GA

YRYS

*KTMI

B(asdNihcsacvrl(AdNGRtarsm

WdllnbtcNe2

CLIN

ICA

L–LIVER

,PA

NCREA

S,A

ND

BILIA

RY

TRA

CT

GASTROENTEROLOGY 2007;132:938 –943

rowth Hormone Reverses Nonalcoholic Steatohepatitis in a Patient Withdult Growth Hormone Deficiency

UTAKA TAKAHASHI,* KEIJI IIDA,* KENTARO TAKAHASHI,* SHIRO YOSHIOKA,* HIDENORI FUKUOKA,*YOKO TAKENO,* MARI IMANAKA,* HITOSHI NISHIZAWA,* MICHIKO TAKAHASHI,* YASUSHI SEO,‡

OSHITAKE HAYASHI,§ TAKUMA KONDO,� YASUHIKO OKIMURA,¶ HIDESUKE KAJI,# RIKO KITAZAWA,**OHEI KITAZAWA,** and KAZUO CHIHARA*

Division of Endocrinology/Metabolism, Neurology, and Hematology/Oncology, Department of Clinical Molecular Medicine; ‡Division of Diabetes, Digestive, andidney Diseases, Department of Clinical Molecular Medicine; and §Division of Molecular Medicine & Medical Genetics, International Center for Medical Research andreatment, Department of Biomedical Informatics, Kobe University Graduate School of Medicine, Kobe; �Kondo Pediatric Clinic, Osaka; ¶Department of Basic Alliededicine, Kobe University School of Medicine, Kobe; #College of Nursing Art and Culture, Akashi; and **Division of Molecular Pathology, Department of Biomedical

nformatics, Kobe University Graduate School of Medicine, Kobe, Japan

wmNeesacfllg

dicmthcrmntawfwqpNr

dtah

ackground & Aims: Nonalcoholic steatohepatitisNASH) is an emerging progressive hepatic diseasend demonstrates steatosis, inflammation, and fibro-is. Insulin resistance is a common feature in theevelopment of NASH. Molecular pathogenesis ofASH consists of 2 steps: triglyceride accumulation

n hepatocytes with insulin resistance and an en-anced oxidative stress caused by reactive oxygen spe-ies. Interestingly, NASH demonstrates a strikingimilarity to the pathologic conditions observed indult growth hormone deficiency (AGHD). AGHD isharacterized by decreased lean body mass, increasedisceral adiposity, abnormal lipid profile, and insulinesistance. Moreover, liver dysfunctions with hyper-ipidemia and nonalcoholic fatty liver diseaseNAFLD) are frequently observed in patients withGHD, and it is accompanied by metabolic syn-rome. Methods: We studied a case diagnosed asASH with hyperlipidemia in AGHD. The effect ofH-replacement therapy on the patient was analyzed.esults: Six months of GH-replacement therapy in

he patient drastically ameliorated NASH and thebnormal lipid profile concomitant with a markededuction in oxidative stress. Conclusions: These re-ults suggest that GH plays an essential role in the

etabolic and redox regulation in the liver.

ith the increasing prevalence of obesity, diabetes,and metabolic syndrome, nonalcoholic fatty liver

isease (NAFLD) has become a common cause of chroniciver disease.1 NAFLD includes both nonalcoholic fattyiver and nonalcoholic steatohepatitis (NASH). The diag-osis is based on the histologic examination of liveriopsy specimens. NASH is characterized by, in additiono steatosis that is seen in NAFLD, mixed inflammatoryell infiltration, hepatocyte ballooning, and fibrosis.2

ASH is a serious condition because it progresses tond-stage liver disease. Obesity, metabolic syndrome, type

diabetes, and hyperlipidemia are frequently associated

ith NAFLD including NASH. Insulin resistance is theost common feature in the development of NAFLD/ASH.3 The cause of NASH appears multifactorial; how-

ver, it has been speculated that the molecular pathogen-sis of NASH consists of 2 steps: first, insulin resistancetatus with an accumulation of fat within hepatocytes,nd, second, mitochondrial reactive oxygen speciesauses lipid peroxidation, cytokine induction, and in-ammation.2 These characteristics show a striking simi-

arity with the pathologic conditions observed in adultrowth hormone deficiency (AGHD).

AGHD is an established clinical entity characterized byecreased lean body mass and bone mineral density,

ncreased visceral adiposity, abnormal lipid profile, de-reased muscle strength and exercise endurance, and di-inished quality of life.4 Recent studies have emphasized

he increased morbidity and mortality of patients withypopituitarism,5,6 which is closely related to GH defi-iency. In a cohort study, AGHD subjects without GH-eplacement therapy showed increased incidences of

yocardial infarction, cereberovascular diseases, malig-ancy, and death as compared with the general popula-ion.6 In contrast, the GH-treated subjects demonstratedn overall malignancy and mortality rate comparableith that of the general population. Moreover, liver dys-

unction with hyperlipidemia7 and NAFLD accompaniedith the metabolic syndrome manifestation were fre-uently observed in AGHD patients.8 In a previous re-ort, one of the AGHD patients was diagnosed withASH by liver biopsy specimens.8 Furthermore, it was

eported that patients with hypothalamic and pituitary

Abbreviations used in this paper: AGHD, adult growth hormoneeficiency; GH, growth hormone; hsCRP, high sensitive C-reactive pro-ein; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic ste-tohepatitis; 8OHdG, 8-hydroxydeoxyguanosine; rhGH, recombinantuman GH; ROS, reactive oxygen species.

© 2007 by the AGA Institute0016-5085/07/$32.00

doi:10.1053/j.gastro.2006.12.024

dpqadbdtst

iwbcrhiwosaltia(ysan

th1tVealtspgdutdclm1na

NNAtch

tewJa

T

GIAALL�ACLHTLLA

IHIO

Naistmpma

r

CLI

NIC

AL–

LIV

ER,

PA

NCREA

S,A

ND

BIL

IARY

TRA

CT

March 2007 GROWTH HORMONE AND NASH 939

ysfunction revealed risk of excessive weight gain, im-aired glucose tolerance, and dyslipidemia with subse-uent development of NAFLD/NASH with a high prev-lence of cirrhosis, increasing their risk for liver-relatedeath.9 Here, we describe a case of AGHD accompaniedy NASH and hyperlipidemia. GH-replacement therapyrastically reversed NASH and hyperlipidemia in the pa-ient. We also analyzed the oxidative stress states in theerum and liver because they were essentially related tohe pathogenesis of NASH and metabolic syndrome.

Case ReportA 31-year-old man was referred to our hospital for

nvestigation of hyperlipidemia and liver dysfunction. Heas delivered in a breech position and with asphyxia. Atirth, his body length and weight had been recorded as 47m and 2490 g, respectively. At 7.5 years of age, he waseferred for an evaluation of growth delay, and, at that time,is height was 104.8 cm (�3.0 SD below the mean accord-

ng to the growth chart for Japanese boys), and his weightas 17.6 kg (�1.7 SD). Routine laboratory analysis ruledut hematologic, liver, and renal diseases. The GH re-ponses to provocative tests with the use of insulin andrginine were extremely low (�0.5 ng/mL [undetectableevels]). He was diagnosed as growth hormone, adrenocor-icotropic hormone, thyroid-stimulating hormone, leutiniz-ng hormone, and follicle-stimulating hormone deficientnd subsequently treated with recombinant human GHrhGH), hydrocortisone, and thyroxine from the age of 7ears. At 13 years, gonadotropin-replacement therapy wastarted, and, at the age of 18 years, GH therapy was stopped,nd administration of hydrocortisone, thyroxine, and go-adotropin therapy was continued.The patient’s physical examination when he was referred

o our hospital at the age of 31 years revealed the following:eight, 165 cm; weight, 60 kg; BMI, 22.0; blood pressure,10/80 mm Hg. He had never consumed alcohol. Labora-ory findings demonstrated mild liver dysfunction with type

hyperlipidemia (Table 1, left); hypertriglyceridemia wasspecially dominant. The serum GH peak levels after insulinnd arginine tests were both �0.15 ng/mL (undetectableevels), respectively. Magnetic resonance imaging of the pi-uitary gland demonstrated a transection in the pituitarytalk, an atrophy of the anterior pituitary gland, and aresence of pseudoposterior pituitary. These findings sug-ested that the patient had an injury in the pituitary stalkuring the delivery in the breech position. An abdominalltrasonography revealed a typical bright liver, indicatinghe presence of fatty liver. No hepatitis virus markers wereetectable. Liver biopsy was performed to determine theause of liver dysfunction and hyperlipidemia. The histo-ogic analysis revealed marked steatosis (33.5%) with inflam-

atory infiltrates and hepatocyte ballooning (Figure 1A andB), and Masson trichrome staining demonstrated perisi-usoidal and pericellular fibrosis in acinar zone 3 (Figure 1C

nd 1D). These findings indicated a pathologic diagnosis of m

ASH. According to the classification by Brunt et al,10

ASH was classified as stage 1 and grade 1. To treat theGHD status and to determine whether GH-replacement

herapy reverses NASH, rhGH was administered daily ac-ording to the general protocol11 after informed consentad been obtained from the patient.

Materials and MethodsBiochemical Analysis and Hormone AssaysSerum glucose, insulin, liver function, high sensi-

ive C-reactive protein (hsCRP), cholesterol, and triglyc-ride levels were determined using standard methodsith the use of automated equipment (Hitachi, Tokyo,

apan) at Kobe University Hospital. Serum levels of GHnd insulin-like growth factor-I were determined by

able 1. Laboratory Studies

Baseline After GH treatment

lucose 93 83 mg/dL (60–110)nsulin 15 4 IU/mLST 88a 24 IU/L (13–31)LT 85a 19 IU/L (8–34)DH 309a 168 IU/L (115–217)AP 60 60 IU/L (35–76)GTP 76a 17 IU/L (9–57)LP 183 121 IU/L (103–321)hE 543a 414 IU/L (187–453)DL cholesterol 143a 144a mg/dL (70–139)DL 31a 42 mg/dL (40–60)G 712a 139 mg/dL (28–149)p(a) 3.8 8.7 mg/dL (0–30)-CAT 180.8a 75.6 IU/L (67.3–108.2)po AI 130 106 mg/dL (123–194)AII 38 27 mg/dL (25–44)B 129a 90 mg/dL (52–105)CII 13.3a 4.3 mg/dL (1.5–4.8)CIII 20.0a 7.6 mg/dL (4.6–11.1)E 11.6a 4.1 mg/dL (1.7–6.3)

GF-I 16.5a 147.0 ng/mL (67–318)OMA IR 3.4a 0.82 (�2.5)

nsulinogenic index 12.2a 3.2 (�0.4)ral glucose-tolerance testInsulin level, IU/mL

Baseline 15 4Peak 234 196120 min 17 11

Glucose level, mg/dLBaseline 93 83Peak 111 142120 min 72 92

OTE. The insulin resistance index, index of homeostasis modelssessment, is calculated as (the fasting glucose level � the fastingnsulin level)/405. (The normal range is less than 2.5.) The insulinecretion index, insulinogenic index, is calculated as the ratio be-ween increases in insulin plasma concentrations during the first 30inutes following an oral glucose load and increases in glucose (G)lasma concentrations over the same period (�I30/�G30). The nor-al range is more than 0.4.

Indicates abnormal values and the parentheses indicate the normalange.

eans of an immunoenzymetric assay (TOSO, Tokyo,

JilhmT(

pi

csG

iyJ(ds

dgtadcer4rtspal

Fliimpscim

Fdfs

CLIN

ICA

L–LIVER

,PA

NCREA

S,A

ND

BILIA

RY

TRA

CT

940 TAKAHASHI ET AL GASTROENTEROLOGY Vol. 132, No. 3

apan) and an immunoradiometric assay (Daiichi Radio-sotope Laboratories, Tokyo, Japan), respectively. Serumevels of tumor necrosis factor (TNF)-� and urine-8-ydroxydeoxyguanosine (urine-8OHdG) were deter-ined by a chemiluminescent immunoassay (Funakoshi,okyo, Japan) and an enzyme-linked immunoassay

NihonYushi, Tokyo, Japan), respectively.

Histologic AnalysisWritten informed consent was obtained from the

atient for the liver biopsies. Biopsy specimens were fixed

igure 1. Pretreatment liver biopsy samples (A–D) and posttreatmentiver biopsy samples (E–H). Before treatment, typical NASH character-stics were observed. These include marked steatosis (33.5%) with annflammatory infiltrate and hepatocyte ballooning (H&E staining; original

agnification, A and E �100; B and F �400) and perisinusoidal andericellular fibrosis in acinar zone 3 demonstrated by Masson trichrometaining (original magnification, C and G �100; F and H �400). Inontrast, steatosis was reduced to 7.2% after treatment, inflammatory

nfiltrate and hepatocyte ballooning disappeared, and fibrosis showed aarked decrease.

n 10% neutral-buffered formalin, embedded in paraffin, l

ut into 4-�m-thick sections, and evaluated by H&Etaining and Masson trichrome staining before and afterH-replacement therapy.

ImmunohistochemistryThe sections were prepared as described above and

ncubated with monoclonal anti-8-hydroxy-2=-deox-guanosine/8-hydroxyguanosine (8OHdG/8OHG; 1:100;aICA, Japan) and antinitrotyrosine monoclonal antibodyMAB5404; 1:10; Chemicon), visualized by using stan-ard immunohistochemical methods, and counter-tained by hematoxylin.

ResultsRhGH was subcutaneously administered at a daily

ose of 3.3 �g/kg at bedtime; subsequently, the dose wasradually increased to 13.3 �g/kg (Figure 2). During thisreatment, no additional therapy for liver dysfunctionnd hyperlipidemia was provided, and the doses of hy-rocortisone, thyroxine, and gonadotropins were nothanged. No adverse effect was observed except for milddema in the leg, which disappeared spontaneously. Se-um insulin-like growth factor-I levels increased from7.3 ng/mL in the baseline to 103.0 ng/mL (normalange, 67–318 ng/mL) after treatment, suggesting thathe dose of replaced GH was at the physiologic level. Ashown in Figure 2, liver dysfunction was rapidly im-roved during the treatment; 6 weeks later, the values ofspartate transaminase, alanine transaminase, and �-gu-utamyl transpeptidase were in the normal range. Re-

igure 2. Clinical course during GH-replacement therapy. The dailyose of rhGH was increased gradually. During the treatment, liver dys-

unction improved to the normal range. Inflammation and oxidativetress markers such as serum levels of hsCRP and TNF-�, and urine

evels of 8OHdG were substantially decreased.

mltvbdcwtdlcita7haarAarcaeFdtm

rtctmtAds

limslavNwoApapgp

Fmsmpmmc(lwilw

CLI

NIC

AL–

LIV

ER,

PA

NCREA

S,A

ND

BIL

IARY

TRA

CT

March 2007 GROWTH HORMONE AND NASH 941

arkably, after 6 months of treatment, the markers foriver function and hyperlipidemia values were restored tohe normal range, but low-density lipoprotein cholesterolalues were not changed (Table 1). Although the totalody fat content obtained by dual x-ray absorptiometryid not change (data not shown), visceral obesity de-reased after treatment (area of visceral adipose tissueas estimated by using computed tomography: before

reatment, 131.6 cm2 and after treatment, 124.0 cm2, ie, aecrease of 5.7% was observed). After treatment, a second

iver biopsy was performed to assess the histologichanges in the liver. Surprisingly, a drastic improvementn NASH was observed as compared with that observed inhe previous biopsy (Figure 1E–H). After treatment, ste-tosis in the liver substantially reduced from 33.5% to.2% (Figure 1A and 1E). The inflammatory infiltrate andepatocyte ballooning completely disappeared (Figure 1Bnd 1F). The perisinusoidal and pericellular fibrosis incinar zone 3 also markedly improved to a trace sur-ounding the central vein (Figure 1C, 1D, 1G, and 1H).dditionally, we analyzed the changes in inflammationnd oxidative status markers during the treatment. Se-um hsCRP has been shown to be a useful marker fororonary disease and inflammation status in obesity,12,13

nd urine-8OHdG level is one of the representative mark-rs for the oxidative status in vivo.14,15 As shown inigure 2, hsCRP and 8OHdG levels were both decreaseduring the treatment. Serum TNF-� levels also decreasedo an undetectable level after treatment (Figure 2). Im-

igure 3. An oxidative stressarker, 8OHdG staining, demon-

trated that, in pretreatment liver,ost of the hepatocytes wereositive for the oxidative stressarker (A); however, after treat-ent, 8OHdG-stained hepato-

ytes were drastically reduced (B).C) A negative control. In normaliver, most of the hepatocytesere negative for 8OHdG stain-

ng. (D) A positive control. In NASHiver, most of the hepatocytesere positive for 8OHdG staining.

unohistochemical analysis using anti-8OHdG antibody T

evealed a large number of 8OHdG-positive hepatocytes;his indicates enhanced oxidative stress (Figure 3A) inontrast to the normal liver (Figure 3C). Notably, afterreatment, the 8OHdG-positive hepatocytes showed a

arked decrease, and the number was comparable withhat present in the normal liver (Figure 3B and 3C).nother method for the detection of markers for oxi-ized protein, nitrotyrosine staining, also demonstratedimilar results (data not shown).

DiscussionWe demonstrated that GH-replacement therapy

ed to dramatic improvement in NASH and dyslipidemian an AGHD patient. It is well-known that, as a case of

etabolic syndrome, dyslipidemia with elevated low-den-ity lipoprotein level, decreased high-density lipoproteinevel, and elevated triglyceride level are frequently associ-ted with AGHD,5,16 and GH-replacement therapy re-erses these abnormalities.16,17 Although the morbidity ofASH in AGHD has not been clarified, liver dysfunctionith hyperlipidemia7 and NAFLD are more frequentlybserved in AGHD patients than in the patient withoutGHD.8 Furthermore, in patients with hypothalamic andituitary dysfunction, the NAFLD development was rel-tively rapid with a high prevalence of cirrhosis. Thislaces the patients at risk for liver-related death,9 sug-esting that NAFLD, especially NASH is a serious com-lication in AGHD patients with panhypopituitarism.

hus far, there has been no useful surrogate marker

iNo

drimcFNt

olsdmthwlplGtrs

cAflNrm

AppptmnftN

1

1

1

1

1

1

1

1

1

1

2

CLIN

ICA

L–LIVER

,PA

NCREA

S,A

ND

BILIA

RY

TRA

CT

942 TAKAHASHI ET AL GASTROENTEROLOGY Vol. 132, No. 3

dentified as a substitute for liver biopsy for diagnosingASH, and this situation hampers an accurate diagnosis

f NASH in NAFLD.Although our study was not randomized and it is

ifficult to completely exclude the possibility that theesults were obtained because of sampling error, themproved liver status appeared to be due to GH-replace-

ent therapy rather than the natural course of recovery,onsidering the poor progressive prognosis of NASH.18

urthermore, in the present case, the effect of GH onASH suggested that the condition in AGHD was essen-

ially related to the onset of NASH.With regard to the pathogenesis in NASH, the presence

f metabolic syndrome with insulin resistance,11 elevatedevels of inflammatory markers,19,20 and oxidativetress21,22 in AGHD patients fulfill the pathogenic con-itions for the onset of NASH. In particular, enhance-ent of oxidative stress is one of the essential factors in

he development of NASH from NAFLD as a secondit.23 Our data demonstrated that, in the AGHD patientith NASH, oxidative stress increased in the serum and

iver. It was shown to decrease drastically after GH-re-lacement therapy, which is concomitant with the histo-

ogic improvement. Although it is not yet clear whetherH replacement exerts a direct or an indirect action on

he reduction in oxidative stress in the liver, previouseports21,22 showed that GH exerted an antioxidativetress effect in vivo.

It has been reported that GH-replacement therapy de-reased the serum levels of hsCRP24 and TNF-�25 inGHD patients. TNF-� plays a key role in causing in-ammation and insulin resistance in the pathogenesis ofASH.26 In the present case, it is speculated that the

eduction in serum levels of TNF-� during the replace-ent therapy exerted a positive effect on the liver status.It is important to elucidate the morbidity of NASH in

GHD because of the poor prognosis of NASH as com-ared with that of NAFLD. Interestingly, decreasedlasma levels of GH in NAFLD patients have been re-orted,27 suggesting that reduced GH levels could causehe onset of general NASH/NAFLD. Although the results

ay not be applicable to most NAFLD patients who areot GH deficient, considering the physiologic role of GH,

or example, having a lipolytic effect in adulthood, GHreatment could be an option for the treatment of someAFLD/NASH patients.

References

1. Adams LA, Lymp JF, St. Sauver J, Sanderson SO, Lindor KD,Feldstein A, Angulo P. The natural history of nonalcoholic fattyliver disease: a population-based cohort study. Gastroenterology2005;129:113–121.

2. Adams LA, Angulo P. Recent concepts in non-alcoholic fatty liverdisease. Diabet Med 2005;22:1129–1133.

3. Marchesini G, Brizi M, Morselli-Labate AM, Bianchi G, BugianesiE, McCullough AJ, Forlani G, Melchionda N. Association of non-alcoholic fatty liver disease with insulin resistance. Am J Med

1999;107:450–455.

4. Adams LA, Angulo P. Treatment of non-alcoholic fatty liver dis-ease. Postgrad Med J 2006;82:315–322.

5. Rosen T, Bengtsson BA. Premature mortality due to cardiovas-cular disease in hypopituitarism. Lancet 1990;336:285–288.

6. Svensson J, Bengtsson BA, Rosen T, Oden A, Johannsson G.Malignant disease and cardiovascular morbidity in hypopituitaryadults with or without growth hormone replacement therapy.J Clin Endocrinol Metab 2004;89:3306–3312.

7. Kaji H, Sakurai T, Iguchi G, Murata M, Kishimoto M, Yoshioka S,Iida K, Okimura Y, Chihara K. Adult growth hormone deficiency inJapan: results of investigation by questionnaire. Endocr J 2002;49:597–604.

8. Ichikawa T, Hamasaki K, Ishikawa H, Ejima E, Eguchi K, NakaoK. Non-alcoholic steatohepatitis and hepatic steatosis in pa-tients with adult onset growth hormone deficiency. Gut 2003;52:914.

9. Adams LA, Feldstein A, Lindor KD, Angulo P. Nonalcoholic fattyliver disease among patients with hypothalamic and pituitarydysfunction. Hepatology 2004;39:909–914.

0. Brunt EM, Janney CG, Di Bisceglie AM, Neuschwander-Tetri BA,Bacon BR. Nonalcoholic steatohepatitis: a proposal for gradingand staging the histological lesions. Am J Gastroenterol 1999;94:2467–2474.

1. Rosen T, Johannsson G, Johansson JO, Bengtsson BA. Conse-quences of growth hormone deficiency in adults and the benefitsand risks of recombinant human growth hormone treatment. Areview paper. Horm Res 1995;43:93–99.

2. Pai JK, Pischon T, Ma J, Manson JE, Hankinson SE, Joshipura K,Curhan GC, Rifai N, Cannuscio CC, Stampfer MJ, Rimm EB.Inflammatory markers and the risk of coronary heart disease inmen and women. N Engl J Med 2004;351:2599–2610.

3. Weiss R, Dziura J, Burgert TS, Tamborlane WV, Taksali SE, YeckelCW, Allen K, Lopes M, Savoye M, Morrison J, Sherwin RS, CaprioS. Obesity and the metabolic syndrome in children and adoles-cents. N Engl J Med 2004;350:2362–2374.

4. Collins AR, Gedik CM, Olmedilla B, Southon S, Bellizzi M. Oxida-tive DNA damage measured in human lymphocytes: large differ-ences between sexes and between countries, and correlationswith heart disease mortality rates. FASEB J 1998;12:1397–1400.

5. Perez DD, Strobel P, Foncea R, Diez MS, Vasquez L, Urquiaga I,Castillo O, Cuevas A, San Martin A, Leighton F. Wine, diet,antioxidant defenses, and oxidative damage. Ann N Y Acad Sci2002;957:136–145.

6. Simpson H, Savine R, Sonksen P, Bengtsson BA, Carlsson L,Christiansen JS, Clemmons D, Cohen P, Hintz R, Ho K, Mullis P,Robinson I, Strasburger C, Tanaka T, Thorner M. Growth hormonereplacement therapy for adults: into the new millennium. GrowthHorm IGF Res 2002;12:1–33.

7. Salomon F, Cuneo RC, Hesp R, Sonksen PH. The effects oftreatment with recombinant human growth hormone on bodycomposition and metabolism in adults with growth hormone de-ficiency. N Engl J Med 1989;321:1797–1803.

8. Younossi ZM, Diehl AM, Ong JP. Nonalcoholic fatty liver dis-ease: an agenda for clinical research. Hepatology 2002;35:746–752.

9. Sesmilo G, Biller BM, Llevadot J, Hayden D, Hanson G, Rifai N,Klibanski A. Effects of growth hormone administration on inflam-matory and other cardiovascular risk markers in men with growthhormone deficiency. A randomized, controlled clinical trial. AnnIntern Med 2000;133:111–122.

0. Sesmilo G, Miller KK, Hayden D, Klibanski A. Inflammatory car-diovascular risk markers in women with hypopituitarism. J Clin

Endocrinol Metab 2001;86:5774–5781.

2

2

2

2

2

2

2

DgvK(

JfDgMS

LIN

ICA

L–LI

VER

,A

NCREA

S,A

ND

BIL

IARY

TRA

CT

March 2007 GROWTH HORMONE AND NASH 943

1. Evans LM, Davies JS, Anderson RA, Ellis GR, Jackson SK, LewisMJ, Frenneaux MP, Rees A, Scanlon MF. The effect of GH replace-ment therapy on endothelial function and oxidative stress in adultgrowth hormone deficiency. Eur J Endocrinol 2000;142:254–262.

2. Kokoszko A, Karbownik M, Lewinski A. Increased lipid peroxida-tion in growth hormone-deficient adult patients. Neurol Endocri-nol Lett 2006;27:225–230.

3. Diehl AM. Lessons from animal models of NASH. Hepatol Res2005;33:138–144.

4. McCallum RW, Sainsbury CA, Spiers A, Dominiczak AF, Petrie JR,Sattar N, Connell JM. Growth hormone replacement reducesC-reactive protein and large-artery stiffness but does not alterendothelial function in patients with adult growth hormone defi-ciency. Clin Endocrinol (Oxf) 2005;62:473–479.

5. Serri O, St-Jacques P, Sartippour M, Renier G. Alterations ofmonocyte function in patients with growth hormone (GH) defi-ciency: effect of substitutive GH therapy. J Clin Endocrinol Metab1999;84:58–63.

6. Hui JM, Hodge A, Farrell GC, Kench JG, Kriketos A, George J.Beyond insulin resistance in NASH: TNF-� or adiponectin? Hepa-

tology 2004;40:46–54. K

7. Lonardo A, Loria P, Leonardi F, Ganazzi D, Carulli N. Growthhormone plasma levels in nonalcoholic fatty liver disease. Am JGastroenterol 2002;97:1071–1972.

Received October 17, 2006. Accepted November 27, 2006.Address requests for reprints to: Yutaka Takahashi, MD, PhD,

ivision of Endocrinology/Metabolism, Neurology, and Hematolo-y/Oncology, Department of Clinical Molecular Medicine, Kobe Uni-ersity Graduate School of Medicine, 7-5-2, Kusunoki-cho, Chuo-ku,obe 650-0017 Japan. e-mail: takahash@med.kobe-u.ac.jp; fax:

81) 78-382-5899.Supported in part by a Grant-in-Aid for Scientific Research from the

apanese Ministry of Education Science, Sports, and Culture; grantsrom the Foundation for Growth Science, Novo Nordisk Growth andevelopment Foundation; and a grant for the 21st Century COE Pro-ram, “Center of Excellence for Signal Transduction Disease: Diabetesellitus as Model” from the Ministry of Education, Culture, Sports,cience, and Technology of Japan.The authors thank Dr Iwakabe for support and Chika Ogata and

ayo Imura for technical assistance.

C P