considered statistically significant.
The characteristics of the 252 patients who underwent liver transplantation during the study period are summarized in Table 1, and the rates of the constituents of metabolic syndrome before and after transplantation are shown in Table 2. Complete data for assessing metabolic syndrome before transplantation were available for only 221 patients. Mean follow-up time after transplantation was 6.2 ± 4.4 years. Compared to the pretransplant period, the posttransplant period was characterized by a statistically significantly higher rate of obesity (BMI > 30 kg/m2), hypertriglyceridemia (>150 mg/dL), HDL cholesterol <40 mg/dL, dyslipidemia, hypertension, and diabetes. The prevalence of metabolic syndrome increased from 5.4% before transplantation (n = 12) to 51.9% after (n = 131).
Table 2. Prevalence of Constituents of Metabolic Syndrome Before and After Liver Transplantation
Table 3 compares the characteristics of the transplant recipients with and without PTMS and the rates of each of the constituents of the metabolic syndrome. Fasting insulin level was available in 90 patients. IR assessed by the HOMA index was observed in 27 patients (30%) and was significantly more prevalent in the PTMS group. Twelve of the 26 tested patients with HCV infection (41.6%) had IR. Risk factors associated with IR on univariate analysis were older age (P = 0.0001), pretransplant BMI > 30 kg/m2 (P = 0.0002), hypertriglyceridemia (P = 0.04), HCV infection (P = 0.02), and treatment with steroids (P = 0.03).
Table 3. Characteristics of Liver Transplant Recipients With and Without Posttransplantation Metabolic Syndrome
Table 4 shows the variables associated with PTMS on univariate analysis. The PTMS group was significantly older than the non-PTMS group and contained more males. Moreover, patients with PTMS had a higher prevalence of diabetes and hypertension, higher levels of serum triglycerides and HDL cholesterol, and a higher BMI before and after transplantation (P < 0.0001).
Table 4. Variables Associated With PTMS on Univariate Analysis
95% Confidence Interval
Obesity (BMI > 30 kg/m2)
Obesity (BMI > 30 kg/m2)
Underlying liver diseases that served as an indication for liver transplantation and were associated with PTMS included HCV infection and cryptogenic cirrhosis; biliary diseases, (primary biliary cholangitis, primary sclerosing cholangitis), autoimmune hepatitis, and HBV infection were not associated with PTMS. Patients with other indications for transplantation had a lower prevalence of PTMS. The use of any type of immunosuppressive drug (cyclosporine, tacrolimus, sirolimus, or steroids) had no effect on the prevalence of PTMS, although there was an association between calcineurin inhibitors (both tacrolimus and cyclosporine) and the prevalence of posttransplant hypertension (P = 0.009 and P = 0.005, respectively) and hypertriglyceridemia (P = 0.004 and P = 0.01, respectively).
On multivariate analysis (Table 5), the variables that independently predicted PTMS were older patient age, presence of pretransplant metabolic derangements, and cryptogenic cirrhosis as the indication for transplantation. Findings for other indications for liver transplantation and for immunosuppressive treatment were not statistically significant.
Table 5. Independent Variables Associated With PTMS on Multivariate Analysis
95% Confidence Interval
Pre-LTx BMI (kg/m2)
Pre-LTx HDL cholesterol
Prevalence of Major Vascular Events
The prevalence of major vascular events in the whole cohort (252 patients) increased from 2.3% before transplantation (6 events) to 10.3% after (26 events), but the difference was not statistically significant (P = 0.06). After transplantation, 3 patients each had 2 events and 20 patients each had one; 17 of the patients with events had PTMS (12.9% of PTMS patients), and 6 of the patients with events did not (4.9% of non-PTMS patients; P = 0.027). The types of events are listed in Table 6. The cumulative incidence of cardiovascular morbidity was statistically higher in patients with PTMS (P < 0.007; Fig. 1). Other variables associated with cardiovascular events on univariate risk analysis were older age (P = 0.0004), waist circumference (P = 0.009), posttransplant BMI > 30 kg/m2 (P = 0.02), hypertension (P = 0.004), and cryptogenic cirrhosis as the cause of end-stage liver disease.
Table 6. Incidence of Major Vascular Events After Transplantation in the Whole Cohort and by Presence of PTMS
Major Vascular Events
All Patients (N = 252)
PTMS (N = 131)
No PTMS (N = 121)
Total no. of patients (%)
Acute coronary syndrome
Transient ischemic attack
Forty-six patients (18.2%) died during follow-up. There was no significant difference in mortality between the patients with (19.0%) or without (17.3%) PTMS (P = 0.7). Death in both groups was related mainly to recurrent liver disease leading to graft failure or immunosuppressant-induced sepsis; only 1 death was due to an acute coronary event.
This study revealed a 59.1% rate of metabolic syndrome in patients after liver transplantation, which is more than twice the estimated age-adjusted prevalence reported in the general Western population (23.7%).23 Furthermore, PTMS is associated with cardiovascular morbidity but not mortality, and it may be predicted by the pretransplant conditions. Rates of hypertension (58.3%), obesity (27.5%), and diabetes (39.6%) after transplantation in our cohort were comparable to those reported in previous studies.24 The rate of dyslipidemia, however, was higher (71% versus 27%-66%).24 The latter finding may be at least partly due to our use of sirolimus (12 patients), the presence of cryptogenic cirrhosis as an indication for liver transplantation (28 patients; most presumably with NAFLD), persistence of preexisting factors after transplantation in patients with NASH-related cirrhosis and undertreatment with lipid-lowering drugs (fibrates or statins in 36 patients). Moreover, the extremely low prevalence rate of PTMS before transplantation (5.4%) might be explained by the hemodynamic and metabolic changes associated with chronic liver disease, leading to peripheral vasodilatation, low arterial blood pressure, and reduced serum cholesterol levels.25
The patients who had PTMS were significantly older than those who did not, reflecting the effect of age on many metabolic disturbances.
On multivariate analysis, cryptogenic cirrhosis was the only underlying liver disease that predicted PTMS. This might be explained by findings that a substantial proportion of patients with cryptogenic cirrhosis have metabolic syndrome or some of its components, suggesting that “cryptogenic” cirrhosis in these cases may be unrecognized NASH.13, 26, 27 As cirrhosis develops, the steatosis and inflammatory features disappear, so the liver biopsy may show cirrhotic-stage disease without other features of NAFLD.13, 26, 27 Accordingly, Malik et al.28 reported that compared to controls, patients undergoing liver transplantation for NASH cirrhosis had a higher BMI and were more likely to be diabetic and hypertensive.
Two of the important findings in our study were the association of HCV infection with PTMS and with IR on univariate analysis. The high (41.6%) rate of IR by HOMA among tested patients with HCV agrees with the recent study of Hanouneh et al.29 wherein rates of fibrosis progression and metabolic syndrome were higher in liver recipients with recurrent HCV infection. IR and diabetes mellitus are the most recognized metabolic derangements to which HCV infection has been linked. Delgado-Borrego et al.11 reported a 77% higher rate of IR by HOMA in HCV-positive than HCV-negative patients in the first year after liver transplantation. In another study, HCV infection proved to be an independent predictor of de novo posttransplant diabetes (OR = 2.6).30 Possible mechanisms underlying the HCV-diabetes link have been proposed: HCV could cause a failure of compensatory hyperinsulinemia by injuring β-cells31 (via a direct cytopathic effect or cytokine-mediated tissue damage), or it may promote or accentuate IR by affecting multiple steps in the insulin-signaling pathways.32
By contrast, the prevalence of PTMS was low in patients with cirrhosis with other indications for transplantation, most of them of biliary origin (primary sclerosing cholangitis, genetic defects). These probably reflect the late involvement of hepatocytes, which are responsible for glucose metabolism.
Although others have suspected a metabolic effect of immunosuppressive drugs, our findings do not support their role in the development of PTMS. However, it should be borne in mind that most of our liver transplant recipients were weaned off corticosteroids within 6 months after transplantation. We also detected a correlation between calcineurin inhibitors and an increased prevalence of some of the posttransplant metabolic derangements (hypertension and hypertriglyceridemia). Previous studies, too, failed to find immunosuppressive drug intake to be a risk factor for PTMS4 or posttransplant diabetes mellitus.11 It is possible that other factors predispose liver transplant recipients to the development of PTMS: the return to normal daily life and free food intake, together with normalization of the hypermetabolic state of advanced liver disease, which can contribute to weight gain and resultant IR6, 9; changes in lipoprotein metabolism induced by the new liver; and the underlying liver disease itself (NAFLD or HCV), as mentioned before.
One limitation of this study is the failure to correct BMI for the amount of ascites when interpreting its impact on PTMS. Be that as it may, the finding that the classical pretransplant metabolic derangements, especially diabetes, predicted PTMS is not surprising. Bianchi et al.33 noted that despite the expected improvement in glucose metabolism after transplantation, in only 14% of cases was pretransplant diabetes cured after transplantation, whereas de novo diabetes developed in 32.5%. This ties in with the contribution to posttransplantation IR and diabetes of calcineurin inhibitors (especially tacrolimus) and steroids, lifestyle modifications, and HCV infection.
Although this study is limited by its retrospective design, it clearly points to the clinical significance of PTMS in liver transplant recipients, which leads to an increase in major vascular events. In our cohort, 17 patients in the PTMS group (12.9%) had major cardiovascular events compared to only 6 (4.9%) in the non-PTMS group (P = 0.027). This finding agrees with the study of Laryea et al.,4 who reported 25 major vascular events affecting 21% of patients. There were significantly more vascular events in patients with posttransplantation metabolic syndrome than in those without (30% versus 8%, P = 0.003). Not surprisingly, cardiovascular morbidity was associated with other posttransplant metabolic derangements, as well as cryptogenic cirrhosis, the prominent underlying liver disease that predicted PTMS.
Most previous studies described similar rates of cardiovascular events on long-term follow-up of liver transplant recipients (24%).34, 35 Neal et al.36 reported that the predicted 10-year risk of coronary heart disease (CHD) increased from 6.9% before transplantation (close to the 7% for the general population) to 11.5% 1 year after. In our study, despite the high incidence of cardiovascular disease, the cardiovascular mortality rate after transplantation was lower than reported by others.17, 18, 35 Among studies of transplant-related versus nontransplant-related patient mortality, Johnston et al.18 found that the relative risk of death from cardiovascular disease was 2.56 in transplant recipients relative to an age-matched nontransplant population. Deaths caused by ischemic heart disease occurred at a median of 27 months after transplantation. The relative risk of ischemic cardiac events was 3.07, and the median times to myocardial infarction and stroke were 32 and 34 months, respectively. Sheiner et al.,35 in a study of 96 patients surviving 5 or more years after transplantation, reported that 6.1% of deaths were due to cardiac disease; overall, 2.2% of patients died of cardiovascular disease. However, Neal et al.36 emphasized the disparity between the high prevalence of cardiovascular risk factors after liver transplantation and the lack of deaths from cardiovascular events. Several possible reasons have been suggested: 1. Tacrolimus is associated with fewer cardiovascular events and lower posttransplant mortality from CHD than cyclosporine.37 However, we did not find immunosuppressive therapy to be a risk factor for cardiovascular events. 2. The presence of CHD or risk factors for CHD before transplantation has been implicated in the development of posttransplantation CHD,18 although we detected a significantly lower prevalence of CHD risk factors before transplantation in our cohort. 3. Patients who undergo liver transplantation are preselected to be at low risk of cardiac disease. Many patients with end-stage liver disease due to NAFLD who are severely obese or have long-term complications of diabetes and/or dyslipidemia, such as cardiovascular and pulmonary diseases, are precluded from liver transplantation. Yet another explanation is the short duration of follow-up in most studies of cardiovascular mortality in this setting.
In conclusion, the prevalence of metabolic syndrome is significantly higher in patients after liver transplantation than the estimated value in the general population. Metabolic syndrome is associated with cardiovascular morbidity but not mortality. Pretransplantation conditions may predict PTMS. Prospective studies are required to identify the significance of metabolic syndrome in liver transplant recipients, and to establish optimal management.