Nutritional and prognostic significance of insulin-like growth factor 1 in patients with liver cirrhosis

TRANSLATED ABSTRACTS OF ARTICLES IN NUTRITION 1997, VOL. 13, NO. 3

SIGNIFICADO NUTRICIONAL Y PRONOSTICO DEL FACTOR DE CRECIMlENTO “INSULINA-LIKE= 1 EN PACIENTES CON CIRROSIS HEPATICA La mayoria de indices nutricionales tienen grandes limitaciones en pacientes con enfermedad hepatica cronica. El factor de crecimiento “insulina-like” 1 (FC-ILl) es regulado por la nut&ion y ha sido propuesto coma indice nutritional. Sin embargo su significado en pacientes con cirrosis hepatica no ha sido investigado. El FC-IL1 &co, al igual que 10s pammetros antropometricos, viscerales e inmunol6gicos tradicionales, fueron evaluados en 64 pacientes cirr6ticos hospitalizados, segnidos clinicamente durante dos atios. El FC-ILls._ no fue diferente en pacientes con o sin signos de desnutricion calorica, definida por valores de la circunferencia muscular de1 brazo (CMB) y/o el pliegue tricipital (PT) menores al percentil 5. Ademh el FC-IL 1z_scoreno se correlaciono con la CMB o el PT. A pesar de su correlation con todas las proteinas viscerales, la reduction de1 FC-IL1 fue mucho mayor y mas kecuente que la de las protemas viscerales. Los con &ores medios por debajo de -2.5, pacientes con FC-ILIZ.,, mostraron menor supervivencia a largo plaza que 10s pacientes con un FCmayor de -2.5 (p
EVALUACION DEL METABOLISM0 PERIOPERATORIO DEL GLICEROL POR LA TECNICA DE TRAZADORES ISOTOPICOS ESTABLES El prop6sito de este estudio fue investigar cambios metabolicos durante y despues de la histerectomia vaginal con miras especificas en el metabolismo de1 glicerol. El metabolismo de1 glicerol y la producci6n de glucosa en el higado fueron medidos antes y despues de la operation usando la tecnica de isotopes estables [(l,l,2,3,3-2Hs)-glicerol,(6,6-zH~)-glucosaJ. Antes, durante y despds de la operation, se determinaron substratos metabolicos (glicerol, acidos gmsos no esterificados, P-hidroxibutirato, glucosa y lactate) y hormonas (insulina, glucag6n, cortisol y catecolaminas). La histerectomia se asocio con un increment0 de1 metabolismo de1 glicerol postoperatorio de 3.56 +1.28 a 6.46k2.44 p molKg.min (~~0.05). Este increment0 estuvo inversamente relacionado con la edad de 10s pacientes (t=O.872, p
Nutrition 13:v-vii, 1997 OElsevier Science Inc. 1997 Printed in the USA. All rights reserved.

EL US0 DE UN SISTEMA EXPERT0 EN LA PRESCRIPCION FORMULAS ENTERALES EN UN HOSPITAL UNIVEZRSITARIO

DE

Un sistema expert0 informatico ha sido desarrollado para la prescription de formulas enterales, basado en las caracteristicas de las necesidades de 10s pacientes. METODOS: Doscientos dote pacientes ingresados en un hospital universitario fueron evaluados de forma prospectiva para comparar la identidad y 10s castes de las formulas enterales prescritas por el sistema expert0 y por 10s climcos de1 hospital. RESULTADOS: Doscientos siete pacientes completaron 10s datos necesarios para su analisis. Hubo un ahorro promedio de 1.18+7.69 WSA por dia para cada paciente usando el sistema expert0 en comparacioo a la formula prescrita por 10s medicos (p==O.O23).CONCLUSIONES: Se deduce que el uso de este programa ahonaria 27564 $USA anualmente en nuestro hospital (con un promedio de 23360 pacientejdias de alimentacion enteral por afio). Concluimos que el uso de un sistema expert0 puede ser rentable en la prescription de formulas enterales para pacientes hospitalizados.

EICOSANOIDES Y EMULSIONES LIPIDICAS EN PACIENTES CON SINDROME DE DISTRES RESPIRATORIO AGUDO A las emulsiones lipidicas se les ha imputado cambios en la fun&n puhnonar. Esos cambios se relacionaron con el efecto fisico de la lipemia inducida por la infusi6n sobre el mtercambio gaseoso. Sin embargo varies estudios en humanos y animales sugieren que el empeoramiento en la timcion puhnonar, observado con la infusion de lipidos, esta mediado por prostaglandinas. Las prostaglandinas se sintetizan enzimaticamente a partir de acidos grasos esenciales. Hemos estudiado 10s efectos de dos emulsiones lipidicas, con diferentes cantidades de acidos grasos esenciales (20% LCT con 55% de acido linoleico y 7% de acido a-linolenico en 100 g de emulsion, y una mezcla fisica de 20% MCT y LCT con 26% de acido linoleico y 4% de acid0 a-linolenico en 100 g de emulsion), sobre 10s niveles plasmaticos de eicosanoides en pacientes con sindrome de distres respiratorio agudo (SDRA). A pew de que 10s niveles plasmaticos de prostanoides en 10s pacientes con SDRA, fueron mas altos que 10s valores de referencia, ninguna de las dos emulsiones lipidicas, administradas a un ritmo de 2 mg/Kg, indujo cambios significativos en 10s eicosanoides con la salvedad de una disminucion en la diferencia de 6-keto prostaglandina F-la entie la circulation sistemica y la arteria puhnonar.

ELSEVlER

0899-9007/97/$17.00 PI1 SO899-9007(97)00036-

1

TRANSLATED

vi

CINC Y CIRROSIS HISTOPATOLOGICA

HEPATICA:

EVALUACION

BIOQUIMICA

ABSTRACTS

E

Se indujo una cirrosis hepatica experimental mediante la administration de tiocetamida. 10s animaIes cirroticos meron divididos en dos grupos, un grupo recibio sulfato de cinc y el otro acme coma control. Los animales tratados con cinc mostraron una restauracion de 10s niveles plasmaticos y hepaticos de cinc y cobre. De la misma forma, 10s niveles plasmaticos de aspmtato aminotransferasa (ASAT), alanino aminitransferasa (ALAT), gamma-glutamil aminotransferasa (GGT) y bilirrubina total, descendieron significativamente. Los estndios al microscopio optic0 mostraron una apariencia normal en la mayoria de 10s hepatocitos de1 grupo tratado con cinc en comparaci6n a 10s animales cirroticos no to&ados. Las cantidades de fibrina, reticulina y colageno, elevadas en 10s Ngados cirr&icos, disminuyeron tras el tratamiento con cinc. La tin&in con PAS mostro la capacidad de 10s hepatocitos para ahnacenar glucogeno tras el tratamiento con cinc. Estos resultados revelan que el cinc podria tener algun efecto beneficioso en el tratamiento de la cirrosis hepatica.

POTENCIAL DEL MONOGLICERIDO Y TRIGLICEFUDO DEL DL3HIDROXlBUTIRATO EN NUTRICION PAEENTERAL: SINTESIS Y ESTUDIOS BIOLOGICOS EN LA RATA Los esteres de cadena cotta de 10s bidos orgkicos se han mostrado prometedores coma potenciales nutrientes en alimemaci6n parenteral. La mayoria de glic6ridos son insolubles en agua pero 10s de 10s cuerpos cet6nicos muestran cierta solubilidad. Es interesante el hecho de que el triglic&ido de1 3hidroxibutirato es soluble en agua mientras que e1 triglickido de1 acetoacetato no lo es. Los mono y trigliceridos de1 DL-3hidroxibutirato fueron smtetizados y ensayados para evahtar su toxicidad y su valor nutritional coma nuttientes parenterales. Ambos compuestos tienen una densidad cal6rica estimada de 19.7 Kj/g (4.7 K&g) y son hidrosolubles. Los compuestos fueron infundidos en ratas durante 7 dias a un ritmo que proporcionaba 113 Kjklia y ademas, se administro una dieta oral hipocalkica. Un grupo control de animales recibio la misma dieta oral durante la infusion isocal6rica de glucosa o suer0 fisiol6gico. Se compararon las entradas, sahdas y balance de nitr6geno, 10s cambios en el peso corporal y el tamago de1 h&ado. Los dos gliceridos de1 3-hidroxibutirato y la infusi6n de ghmosa, mostraron similar retenci6n de nitr6geno, cambios en el peso corporal y tamat de1 higado. Las ratas que recibieron suer0 fisiol6gico retuvieron menos nitr6geno y mostraron menor peso corporal e higados de menor tamaf~o. Los datos demostraron que 10s gliceridos de1 DL-3hidroxibutirato no son tkicos, aportan energia por via intravenosa y pod&n ser utilizados coma nutrientes para alimentaci6n parenteral.

OF

ARTICLES

IN NUTRITION,

VOL.

13, NO.

3

LTNS~-~KEGR~~IHFA~RI : MARQWR NUmlTiONNELm PRONOSKAUCOURSDEL4ClRRHOSE. L.aplupart des param&% classique de l’&hiation nutritiomselle sent d’un in&&t limit6 chez le patient cirrhotique. L’IGF-I eat r6gul6 par la mnrition et pourrait done constimer un maqueur de F&at nutritionnel Cependant, son utilisation chez le cirrhotique n’a pas W &ahu?e. LTGF-I skique et lea valeurs de diffkents paran&tms antbropoi&riques et biologiqw ant W d&ermin& chez 64 patients cirrhotiques hospitalis& avec un suivi climque de deux ans. I_e Z score de I’IGF-I etait de l’ordm de -2,16 f 1,08 et &it corrt% n&gativement au score de Child Pugh @ < 0,Ol) qui est le plus fiable pour apprkia la s&kite de l’atteinte h6patiqw. Le Z score de l’IGF-I itWait pas diff&ent entm les patienta avec et saris sigrbss de maltmtrition &rg&ique d&inie par des valeurs de circonfQence muscuke brwhiale (CMB) etkm du pli cutan tricipital (PCT) infkrleurea au Seme percentile. II ny avait pas de corr6latkm entns le Z score de I’IGF-I et la CMB ou le PC!T. Quoique corn% avec les dif&entea protimes v&c&ales, la reduction de I’IGFI &ait plus importante et plus ik@ente que celle des prottkes visc&alea. I_es patients ayant un Z score de ITGF-I in&km a la valeur nkhane (-25) pi&mtaient un tanxdesurvieBlongtenneplusfaiblequeceuxavecunZscore~~g -2,s (p< 0,Ol). Ces r&hats indiquent que I’IGF-I skique n’est pas corn% avec la malnutrition &erg&qua chez les patients cirrhotiques, mais qu’il pnkente un bon caract& predictif en termes de suivie et comtitue un marqueur p&rice des alt&ations h6patiques. De nomhreux facteurs, pour la plupart r&s a la &kit6 de l’atteinte h@atique, pourraient contribuer ii la diminution & la concentration de I’IGF-I. Cette pathogekie complexe est sans dome ii l’origiw ducamcbkepr&iictifdecepami&re.

EVALUAnoNP~O~~W~~USMEDU~Y~~AL'AIDE lmOlwESsTABLEs. Le but de ce travail &sit d%vahmr les modifications n&aboliques pendant et ap&s hystkectomie en s’inmmssant plus particuli&ement au &&olisnxs du glycerol. Sept patientes pr&sentam un myoma u&in h&tin mais sans autres affections ont et6 inclues dam cette &de. Le renouvellement du gIyc&ol et Ia production h6patique de glwose ant W mestir& avant et ap&s intervention a l’aide d’isotopes stables ([1,1,2,3,3&5]-glyckrol, [6,6-%2]-g1ucose). Les concentrations de substrats (glycerol, acmes gras libres, 8-hydroxybutyram glucose, lactate) et d’hormones (insulina, glucagon, cortisol, catkholamines) ont et6 d&ermh& avant, pendant et apres lTnteivention. L’hystkectomie etait associ6e a une augmentation postopkatoire du renouvellement du glyc&ol(6,46 f 244 vs 356 f 1.28 pmol/kg.mim p < 0.05). L’amplitude de l’augmentation &ait corr&e n4gativement avec 1’8ge (r = 0,872, p < 0,05). Les com%ntrations de glyckol tendaient it augmenter mais de man&e non significative en p&iop&toim. La production h6patique de glucose et la glyc&ie augmentaient en postopkatoire, respectivement de 9,75 f 1,61 a 12,79 f 1.15 pmolkgmin @ < 0,051 et de 46 f 0,9 a 6,2* 0.9 mmoY1 (p < 0,05). Les concentrations de untisol et de catkcholaaktes augmentaiem pendant et apr&sl’intervention, t&is qua cek dksuhne et de ghtcagon ne vatiaient pas. L’acc&ration de la lipolyse api% hystkectomie n’&ait pas identifiable par la settIe mesum des cowentmtions plasmatiques de glyc&ol. Ces rkultats montrent que Putilisation des isotopes stables permet une meilkum &&ration dea flux mkaboliques que des simples dosages statiques de substrats plasmatiques, en particulier en p6riode p&iof&atoire.

APPLIED

NUTRITIONAL

Nutrition Vol. 13, No. 3, 1997

INVESTIGATION

Nutritional and Prognostic Significance of Insulin-Like Growth Factor 1 in Patients With Liver Cirrhosis LORENZA CAREGARO, MD, FRANCA ALBERINO, MD, PIER0 AMODIO, MD, CARLO MERKEL, MD, PAOLO ANGELI, MD, MARIO PLEBANI, MD,* MASSIMO BOLOGNESI, MD, AND ANGELO GATTA, MD From the Department of Clinical and Experimental Medicine, and “Central Laboratory, University of Padua, Padua, Italy Date accepted: 3 1 May 1996 ABSTRACT

Most of the traditional parameters for nutrition assessment have important limitations in patients with chronic liver disease. Insulin-like growth factor 1 (IGF-1) has been found to be regulated by nutrition and proposed as a nutritional marker. Its nutritional significance in patients with liver cirrhosis, however, has not been investigated. Serum IGF-I as well as traditional anthropometric, visceral, and immunologic parameters were evaluated in 64 hospitalized cirrhotics, followed up clinically for 2 y. IGF- l,.,,,, averaged -2.16 5 I .08 and inversely correlated with Child-Pugh score (P < was not different in patients 0.01)) the most reliable composite score reflecting the severity of liver disease. IGF-1 I’,LorC with or without signs of energy malnutrition, as defined by values of midarm muscle circumference (MAMC) and/or did not correlate with MAMC or TSF. Despite its correlation triceps skinfold (TSF) <5th percentile. Moreover, IGF- 1z_sco~~ with all viscera1 proteins. the reduction of IGF-1 was much greater and more frequent than that of visceral proteins. Patients with IGF-l..,,,,, < median values (-2.5 ) showed lower long-term survival rates compared with patients with IGF-l,_,,,, > -2.5 (P < 0.01 ). These data indicate that serum IGF-I is not related to energy malnutrition in cirrhotic patients, while it appears to be a good predictor of survival and an early marker of liver dysfunction. Multiple factors, most of which are related to the severity of the liver disease, may contribute to the reduction of IGF-1. This multifactorial pathogenesis probably accounts for its prognostic significance. Nurvition 1997; 13: 185 190. OElsevier Science Inc. 1997 Key words: insulin-like growth factor 1, liver cirrhosis, nutrition, survival

INTRODUCTION

Protein energy malnutrition, a common finding in patients with liver disease,‘-’ has important clinical implications. A poor nutritional status negatively influences surviva13,4while appropriate nutritional intervention has been found to improve liver function and survival.5~6 Though the importance of nutritional factors has been greatly emphasized in the last decade, most of the traditional parameters for nutrition assessment have important limitations in these patients.2.7 Body weight is an insensitive indicator because of apparent or subclinical salt and water retention, while the nutritional significance of visceral proteins is limited by their dependence on liver function. Alcohol abuse per se also affects hepatic protein synthesis8 Moreover, the complex immunologic alterations, charac-

teristic of both alcoholic” and nonalcoholic liver disease,’ may influence immunologic parameters. Skinfold anthropometry is considered-at present-the most reliable clinical measure of nutritional status in patients with chronic liver disease.” However, it must be taken into account that malnutrition may be greatly underestimated when diagnosed by anthropometry only, as the visceral compartment is not evaluated. In addition, anthropometric parameters are not adequate for monitoring nutrition therapy as they are insensitive to short-term changes in nutrition. There is, therefore, a need for better methods for nutrition assessment, particularly for the evaluation of visceral compartment. Insulin-like growth factor 1 (the IGF-1, or somatomedin C) is a 7649 single-chain peptide that mediates most of the metabolic effects of growth honnone (GH) , and plays an important

This work was supported in part by grants from Minister0 dell’Universit8 e della Ricerca Scientifica e Tecnologica (MURST 60%). Correspondence to: Lorenza Caregaro, MD, Istituto di Medicina Clinica, Universit5 di Padova, Policlinico, Via Giustiniani 2, 35 128 Padua, Italy.

Nutrition 13:185-190, 1997 OElsevier Science Inc. 1997 Printed in the USA. All rights reserved

ELSEVIER

0899-9007/97/$17.00 PI1 SO899-9007(96)00399-

1

186

INSULIN-LIKE

GROWTH

role in the stimulation of biological growth. Its production depends on GH, to which it is linked by a feedback mechanism,l’,” IGF-1 is also regulated by nutrition independently of GH.‘j Recently, IGF-I has been found to be a sensitive marker of malnutrition.‘4.‘5 Moreover, because of its very short halflife ” and lack of significant influences of stress, ” IGF- 1 has been suggested to be adequate for monitoring nutrition therapy.” Though low levels of IGF-1, repeatedly reported in cirrhotic patients, have been attributed to liver function impairment, “~‘t’the relationships between IGF- 1 and nutritional alterations characteristic of cirrhosis have not been investigated. The present study aimed to evaluate the nutritional and clinical significance of IGF-1 in liver cirrhosis by comparing IGF-1 with the traditional anthropometric, visceral, and immunologic parameters. METHODS AND MATERIALS Patients Between March 1990 and October 1993, 64 hospitalized cirrhotic patients were included in the study. Forty-six patients were male and 18 female, their ages ranging from 32 to 75 y (mean 57 t- 9). The diagnosis of cirrhosis was based on clinical, laboratory, and ultrasonographic criteria, and histologically confirmed in 30 patients. Impaired blood coagulation prevented liver biopsy in the other 34 subjects. The etiology of liver cirrhosis was alcohol-related in 42 subjects and virus-related (hepatitis B or C) in the other 22 individuals. Thirty-eight of the 42 patients with alcoholic cirrhosis (90.4%) were abstinent from alcohol for at least 3 mo at the time of the study. Cirrhotics with diabetes mellitus, hepatocellular carcinoma, anasarca, or infectious disease were excluded. The study population was divided into three groups according to the severity of the liver disease assessed by Child-Pugh criteria,” which attribute an arbitrary score to serum albumin, bilirubin, prothrombin activity, and to the presence of ascites and encephalopathy. This classification is considered at present the most reliable composite score reflecting the severity of liver disease.” Seventeen patients were graded as A (26.6%), 21 as B (32.80/c), and 26 as C class (40.6%). Clinical parameters of cirrhotic patients are reported in detail in Table I. Informed consent was obtained from all subjects. The protocol was in accord with the Helsinki Declaration of the 1975 as revised in 1983.

TABLE

I.

CLINICAL PARAMETERS OF CIRRHOTIC PATIENTS Sex Age (Y)” Etiology (% of patients) Alcoholic Virus-related Ascites (“lopresent) Esophageal varices (c/c present) Child classh (% of patients) Child A Child B Child C “Mean 2 SD. h Calculated as described by Pugh et al.”

46M, 18F 57.3 -+ 9.5 (range: 32-75) 42164 (65.7%) 22164 (34.3%) 40/64 (62.5%) 39/64 (60.9%) 17/64 (26.6%) 21/64 (32.8%) 26164 (40.6%)

FACTOR

1 AND NUTRITIONAL

STATUS

IN CIRRHOSIS

Study Protocol and Determinations Patients were enrolled between 7 and 15 d after admission to permit diagnostic evaluation and stabilization of the treatment. During this period a standard diet (35 kcal/kg body weight; 15% protein, 30% lipids, 55% carbohydrates) was offered. Sodium intake was restricted (40 mmol/d) only in ascitic patients. An approximate estimate of intake of energy and essential nutrients was derived from a 24-h dietary recall by an experienced dietitian during the prestudy period. Anthropometric and biochemical parameters were evaluated on the same day. Anthropometric evaluation included the measure of body weight, height, triceps skinfold (TSF) and midupper arm circumference (MAC) Body weight was expressed as a percentage of ideal body weight (lBW%) calculated from height and sex using the 1984 Frisancho tables.“’ Values ~90% were considered to be subnormal. TSF and midarm muscle circumference (MAC) were measured as indexes of body fat and muscle protein compartment, respectively. TSF was measured by the same observer, with a Holtain caliper (Holtain, Ltd., Crymych, Dyfed, UK) at the midpoint between the acromion and the olecranon of the nondominant arm. To minimize intraoperator variability, the average of three consecutive measurements was recorded. MAMC was calculated by the following formula: MAMC

(cm) = MAC (cm) - 10.314 x TSF (mm)],

where MAC was mid upper-arm circumference, measured with a tape at the same site of TSF. The percentiles of MAMC and TSF were calculated from standard tables based on age and sex.2.1 Observed values of MAMC and TSF were also normalized to percentage of 50th percentile derived from the same tables (MAMC% and TSF%). Serum albumin, transthyretin (prealbumin), transferrin, and retinol-binding protein were determined by immunodiffusion techniques using commercial kits (Behring Institute, Scoppito, Italy). Plasma IGF-1 was analyzed by IRMA (DSL, Webster, USA), after acid-ethanol extraction from its binding proteins. Blood samples were collected in all patients between 0800 and 0900 after an overnight fast. Because IGF-I varies with age and sex, we used sex- and age-adjusted normalizing z-score transformations derived from a normal population ( 135 healthy subjects, matched for sex and age with cirrhotic patients; mean age = 53 y, range = 28-74 y). Immune function was assessed by total lymphocyte count and skin test response. The latter was evaluated by intradermal injection of seven antigens (Multitest IMC, Merieux Institute, Rome, Italy). A positive response was defined as a cutaneous induration at 48 h ~2 mm. Patients’ reactivity to skin tests was classified as: normal (two or more positive responses), hypoergy (only one positive response) or anergy (no positive response). A skin test score was also calculated as the sum of the mean induration diameters with all tested antigens. A single observer administered all antigens and recorded all readings. Patients were followed up clinically every 6 mo for at least 24 mo or until death. During the follow-up period, nine patients underwent orthotopic liver transplantation and one developed a hepatocarcinoma. These patients were excluded from the analysis of survival. Statistical Analysis Data are listed as mean I SD. Circulating concentrations of IGF-1 are expressed both as z-score and absolute values. Comparison between means was tested by Student’s t test for

INSULIN-LIKE

GROWTH

FACTOR

1 AND

NUTRITIONAL

STATUS

TABLE NUTRITIONAL All (n = 64) Body weight % IBW” MAMC % of standardb TSF % of standard” Albumin (35-50 g/L) Transthyretin (I 00-400

mg/L)

Retinol-binding protein (30-60 mg/L) Transferrin (2-4 g/L) Lymphocyte countC (1000-4800 x 106/L) Skin test scorec (24 mm) IGF- 1 (ng/mL) IGF- 1 z-score

97.33 89.94 86.58 32.57

2 2 2 2

17.77 15.47 51.95 7.27

PARAMETERS

Child A (n = 17) 99.65 96.53 101.94 39.47

t ? -t 2

II. OF CIRRHOTIC

Child B (n = 21)

17.68 15.88 63.43 5.35

187

IN CIRRHOSIS

98.86 94.81 95.57 33.57

t 1-+ i

19.57 13.34 43.31 5.94

PATIENTS P (A vs B)

P (B vs C)

P (A vs C)

NS NS NS <0.005

NS <0.005 NS
NS <0.005 NS
67.18 5 20.46

NS



Child C (n = 26) 94.16 81.69 69.27 27.26

2 ? + 2

16.80 13.36 46.74 4.92

102.10 -t 49.59

133.62 + 47.97

17.96 t 9.27 2.20 * 0.97

22.08 2 8.34 2.61 2 0.87

19.48 +- 11.89 2.68 -+ 0.87

13.91 t 4.98 1.63 + 0.82

NS NS

NS <0.005

<0.005 <0.005

1148 ? 632

1078 t 481

1104 z 595

1233 ? 756

NS

NS

NS

7.26 5 7.27 42.1 ? 33.4 -2.16 -+ 1.08

8.37 ? 6.99 80.0 2 63.3 -1.67 -c 1.33

5.20 ? 4.85 62.7 ? 57.6 -1.91 2 1.16

8.38 ? 9.08 32.6 2 19.4 -2.67 2 0.46

NS NS NS

NS <0.02 <0.005

NS
Data are mean + SD. Normal range for the laboratory in parentheses. ’ Calculated from ideal weight for height.23 h Calculated from reference 24. ‘Number of tested patients = 56. IBW. ideal body weight; MAMC, midarm muscle circumference;

116.77 t 51.64

TSF, triceps skinfold.

unpaired data, and comparison between frequencies by corrected x2 regression analysis. Significance was accepted at the 5% probability level, using two-sided test. Survival curves were computed by the Kaplan-Meier method, and comparison of survival distribution was made with the log-rank test. RESULTS

Nutrition Assessment Only 7 of the 64 cirrhotics (10.9%) were considered to have an inadequate energy intake, as their estimated intake during the prestudy period was <125 kJ/kg body weight. Though resting energy expenditure was not measured by indirect calorimetry in our patients, this energy intake is considered to satisfy the daily energy requirements of the majority of stable cirrhotic patients.2.25 Nutritional parameters and IGF-1 values of cirrhotic patients, divided into the three Child classes according to the severity of the liver disease, are reported in detail in Table II. Severe energy malnutrition, as defined by values of MAMC and/or TSF <5th percentile, was found in 23 of the 64 studied patients (35.9%). Forty-four of the 64 patients (68.7%) showed low values of at least two of the studied visceral proteins. In all but two patients <5th percentile for MAMC and/or TSF (21/64, 32.8%), we found low values of at least two serum proteins, while a reduction of serum proteins without significant anthropometric abnormalities was observed in 23 of the 64 patients (35.9%). Anthropometric, biochemical, and immunologic parameters were not statistically different in patients with alcoholic and virus-related cirrhosis. MAMC% and TSF% correlated with each other (I = 0.32, P < 0.01). MAMC%-but not TSF%-correlated with albumin (Y = 0.35, P < 0.01 ), transthyretin (r = 0.33, P < O.OS),

and transferrin (r = 0.28, P < 0.05), while no correlation was found between retinol-binding protein and anthropometric parameters (r = 0.10 for MAMC% and r = 0.17 for TSF%). Fifteen of the 64 patients (23.4%) were anergic and 26 hypoergic (46.4%). Lymphocyte count was < 1000/mm3 in 23 patients (35.9%). The skin test score was correlated with lymphocyte count (r = 0.43, P < 0.001) but not with visceral proteins (r = -0.10 with albumin, r = -0.005 with transthyretin, r = -0.19 with transferrin, and r = -0.10 with retinolbinding protein) or anthropometric parameters (r = 0.01 with TSF% and r = 0.05 with MAMC%). Nutrition

Related to the Severity of Liver Disease

As shown in Table II, all visceral proteins progressively decreased from Child A to Child C class. The same trend was also observed for anthropometric parameters (MAMC% and TSF% ) , but the difference among groups was statistically significant only for MAMC%. Moreover, MAMC% but not TSF% inversely correlated with Child-Pugh score (r = -0.38, P < 0.001 ), confirming the greater clinical significance of muscle mass depletion.3.26 Lymphocyte count and skin test score were not statistically different in the three Child classes. In addition, neither lymphocyte count nor skin test score correlated with ChildPugh score. IGF-1 was < -2 in 47 of 64 cirrhotic patients (73.4%): IGF- 1z-scorn 10117 in the Child A group (X8%), 14/21 in the Child B group (66.6%), and 23/26 in the Child C group (88.4%). As shown in Figure 1, IGF-l,.,,,, progressively decreased from Child A to Child C class. IGF-1 was not significantly different in patients with alcoholic and virus-related cirrhosis. It averaged 51.42 5 42.36 ng/mL in patients with alcohol-related cirrhosis

188

INSULIN-LIKE

GROWTH

IGF-1 z score 1 0.5 0 -0.5

-1 -1.5 -2 -2,5 -3 -3,5 -4

All

Child A

Child B

Child C

FIG. 1. IGF-I z-score in cirrhotic patients divided according severity of the liver disease assessed by Child-Pugh criteria.” IGF-1 z-score was significantly lower in Child C compared with B (P < 0.005) and Child A patients (P < 0.001). Vertical bars sent mean 2 1 SD. IGF-1, insulin-like growth factor 1.

to the Mean Child repre-

and 62.50 2 65.31 ng/mL in those with virus-related liver disease. Mean z-score was -2.23 t- 0.91 and -2.02 t 1.35, respectively. IGF-1 was lower, but without statistical significance, in patients <5th percentile for MAMC and/or TSF compared with patients >5th percentile (-2.37 ? 0.62 compared with -2.07 ? 1.21). Moreover, no correlation was found between IGF-1 and anthropometric parameters (r = 0.13 between IGF-l,_,,,, and MAMC%, r = 0.03 between IGF-l,_,,,,, and TSF%) . IGFwas strictly correlated with all visceral proteins (P < 1L.SCOTC 0.001 with albumin and transthyretin, P < 0.01 with retinolbinding protein and transferrin) . Nevertheless, the reduction of IGF-l..,,,,, was much greater and more frequent than that of serum proteins. As shown in Figure 2, IGF-1 was reduced also in a significant number of patients with normal values of albumin and transthyretin. IGF- 1Z_scOTe inversely correlated with Child-Pugh score (r = -0.38, P < 0.01) while the correlation with prothrombin activity, another index of liver synthetic capacity, was poor (r = 0.26, P < 0.05). IGF-I,.,,,,, did not correlate with immunologic parameters (r = 0.12 with lymphocyte count, r = 0.02 with skin test score). The survival analysis was performed by dividing cirrhotics into two groups according to the median IGF-l,.,,, (-2.5). Patients < -2.5 showed significantly lower survival rates with IGF- 1L~SmTe at 2 y (P < 0.01) compared with patients with IGF-l,.,,,, > -2.5 (Fig. 3). A comparable difference on survival cumulative rate (P < 0.01) was observed when patients were divided according to the median albumin z-score ( -2.l), while no difference was evidenced when they were divided according to the median prothrombin activity z-score (P = 0.11)

FACTOR

1 AND

NUTRITIONAL

STATUS

IN CIRRHOSIS

limited in patients with liver disease. All four circulating plasma proteins utilized to monitor visceral protein stores are synthesized mainly by liver cells. They also reflect alterations of water and sodium balance, and the acute phase response. In addition, major defects in immune function such as low lymphocyte count and alterations of cell-mediated immunity, which are generally associated with visceral protein depletion, *’ may occur independently of malnutrition in patients with both alcoholic and nonalcoholic liver disease.2~3~9~28 The lack of correlation between immunologic and nutritional parameters in our patients supports this concept. Skinfold anthropometry is considered a more reliable marker of malnutrition compared with visceral proteins and immunologic parameters, as it is not influenced by liver disease and its complications. However, when malnutrition is diagnosed by anthropometry only, a subgroup of patients with protein depletion, but without clinical features of energy depletion, remains unrecognized. A more precise diagnostic evaluation of malnutrition is of primary importance in improving the impact of nutrition therapy on complications and survival. In fact it has been recently demonstrated that only mildly or moderately malnourished patients, but not severely malnourished patients with chronic alcoholic liver disease, may benefit from nutritional support.’ Theoretically speaking, IGF-1 would be a better nutritional

IGF-1 z score I I

2

I I I I

I 0

’

.I

* I .I

-1 _2 -

-

-

-

-

-

-

-

.

-3

??

-

f

. .

. . mm. , ..I.. .. . .

.

0

5

10

15

20

serum 3

-mT ‘%A

25

-

. -

.

-

-

-

.

I.

.

-4

.

*I

. ,*_.;“lrn&

.

.

..

I 35

30

albumin

40

45

50

55

(g/l)

I

I

2

.

.

I I

1

I I

0

I

-1

I I I

.

.

??

.

y.

.

DISCUSSION In keeping with recent studies, our data confirm that nutritional disorders are very common in cirrhotic patients, and related to the severity of the liver disease.‘.3.5 As previously demonstrated, 3 the prevalence of alterations of visceral proteins is extremely high while anthropometric depletion is less frequent, but almost constantly associated with protein depletion. Unfortunately the nutritional significance of serum protein is

J

50

100

serum

150

transthyretin

200

250

300

(mg/l)

FIG. 2. Relationship between plasma IGF-1 z-score and serum albumin and transthyretin in cirrhotic patients. Dashed lines represent the limits of normal values. IGF- I, insulin-like growth factor 1.

INSULIN-LIKE

GROWTH

FACTOR

1 AND

----~ ii!

NUTRITIONAL

survival probability 1

I --

L---------------

0.a

I

p-=0.01

0.f

0.4

,l

6

12

16

24

time (months) FIG. 3. Survival probability at 2 y in cirrhotic patients divided into two groups according to the median IGF-1 z-score ( -2.5). IGF- I, insulinlike growth factor I. index than visceral proteins in cirrhotic patients, as it is synthetized not only in the liver but also in other organs,” and it is not influenced by acute phase response.” Because IGF-1 has important anabolic effects, particularly on muscle and skeletal tissues, the low plasma values, repeatedly reported in cirrhosis, “-*’ have been suggested to reflect lean body mass depletion. In our patients, however, we did not find any correlation between IGF-1 and anthropometric parameters, nor any significant difference in IGF1 between patients with or without signs of energy malnutrition. Therefore, in spite of its known anabolic properties, IGF-1 does not reflect the presence or the severity of energy malnutrition in patients with chronic liver disease. IGF-1 was strictly correlated with all visceral proteins in our patients and showed a progressive decrease with the worsening of the liver disease. These observations may be interpreted as confirming the dependence of IGF-1 on liver synthetic capacity. Other data, however, indicate that the low circulating concentrations of IGF-1 in cirrhosis reflect other alterations in addition to hepatic synthetic capacity. IGF-1 was very low also in a large percentage of compensated cirrhotic patients (Child A group) with normal values of albumin and transthyretin. In addition, as previously reported by Untermann et al.14 in critically ill patients, the percentage reduction of IGF-1 was much greater than that of visceral proteins. Finally, the correlation of IGF-1 with prothrombin activity, another index of liver synthetic capacity, was poor. Though further studies are needed to confirm this interpretation, it is tempting to speculate that IGF-1 represents a more sensitive and early indicator of liver dysfunction compared with other clinical parameters. A number of studies have demonstrated that IGF-1 is related to nutritional intake, particularly protein intake, and nitrogen balance.‘“.” Though a precise evaluation of nutritional intake was beyond the scope of the present study, the approximate estimate of energy intake revealed an inadequate intake in only a minority of our patients while low serum IGF-1 concentration was quite a universal finding. It seems unlikely, therefore, that low serum IGF1 merely reflects an inadequate energy or protein intake. A pattern characterized by low IGF-1, associated with high basal GH and

STATUS

189

IN CIRRHOSIS

acquired GH resistance, has been described in hypercatabolic patients.‘” In this setting the IGF-l-mediated anabolic effects of GH are reduced, while IGF- 1 treatment can reverse the negative nitrogen balance. A hypercatabolic state may occur in liver cirrhosis, particularly during periods of poor nutritional intake, intercurrent illnesses, and complications. The majority of stable cirrhotics, however, exhibit a positive nitrogen balance.30 Therefore a state of negative nitrogen balance could account for the reduction of IGF-1 only in a minority of patients. Further studies, however, are needed to evaluate the relationships between abnormalities of GH/IGF- 1 axis and the complex changes of protein metabolism characteristic of cirrhosis.3o The relationships with glucose metabolism alterations commonly found in cirrhotic patients (peripheral insulin resistance. hyperinsulinemia, and defective nonoxidative glucose metabolism)3’ have to be defined too. As insulin contributes with GH to the generation of IGF-I from the liver, the shunting of portal blood, which contains insulin, away from the liver has also been implicated in the low IGF-1 levels of cirrhotic patients. Recently, Assad et al.l9 evaluated GH/IGF-1 dynamics in two different forms of chronic liver disease--alcoholic cirrhosis and schistosomal hepatic fibrosis. While portal hypertension and portosystemic shunting are characteristic of both, liver parenchymal involvement is minimal in schistosomal hepatic disease. The observation of the same hormonal pattern, i.e., very low serum IGF-1 despite elevated GH, in both groups of patients has been interpreted as supporting the hypothesis that the reduction of IGF-1 is related to the degree of portal hypertension and portosystemic shunting more than to the degree of liver function impairment. Hyperestrogenism, another complication of liver cirrhosis, would also contribute to reduce circulating IGF- 1, as it has been shown that estrogen administration in women decreases serum IGF1, probably by inhibiting GH-stimulated hepatic IGF-1 synthesis.” Though the involved pathophysiologic mechanisms are not clear, it may be argued that the reduction of IGF-1 represents the result of multiple alterations that may act synergistically in patients with chronic liver disease. These alterations include liver function impairment, malnutrition, abnormalities of protein and glucose metabolism, portal hypertension, and hyperestrogenism. The influence of multiple factors, most of which are related to the severity of the liver disease, could account for the high prognostic significance of IGF- 1. The observation that IGF- 1 is a good predictor of survival in cirrhotic patients is clinically important, as the accuracy of the commonly used tests is poor and there is a need for a better prognostic evaluation. In conclusion, our data indicate that IGF-1 does not reflect the presence or the severity of energy malnutrition in cirrhotic patients. The great reduction of IGF-1 in the majority of cirrhotic patients, also in those with compensated liver disease and normal serum concentration of albumin and other proteins, together with the finding of an inverse correlation with Child-Pugh score, may be interpreted as indicating that IGF- 1 is an early indicator of hepatic dysfunction. Multiple factors directly or indirectly related to hepatic disease may act synergis&lly on circulating IGF-1 concentrations, accounting for its clinical and prognostic significance. Future studies will contribute to better defining the complex relationships between IGF-1 and liver disease. In clinical practice, measurements of IGF1 may be useful in improving the prognostic accuracy of the commonly used tests and formula? in patients with chronic liver disease. SUMMARY

LOW serum IGF-1, characteristic of cirrhotic patients, does not reflect the presence or the severity of energy malnutrition, while it appears to be a good predictor of survival and an early indicator of liver dysfunction. Its high predictive value on survival may be used to improve the commonly used prognostic formulas.

INSULIN-LIKE

GROWTH

FACTOR

1 AND

NUTRITIONAL

STATUS

IN CIRRHOSIS

REFERENCES 1. Mendenhall CL, Anderson S, Weesner RE, Goldberg SJ, Crolic KA. Protein-calorie malnutrition associated with alcoholic hepatitis. Veterans Administration Cooperative Study Group on Alcoholic Hepatitis. Am J Med 1984;76:211 2. Morgan MY. Nutritional aspects of liver and biliary disease. In: McIntyre N, Benhamou JP, Bircher J, Rizzetto M, Rodes J, eds. Oxford textbook of clinical hepatology. Oxford Medical Publications, Oxford, 1991:1339 3. Caregaro L, Alberino F, Amodio P, et al. Malnutrition in alcoholic and virus-related cirrhosis. Am J Clin Nutr 1996;63:602 4. Mendenhall CL, Tosch T, Weesner RE, et al. VA cooperative study on alcoholic hepatitis II: prognostic significance of protein-calorie malnutrition. Am J Clin Nutr 1986;43:213 5. Mendenhall C, Moritz TE, Roselle GA, et al. A study of oral nutritional support with oxandrolone in malnourished patients with alcoholic hepatitis: results of a Department of Veterans Affairs Cooperative Study. Hepatology 1993; 17:564 6. Mezey E, Caballeria J, Mitchell MC, Pares A, Herlong HF, Rod&s J. Effect of parenteral aminoacid supplementation on short-term and long-term outcomes in severe alcoholic hepatitis: a randomized controlled clinical trial. Hepatology 1991; 14:1090 7. Merli M, Romiti A, Riggio 0, Capocaccia L. Optimal nutritional indexes in chronic liver disease. JPEN 1987; 11: 130 8. Baraona E, Pikkarainen P, Salaspuro M, Finkelman F, Leiber CS. Acute effects of ethanol on hepatic protein synthesis and secretion in the rat. Gastroenterology 1980;79:104 reactions in alcoholic liver disease. 9. Paronetto F. Immunological Semin Liver Dis 1993; 13:183 10. Hehir DJ, Jenkins RL, Bistrian BR, Wagner D, Blackbum GL. Nutrition in patients undergoing orthotopic liver transplantation. JPEN 1985;9:695 11. Holly JMP, Wass JAH. Insulin-like growth factors; autocrine, paracrine or endocrine? New perspectives of the somatomedin hypothesis in the light of recent developments. J Endocrinol 1989; 122:611 12. Hartman ML, Veldhuis JD, Thomer MO. Normal control of growth hormone secretion. Horm Res 1993;40:37 LE, Clemmons DR. Changes in plasma 13. Isley WL, Underwood somatomedin-C in response to diets with variable protein and energy content. JPEN 1984; 8:407 14. Untermann TG, Vazquez RM, Slas AJ, Martyn PA, Phillips LS. Nutrition and somatomedin. XIII. Usefulness of somatomedin-C in nutritional assessment. Am J Med 1985;78:228 15. Burgess EJ. Insulin-like growth factor 1: a valid nutritional indicator during parenteral feeding of patients suffering an acute phase response. Ann Clin Biochem 1992;29:137 16. Moller S, Gronbmk M, Main K, Becker U, Skakkebrek NE. Urinary

(For an additional

perspective,

17.

18. 19.

20.

21.

22.

23.

24. 25. 26.

27.

28.

29.

30. 3 1.

growth hormone (U-GH) excretion and serum insulin-like growth factor 1 (IGF-1) in patients with alcoholic cirrhosis. J Hepatol 1993; 17:315 Wu JC, Daughaday WH, Lee SD, et al. Radioimmunoassay of serum IGF-1 and IGF-II in patients with chronic liver diseases and hepatocellular carcinoma with or without hypoglycemia. J Lab Clin Med 1988; 112:58 Eriksson S. Insulin-like growth factor-l in chronic liver disease. J Int Med 1994;235:93 Assad SN, Cunningham GR, Samaan NA. Abnormal growth hormone dynamics in chronic liver disease do not depend on severe parenchymal disease. Metabolism 1990;39:349 Buzzelli G, Dattolo P, Pinzani M, Brocchi A, Roman0 S, Gentilini P. Circulating growth hormone and insulin-like growth factor-I in nonalcoholic liver cirrhosis with or without superimposed hepatocarcinoma: evidence of an altered circadian rhythm. Am J Gastroenterol 1993; 88: 1744 Pugh RNH, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60:646 Infante-Rivard C, Esnaola S, Villeneuve JP. Clinical and statistical validity of conventional prognostic factors in predicting short-term survival among cirrhotics. Hepatology 1987;7:660 Frisancho AR. New standards of weight and body composition by frame size and height for assessment of nutritional status of adults and the elderly. Am J Clin Nutr 1984;40:808 Frisancho AR. New norms of upper limb fat and muscle areas for assessment of nutritional status. Am J Clin Nutr 1981;34:2540 Nompleggi DJ, Bonkovsky HL. Nutritional supplementation in chronic liver disease. Hepatology 1994; 195 18 Abad-Lacruz A, Cabrt E, Gonzalez-Huix F, et al. Routine tests of renal function, alcoholism, and nutrition improve the prognostic accuracy of Child-Pugh score in nonbleeding advanced cirrhotics. Am J Gastroenterol 1993;88:382 Silberman H. Evaluation of nutritional status. In: Silberman H, ed. Parenteral and enteral nutrition, 2nd ed. Norwalk, CT, and San Mateo, CA: Appleton Lange, 1989: 19 Mills PR, Shenkin A, Anthony RS, et al. Assessment of nutritional status and in vivo immune response in alcoholic liver disease. Am J Clin Nutr 1983;38:849 Bentham J, Rodriguez-Amao J, Ross RJM. Acquired growth hormone resistance in patients with hypercatabolism. Horm Res 1993;40:87 McCullough AJ, Tavill AS. Disordered energy and protein metabolism in liver disease. Semin Liv Dis 1991; 11:265 Mtiller MJ, Baker KHW, Selberg 0. Metabolism of energy-yielding substrates in patients with liver cirrhosis. Clin Invest 1994;72:568

see Editorial

Comment

on page 231).