Metabolic syndrome in liver transplantation: Relation to etiology and immunosuppression

ORIGINAL ARTICLE

Metabolic Syndrome in Liver Transplantation:Relation to Etiology and ImmunosuppressionGiampaolo Bianchi, Giulio Marchesini, Rebecca Marzocchi, Antonio D. Pinna, and Marco ZoliDipartimento di Medicina Interna and Centro Trapianti di Fegato e Multiorgano, Alma Mater Studiorum,Universita di Bologna, Bologna, Italy

Excessive weight gain, hypertension, hyperlipidemia, and diabetes are frequently observed in patients having undergone livertransplantation (LTx). These alterations are probably multifactorial in origin, and cluster to generate a metabolic syndrome(MS), increasing the risk of cardiovascular events. We assessed the prevalence of MS (National Cholesterol EducationProgram-Adult Treatment Panel III criteria) in 296 LTx patients in the course of regular follow-up, at least 6 months aftertransplantation (median, 38 months). Several pre-LTx and post-LTx data were collected to identify the factors associated withthe presence of MS. In a subset of 99 patients, insulin resistance was measured by the homeostasis model assessment. Highblood pressure was present in 53% of cases, hyperlipidemia in 51%, high glucose in 37%, and enlarged waist circumferencein 32%. Overall, MS (defined as 3 or more of the above features) was present in 44.5% of cases. Insulin resistance(homeostasis model assessment � 2.7) was observed in 41% of cases. Hypertension and hyperlipidemia were more frequentin subjects on cyclosporine than in tacrolimus-treated cases, whereas the type of immunosuppressive drug had no effect onthe prevalence of diabetes, enlarged waist, and MS. In a logistic regression analysis, only pre-LTx body mass index (odds ratio,1.20), body mass index increase (odds ratio, 1.18), and pre-LTx diabetes (odds ratio, 2.36) predicted MS; age, gender, etiologyof liver disease, time from LTx, type of immunosuppressive drug, and previous hepatocellular carcinoma were removed fromthe model. Disorders related to MS are frequent in LTx patients, and are related to both pre-LTx conditions and to weight gain.Weight control is mandatory in LTx patients to prevent risk factors of premature atherosclerosis. Liver Transpl 14:1648-1654, 2008. © 2008 AASLD.

Received March 25, 2008; accepted June 9, 2008.

The survival and quality of life of patients after livertransplantation (LTx) has markedly improved duringthe last decades, but the survival curve has now flat-tened.1 The return to normal daily life and home andworking activities, together with normalization of thehypermetabolic state of advanced liver failure2 and freefood intake, are accompanied by a progressive weightgain.3 In turn, this heralds the development of hyper-tension, changes in blood lipid profile with hyperlipid-emia, and altered glucose regulation.4,5 The origin ofthese alterations, apparently related to insulin resis-

tance (IR) and characterizing the metabolic syndrome(MS), is under debate.

In a very few cases, the occurrence of posttransplan-tation diabetes mellitus (PTDM) or full-blown post-transplantation metabolic syndrome (PTMS) might bethe effect of recurrent liver disease, caused by etiologicfactors also involved in the original liver cell failure,6 vianonalcoholic steatohepatitis (NASH) and cryptogeniccirrhosis,7 and encouraged by unhealthy lifestyles. Inthe vast majority of cases PTDM and PTMS might belifestyle-related and induced by post-LTx immunosup-

Abbreviations: ATPIII, National Cholesterol Education Program-Adult Treatment Panel III; BMI, body mass index; CI, confidenceinterval; DM, diabetes mellitus; F, female; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HDL,high-density lipoprotein; HOMA, homeostasis model assessment; IR, insulin resistance; LTx, liver transplantation; M, male; MS,metabolic syndrome; NASH, nonalcoholic steatohepatitis; PTDM, posttransplantation diabetes mellitus; PTMS, posttransplantationmetabolic syndrome.Address reprint requests to Giampaolo Bianchi, M.D., Dipartimento di Medicina Interna, Cardioangiologia, Epatologia Universita di Bologna,Policlinico S. Orsola-Malpighi, Via Massarenti 9, I-40138 Bologna, Italy. Telephone: 39-0516364825; FAX: 39-0516363803;E-mail: [email protected]

DOI 10.1002/lt.21588Published online in Wiley InterScience (www.interscience.wiley.com).

LIVER TRANSPLANTATION 14:1648-1654, 2008

© 2008 American Association for the Study of Liver Diseases.

pression.8 In keeping with this hypothesis and the met-abolic effects of calcineurin inhibitors,9 these featuresappear independently of etiology (viral, cholestatic, au-toimmune, secondary to iron or copper overload, as wellas cryptogenic cirrhosis) and even in subjects submit-ted to LTx for noncirrhotic liver disease.

Whatever the origin, the presence of the PTMS hasbeen associated with an increased prevalence of cardio-vascular disorders, which becomes a prevailing causeof morbidity/mortality in long-term LTx survivors inmany outcome studies10 and represents a relevant clin-ical issue,11 also favoring graft loss.12

However, only a few studies have systematically re-ported data on the markers of MS before and aftertransplantation in large LTx series, and on risk factorsfor the development of metabolic alterations. We carriedout a cross-sectional study in a large cohort of patientshaving received transplants, including a systematicanalysis of their pre-LTx clinical and biochemical con-dition, to verify the relative role of immunosuppression,etiology, and underlying metabolic factors in the devel-opment of PTMS.

PATIENTS AND METHODS

Patients

During the period of May 1, 2004 to December 21,2006, we retrospectively analyzed the data of 296 con-secutive patients (196 males, 100 females) who hadreceived a liver transplant from a cadaver donor 6 ormore months earlier, in the Transplant Unit of the S.Orsola-Malpighi Hospital, and seen as outpatients dur-ing regular follow-up visits. The standard immunosup-pressive therapy consisted of a calcineurin inhibitor. Allpatients had a piggy-back allograft implantation andreceived identical intraoperative and postoperativecare. Nearly all subjects received a calcineurin inhibitordrug (191 cyclosporine, 105 tacrolimus), associatedwith tapered corticosteroids. A small group of subjectson sirolimus (4 subjects) was not considered.

The majority of cases (over 40%) had undergonetransplantation because of liver failure caused by thehepatitis C virus (HCV); over 20% had hepatitis B virus(HBV)-related liver disease; and another 20% had liverfailure caused by a variety of genetic or biliary condi-tions, and were merged into a miscellaneous group forstatistical analyses (Table 1). This group also included25 cases with biliary cirrhosis (either primary biliarycirrhosis or primary sclerosing cholangitis) and 10cases with cryptogenic cirrhosis, potentially an expres-sion of advanced nonalcoholic fatty liver disease. Thedata for these subgroups will be presented in the detailswhenever relevant.

In a few cases, liver disease had a multifactorial ori-gin. Four cases in which HCV was associated with ex-cessive alcohol drinking were included in the HCVgroup, and similarly, six cases with HBV and excessivealcohol drinking were included in the HBV group, giventhat in all cases stopping alcohol abuse was a prereq-uisite for entering the LTx waiting list. Finally, 14 cases

with combined HBV and HCV infection were included inthe HCV group, given that the metabolic impact of HCVinfection is much higher than that ascribed to HBV.

After LTx, methylprednisolone had been gradually ta-pered down and usually discontinued after 6 months.Only 9 subjects, having undergone transplantation forliver disease of autoimmune origin or with other clinicalmarkers of autoimmune disorders, were continuingmethylprednisolone at low doses (2.5 to 5 mg/day) atthe time of the study.

During a preplanned outpatient follow-up visit at theLiver Transplantation Center of Azienda Policlinico S.Orsola-Malpighi, consecutive patients had an extendedclinical examination, including anthropometric assess-ment and registration of height and weight.

Patients were also interviewed about their weight his-tory, including weight at the age of 20 years and detailsof pre-LTx body weight, and weight at time of inclusionon the transplantation list (registered in the database),to define the time of onset of weight changes and/oroverweight. We also recorded from the database pre-LTx data, including etiology of liver disease, presence/absence of hepatocellular carcinoma, and pre-LTx dia-betes mellitus. Finally, all subjects had a thoroughassessment of clinical and biochemical data reflectingthe features of the MS. All these data are reported inTable 1.

Anthropometric Assessment

Waist circumference was measured at the umbilicuswith the patient standing, as an index of central fataccumulation. Body mass index (BMI) was calculatedas the ratio between patient weight (in kilograms) andthe square of patient height (in meters).

Laboratory Assays

Fasting venous blood samples were collected and as-sayed for lipid profile (total cholesterol, high-densitylipoprotein (HDL)-cholesterol, and triglycerides), glu-cose, and creatinine on the same day of blood sampling.In an unselected, consecutive group of 99 subjects,insulin was also measured in aliquots of serum frozenat �80°C with an electrochemoluminescence immuno-assay (ECLIA; Insulin Elecsys Roche Diagnostics), withan analytical sensitivity of 0.2 mU/L.

Insulin Sensitivity and IR

An estimate of IR was calculated using homeostasismodel assessment (HOMA) according to the formula:HOMA-IR � fasting insulin (mU/mL) � fasting glucose(mmol/L), divided by 22.5.13 The cutoff for the diagno-sis of IR was set at a value of 2.7, which was previouslydetermined in our laboratory.14

Assessment of the MS

We investigated the presence of PTMS according to theproposal of the National Cholesterol Education ProgramAdult Treatment Panel III (ATPIII),15 which was revised in

METABOLIC SYNDROME IN LIVER TRANSPLANTATION 1649

LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases

accordance with the new standard of normal fasting glu-cose.16 According to this proposal, the presence of MSmay be diagnosed in subjects having at least 3 of thefollowing 5 criteria: (1) waist circumference �102 cm inmen and �88 cm in women; (2) blood glucose �100

mg/dL (5.6 mmol/L) or treated for diabetes; (3) triglycer-ides �150 mg/dL (1.7 mmol/L) or treated with fibrates;(4) HDL-cholesterol �40 mg/dL (1.0 mmol/L) in malesand �50 mg/dL (1.3 mmol/L) in females; and (5) bloodpressure �130/85 mm Hg or treated for hypertension.

TABLE 1. Clinical and Laboratory Data of the Subjects Enrolled in the Study

All (n � 296)

Cyclosporine-Treated

(n � 191) Tacrolimus-Treated (n � 105) P

Age (years) 56 (13) 57 (13) 55 (13) 0.070Gender (M/F) 196/100 133/58 63/42 0.094Time from LTx

(months)38 (55) 51 (65) 26 (31) �0.001

Etiology of liver disease 0.506Alcohol-related (%) 14.4 (10.7–18.7) 15.8 (11.1–21.4) 12.5 (7.1–19.7)HBV-related (%) 22.5 (18.0–27.4) 23.2 (17.5–29.4) 21.2 (14.0–29.4)HCV-related (%) 43.0 (37.3–48.5) 43.2 (36.1–50.0) 41.3 (31.9–50.5)Miscellaneous (%) 20.1 (15.8–24.9) 17.9 (12.9–23.7) 25.0 (17.3–33.6)

Previous HCC (%) 19.1 (14.9–23.8) 21.6 (16.1–27.7) 11.5 (6.4–18.5) 0.047Current alcohol

drinkers (%)*1.3 (0.4–3.2) 2.1 (0.7–4.9) — 0.172

Current smokers (%) 23.8 (17.8–30.4) 23.4 (16.0–31.7) 21.0 (12.1–31.7) 0.719Diabetes (%) 37.8 (32.3–43.3) 34.2 (27.6–40.9) 44.2 (34.6–53.3) 0.117Pre-LTx diabetes (%) 9.8 (6.8–13.5) 11.1 (7.1–16.0) 7.7 (3.6–16.9) 0.472Post-LTx diabetes (%) 29.4 (24.3–34.6) 25.3 (19.4–31.6) 36.5 (27.5–45.6) 0.058HOMA (%)† 3.09 [2.30] 2.74 [1.91] 3.61 [2.73] 0.104HOMA � 2.7 (%)† 41.4 (31.5–51.3) 37.3 (24.6–50.0) 47.5 (31.3–63.7) 0.311BMI (kg/m2) 25.9 [4.1] 26.4 [4.0] 24.9 [4.2] 0.003BMI at 20 years old

(kg/m2)23.3 [4.0] 23.3 [4.1] 23.0 [3.4] 0.577

Pre-LTx BMI (kg/m2) 24.7 [3.6] 25.2 [3.6] 23.9 [3.5] 0.005Pre-LTx weight class,

normal/overweight/obesity (%)

42/43/15 37/45/18 51/41/8 0.015

BMI at 20 years old �30 kg/m2 (%)‡

8.1 (4.3–13.5) 7.1 (2.9–13.8) 8.3 (2.7–18.2) 0.789

Dyslipidemia 50.7 (44.8–56.2) 56.8 (49.5–63.4) 37.5 (28.4–46.6) 0.003High waist

circumference(M: �102 cm;F: �88) (%)

32.4 (27.2–37.8) 36.3 (29.6–43.1) 26.9 (18.9–35.6) 0.132

High blood pressure(�130/85 mm Hg)(%)

52.7 (46.8–58.2) 58.9 (51.6–65.4) 41.2 (31.7–50.4) 0.005

High blood glucose(�100 mg/dL) (%)

59.8 (54.0–65.1) 57.9 (50.5–64.4) 62.7 (52.6–71.1) 0.408

High triglycerides(�150 mg/dL) (%)

36.8 (31.4–42.3) 40.5 (33.5–47.4) 29.4 (21.0–38.4) 0.060

Low HDL–cholesterol(M: �40 mg/dL; F:�50) (%)

50.0 (44.2–55.5) 50.5 (43.2–57.3) 49.0 (39.1–58.1) 0.902

Number of MS criteria 0.2410–2 (%) 55.4 (49.5–60.8) 52.1 (44.8–58.8) 60.8 (50.6–69.2)3 (%) 26.0 (21.2–31.1) 26.3 (20.3–32.7) 25.5 (17.6–34.2)4–5 (%) 18.6 (14.4–23.2) 21.6 (16.1–27.7) 13.7 (8.0–21.2)

NOTE: Data are reported as the mean (standard deviation), median (interquartile range), or percent prevalence (95% CI) of theobserved condition.Abbreviations: F, female; HCC, hepatocellular carcinoma; M, male.*Data were available for 107 cyclosporine-treated and 62 tacrolimus-treated cases.†Data were available for 59 cyclosporine-treated and 40 tacrolimus-treated cases.‡Data were available for 85 cyclosporine-treated and 48 tacrolimus-treated cases.

1650 BIANCHI ET AL.


Statistical Analysis

Clinical and laboratory data were entered into an elec-tronic spreadsheet. Statistical analysis was performedwith the statistical package SPSS for Windows version8.0 (SPSS Inc., Chicago, IL). Data in the text and intables are presented as means � standard deviation oras prevalence [95% confidence intervals (CIs)]. Differ-ences among the groups were checked for significanceby means of analysis of variance or Mann-Whitney Utest. The chi-square test, or Fischer’s exact test whennecessary, were used to test for significance of the prev-alence of clinical, laboratory, and instrumental abnor-malities. Possible correlations among variables weretested for significance with multiple regression or step-wise logistic regression analysis. In the assessment offactors associated with PTMS, the following variableswere tested: age, etiology, gender, time from LTx, pre-LTx diabetes, pre-LTx BMI, 20-year BMI, and type ofimmunosuppression. P values � 0.05 (2-tailed analy-ses) were considered statistically significant.

RESULTS

The final data set included 296 LTx subjects (196 malesand 100 females), aged 18 to 77 years (median, 56);their pertinent data are reported in Table 1. The largemajority of cases had undergone transplantation forcirrhosis of viral etiology (HBV, HCV, or combined) from7 months to over 17 years before the study.

The median age was different according to the etiol-ogy of liver disease (Kruskal-Wallis test: 29.14; P �0.001). High blood pressure was present in 156 sub-jects (52.7%), hyperlipidemia in 150 (50.7%), diabetesin 112 (37.8%), and enlarged waist circumference in 96(32.4%). Finally, 170 cases (57.4%) were overweight orobese at the time of survey.

The ATPIII diagnostic criteria of MS were fulfilled in132 subjects (Table 1). Elevated BMI and hypertensionwere significantly more frequent in LTx subjects oncyclosporine as an immunosuppressive drug, whereasthe overall prevalence of diabetes was nonsignificantlylower. However, the development of diabetes after LTxwas more common in subjects on tacrolimus.

No significant differences were observed in the prev-alence of PTMS in relation to immunosuppressive treat-

ment (cyclosporine: 47.6%; tacrolimus: 39.0%; P �0.193).

No differences in the prevalence of PTMS (�2 � 1.73;P � 0.631) and its individual features were observed inrelation to the etiology of liver disease, with the notableexception of high blood glucose levels, which were morecommon in subjects with cirrhosis of alcoholic origin(�2 � 9.94; P � 0.019) (Table 2). However, the preva-lence of diabetes (�2 � 6.55; P � 0.088) and the time ofdiabetes onset in cases with PTDM (�2 � 10.31; P �0.112) were also similar between groups. Subjects withPTMS were not older than subjects without MS (Mann-Whitney U test: P � 0.361). Subjects belonging to themiscellaneous etiology group were moderately younger.This group was very heterogeneous, and subjects withbiliary cirrhosis were characterized by the lower preva-lence of pre-LTx diabetes mellitus (DM) (3.8%) amonggroups. The prevalence of diabetes in cryptogenic cir-rhosis was only 10%.

With very few exceptions (9 cases), the subjects whowere overweight or obese in the pre-LTx phase were alsooverweight or obese in the post-LTx assessment, but 68more cases who had normal BMI before LTx became over-weight in the post-LTx phase (Table 3), with no differencesin relation to etiology of liver disease. Obesity (BMI �30kg/m2) was present in 43 cases (15%). In addition, dia-betes, which was rare in the pre-LTx phase, increased by300% in the post-LTx survey, particularly in subjects withliver disease of alcoholic and HCV etiology. Only 4 caseswith pre-LTx diabetes had normal glucose regulation inthe post-LTx survey (Table 4).

In a forward-looking logistic regression to define thevariable/variables independently associated withPTMS, the model identified pre-LTx BMI [odds ratio(OR), 1.20; 95% CI, 1.12-1.30; P � 0.001], changes inBMI (OR, 1.18; 95% CI, 1.09-1.30; P � 0.001), and thepresence of pre-LTx diabetes (OR, 2.36; 95% CI, 1.01-5.56; P � 0.048) as the sole significant variables. Age,gender, etiology of liver disease, time from LTx, type ofimmunosuppressive drug, and previous hepatocellularcarcinoma were rejected from the model.

Fasting insulin was available in 99 cases. IR, as as-sessed by HOMA, was observed in 41 subjects (41.4%),with no differences in relation to the type of immuno-suppressive drug.

TABLE 2. Age, Time Since Transplantation, and Prevalence of the Metabolic Syndrome and Its Individual Features

in Relation to the Etiology of Liver Disease

Alcoholic

(n � 43)

HBV-Related

(n � 67)

HCV-Related

(n � 126)

Others

(n � 60) P

Age (years) 53.7 � 7.8 52.5 � 8.3 57.5 � 7.8 48.7 � 13.6 �0.001Time from LTx (months) 43.0 � 38.0 50.6 � 37.4 46.0 � 36.2 61.4 � 50.0 0.423Enlarged waist (%) 41.9 (27.4–55.6) 32.8 (22.1–44.0) 29.4 (21.8–37.4) 31.7 (20.6–43.4) 0.339High glucose (%) 69.8 (53.7–80.5) 47.8 (35.6–58.8) 66.7 (57.6–73.9) 51.7 (38.5–63.0) 0.019High blood pressure (%) 55.8 (40.0–68.5) 53.7 (41.2–54.4) 54.0 (44.9–62.0) 46.7 (33.9–58.3) 0.764High triglycerides (%) 39.5 (25.4–53.3) 34.3 (23.4–45.5) 34.1 (26.1–42.4) 43.3 (30.8–55.1) 0.615Low HDL–cholesterol (%) 51.2 (35.7–64.3) 58.2 (45.5–68.6) 50.8 (41.8–59.0) 38.3 (26.4–50.2) 0.163Metabolic syndrome (%) 51.2 (35.7–64.3) 44.8 (32.8–55.9) 45.2 (36.4–53.5) 38.3 (26.4–50.2) 0.631



DISCUSSION

The clinical and biochemical features of MS are highlyprevalent in LTx patients, independently of etiology ofpre-LTx liver disease and type of immunosuppressivetreatment. The metabolic changes associated with MSand IR promote the development of atherosclerosis andrepresent risk factors for premature coronary heart dis-ease, which was not present before LTx, and an earlyrecognition of MS is important for the prevention ofcardiovascular events.

The prevalence of posttransplantation MS is a matter ofconcern in the transplantation area: the presence of MS isone more risk factor to be considered in the pretransplan-tation workup,11 but after transplantation the risk ap-pears to be particularly elevated, particularly in subjectshaving undergone transplantation for HCV-related dis-ease,17,18 increasing the risk by a factor of 2 or more.19

The relationship between HCV infection and IR/diabetes is well-known in liver disease patients,20

and has been related to different mechanism(s) in-volving either a direct effect of the virus on insulinsignaling21 or simply favoring the phenotypic expres-sion of classical diabetes risk factors.22 Whatever themechanism, the presence of PTDM remains a nega-tive prognostic factor in the long term,23 although it isrespondent to oral hypoglycemic agents.4 It has alsobeen demonstrated that PTMS and nonalcoholic fattyliver disease in HCV patients is associated with a highrisk of fibrosis progression.24 By contrast, the prev-alence of PTMS was very low in cirrhosis of biliaryorigin, in keeping with the late involvement of hepa-tocytes, which are directly responsible for glucosemetabolism.

The prevalence of PTMS was ascertained in the well-

TABLE 3. Prevalence of Overweight/Obese Patients According to the Etiology of Liver Disease

Group

Pre-LTx Post-LTx Pre-LTx/Post-LTx Changes

Pre-LTx/

Post-LTx

Difference,

P

Normal

Weight

(n � 185)

Overweight/

Obese

(n � 111)

Normal

Weight

(n � 126)

Overweight/

Obese

(n � 170)

Became

Overweight/

Obese

(n � 68)*

Regressed to

Normal

Weight

(n � 9)†

All cases (n � 296) (%) 185 (62.5) 111 (37.5) 126 (42.6) 170 (57.4) 68 (36.8) 9 (8.1) � 0.001Alcohol (n � 43) (%) 25 (58.1) 18 (41.9) 15 (34.9) 28 (65.1) 11 (44.0) 1 (5.6) 0.051HBV (n � 67) (%) 39 (58.2) 28 (41.8) 25 (37.3) 42 (62.7) 14 (35.9) 0 0.025HCV (n � 126) (%) 75 (59.5) 51 (40.5) 48 (38.1) 78 (61.9) 33 (44.0) 6 (11.8) 0.001Other (n � 60) (%) 46 (76.7) 14 (23.3) 38 (63.3) 22 (36.7) 10 (21.7) 2 (14.3) 0.163P across groups Chi-square � 6.489;

P � 0.090Chi-square � 13.401;

P � 0.003Chi-square

� 2.414;P � 0.491

Chi-square� 3.484;

P � 0.323

NOTE: Data are reported as the number (percent prevalence) of observed disorders.*Data are expressed as the percentage of pre-LTx normal weight.†Data are expressed as the percentage of pre-LTx overweight/obesity.

TABLE 4. Prevalence of Diabetes in Relation to the Etiology of Liver Disease

Group

Pre-LTx Post-LTx

Pre/Post-LTx Changes

Pre/Post

Difference,

P

No Diabetes

(n � 267)

Diabetes

(n � 29)

No Diabetes

(n � 184)

Diabetes

(n �

112)

Developed

Diabetes

(n � 87)*

Regressed

to No

Diabetes

(n � 4)†

All cases (n � 296) (%) 267 (90.2) 29 (9.8) 184 (62.2) 112 (37.8) 87 (32.5) 4 (14.3) �0.001Alcohol (n � 43) (%) 36 (83.7) 7 (16.3) 23 (53.5) 20 (46.5) 14 (38.9) 1 (14.3) 0.005HBV (n � 67) (%) 60 (89.6) 7 (10.4) 46 (68.7) 21 (31.3) 15 (25.0) 1 (14.3) 0.006HCV (n � 126 (%) 114 (90.5) 12 (9.5) 73 (57.9) 53 (42.1) 43 (34.1) 2 (18.2) �0.001Other (n � 60) (%) 57 (95.0) 3 (5.0) 42 (70.0) 18 (30.0) 15 (25.0) 0 �0.001P across groups � Chi-square � 3.649;

P � 0.302Chi-square � 5.101;

P � 0.165Chi-square �

5.092;P � 0.532

—

NOTE: Data are reported as the number (percent prevalence) of observed disorders.*Data are expressed as the percentage of pre-LTx no diabetes.†Data are expressed as the percentage of pre-LTx diabetes.

1652 BIANCHI ET AL.


accepted revision of the ATPIII criteria.16 More recently,the International Diabetes Federation proposed a dif-ferent classification having visceral obesity as a pivotalfeature.25 Using these criteria, the prevalence of PTMSdid not change remarkably (43.9% in the whole sampleversus 44.5% by ATPIII criteria). These figures are def-initely higher compared to reports in the Italian popu-lation in the age range between 45 and 65 years (around30%-35%),26 and is definitely surprising consideringthat the prevalence is expected to be extremely low inthe pre-LTx period. Data were not available for definingpre-LTx MS in our population, and any inference issubject to criticism.27 However, the figure should bevery low, considering the hypotension due to vasodila-tion and the low lipid levels of advanced liver failure.

PTDM and PTMS were significantly associated withclassical pre-LTx risk factors, namely diabetes andBMI, but no association was found with the type ofimmunosuppression. The glucocorticoid-sparing ac-tion of calcineurin inhibitors was originally welcomed inthe hope to reduce the risk of diabetes, but soon itbecame evident that both cyclosporine and tacrolimushad a diabetogenic activity, impairing insulin sensitiv-ity.28 Only later was it shown that calcineurin inhibi-tors also reduce insulin secretion,29 and that tacroli-mus was more diabetogenic than cyclosporine,30,31

mainly in the early post-LTx phase,32 because of itsmultiple sites of action.33 However, after controlling fortime from liver transplantation, the type of immuno-suppression did not qualify as prognostic factor forPTDM or PTMS. Once DM is established in the pre-transplantation period, LTx does not always produce acomplete clinical regression. Tueche34 reported a denovo incidence of DM in 26/108 (24%) transplant pa-tients at 6 months, whereas only 8/35 (23%) werecured of their pre-LTx DM. Similarly, in our long-termanalysis, we found that LTx cured pre-LTx DM in only14% of cases, in keeping with an established b-celldefect,2 and PTDM developed in 32.5% of cases. Thishighlights the need for careful surveillance and com-prehensive nutritional education to prevent DM andDM complications in the early and late post-LTx phase.

From a clinical perspective, the presence of PTMS isassociated with an increased risk of major vascularevents in the LTx population.5 Bianchi et al.35 reportedan association between high levels of homocysteine andarterial events in LTx patients, but their potential neg-ative effect in subjects with MS is doubtful.36-39 How-ever, in LTx the actual risk of cardiovascular eventsshould be lower than in the general population, consid-ering the a priori exclusion from transplant proceduresof all cases at high vascular risk.

In our series, very few subjects had undergone LTxbecause of metabolic liver disease progressing to end-stage cryptogenic cirrhosis, in contrast with series re-ported in other countries.40 Only 10 cases in the mis-cellaneous group could be ascribed to this etiology, andthe prevalence of PTDM and MS in this cohort was low(3 cases for both; 30%), but in 1 case diabetes wasalready present before transplantation. Patients under-going LTx for cirrhosis resulting from NASH may have

poorer outcomes, despite careful selection of candi-dates. It is very likely that many potential LTx candi-dates with NASH are excluded from surgery due tocomorbid conditions related to MS, thus explainingtheir low prevalence in countries with a high prevalenceof viral and alcoholic disease.41 Their clinical outcomesappear to be similar to those observed in other patientgroups42 or even better.43

This study has limitations. The retrospective collec-tion of a few data may have introduced a recall bias, butmost data were recorded in the original database, andpre-LTx weight or the presence of diseases could easilybe remembered. With this caveat, we conclude that therisk factors of PTMS include pre-LTx conditions, as wellas post-LTx weight gain and post-LTx obesity; the latterwas reported to account for both cardiovascular riskand risk of graft loss in different settings.3,5,12,44 Be-havioral treatment toward a healthy lifestyle and pre-venting obesity is mandatory to reduce the burden ofcardiovascular events in LTx patients, and there is nowevidence that this is feasible and effective.45 This retro-spective analysis is the basis for follow-up and inter-vention in our post-LTx population.

REFERENCES

1. Futagawa Y, Terasaki PI, Waki K, Cai J, Gjertson DW. Noimprovement in long-term liver transplant graft survival inthe last decade: an analysis of the UNOS data. Am J Trans-plant 2006;6:1398-1406.

2. Perseghin G, Mazzaferro V, Sereni LP, Regalia E, BenediniS, Bazzigaluppi E, et al. Contribution of reduced insulinsensitivity and secretion to the pathogenesis of hepatoge-nous diabetes: effect of liver transplantation. Hepatology2000;31:694-703.

3. Richards J, Gunson B, Johnson J, Neuberger J. Weightgain and obesity after liver transplantation. Transpl Int2005;18:461-466.

4. Jindal RM. Posttransplant diabetes mellitus—a review.Transplantation 1994;58:1289-1298.

5. Laryea M, Watt KD, Molinari M, Walsh MJ, McAlister VC,Marotta PJ, et al. Metabolic syndrome in liver transplantrecipients: prevalence and association with major vascularevents. Liver Transpl 2007;13:1109-1114.

6. Angelico F, Del Ben M, Francioso S, Hurtova M, Battista S,Palmieri GP, et al. Recurrence of insulin resistant meta-bolic syndrome following liver transplantation. Eur J Gas-troenterol Hepatol 2003;15:99-102.

7. Ong J, Younossi ZM, Reddy V, Price LL, Gramlich T, Mayes J,et al. Cryptogenic cirrhosis and posttransplantation nonal-coholic fatty liver disease. Liver Transpl 2001;7:797-801.

8. Hjelmesaeth J, Asberg A, Muller F, Hartmann A, Jenssen T.New-onset posttransplantation diabetes mellitus: insulin re-sistance or insulinopenia? Curr Diab Rev 2005;1:1-10.

9. Marchetti P, Navalesi R. The metabolic effects of cyclo-sporin and tacrolimus. J Endocrinol Invest 2000;23:482-490.

10. Perkins JD. Influence of obesity with concurrent diseaseson liver transplant survival. Liver Transpl 2007;13:928-929.

11. Perkins JD. Metabolic syndrome: a new view of some fa-miliar transplant risks. Liver Transpl 2006;12:485-486.

12. Ducloux D, Kazory A, Simula-Faivre D, Chalopin JM. One-year post-transplant weight gain is a risk factor for graftloss. Am J Transplant 2005;5:2922-2928.

13. Matthews DR, Hosker JP, Rudenski AS, Naylor BA,



Treacher DF, Turner RC. Homeostasis model assessment:insulin resistance and beta-cell function from plasmafasting glucose and insulin concentrations in man. Diabe-tologia 1985;28:412-419.

14. Bugianesi E, Pagotto U, Manini R, Vanni E, Gastaldelli A,De Iasio R, et al. Plasma adiponectin in nonalcoholic fattyliver is related to hepatic insulin resistance and hepatic fatcontent, not to liver disease severity. J Clin EndocrinolMetab 2005;90:3498-3504.

15. Expert panel on detection evaluation and treatment ofhigh blood cholesterol in adults. Executive summary ofthe third report of the National Cholesterol Education Pro-gram (NCEP) expert panel on detection, evaluation, andtreatment of high blood cholesterol in adults (Adult Treat-ment Panel III). JAMA 2001;285:2486-2497.

16. Grundy SM, Cleeman JI, Daniels SR, Donato KA, EckelRH, Franklin BA, et al. Diagnosis and management of themetabolic syndrome: an American Heart Association/Na-tional Heart, Lung, and Blood Institute Scientific State-ment. Circulation 2005;112:2735-2752.

17. Baid S, Cosimi AB, Farrell ML, Schoenfeld DA, Feng S,Chung RT, et al. Posttransplant diabetes mellitus in livertransplant recipients: risk factors, temporal relationshipwith hepatitis C virus allograft hepatitis, and impact onmortality. Transplantation 2001;72:1066-1072.

18. Bigam DL, Pennington JJ, Carpentier A, Wanless IR,Hemming AW, Croxford R, et al. Hepatitis C-related cir-rhosis: a predictor of diabetes after liver transplantation.Hepatology 2000;32:87-90.

19. Khalili M, Lim JW, Bass N, Ascher NL, Roberts JP, Ter-rault NA. New onset diabetes mellitus after liver trans-plantation: the critical role of hepatitis C infection. LiverTranspl 2004;10:349-355.

20. Mehta SH, Brancati FL, Sulkowski MS, Strathdee SA,Szklo M, Thomas DL. Prevalence of type 2 diabetes melli-tus among persons with hepatitis C virus infection in theUnited States. Ann Intern Med 2000;133:592-599.

21. Shintani Y, Fujie H, Miyoshi H, Tsutsumi T, Tsukamoto K,Kimura S, et al. Hepatitis C virus infection and diabetes:direct involvement of the virus in the development of in-sulin resistance. Gastroenterology 2004;126:840-848.

22. Petit JM, Bour JB, Galland-Jos C, Minello A, Verges B,Guiguet M, et al. Risk factors for diabetes mellitus andearly insulin resistance in chronic hepatitis C. J Hepatol2001;35:279-283.

23. Montori VM, Basu A, Erwin PJ, Velosa JA, Gabriel SE,Kudva YC. Posttransplantation diabetes: a systematic re-view of the literature. Diabetes Care 2002;25:583-592.

24. Mindikoglu AL, Regev A, Casanova-Romero PY, BejaranoPA, Martinez EJ, Tzakis AG, et al. The impact of nonalco-holic fatty liver disease and the metabolic syndrome onprogression of fibrosis in patients with recurrent HCVafter liver transplantation. Transplant Proc 2006;38:1440-1444.

25. Alberti KG, Zimmet P, Shaw J. The metabolic syn-drome—a new worldwide definition. Lancet 2005;366:1059-1062.

26. Miccoli R, Bianchi C, Odoguardi L, Penno G, Caricato F,Giovannitti MG, et al. Prevalence of the metabolic syn-drome among Italian adults according to ATP III definition.Nutr Metab Cardiovasc Dis 2005;15:250-254.

27. Liangpunsakul S, Chalasani NP. Metabolic syndrome fol-lowing liver transplantation is associated with major vas-cular events. Liver Transpl 2007;13:1078-1079.

28. Menegazzo LA, Ursich MJ, Fukui RT, Rocha DM, Silva ME,Ianhez LE, et al. Mechanism of the diabetogenic action ofcyclosporin A. Horm Metab Res 1998;30:663-667.

29. Heisel O, Heisel R, Balshaw R, Keown P. New onset dia-betes mellitus in patients receiving calcineurin inhibitors:

a systematic review and meta-analysis. Am J Transplant2004;4:583-595.

30. Vincenti F, Tuncer M, Castagneto M, Klinger M, Friman S,Scheuermann EH, et al. Prospective, multicenter, random-ized trial to compare incidence of new-onset diabetes melli-tus and glucose metabolism in patients receiving cyclospor-ine microemulsion versus tacrolimus after de novo kidneytransplantation. Transplant Proc 2005;37:1001-1004.

31. van Duijnhoven EM, Christiaans MH, Boots JM, NiemanFH, Wolffenbuttel BH, van Hooff JP. Glucose metabolismin the first 3 years after renal transplantation in patientsreceiving tacrolimus versus cyclosporine-based immuno-suppression. J Am Soc Nephrol 2002;13:213-220.

32. David-Neto E, Lemos FC, Fadel LM, Agena F, Sato MY,Coccuza C, et al. The dynamics of glucose metabolismunder calcineurin inhibitors in the first year after renaltransplantation in nonobese patients. Transplantation2007;84:50-55.

33. Uchizono Y, Iwase M, Nakamura U, Sasaki N, Goto D, IidaM. Tacrolimus impairment of insulin secretion in isolatedrat islets occurs at multiple distal sites in stimulus-secre-tion coupling. Endocrinology 2004;145:2264-2272.

34. Tueche SG. Diabetes mellitus after liver transplant newetiologic clues and cornerstones for understanding.Transplant Proc 2003;35:1466-1468.

35. Bianchi G, Nicolino F, Passerini G, Grazi GL, Zappoli P,Graziani R, et al. Plasma total homocysteine and cardio-vascular risk in patients submitted to liver transplanta-tion. Liver Transpl 2006;12:105-111.

36. Bjorck J, Hellgren M, Rastam L, Lindblad U. Associationsbetween serum insulin and homocysteine in a Swedishpopulation-a potential link between the metabolic syn-drome and hyperhomocysteinemia: the Skaraborg project.Metabolism 2006;55:1007-1013.

37. Godsland IF, Elkeles RS, Feher MD, Nugara F, RubensMB, Richmond W, et al. Coronary calcification, homocys-teine, C-reactive protein and the metabolic syndrome inType 2 diabetes: the Prospective Evaluation of DiabeticIschaemic Heart Disease by Coronary Tomography (PRE-DICT) Study. Diabet Med 2006;23:1192-1200.

38. Hajer GR, van der Graaf Y, Olijhoek JK, Verhaar MC,Visseren FL. Levels of homocysteine are increased in met-abolic syndrome patients but are not associated with anincreased cardiovascular risk, in contrast to patientswithout the metabolic syndrome. Heart 2007;93:216-220.

39. Junyent M, Zambon D, Gilabert R, Nunez I, Cofan M, RosE. Carotid atherosclerosis and vascular age in the assess-ment of coronary heart disease risk beyond the Framing-ham Risk Score. Atherosclerosis 2008;196:803-809.

40. Burke A, Lucey MR. Non-alcoholic fatty liver disease, non-alcoholic steatohepatitis and orthotopic liver transplanta-tion. Am J Transplant 2004;4:686-693.

41. Fracanzani AL, Burdick L, Raselli S, Pedotti P, Grigore L,Santorelli G, et al. Carotid artery intima-media thickness innonalcoholic fatty liver disease. Am J Med 2008;121:72-78.

42. Contos MJ, Cales W, Sterling RK, Luketic VA, ShiffmanML, Mills AS, et al. Development of nonalcoholic fatty liverdisease after orthotopic liver transplantation for crypto-genic cirrhosis. Liver Transpl 2001;7:363-373.

43. Heneghan MA, Zolfino T, Muiesan P, Portmann BC, RelaM, Heaton ND, et al. An evaluation of long-term outcomesafter liver transplantation for cryptogenic cirrhosis. LiverTranspl 2003;9:921-928.

44. Munoz SJ, Elgenaidi H. Cardiovascular risk factors afterliver transplantation. Liver Transpl 2005;11:S52–S56.

45. Krasnoff JB, Vintro AQ, Ascher NL, Bass NM, Paul SM,Dodd MJ, et al. A randomized trial of exercise and dietarycounseling after liver transplantation. Am J Transplant2006;6:1896-1905.

1654 BIANCHI ET AL.


Documents

Metabolic syndrome in liver transplantation: Relation to etiology and immunosuppression