Increased Circulating Leptin in Alcoholic Cirrhosis: Relation to Release and Disposal

Transcription

1 Increased Circulating Leptin in Alcoholic Cirrhosis: Relation to Release and Disposal JENS H. HENRIKSEN, 1 JENS JUUL HOLST, 2 SØREN MØLLER, 1 KIM BRINCH, 1 AND FLEMMING BENDTSEN 3 Abbreviations: BMI, body mass index; BMI, ascitic-free body mass index; NS, not significant. From the 1 Departments of Clinical Physiology and 3 Gastroenterology, Hvidovre Hospital; and the 2 Department of Medical Physiology, The Panum Institute, University of Copenhagen, Copenhagen, Denmark. Received November 30, 1998; accepted March 5, Supported by The John and Birthe Meyer Foundation and an H:S research grant. Address reprint requests to: Jens H. Henriksen, M.D., Department of Clinical Physiology & Nuclear Medicine, 239, Hvidovre Hospital, University of Copenhagen, DK-2650 Hvidovre, Denmark. jens.h.henriksen@hh.hosp.dk; fax: Copyright 1999 by the American Association for the Study of Liver Diseases /99/ $3.00/ Leptin is a cytokine peptide that decreases appetite and thereby food intake and increases energy expenditure. It is produced in fat cells, but recent animal experiments have shown expression of leptin in modified stellate hepatic cells. Because a change in circulating leptin in cirrhosis could be caused by an altered production rate, altered disposal rate, or both, the present study was undertaken to identify regions of leptin overflow into the blood stream and regions of leptin extraction. Patients with alcoholic cirrhosis (n 16) and control patients without liver disease (n 12) were studied during catheterization with elective blood sampling from different vascular beds. Blood samples for leptin determination (radioimmunoassay) were taken simultaneously from artery/hepatic vein, artery/renal vein, artery/iliac vein, and artery/cubital vein. Patients with cirrhosis had significantly increased circulating leptin (7.3 vs. control 2.6 ng/ml, P F.002) that correlated directly to ascitic-free body mass index (r 0.71, P F.005). A significant renal extraction ratio of leptin was observed in control patients (0.16) and in patients with cirrhosis (0.07), but the latter value was significantly lower than in the control patients ( 44%, P F.05) and inversely correlated to serum creatinine (r 0.60, P F.05). A significant, but equal, hepatosplanchnic extraction of leptin was observed in cirrhotic patients and control patients (0.08 vs. 0.07). In patients with cirrhosis a significant cubital venous-arterial difference in leptin was observed, but not in control patients. The iliac venous/arterial leptin ratio was significantly above 1.0 in both groups and of similar size (1.16 vs. 1.15), but a higher difference in concentration was found in the cirrhotic patients ( 33%, P F.05). The spillover rates of leptin in cirrhotic patients may be even higher than estimated from the increased systemic veno-arterial gradients. In conclusion, the elevated circulating leptin in patients with cirrhosis is most likely caused by a combination of decreased renal extraction and increased release from subcutaneous abdominal, femoral, gluteal, retroperitoneal pelvic, and upper limb fat tissue areas. The hepatosplanchnic bed drained through hepatic veins could not be identified as a source of increased circulating leptin in cirrhosis, but a contribution by the portosystemic collateral flow cannot be excluded. (HEPATOLOGY 1999;29: ) Leptin is a cytokine-type peptide hormone (16 kd) that is produced in fat cells. 1,2 It was discovered in 1994 by Friedman et al., 1 and several studies have substantiated that leptin decreases appetite and thereby food intake and increases the expenditure of energy in normal animals and humans. 2-7 It appears to be important in the regulation of metabolism, insulin secretion and sensitivity, and body weight. 6,8-12 In normal humans, the circulating level of leptin is higher in women than in men, 6,7 and leptin production in fat cells is affected by glucose and fat metabolism Circulating leptin is related to the adipose tissue mass, and in most cases high leptin levels correlate directly with the body mass index (BMI). 2,7 This is most strikingly observed in obese subjects. 7 However, an exception is patients with renal insufficiency, in which high circulating levels of leptin are seen in the absence of excess body fat Recently, an increased circulating level of leptin was reported in women with alcoholic cirrhosis, but not in men. 19 However, when expressed per unit fat mass, leptin was elevated in both women and men with cirrhosis. 19 In addition, recent animal experiments have indicated that, after liver injury, hepatic stellate cells transdifferentiate with the occurrence of leptin messenger RNA. 20 Moreover, leptin seems to be related to certain types of hepatic steatosis. 11 As changed circulating leptin in cirrhosis could be caused by an increased production rate, a decreased disposal rate, or both, the present study was undertaken to identify vascular regions of leptin overflow into the blood stream and regions of leptin extraction. We studied patients with alcoholic cirrhosis and control patients without liver disease during liver vein catheterization with elective blood sampling in different vascular beds. Moreover, the circulating level of leptin was related to body size, disease stage, and splanchnic and systemic hemodynamics. PATIENTS AND METHODS Study Population. The study population comprised 16 patients with cirrhosis referred to hemodynamic investigation to diagnose and quantify portal hypertension. All patients had biopsy-verified cirrhosis. The age range was 37 to 71 years with an average of 53 years. All had a history of high alcohol intake, ie, a consumption exceeding 50 g per day for more than 5 years. None of the patients had experienced recent gastrointestinal bleeding or had encephalopathy above grade I. All patients were abstaining from alcohol and had

2 HEPATOLOGY Vol. 29, No. 6, 1999 HENRIKSEN ET AL TABLE 1. Clinical, Biochemical, and Hemodynamic Data on 16 Patients With Alcoholic Cirrhosis and 12 Control Patients Without Liver Disease Undergoing Hemodynamic Investigation Control Patients Cirrhotic Patients Age (years) Body height (cm) Body weight (kg) * Ideal body weight (kg) Sex (M/F) 7/5 10/6 BMI (kg/m 2 ) (23 0.9) Child-Turcotte class A/B/C 3/6/7 Hemoglobin ( , mmol/l) P-glucose ( mmol/l) S-bilirubin (2-17, µmol/l) S-aspartate aminotransferase (10-40, U/L) S-alkaline phosphatase (50-275, U/L) S-sodium ( , mmol/l) S-creatinine (49-121, µmol/l) S-albumin ( , µmol/l) Coagulation factor 2, 7, 10 ( , index) Immunoreactive insulin (50-180, pmol/l) Hepatic blood flow (L/min) ICG clearance ( , ml/min) Galactose elimination capacity (M 1.7, F 1.4 mmol/min) Hepatic venous pressure gradient (0-5, mm Hg) Cardiac output (L/min) Plasma volume (L) Mean arterial pressure (mm Hg) Systemic vascular resistance (dyn sec/cm 5 ) 1, NOTE. Data expressed as mean SEM. *P.05. P.01. BMI. P.005 compared to control patients. no signs of withdrawal symptoms at the time of the study. The clinical and biochemical characteristics are summarized in Table 1. According to the modified Child-Turcotte classification, three patients were in class A, six in class B, and seven in class C. Three patients had signs of early hepatic nephropathy. None of the patients had signs of heart failure, organic renal disease, diabetes, cancer, or any other major disease. Ultrasonography showed the presence of ascites in nine patients, for which they were receiving diuretics (spironolactone 100 mg and furosemide 40 to 80 mg per day) and had been put on a sodium-restricted diet of 40 mmol per day. None of the patients received any other medical treatment. The diet of the patients without fluid retention was not restricted. BMI was determined as the actual body weight relative to the square of the body height (BMI, kg/m 2 ) and by applying ascitic-free body weight obtained either by subtraction of the weight of the ascitic fluid removed by paracentesis, or as the body weight determined later during successful diuretic treatment (BMI, kg/m 2 ). There were 12 control patients with minor disorders (suspicion of intestinal ischemia, although unconfirmed, irritable bowel syndrome, postcholecystectomy pain) who underwent a diagnostic catheterization. The age range was 45 to 79 years with an average of 57 years. All the patients consented to participate in the study, which was approved by the Ethics Committee for Medical Research in Copenhagen and performed in accordance with the guidelines established in The Helsinki Declaration II. No complications or side effects were encountered during the study. Catheterization. Patients with cirrhosis and control patients underwent liver vein catheterization, which was performed in the morning after an overnight fast and at least 1 hour in the supine position as described elsewhere. 21 In brief, a Cournand catheter (7F) or Swan-Ganz catheter (7F) was guided under local anesthesia to the right hepatic and renal veins through the femoral route under fluoroscopic control. A small indwelling polyethylene catheter (5F) was introduced into the femoral artery by the Seldinger technique. Simultaneous blood samples for leptin determination were taken from the following locations: artery/hepatic vein, artery/renal vein, artery/iliac vein, and artery/cubital vein, as described earlier. 21,22 The hepatosplanchnic disposal rate of leptin was estimated as (C a C hv ) multiplied by hepatic plasma flow, in which C a and C hv are leptin concentrations in arterial and hepatic venous plasma, respectively. The hepatosplanchnic and renal extraction ratios were determined as E (C a C hv )/C a and E (C a C rv )/C a, respectively, in which C rv is the leptin concentration in renal venous plasma. The hepatosplanchnic clearance of leptin was estimated as the extraction ratio multiplied by the hepatic plasma flow. Hepatic plasma and blood flow was measured by the indocyanine green constant infusion technique, as described elsewhere. 23 The clearance of indocyanine green was determined as the infusion rate divided by the arterial indocyanine green concentration. 23 The galactose elimination capacity (estimate of functional hepatocyte mass) was determined according to Tygstrup. 24 Cardiac output and plasma volume were measured by the indicator dilution technique after a bolus injection of 150 KBq of 125 I-labelled human serum albumin into the right atrium, as described earlier. 25 Wedged and free hepatic vein pressures were determined in different vessels and the mean values of repetitive measurements were used. Systemic vascular resistance was determined as (mean arterial pressure minus right atrial pressure) divided by cardiac output. 26 Relation to Time of Blood Sampling. To test a possible effect of time of blood sampling on the circulating level of leptin, several arterial samples were taken in the time interval to hours, and the relative leptin concentration versus time was evaluated by regression analysis. Relation to Food Consumption. To test the potential effect of food consumption, a test meal (consisting of 4,000 KJ) was given to six control patients followed by simultaneous arterial and hepatic venous sampling for leptin determination at intervals of 15 minutes up to 45 minutes. Biochemical Analysis. Routine biochemical tests were performed in an autoanalyzer (SMAC, Technicon Instruments, Tarrytown, NY) and a Beckman Glucose Analyzer 1 (Beckman Instruments Inc., Fullerton CA). Immunoreactive insulin in plasma was measured according to the principles described by Albano et al. 27 against standards of human insulin. The tracer was human insulin monoiodinated in position A14 (a gift from Novo-Nordisk A/S, Bagsværd, Denmark). The guinea pig antibody employed (code nr 2004) crossreacts with proinsulin and splitinsulins. Between-assay coefficient of variation was below 5%, and the detection limit was below 5 pmol/l. Blood samples for leptin analysis were collected in prechilled tubes (6 IU heparin/500k IU Approtinin per ml), centrifuged at 4 C, and stored at 25 C until analyzed. The analysis of leptin was performed by a radioimmunoassay as described by Ma et al. 28 with the commercially available radioimmunoassay kit for human leptin (catalogue #HL-81K) from Linco Research, Inc. (St. Louis, MO). The detection limit in our hands was 0.5 ng/ml. Within- and between-assay coefficients of variation ranged from 3.4% to 6.2% and 3.6% to 8.3%, respectively, at a concentration range between 1.9 and 25.6 ng/ml. The assay was linear in the range 0.5 to 100 ng/ml. For the batches utilized in the present study, all high and low control samples provided by the manufacturer conformed with expected values and the standard curves showed ED20, ED50, and ED80 values, as indicated by the manufacturer.

3 1820 HENRIKSEN ET AL. HEPATOLOGY June 1999 Statistical Evaluation. Data are expressed as mean and standard error of the mean. Statistical analysis was performed by unpaired/ paired Student s t test or Mann-Whitney/Wilcoxon rank tests. Correlations between independent variables were performed with the Pearson regression test (method of least squares) or by the Spearman s rank correlation test. P.05 was considered to be significant. RESULTS Circulating Leptin. As shown in Fig. 1, patients with cirrhosis had significantly increased circulating levels of leptin ( vs. control patients ng/ml, P.002). When the women and men were tested separately, significant differences were still present (P.005 and P.05, respectively). There was a tendency towards higher concentrations in women than in men (8.7 vs. 6 ng/ml, not significant [NS]). When the three patients with hepatic nephropathy were excluded, circulating leptin was still significantly elevated ( ng/ml, P.01). Relation to Body Size. Body height and ideal body weight were similar in the patients with cirrhosis and the control patients (Table 1), but there was a somewhat higher body weight in the former group (P.05), in part because of the presence of ascites. BMI was significantly higher in the cirrhotic patients than in the control patients (25 vs. 20 kg/m 2, P.01), but not significantly different when estimated from ascitic-free body mass index (BMI ) (BMI 23 vs. FIG. 2. Relation between circulating leptin and BMI in control patients (open circles: r 0.84, P.001) and BMI in patients with cirrhosis (closed circles: r 0.71, P.005). FIG. 1. Circulating leptin in control patients (n 12) and patients with cirrhosis (n 16). When subdivided according to gender the difference is still significant (women P.005; men P.05). 20 kg/m 2, NS). Significant direct correlations between BMI and circulating leptin were observed in both groups (cirrhotic patients: BMI r 0.56, P.05, BMI r 0.71, P.005; control patients: r 0.84, P.001), see Fig. 2. Circulating leptin was related to body weight in the control patients (r 0.65, P.05), but not in the cirrhotic patients (r 0.25, NS). Relation to Biochemistry and Hemodynamics. The biochemical and splanchnic and systemic hemodynamic variables are summarized in Table 1. Circulating leptin had a significant, or borderline significant, direct relation to plasma glucose (r 0.63, P.05), immunoreactive insulin (r 0.48, P.06), and wedged hepatic vein pressure (r 0.45, P.07). No significant correlations were found between circulating leptin on the one hand and the Child score and other hemodynamic and biochemical variables on the other (r 0.33 to 0.39, NS). Relation to Time of Blood Sampling and Food Consumption. Circulating leptin did not change significantly with time during the investigation period from to hours (control patients: r 0.13, NS; patients with cirrhosis: r 0.2, NS), see Fig. 3. The within-control coefficient of variation in this period was 6.8% and the within-patients with cirrhosis coefficient of variation was 8.4% (NS). In the six control patients receiving a test meal, no significant change was observed in mean arterial leptin (coefficient of variation 6.1%, NS) and mean hepatic venous leptin (11%, NS), see Table 2.

4 HEPATOLOGY Vol. 29, No. 6, 1999 HENRIKSEN ET AL FIG. 4. Venous-arterial concentration differences of leptin in different vascular beds in control patients and patients with cirrhosis. (Concentration differences significantly different from zero: *P.05, **P.01). FIG. 3. (A) Relative arterial leptin concentration versus sampling time in control patients (r 0.13, ns). B: patients with cirrhosis (r 0.2, ns). TABLE 2. Arterial and Hepatic Venous Plasma Concentration of Leptin in Six Control Subjects Before and After a Test Meal After Test Meal Before Test Meal 15 min 30 min 45 min Artery NS Hepatic vein NS NOTE. Test meal consisted of 4,000 kj. Concentration of leptin is expressed as ng/ml. Data expressed as mean SEM. Abbreviation: NS, not significant. Leptin in Different Vascular Beds. As shown in Fig. 4, highly significant differences were seen in the leptin concentrations in different vessels. Thus, in control patients, as well as in cirrhotic patients, a net leptin disposal was observed in the kidney and hepatosplanchnic bed. In contrast, a highly significant leptin overflow was observed in the iliac vein, and in the cirrhotic patients, also in the cubital vein. Kidney. A significant arteriorenal venous net extraction of leptin (E , P.005) was seen in the control patients, see Fig. 5. The arteriorenal venous extraction was also significant in the patients with cirrhosis (E , P.01), but this value was substantially lower than in the control subjects ( 44%, P.05), and inversely correlated to serum creatinine (r 0.60, P.05). The difference in the arteriorenal venous concentration was also smaller in the cirrhotic patients than in the control patients, but the difference between the groups did not reach statistical significance (0.38 vs ng/ml, NS). Hepatosplanchnic System. A significant arteriohepatic venous net extraction of leptin was observed in the control patients (E , P.05) and in the patients with cirrhosis (E , P.01), see Table 3 and Fig. 6. Hepatosplanchnic extraction ratios, as well as hepatosplanchnic clearances, of leptin were similar in the two groups and no correlation was found to wedged hepatic vein pressure (r 0.14, NS). However, because of the higher circulating level of leptin in the cirrhotic patients, the actual hepato-

5 1822 HENRIKSEN ET AL. HEPATOLOGY June 1999 FIG. 5. Relation between the renal extraction ratio of leptin [(C A C rv )/ C A, see text] and s-creatinine in control patients (open circles: r 0.21, NS) and patients with cirrhosis (closed circles: r 0.60, P.05). The renal extraction ratio of leptin in cirrhotic patients is significantly lower than in control patients vs. ( vs , P.05). TABLE 3. Net Hepatosplanchnic Extraction Ratio, Clearance, and Disposal Rate of Leptin in Control Patients and Patients With Cirrhosis Control Patients Cirrhotic Patients P Extraction ratio NS Clearance (ml/min) NS Disposal rate (ng/min) NOTE. Data expressed as mean SEM. Abbreviation: NS, not significant. FIG. 6. Venous/arterial concentration ratios of leptin in different vascular beds in control patients and patients with cirrhosis. (Ratios significantly different from 1.00: *P.05, **P.01, ***P.001). splanchnic net removal of leptin was higher in the hepatosplanchnic bed drained through the hepatic veins in these patients than in the control patients ( vs ng/min, P.05). Iliac Vein. As shown in Fig. 4, the leptin concentration in the iliac vein was significantly higher than that in the artery both in the control patients (P.05) and in the cirrhotic patients (P.005). The iliac vein/arterial concentration ratio of leptin was similar in the two groups ( vs , NS), but the difference in the iliac venousarterial concentration was higher in the cirrhotic patients than in the control patients ( vs ng/ml, P.05), see Figs. 4 and 6. Cubital Vein. No significant cubital venous-arterial difference was observed in the control patients ( ng/ml, NS), see Fig. 4. However, in the patients with cirrhosis, the cubital venous/arterial concentration ratio of leptin was almost similar to that of the iliac venous/arterial ratio, and a significant difference in cubital venous arterial concentration ( ng/ml, P.05) and concentration ratio ( , P.01) were found in these patients, see Figs. 4 and 6. DISCUSSION The present study shows that: (1) patients with alcoholic cirrhosis have increased circulating leptin, irrespective of gender. (2) A significant leptin extraction occurs in the hepatosplanchnic bed and in the kidney in both control patients and patients with cirrhosis, but the renal extraction is significantly smaller in the latter group. (3) A significant spillover of leptin into the iliac vein occurs in both control patients and cirrhotic patients, the latter also having a significant spillover into the peripheral (cubital) vein. Circulating (arterial) leptin was substantially increased in the patients with cirrhosis, with an average value more than twice that of the control patients. It is generally accepted that women have higher circulating leptin than men because of their higher mass of adipose tissue. 2,6,7 In a recently published study on serum leptin in cirrhosis, substantially higher values of leptin were found in cirrhotic women than in men. 19 A tendency toward higher values in cirrhotic women was also present in our study, but it did not reach statistical significance, probably because of the small number, when subdividing according to gender. In the study by McCullough et al. 19 cubital venous samples were applied. As we found an elevated cubital venous/arterial leptin ratio, especially in the women, arterial sampling may contribute to the somewhat smaller gender difference found in our study.

6 HEPATOLOGY Vol. 29, No. 6, 1999 HENRIKSEN ET AL In the control patients, we found a direct relation between circulating leptin and body weight and BMI, but in the cirrhotic patients only a relation to BMI could be shown. However, BMI is not likely to reflect adipose tissue specifically even when adjusted for ascitic fluid, as it is generally accepted that cirrhotic patients are wasted and may be overhydrated. Thus, this relation may at least in part be a reflection of the severity of the disease. We did not determine body composition, but in the study of McCullough et al. 19 higher values of circulating leptin were found in cirrhotic patients, also when adjusted for reduced adipose tissue mass. Circulating leptin is not a static reflection of body fat stores as leptin expression and secretions are affected by actual metabolic and physiological signals, such as blood glucose, fat acids, and steroid hormones. 2 It has recently been shown that plasma leptin concentrations decrease with fasting and increase with food consumption. 29,30 Moreover, there may be a circadian rhythm. 31,32 In the present study, circulating leptin did not change during the catheterization procedure, and both our control patients and cirrhotic patients were fasting. We did not find that food ingestion played a significant role in a subgroup of control patients. Thus, the present differences in vascular concentrations are not likely to be caused by a different food intake before the study or by time-dependent variations. Apart from obesity, patients with diabetes mellitus and end-stage renal disease have elevated levels of plasma leptin ,33 Recent studies of the distribution of the leptin receptor in tissues have identified expression of this receptor in lung, kidney, pancreatic cells, and hepatocytes. 34,35 Plasma half-life and animal experiments suggest that leptin and other peptide hormones are degraded by proximal renal tubular cells. 36,37 However, nothing is known of the relation of leptin disposal to leptin receptors, several isoforms of which are present, e.g., long and short leptin receptor isoforms. Extraction and spillover of leptin into different vascular beds have not been described in humans before. The multiorgan vein catheterization technique with simultaneous sampling from organ inlet (artery) and outlet (vein) is a wellestablished method of identifying regions of disposal and production/release/secretion of bioactive substances. 21,23,38 A significant leptin extraction was identified in both normal and cirrhotic kidneys. The renal extraction ratio of leptin in cirrhotic patients was only half that found in control patients. Renal plasma flow was not determined in the present study, but it is well known that patients with cirrhosis have decreased renal perfusion, even when serum creatinine is normal. 39,40 The renal plasma clearance (extraction ratio multiplied by renal plasma flow) may therefore be even lower in patients with cirrhosis. Thus, decreased renal disposal of leptin may contribute to the elevated level of circulating leptin observed in patients with cirrhosis. The presence of inverse correlations to serum creatinine supports this interpretation. Recent animal experiments have suggested that modified hepatic stellate cells may express leptin. 20 It could therefore be expected that the cirrhotic liver produces leptin and contributes to the elevated circulating leptin with a significant hepatic venous spillover. However, we found no indication of a hepatosplanchnic net production of leptin in patients with cirrhosis. On the contrary, we found a net extraction of a magnitude similar to that in the control patients. Thus, increased hepatic production of leptin is not the cause of elevated circulating leptin in cirrhosis. Leptin produced in stellate cells may act locally in the liver rather than at distant sites. Therefore, the present results do not contradict the study of Potter et al. 20 It is possible that, in a serial system as complex as the hepatosplanchnic system, both production and disposal of bioactive substances may take place within the system (e.g., insulin, glucagon, vasoactive intestinal polypeptide). This may also apply to leptin, e.g., from omental fat tissue and gastric fundic glandular cells. 11,29,30 Portosystemic shunting was not quantified in the present study. As collateral veins (azygos vein) were not catheterized, it cannot be completely excluded that increased gastrointestinal secretion of leptin into the circulation takes place, but the finding of a normal hepatosplanchnic extraction and clearance of leptin speaks against this possibility. On the other hand, the borderline significant direct relation between circulating leptin and wedged hepatic vein pressure may suggest a relation to the deranged hepatosplanchnic hemodynamics. We identified a significant spillover of leptin into the iliac veins of both control patients and patients with cirrhosis. In fact, the iliac venous/arterial ratio was similar in both groups, but a higher difference in concentration was found in the cirrhotic patients. This suggests that a substantial spillover of leptin coming from abdominal, retroperitoneal pelvic, gluteal, and femoral fat tissue contributes appreciably to the elevated circulating level of leptin. In addition, some portosystemic collateral flow may go to the iliac veins and potentially contribute to the elevated circulating leptin. Superselective catheterization of abdominal adipose tissue veins 41,42 was beyond the possibilities in the present study, but the concentration of leptin would most likely have been even higher here. As the cirrhotic patients were hyperkinetic, as evidenced by their significantly increased plasma volume and cardiac output and decreased systemic vascular resistance, the finding of increased systemic venoarterial gradients may cover even higher production rates in cirrhotic patients. Moreover, in cirrhotic patients leptin may come from more widespread areas, as indicated by the presence of a significant difference in the cubital venous-arterial concentration in these patients. Leptin is expressed, synthesized, and secreted by fat cells. 6,7,31 Major secretory stimuli are cytokines (i.e., interleukin-1 and tumor necrosis factor ), corticosteroids, sex steroids, and insulin. 12,14,32,43,44 A number of these stimuli are present in excess in cirrhosis, and we found, as could be expected, 6,13 a direct relation to fasting plasma insulin and glucose. The elevated leptin could contribute to the anorexia, insulin resistance, hypermetabolism, and atrophy found in cirrhotic patients. Clarification of these aspects must await future investigations. In conclusion, the elevated circulating levels of leptin in patients with alcoholic cirrhosis is most likely caused by a combination of decreased renal disposal and increased release from subcutaneous abdominal, femoral, gluteal, retroperitoneal pelvic, and upper limb areas. We did not identify the hepatosplanchnic area as a source of increased circulating leptin in cirrhosis, although a contribution by the portosystemic collateral blood flow cannot be excluded. The pathophysiological consequences of elevated circulating leptin (decreased appetite and thus food intake, increased energy expenditure, insulin resistance, severity of steatosis) will require further studies.

7 1824 HENRIKSEN ET AL. HEPATOLOGY June 1999 Acknowledgment: The authors wish to express their gratitude to Hanne B. Hansen, Bente Henriksen, Kirsten Nielsen, and Rosemary Sørensen for their skillful assistance. REFERENCES 1. Friedman JM. Leptin, leptin receptors and the control of body weight. Eur J Med Res 1997;2: Frühbeck G, Jebb SA, Prentice AM. Leptin: physiology and pathophysiology. Clin Physiol 1998;18: Tartaglia LA, Dembski M, Weng X, Deng N, Culpepper J, Devos R, Richards GJ, et al. Identification and expression cloning of a leptin receptor OB-R. Cell 1995;83: Caro JF, Sinha MK, Kolaczynski JW, Zhang PL, Considine RV. Leptin: the tale of an obesity gene. Diabetes 1996;45: Vaisse C, Halaas JL, Horvath CM, Darnell JE, Stoffel M, Friedman JM. Leptin activation of stat 3 in the hypothalamus of wild-type and ob/ob mice but not db/db mice. Nat Genet 1996;14: Kennedy A, Gettys TW, Watson P, Wallace P, Ganaway E, Pan Q, Garvey WT. The metabolic significance of leptin in humans: gender-based differences in relationship to adiposity, insulin sensitivity, and energy expenditure. J Clin Endocrinol Metab 1997;82: Considine RV, Sinha MK, Heiman ML, Kriauciunas A, Stephens TW, Nyce MR, Ohannesian JP, et al. Serum immunoreactive-leptin concentrations in normal weight and obese humans. N Engl J Med 1996;334: Campfield LA, Smith FJ, Guisez Y, Devos R, Burn P. Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 1995;269: Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT, Rabinowitz D, Lallone RL, et al. Weight-reducing effects of the plasma protein encoded by the obese gene. Science 1995;269: Saladin R, De Vos P, Guerre-Millo M, Leturque A, Girard J, Staels B, Auwerx J. Transient increase in obese gene expression after food intake or insulin administration. Nature 1995;377: Kaplan LM. Leptin, obesity, and liver disease. Gastroenterology 1998;115: Grunfeld C, Zhao C, Fuller J, Pollock A, Moser A, Friedman J, Feingold KR. Endotoxin and cytokines induce expression of leptin, the OB gene product, in hamsters. A role for leptin in the anorexia of infection. J Clin Invest 1996;97: Kolacynski JW, Nyce MR, Considine RV, Boden G, Nolan JJ, Henry R, Mudallar SR, et al. Acute and chronic effects of insulin on leptin production in humans in vivo and in vitro. Diabetes 1996;45: Masuzaki H, Ogawa Y, Hosoda K, Miyawaki T, Hanaoka I, Hiraoka J, Yasuno A, et al. Glucocorticoid regulation of leptin synthesis and secretion in humans: elevated plasma leptin levels in Cushing s syndrome. J Clin Endocrinol Metab 1997;82: Halleux CM, Servais I, Reul BA, Detry R, Brichard SM. Multihormonal control of ob gene expression and leptin secretion from cultured human visceral adipose tissue: increased responsiveness to glucocorticoids in obesity. J Clin Endocrinol Metab 1998;83: Merabet E, Dagogo-Jack S, Coyne DW, Klein S, Santiago JV, Hmiel SP, Landt M. Increased plasma leptin concentration in end-stage renal disease. J Clin Endocrinol Metab 1997;82: Howard JK, Lord GM, Clutterbuck EJ, Ghatei MA, Pusey CD, Bloom SR. Plasma immunoreactive leptin concentration in end-stage renal disease. Clin Sci 1997;93: Shoji T, Nishizawa Y, Emoto M, Maekawa K, Hiura Y, Tanaka S, Kawagishi T, et al. Renal function and insulin resistance as determinants of plasma leptin levels in patients with NIDDM. Diabetologia 1997;40: McCullough AJ, Bugianesi E, Marchesini G, Kalhan SC. Genderdependent alterations in serum leptin in alcoholic cirrhosis. Gastroenterology 1998;115: Potter JJ, Womack L, Mezey E, Anania FA. Transdifferentiation of rat hepatic stellate cells results in leptin expression. Biochem Biophys Res Commun 1998;244: Henriksen JH, Ring-Larsen H, Kanstrup I-L, Christensen NJ. Splanchnic and renal elimination and release of catecholamines in cirrhosis. Evidence of enhanced sympathetic nervous activity in patients with decompensated cirrhosis. Gut 1984;25: Bentsen KD, Henriksen JH, Boesby S, Jensen LT, Hørslev-Petersen K, Lorenzen I. Hepatic extraction of the aminoterminal propeptide of type III procollagen before and after bile duct ligation in pigs. Liver 1991;11: Henriksen JH, Winkler K. Hepatic blood flow determination a comparison of 99m Tc-diethyl-IDA and indocyanine green as hepatic blood flow indicators in man. J Hepatol 1987;4: Tygstrup N. Determination of the hepatic galactose elimination capacity after a single intravenous injection in man. Acta Physiol Scand 1963;58: Henriksen JH, Bendtsen F, Sørensen TIA, Stadeager C, Ring-Larsen H. Reduced central blood volume in cirrhosis. Gastroenterology 1989;97: Møller S, Wiinberg N, Henriksen JH. Non-invasive 24-hour ambulatory arterial blood pressure monitoring in cirrhosis. HEPATOLOGY 1995;22: Albano JDM, Ekins RP, Maritz G, Turner RC. A sensitive precise radioimmunoassay of serum insulin relying on charcoal separation of bound and free hormone moieties. Acta Endocrinol 1972;70: Ma Z, Gingerich RL, Santiago JV, Klein S, Smith CH, Landt M. Radioimmunoassay of leptin in human plasma. Clin Chem 1996;42: Bado A, Levasseur S, Attoub S, Kermorgant S, Laigneau J-P, Bortoluzzi M-N, Moizo L, et al. The stomach is a source of leptin. Nature 1998;394: Barrachina MD, Martinez V, Wang L, Wei JT, Taché Y. Synergistic interaction between leptin and cholecystokinin to reduce short-term food intake in mice. Proc Natl Acad Sci U S A 1997;94: Sinha MK, Ohannesian JP, Heiman ML, Kriauciunas A, Stephens TW, Magosin S, Marco C, et al. Nocturnal rise of leptin in lean, obese and non-insulin-dependent diabetes mellitus subjects. J Clin Invest 1996;97: Sinha MK. Human leptin: the hormone of adipose tissue. Eur J Endocrinol 1997;136: Hanley AJG, Harris SB, Gao XJ, Kwan J, Zinman B. Serum immunoreactive leptin concentrations in a Canadian aboriginal population with high rates of NIDDM. Diabetes Care 1997;20: Lee GH, Proenca R, Montez JM, Carroll KM, Darvishzadeh JG, Lee JI, Friedman JM. Abnormal splicing of the leptin receptor in diabetic mice. Nature 1996;379: Tartaglia LA. The leptin receptor. J Biol Chem 1997;272: Cumin F, Baum H-P, Levens N. Leptin is cleared from the circulation primarily by the kidney. Int J Obesity 1996;20: Cumin F, Baum H-P, De Gasparo M, Levens N. Removal of endogenous leptin from the circulation by the kidney. Int J Obesity 1997;21: Guevara M, Gines P, Jimenez W, Sort P, Fernandez-Esparrach G, Escorsell A, Bataller R, et al. Increased adrenomedullin levels in cirrhosis: relationship with hemodynamic abnormalities and vasoconstrictor systems. Gastroenterology 1998;114: Ring-Larsen H. Hepatic nephropathy related to haemodynamics. Liver 1983;3: Epstein M. Hepatorenal syndrome. In Epstein M (ed): The Kidney in Liver Disease (ed 4). Philadelphia: Hanley & Belfus, Inc., 1996, pp Lonnroth P, Jansson PA, Smith U. Adipose tissue metabolism in humans determined by vein catheterization and microdialysis techniques. Am J Physiol 1994;267:E1035-E Simonsen L, Bülow J, Madsen J. Adipose tissue metabolism in humans determined by vein catheterization and microdialysis techniques. Am J Physiol 1994;266:E357-E Friedman JM. Leptin, leptin receptors, and the control of body weight. Nutr Rev 1997;56:S38-S Hafner SM, Miettinen H, Karhapaa, Mykkanen Æ, Laakso M. Leptin concentrations, sex hormones, and cortisol in non-diabetic men. J Clin Endocrinol Metab 1997;82: