The findings of this study were presented in part at the annual meetings of the Canadian Association for the Study of the Liver (CASL) in Banff, Alberta, Canada, in March 2005 and the European Association for the Study of the Liver (EASL) in Vienna, Austria, in May 2006.
The advent of highly effective antirejection medication has ushered in an era of long-term patient survival in liver transplantation. Calcineurin inhibitors such as cyclosporine and tacrolimus are now widely used in transplant centers. The introduction of nonnephrotoxic drugs such as mycophenolate mofetil and sirolimus provide potential new immunosuppressive strategies to further improve long-term results.
These immunosuppression drugs can be associated with metabolic derangements that are now well-recognized side effects. Notably, cyclosporine and tacrolimus are associated with hypertension, hyperglycemia, and dyslipidemia; also, sirolimus may contribute to hyperlipidemia in these solid-organ transplant recipients. These adverse drug reactions often lead to dose adjustments, discontinuation of therapy, or often addition of medications for treatment of these adverse events.
Along with obesity, hypertension, hyperglycemia, and dyslipidemia are the main features of metabolic syndrome.1 There has been much interest in metabolic syndrome recently, notably in the cardiovascular literature. Metabolic syndrome is related to defects in insulin sensitivity, predisposing subjects to the development of type 2 diabetes mellitus2 and cardiovascular disease3 in the nontransplantation population. Although the presence of any of the 4 cardinal features of metabolic syndrome in an individual is a risk factor for cardiac morbidity,4 it has been well documented that the presence of 3 or more is associated with a higher incidence of cardiovascular complications.5, 6
As posttransplantation survival improves, cardiovascular disease has emerged as the leading cause of non-graft-related deaths.7 In a report that described the cause of death in 299 adult liver transplant recipients who lived for more than 3 yr, 8 of 38 deaths (21%) were due to cardiovascular complications.8 Similarly, in a study of 542 patients who survived at least 1 yr after liver transplantation, cardiovascular events accounted for 18 of 43 (42%) of non-graft-related deaths.9
Cardiovascular disease and diabetes are commonly the cause of organ failure in cardiac and renal transplant candidates. This has not been true of most liver patients. However, posttransplantation, they will often have 1 or more metabolic disturbances either unmasked or caused by antirejection medications. In fact, some patients will meet the criteria for metabolic syndrome. The goal of this study is to determine the prevalence of metabolic syndrome post–liver transplant, the risk factors for developing the condition and whether posttransplantation metabolic syndrome (PTMS) carries the same increased risk of cardiovascular disease as metabolic syndrome in the general population.
PTMS, posttransplantation metabolic syndrome.
PATIENTS AND METHODS
The Multi-Organ Transplant Program is based at the Queen Elizabeth II Health Sciences Center in Halifax, NS, Canada. It provides full transplant care for a population of over 2.5 million inhabiting the 4 eastern Canadian Atlantic provinces. All monitoring and adjustment of immunosuppression therapy is made centrally by our Liver Transplant Office. The Office maintains full patient care records including current medications, recent blood work, clinical parameters including height, weight, and blood pressure, and any major complications (renal failure, cardiovascular events, malignancies, and osteoporosis) that have arisen during follow-up.
Transplant related data were extracted from the Transplant Atlantic Registry. To be included in the study, patients had to have received a transplant between January 1998 and July 2004, and to have been followed by the Liver Transplant Office in Halifax, NS until the end of December 2005. All patients had to have survived at least 3 months post–liver transplantation. Until June 2001, all liver transplantations were conducted at our Center in Halifax, NS using immunosuppression protocol based on low-dose tacrolimus and sirolimus with tapering dose of prednisone.10 Between July 2001 and December 2005, due to loss of transplant surgeons from our Program, transplant operations were completed by the Multi-Organ Transplant Program in London, Ontario, Canada, using immunosuppression protocol based on tacrolimus and tapering dose of prednisone with or without mycophenolate mofetil.
The following data were collected: age, gender, indication for transplantation, date of transplant, length of follow-up, pretransplantation weight, and posttransplantation body mass index. Most recent values for the following serum levels were also compiled: fasting glucose, random glucose, total cholesterol, low-density lipoprotein, high-density lipoprotein, triglycerides, hemoglobin A1C, creatinine, and hemoglobin. Information on medications prescribed with dosage was also collected, focusing on antirejection drugs including prednisone, hypoglycemic agents including insulin, antihypertensive drugs, and cholesterol- or triglyceride-lowering agents.
Definition of Metabolic Syndrome
Metabolic syndrome was defined using the Guidelines from the 2001 National Cholesterol Education Program - Adult Treatment Panel III.11 Metabolic syndrome was said to be present if patients had at least 3 of the following:
1Abdominal obesity (waist-to-hip ratio >0.90, body mass index ≥30 kg/m2, or a waist girth = 102 cm for men and 88 cm for women);
2Triglycerides ≥1.7 mmol/L;
3High-density lipoprotein <1.0 mmol/L in men and <1.3 mmol/L in women;
4Blood pressure ≥130/85 mmHg;
5Fasting plasma glucose ≥5.6 mmol/L.
Acceptable fasting glucose level cutoff was lowered from 6.2 to 5.6 mmol/L as recommended by a 2004 report from the National Heart, Lung, and Blood Institute and the American Heart Association.12 This is in accordance with a 2003 recommendation from the American Diabetes Association.13
For the purposes of our study, we also used clinical surrogates in diagnosing metabolic syndrome; patients met criteria for abnormal fasting glucose if they were on hypoglycemic agents including insulin. Use of antihypertensive agents was also sufficient to be considered hypertensive. Lipid lowering agents were initiated only in those with documented fasting serum low-density lipoprotein or triglycerides >2.5 mmol/L for more than 6 months in spite of appropriate dietary recommendations.
Definition of Major Vascular Events
All major vascular events including stroke, transient ischemic attack, myocardial infarction, acute coronary syndrome, and sudden cardiac death were compiled. All coronary heart diseases were identified by electrocardiography, coronary angiography, or coronary revascularization. Ischemic stroke events (thrombotic or cardioembolic) were classified using criteria adapted from the National Survey of Stroke.14 Most of these events were obtained from forwarded discharge summaries or clinic notes. Patients were not systematically screened for asymptomatic coronary heart disease or cerebrovascular disease post–liver transplantation.
Descriptive statistics (mean, standard deviation, range, and proportions) were used to describe each variable. Comparisons between groups were made with the χ2 test. Prevalence rates were derived by dividing the number of events by person-time at risk. To model the relationship between the metabolic syndrome and time to incidence of major vascular events, we used the nonparametric distribution analysis with right censoring for death and created a cumulative failure (major vascular events) probabilities plot. Statistical significance was demonstrated by a P value < 0.05. Statistical analysis was performed with Minitab version 14 (Minitab, State College, PA).
Prevalence of PTMS
A total of 148 patients received transplantation during the study period for both acute liver failure and end-stage liver disease due to chronic liver disease. Out of 132 patients who survived beyond 3 months following liver transplant, 119 had been followed up for at least 18 months. Complete data for assessment of metabolic syndrome was not available for 1 patient, leaving 118 patients in our cohort. The patient characteristics are summarized in Table 1. Following liver transplantation there was significant increase in individuals with hypertension from 10 to 62% (from 12 to 72 patients), dyslipidemia from 3 to 46% (from 4 to 54 patients), and diabetes mellitus from 13 to 61% (from 15 to 73 patients). Table 2 summarizes the prevalence of PTMS and its 5 components after liver transplantation. Metabolic syndrome occurred post–liver transplantation in 69 of 118 (58%) patients.
Table 1. Characteristics of Liver Transplant Recipients With Descriptive Analysis of Those With PTMS and Those Without by Univariate Analysis
Abbreviations: HDL, high density lipoprotein, BMI, body mass index;.
Mean ± standard deviation.
Mean ± standard error of the mean.
54 ± 11
57 ± 1
49 ± 2
Indication for liver transplantation (%)
Fulminant liver failure
Primary biliary cirrhosis
Primary sclerosing cholangitis
Hepatitis C infection
Time from transplant (months)
58 ± 21
58 ± 3
60 ± 3
Pretransplantation weight (kg)
80 ± 19
86 ± 2
73 ± 2
Posttransplantation weight (kg)
83 ± 21
90 ± 3
74 ± 2
Posttransplantation BMI (kg/m2)
29 ± 6
31 ± 1
25 ± 1
Fasting glucose (mmol/L)
6.6 ± 2.3
7.4 ± 0.3
5.4 ± 0.2
Hemoglobin A1c (%)
6.0 ± 1.2
6.4 ± 0.2
5.4 ± 0.1
1.15 ± 0.40
1.0 ± 0.3
1.4 ± 0.4
2.72 ± 2.89
3.6 ± 0.4
1.4 ± 0.1
124 ± 86
132 ± 11
114 ± 11
132 ± 18
131 ± 2
134 ± 2
Table 2. Prevalence of Metabolic Derangements Included in the Definition of PTMS by Category at the End of the Study Period
Total (N = 118) (%)
PTMS (N = 69) (%)
No PTMS (N = 49) (%)
Abbreviations: HDL, high density lipoprotein; BMI, body mass index.
BMI >30 kg/m2
Overall, by univariate analysis, patients with PTMS and without had similar follow-up times from transplantation, creatinine levels, and hemoglobin levels. Patients with PTMS were significantly older with higher average pretransplantation weight, posttransplantation body mass index, fasting glucose, triglycerides, and hemoglobin A1C. Weight gain after liver transplantation was not a factor in the development of PTMS in this patient population. Many patients had ascites before liver transplantation leading to overestimation of pretransplantation weight since information regarding body composition was lacking.
As shown in Table 1, the patients seemed to segregate into 3 groups of underlying diseases requiring liver transplantation: those associated with an increased prevalence of PTMS—alcohol-related liver disease (74%), cryptogenic cirrhosis (90%), and hepatitis C infection (79%); those associated with a lower prevalence of PTMS—fulminant liver failure (29%) and primary sclerosing cholangitis (18%); and those with an even chance of developing PTMS or not—primary biliary cirrhosis (47%) and other indications (50%).
The use of tacrolimus, mycophenolate, sirolimus, and prednisone were not significantly different between groups. The mean daily dose (± standard deviation) of cyclosporine (Neoral) was 133 ± 29 mg (divided into a twice daily dosage), tacrolimus 4 ± 3 mg (divided into a twice daily dosage). The 32% of patients on mycophenolate mofetil were on total daily dose of 915 ± 325 mg; while the 44% of patients on sirolimus were taking only 2 ± 1 mg dosage. The mean daily dose of prednisone was 2 ± 5 mg with 70% off corticosteroids. The mean (± standard error of the mean) dose of prednisone among those with diabetes vs. no diabetes was no different with 1 ± 3 mg vs. 2.5 ± 6 mg, respectively (P value = 0.183).
In multivariate analysis of patient characteristics including age, gender, indication for transplantation, comorbidities, and immunosuppressant drugs, only age and indication for transplantation were identified as independent variables associated with PTMS. Logistic regression analysis for PTMS found having alcohol-related liver disease, cryptogenic cirrhosis, or hepatitis C infection to be the only statistically significant variable with odds ratio of 3.43 (95% confidence interval 1.90-6.21; P < 0.001). Whereas age (odds ratio 1.04), female gender (odds ratio 0.67) and immunosuppressant drugs were not significant (odds ratio for calcineurin inhibitors, sirolimus, mycophenolate mofetil, and corticosteroids was 0.38, 1.81, 0.72, and 1.55, respectively)
Prevalence of Major Vascular Event
Over an average follow-up period of 58 months (range, 20 to 95 months), 25 major vascular events occurred, as shown in Table 3. Of those 25 events, 21 occurred in the 69 patients with PTMS (30%) and 4 occurred in the 49 patients without PTMS (8%). The cumulative incidence of major vascular events was statistically more frequent in the PTMS group, with a P value of 0.003, or a χ2 value of 9.036 (Fig. 1).
Table 3. Incidence of Major Vascular Events During the Study Period Overall and for PTMS Patients and Those Without PTMS
During the follow-up period, 7 (6%) died, 4 after myocardial infarction, 1 from esophageal cancer, 1 from rapid recurrence of hepatitis C with decompensated liver disease, and 1 following community-acquired pneumonia. When compared, there was no statistically significant difference in death from all cause between PTMS group and those without PTMS, 9% vs. 2%, respectively (χ2 value of 2.274, P value = 0.132).
Although this study is descriptive in nature, it appears that metabolic disturbances, such as hypertension, dyslipidemia, hyperglycemia, and obesity, are almost universal in patients post–liver transplant. In addition, PTMS was very common in our cohort of liver transplantation patients. This may in part reflect a higher prevalence of metabolic syndrome in our catchment area. However, when compared to previously reported studies,15, 16 the prevalence of hypertension,17 hyperlipidemia,18 and weight gain19 in our liver transplantation population is comparable. The prevalence of diabetes in our transplant recipients was higher than what has been described. We believe the adoption of a lower threshold for fasting blood glucose as per recent recommendations is at least in part responsible for this discrepancy.12, 13 In spite those variations, it is clear that liver transplantation patients are at increased risk for metabolic syndrome but the etiologic relations remain unclear.20
Our hypothesis was that PTMS was due to well-known metabolic side effects of immunosuppressant agents. Although there is association of hypertension with calcineurin inhibitors, hyperglycemia with tacrolimus, hypercholesterolemia with cyclosporine, and hypertriglyceridemia with sirolimus, our univariate analysis did not show any specific association with PTMS. The lack of association with tacrolimus or sirolimus may also be related to smaller doses of these drugs in our patient population.10 The effect of cyclosporine could not specifically be studied as only 3 patients were on that medication at the time of data collection.
We also did not find significantly higher doses of prednisone in those with metabolic syndrome. This may reflect that most of our patients were weaned off corticosteroids soon after transplantation, in a timeframe that may not have allowed metabolic syndrome to develop. Furthermore, the use of pulse steroids for cellular rejection in our patients on combination of tacrolimus and sirolimus is rare (6%) as previously reported.10
PTMS patients were significantly older than those without metabolic syndrome, reflecting the age-related nature of many metabolic disturbances. In spite of overestimation of pretransplantation weight due to fluid retention and ascites, the finding that patients that developed PTMS had higher pretransplantation weight is interesting but will need a more accurate measurements of body composition.
Although this study is a retrospective chart review and therefore limited, the dramatically greater number of major vascular events in patients with PTMS vs. those without is a significant finding. With the exception of one,15 previous studies have also described similar rates of cardiovascular events during long-term follow-up of liver transplant recipients.8, 9, 21, 22 Furthermore, the high incidence of these events in a relatively young population hints at an acceleration of the pathophysiology of cardiac and cerebral vascular disease in these patients. This may be related to the impact of immunosuppression on the underlying liver disease such as nonalcoholic steatohepatitis and chronic hepatitis C virus infection.23
Whether the clinical relevance of PTMS in liver transplant patients is the same as that of the general population with metabolic syndrome is an important question. Furthermore, one wonders if aggressive protocols of detection and treatment of PTMS could lead to fewer cardiovascular events in these patients. Devising prospective studies to answer these questions may lead to real changes in the natural history of liver transplantation.
We thank the liver transplant nurses of Multi-Organ Transplant Programs in Halifax, NS, and London, ON for their diligence in collecting and documenting everything that happens to our patients. We extend special thanks to Ms Jo-Anne Douglas, research database consultant at Dalhousie University Department of Medicine, for getting us on the right footing at the start of the project.