Physical activity and metabolic syndrome in liver transplant recipients

Authors


  • There is no financial or grant support to report for this work, and none of the authors have any relevant conflicts of interest to disclose.

Address reprint requests to Eric R. Kallwitz, M.D., Division of Hepatology, Loyola University Medical Center, 2160 South First Avenue, Maywood, IL 60153. Telephone: 855-483-7362; FAX 708-216-6299; E-mail: ekallwitz@lumc.edu

Abstract

There is a high prevalence of metabolic syndrome in liver transplant recipients, a population that tends to be physically inactive. The aim of this study was to characterize physical activity and evaluate the relationship between physical activity and metabolic syndrome after liver transplantation. A cross-sectional analysis was performed in patients more than 3 months after transplantation. Metabolic syndrome was classified according to National Cholesterol Education Panel Adult Treatment Panel III guidelines. Physical activity, including duration, frequency, and metabolic equivalents of task (METs), was assessed. The study population consisted of 204 subjects, with 156 more than 1 year after transplantation. The median time after transplantation was 53.5 months (range = 3-299 months). The mean duration of exercise was 90 ± 142 minutes, and the mean MET score was 3.6 ± 1.5. Metabolic syndrome was observed in 58.8% of all subjects and in 63.5% of the subjects more than 1 year after transplantation. In a multivariate analysis involving all subjects, metabolic syndrome was associated with a time after transplantation greater than 1 year [odds ratio (OR) = 2.909, 95% confidence interval (CI) = 1.389-6.092] and older age (OR = 1.036, 95% CI = 1.001-1.072). A second analysis was performed for only patients more than 1 year after transplantation. In a multivariate analysis, metabolic syndrome was associated with lower exercise intensity (OR = 0.690, 95% CI = 0.536-0.887), older age (OR = 1.056, 95% CI = 1.014-1.101), and pretransplant diabetes (OR = 4.246, 95% CI = 1.300-13.864). In conclusion, metabolic syndrome is common after liver transplantation, and the rate is significantly higher in patients more than 1 year after transplantation. The observation that exercise intensity is inversely related to metabolic syndrome after transplantation is novel and suggests that physical activity might provide a means for reducing metabolic syndrome complications in liver transplant recipients. Liver Transpl 19:1125–1131, 2013. © 2013 AASLD.

Abbreviations
CI

confidence interval

HCC

hepatocellular carcinoma

HCV

hepatitis C virus

HDL

high-density lipoprotein

MET

metabolic equivalent of task

MSC

metabolic syndrome complication

NAFLD

nonalcoholic fatty liver disease

OR

odds ratio

Liver transplantation provides lifesaving therapy for patients with acute or chronic liver failure and hepatocellular carcinoma. However, liver transplant recipients face long-term complications, including metabolic abnormalities and an increased risk of cardiovascular disease.[1-4] Metabolic syndrome affects 45% to 58% of liver transplant patients, and the prevalence is particularly high in patients with metabolic syndrome in the pretransplant period.[1-4]

The causes and consequences of posttransplant metabolic syndrome are well recognized. Calcineurin inhibitors and corticosteroids are linked to diabetes, hypertension, and dyslipidemia.[5, 6] Many liver transplant recipients are sedentary[7, 8] and consume excess calories,[9] and these factors contribute to the development of obesity and insulin resistance. Posttransplant metabolic abnormalities are associated with cardiovascular disease, which is the third leading cause of long-term mortality after liver transplantation.[10]

In the nontransplant setting, lifestyle interventions have been shown to prevent metabolic syndrome in at-risk individuals. For example, a lifestyle intervention aimed at reducing and maintaining a 7% weight loss, coupled with exercise for more than 150 minutes per week, resulted in a 41% decrease in the incidence of metabolic syndrome.[11] There are limited data for assessing the impact of physical activity on metabolic syndrome in liver transplant recipients. In one report,[8] higher activity levels were associated with lower rates of hypertension and a lower body mass index. Further data are needed to determine whether physical activity can prevent or reverse metabolic syndrome in liver transplant recipients, who often have limited exercise tolerance and must remain on immunosuppressive medications.

The current study evaluated exercise intensity and duration in a group of liver transplant patients who were at various times after transplantation. The aim was to assess the relationship between exercise habits and the presence of metabolic syndrome in liver transplant patients.

PATIENTS AND METHODS

The protocol was approved by the institutional review board of the University of Illinois at Chicago. The subjects consisted of a convenience sample of patients followed at the University of Illinois Medical Center for post–liver transplant care. Patients who presented to the transplant clinic and were more than 3 months from liver transplantation were enrolled between December 2009 and December 2011.

Data Collection

The study was cross-sectional in design. Data were collected at a single time point after consent was obtained during a routine, scheduled posttransplant clinic visit. The demographic data consisted of the date of birth, sex, and ethnicity. Age was analyzed at the time of enrollment, not at the time of transplantation. Clinical information included the date of liver transplantation, the indication for liver transplantation, and whether or not the recipient had pretransplant diabetes. Subjects were categorized in a dichotomous fashion with respect to the following transplant indications: hepatitis C virus (HCV), alcohol, and nonalcoholic fatty liver disease (NAFLD). Subjects with more than 1 potential indication (ie, HCV and prior alcohol abuse) were classified according to the main indication for transplantation. Other transplant indications were not separately analyzed because they were small in number. Participants also were dichotomized according to the presence or absence of hepatocellular carcinoma (HCC).

Anthropometric, Metabolic, and Physical Activity Data

Height, weight, waist circumference, and blood pressure were measured during the study visit. The waist circumference was assessed at the midway point between the lowest rib and the iliac crest. One set of laboratory measurements, including fasting glucose, triglycerides, and high-density lipoprotein (HDL) cholesterol, was recorded for each study participant from testing performed on the day of the study visit or from values obtained within 12 months of consent. Metabolic syndrome and each of its components were defined in accordance with recommendations from the National Cholesterol Education Panel Adult Treatment Panel III.[12] Metabolic syndrome was defined as the presence of 3 or more components of metabolic syndrome. Physical activity was estimated through an interview conducted by a member of the research team. Patients were instructed to report only dedicated exercise physical activity and not to include other leisure time hobbies or occupational physical activity. They provided the number of exercise sessions performed per week and the average minutes per session. In addition, by selecting from a chart showing various types of exercise activity, subjects estimated the average intensity of their physical activity in metabolic equivalents of task (METs).[13] As a descriptive measure, patients were also evaluated as to whether they met guidelines proposed by the Centers for Disease Control and Prevention for physical activity.[14]

Immunosuppression

By protocol, the induction immunosuppression consisted of methylprednisolone and basiliximab followed by a rapid prednisone taper. Mycophenolate was typically discontinued 6 to 12 months after transplantation. A calcineurin inhibitor was started on postoperative day 2 and was continued as maintenance therapy with target trough levels of 4 to 6 ng/mL for tacrolimus and 100 to 120 ng/mL for cyclosporine. Patients who received a liver-kidney transplant or underwent transplantation for autoimmune hepatitis generally were maintained on long-term, low-dose prednisone (typically 5 mg daily). The primary maintenance immunosuppressive agent was recorded, as was the use of long-term, low-dose prednisone and any bolus corticosteroid treatment for acute cellular rejection.

Statistical Methods

Descriptive statistics were evaluated both to assess the normalcy of data and to evaluate outlying values. Categorical variables were compared with Pearson's chi-square testing. Among continuous variables, age and METs were normally distributed and were analyzed with the Student t test, whereas exercise per week (minutes) was not normally distributed and was analyzed with the Mann-Whitney U test. An analysis of variance was used to compare means among more than 2 groups. Factors associated with posttransplant metabolic syndrome in the univariate analysis (P < 0.1) were entered as independent variables into a binary logistic regression analysis with posttransplant metabolic syndrome as the dependent variable. Data analysis was performed with IBM SPSS Statistics 20.

RESULTS

Two hundred thirty-eight patients consented to participate. The study population consisted of 204 subjects with complete metabolic syndrome data. There were no differences in age, ethnicity, or indication for transplantation between patients with complete metabolic data who were included in the analysis and patients who were excluded because of incomplete metabolic data. The participants included 112 men and 92 women with a mean age of 57 ± 10 years. The ethnic composition was 44.6% non-Hispanic white, 26.5% Hispanic, 24.0% African American, and 4.9% Asian or other. The indications for transplantation included HCV (39.7%), alcohol (25.4%), and NAFLD (10.3%). In addition, 20.6% of the recipients had HCC at the time of transplantation. Pretransplant diabetes was present in 21% of the cases. The maintenance immunosuppression consisted of tacrolimus (78.9%), cyclosporine A (17.6%), or a regimen not based on calcineurin inhibitors (3.4%). Long-term, low-dose prednisone was provided to 16.2% of the recipients, and 14.2% received bolus corticosteroids for acute cellular rejection. The median time from transplantation was 53.5 months (range = 3-299 months). One hundred fifty-six subjects (76%) were more than 1 year from transplantation at the time of their evaluation.

Fifty-two percent of the liver transplant recipients (107/204) reported performing no formal exercise. Among patients who exercised, the mean duration of exercise was 90 ± 142 minutes. Twenty-four percent reported exercising more than 150 minutes per week. The mean MET score was 3.6 ± 1.5. Exercise intensity was strongly correlated with the duration of exercise (r = 0.463, P < 0.001).

Metabolic Syndrome

Metabolic syndrome was observed in 58.8% of the liver transplant recipients. The prevalences of the components of metabolic syndrome were as follows: 59.3% for impaired fasting glucose, 82.8% for elevated blood pressures, 39.2% for hypertriglyceridemia, 51.0% for decreased HDL, and 50.0% for increased waist circumference. Associations between demographic and clinical factors and metabolic syndrome are shown in Table 1. Persons with metabolic syndrome were older and were more frequently more than 1 year after transplantation and on cyclosporine A versus tacrolimus. There were strong trends toward higher rates of pretransplant diabetes and lower exercise intensity in those with metabolic syndrome. As shown in Table 1, there was no association between metabolic syndrome and sex, ethnicity, indication for transplantation, previous treatment for acute cellular rejection, chronic low-dose prednisone use, or exercise duration per week.

Table 1. Univariate and Multivariate Analyses of Factors Evaluated for Assessing Relationships With Metabolic Syndrome After Liver Transplantation
VariableUnivariate AnalysisMultivariate Analysis
Metabolic Syndrome PresentMetabolic Syndrome AbsentP ValueB Coefficient/Standard ErrorP ValueOR (95% CI)
  1. a

    The data are presented as means and standard deviations.

Time after transplantation: >1 year/<1 year (n/n)99/2157/270.0211.068/0.3770.0052.909 (1.389–6.092)
Age (years)a58.9 ± 9.654.6 ± 9.00.0010.035/0.0180.0451.036 (1.001–1.072)
Pretransplant diabetes (%)29.517.30.050.751/0.3920.062.119 (0.982–4.572)
METsa3.4 ± 1.63.8 ± 1.50.06−0.188/0.1080.070.828 (0.677–1.014)
Maintenance immunosuppression: tacrolimus/cyclosporine A (n/n)88/2773/90.030.410/0.3260.211.506 (0.795–2.855)
Sex: female (%)42.548.80.37   
Ethnicity: non-Hispanic white/African American/Hispanic/other (n/n/n/n)51/26/34/940/23/20/10.20   
Indication: NAFLD (%)10.89.50.76   
Indication: alcohol (%)25.825.00.89   
Indication: HCV (%)42.538.10.53   
Presence of HCC (%)21.819.70.72   
Treatment for acute cellular rejection (%)12.617.30.36   
Chronic prednisone use (%)15.017.90.59   
Exercise duration per week (minutes)a77.4 ± 115.7107.3 ± 172.90.69   

As shown in Table 2, impaired fasting glucose and an elevated waist circumference were associated with lower levels of exercise intensity. When the exercise duration was evaluated, we found a trend toward a lower mean duration of exercise in those with hypertriglyceridemia.

Table 2. Exercise Activity and Metabolic Syndrome Component (MSC) After Liver Transplantation in All Subjects
Metabolic FeatureMETsP ValueExercise per Week (minutes)P Value
MSC PresentMSC AbsentMSC PresentMSC Absent
Impaired fasting glucose3.4 ± 1.53.9 ± 1.60.0289.7 ± 134.689.6 ± 154.00.39
Elevated blood pressure3.5 ± 1.63.9 ± 1.40.1881.2 ± 127.7130.8 ± 196.00.37
Hypertriglyceridemia3.4 ± 1.73.7 ± 1.50.1664.1 ± 103.9106.3 ± 160.80.08
Decreased HDL3.4 ± 1.63.8 ± 1.60.1174.1 ± 128.2104.6 ± 154.60.22
Increased waist circumference3.4 ± 1.63.8 ± 1.60.0371.3 ± 103.2108.1 ± 171.60.55

A multivariate analysis was performed to assess the association between age, pretransplant diabetes, immunosuppression, a time after transplantation greater than 1 year, METs, and metabolic syndrome. Older age and a time after transplantation greater than 1 year remained significantly associated with metabolic syndrome (Table 1). There was a trend toward an association between exercise intensity and pretransplant diabetes and metabolic syndrome (Table 1).

Metabolic Syndrome in Recipients More Than 1 Year After Transplantation

A further analysis was performed for the 156 subjects who were more than 1 year after transplantation. Metabolic syndrome was present in 63.5% of the members of this group. The prevalence of impaired fasting glucose was 58.3%, the prevalence of elevated blood pressure was 85.3%, the prevalence of hypertriglyceridemia was 44.9%, the prevalence of decreased HDL was 50.6%, and the prevalence of increased waist circumference was 51.9%.

Slightly more than half of the patients (51%) who were more than 1 year from transplantation reported engaging in dedicated exercise activity, with 27% of the total population performing more than 150 minutes per week. Among the patients who exercised, the mean duration was 94.8 ± 139.1 minutes per week, and the mean intensity was 3.7 ± 1.6 METs. Associations between demographic and clinical variables and metabolic syndrome in subjects more than 1 year after transplantation are shown in Table 3. Persons with metabolic syndrome were older, had a higher rate of pretransplant diabetes, were more likely to be on cyclosporine A–based immunosuppression, and exercised at a significantly lower intensity. Moreover, there was an inverse relationship between the mean MET score and the number of metabolic abnormalities present (Fig. 1). As shown in Table 3, there were no associations between metabolic syndrome and the mean time after transplantation, sex, ethnicity, transplant indication, treatment for acute cellular rejection, chronic low-dose prednisone use, or exercise duration per week. Table 4 shows further associations between components of metabolic syndrome and exercise parameters.

Figure 1.

Among patients more than 1 year after transplantation, the mean MET scores were lower for those with more features of metabolic syndrome (P = 0.002). Errors bars indicate ±1 standard deviation.

Table 3. Univariate and Multivariate Analyses of Factors Evaluated for Assessing Relationships With Metabolic Syndrome in Recipients More Than 1 Year After Transplantation
VariableUnivariate AnalysisMultivariate Analysis
Metabolic Syndrome PresentMetabolic Syndrome AbsentP ValueB Coefficient/Standard ErrorP ValueOR (95% CI)
  1. a

    The data are presented as means and standard deviations.

Age (years)a59.6 ± 9.653.5 ± 9.90.0010.055/0.1290.0091.056 (1.014–1.101)
Pretransplant diabetes (%)28.37.40.0031.446/0.6040.024.246 (1.300–13.864)
METsa3.4 ± 1.64.3 ± 1.4<0.001−0.371/0.1290.0040.690 (0.536–0.887)
Maintenance immunosuppression: tacrolimus/cyclosporine A (n/n)70/2549/60.030.360/0.3760.341.433 (0.686–2.994)
Time after transplantation (months)a82.0 ± 49.884.0 ± 63.20.84   
Sex: female (%)42.454.40.15   
Ethnicity: non-Hispanic white/African American/Hispanic/other (n/n/n/n)40/21/31/728/15/14/00.23   
Indication: NAFLD (%)7.110.50.45   
Indication: Alcohol (%)26.321.10.47   
Indication: HCV (%)44.436.80.35   
Presence of HCC (%)19.418.50.90   
Treatment for acute cellular rejection (%)13.316.40.60   
Chronic prednisone use (%)17.217.50.95   
Exercise duration per week (minutes)a80.4 ± 116.8119.9 ± 169.30.31   
Table 4. Exercise Activity and MSC in Recipients More Than 1 Year After Transplantation
Metabolic FeatureMETsP ValueExercise per Week (minutes)P Value
MSC PresentMSC AbsentMSC PresentMSC Absent
Impaired fasting glucose3.5 ± 1.64.0 ± 1.70.0999.3 ± 137.988.5 ± 141.60.25
Elevated blood pressure3.6 ± 1.74.1 ± 1.40.1687.7 ± 131.1135.9 ± 176.80.16
Hypertriglyceridemia3.5 ± 1.63.9 ± 1.60.1170.9 ± 109.1114.3 ± 157.40.06
Decreased HDL3.3 ± 1.64.1 ± 1.60.00469.3 ± 109.7121.0 ± 160.50.03
Increased waist circumference3.3 ± 1.64.2 ± 1.60.0375.6 ± 100.9115.7 ± 169.40.38

A multivariate analysis was performed to further assess the relationships between age, pretransplant diabetes, immunosuppression, and METs and metabolic syndrome. Older age, pretransplant diabetes, and higher METs remained associated with metabolic syndrome in the cohort of patients more than 1 year after transplantation (Table 3).

DISCUSSION

Metabolic complications are an important long-term health concern for transplant recipients, and the prevalence of metabolic syndrome after transplantation is much higher than that in the general population.[1-4, 15] Data from the current, single-center study show that the rate of metabolic syndrome was high during the first year after transplantation (43.8%) and was significantly greater thereafter (63.5%). Liver transplant recipients were found to have higher standard prevalence ratios of both diabetes and hypertension, whereas the rates of obesity and hyperlipidemia were similar to those in the general population.[16] Calcineurin inhibitors contribute to the higher risk for diabetes and hypertension[17] but are generally needed to maintain graft function. Therefore, lifestyle modification represents the most pragmatic approach to achieving risk reduction for metabolic syndrome.

What is new and particularly important in the present study is the data regarding exercise activity and the relationship between exercise and metabolic disease. The most discouraging findings are that more than one-half of the recipients reported no dedicated exercise-related physical activity, and only 24% performed the recommended amount of more than 150 minutes per week. This is substantially less than the 49% of the United States population that reported meeting aerobic activity guidelines in 2011.[18] An additional novel finding is the significantly increased rate of metabolic syndrome in recipients more than 1 year after transplantation. Similarly to other published data,[1, 3, 4] when the absolute number of days after transplantation was evaluated, the time from transplantation was not associated with metabolic syndrome. The lower rate of metabolic syndrome within the first year is an important observation and suggests that an intervention successful at reducing rates of metabolic syndrome in liver transplant recipients should be considered early after surgery.

The beneficial associations observed between exercise and lower rates of metabolic abnormalities are quite encouraging. Exercise intensity seemed to be more important than duration, and the effect was stronger in patients who were more than 1 year from transplantation. It is possible that exercise intensity was more accurately reported than duration and could make a better screening tool for assessing physical activity and the risk of metabolic syndrome in a clinical setting. An alternative explanation is that exercise intensity provided a better gauge of overall health and fitness. Overall, these findings suggest that a structured exercise program might help to prevent or reduce metabolic abnormalities in liver transplant recipients.

To date, there are limited data on behavioral interventions for reducing weight gain and metabolic syndrome after transplantation despite an abundance of data showing that the majority of patients gain weight within the first year after transplantation.[19, 20] In one report,[7] structured exercise programs were shown to increase fitness for at least the first 6 months after transplantation. A randomized intervention study showed that a diet and exercise program was associated with an overall improvement in exercise capacity and a trend toward less fat mass when adherent patients were compared to nonadherent ones.[21] However, adherence rates were low, and the impact of exercise on metabolic syndrome was not evaluated. These data suggest that an optimally effective posttransplant intervention would require a multidisciplinary team approach, including dietary, exercise, and behavioral interventions with frequent prompting and monitoring of adherence.

The current study has limitations. The study population was a convenience sample, and the data were collected in a cross-sectional fashion. In addition, the population was heterogeneous with a wide range of times after transplantation. Causality could not be determined from the results, and the impact of a new diagnosis of a metabolic complication or physician counseling, which may have occurred before data collection, could not be assessed. The study relied on self-reporting of exercise intensity, and this has been shown to overestimate actual intensity in obese populations.[22] Other studies have found cryptogenic cirrhosis and HCV cirrhosis to be risk factors for posttransplant metabolic syndrome,[3, 4] and these associations were lacking in the present analysis. The current study classified patients according to the main indication for transplantation and, therefore, did not assess for the coexistence of NAFLD in persons with other causes of liver disease; this potentially dampened associations. Strengths of the study include the large sample size, the similar proportions of men and women, the ethnic diversity, and the ranges of ages and times from transplantation.

Metabolic syndrome remains a major problem for liver transplant recipients. The prevalence of metabolic syndrome in a population of transplant recipients in the United States was found to be quite high, and this result is consistent with other studies in North American populations.[4] The early onset of metabolic syndrome in transplant recipients indicates the importance of early intervention during the first year after surgery. The relative benefits of exercise duration and exercise intensity require further study. Because the prevalence of metabolic syndrome is reaching epidemic levels in transplant recipients, an effective intervention is greatly needed.

Ancillary