0.34 Perioperative TPN in malnourished patients with gastrointestinal cancer: a randomized clinical trial F. Bozzetti, C. Gavazzi, L. Cozzaglio, R. Mice~i, N. Rossi, L. Mariani and S. Piacenza Istituto Nazioinale Tumori, Milano, Italy, Aim of the study: To evaluate the impact of TPN on post-operative mor- bidity and mortality in malnourished patients affected by gastrointestinal cancer. Patients and methods: Ninety malnourished patients (weight loss _>10% of usual weight), affected by gastric or colon-rectum cancer, were stratified by tumour localization and age (_<or >65 years), and were ran- domized in two groups. TPN group (43 patients) received TPN through a central venous catheter for 10 days before surgery and up to onset of bowel movements. Non-protein calories were planned as 150% of rest- ing energy expenditure, with lipids accounting for 30% of non-protein calories and Kcal:N ratio of 150:1. Control group (CTR; 47 patients) fol- lowed a standard oral diet before surgery, and received a hypocaloric parenteral nutrition post-operatively. Morbidity (mild and severe compli- cations) and hospital mortality were recorded for each patient. Severe complications were those requiring admission to ICU or a second surgi- cal procedure. Results: On the whole, post-operative complications were 57 in 27 patients of CTR group and 25 in 16 patients of TPN group. The fre- quency of patients with mild and severe complications was 32% and 25% in CTR group, 25% and 12% in TPN group. The difference between the two groups was statistically significant at the Mann- Whitney test (P = 0.013). Such difference was mainly due to non-infec- tious complications. Five patients in the CTR group died after severe complications whereas none in TPN group (P = 0.035 at the Fisher's exact test). Conclusions: Perioperative TPN in malnourished patients, affected by gastrointestinal cancer reduces post-operative morbidity and mortality. Data on morbidity are in agreement with the findings observed by the Maastricht Trial 1 in a subset of cancer patients with weight loss _>10%. Reference: 1. Von Meyenfeldt M F, et al. Clin Nutr 1992; 11 : 180-186. 0.35 Increased rates of gluconeogenesis from alanine in weight-losing lung cancer patients S. Halfwerk, P. C. Dagnelie, J. W. O. van den Berg, C. H. K. Hordijk-Luijk, J. L. D. Wattimena, G. R. Swart and J. H. P. Wilson Institute of Internal Medicine II, Erasmus University of Rotterdam, The Netherlands. Introduction: Cancer cachexia is characterized by increased loss of muscle mass and body protein. One of the suggested mechanisms of this phenomenon involves elevated utilization of muscle-derived amino acids, alanine especially, for hepatic glucose synthesis (gluconeogenesis). Gluconeogenesis from alanine has never been quantitatively assessed in patients with specific tumour types. Aim of the study: To compare gluconeogenesis from alanine in lung cancer patients with/without weight loss and healthy subjects. Methods: We studied weight-stable (WS, n = 8) and weight-losing (WL, n = 5) lung cancer patients, without liver metastases or metabolic disease. Healthy subjects were included as controls (n = 10). After an overnight fast, turnover measurements were performed in steady state (60-90 minutes) using a primed-constant infusion of [6,6-2H2]-glucese and [3-13C]-alanine. Results: WL patients had lost an average of 15% (range 8-22%) of pre- illness weight over the previous 6 months. Alanine flux and gluconeogen- esis from alanine were markedly elevated in WL lung cancer patients when compared with WS patients and healthy controls (Table 1). Glucose flux was also significantly higher in WL patients than in control subjects. Table 1 : Turnover rates (mmol/kg.h) in lung cancer patients and controls (mean _+SEM). Control (n = 10) Ca-WS(n = 8) Ca-WL (n = 5) Glucose 0.46 -+ 0.06 0.57 ± 0.03 0.68 -+ 0.4 ~ Alanine 0.39 -+ 0.03 0.32 +_ 0.04 0.52 _+ 0.05 ab Gluconeogenesis from alanine 0.25_+ 0.04 0.24-+ 0.04 0.41 _+ 0.05 ab Note: a) Significantlydifferent from controls;b) from WS (P < 0.05, t-test). Conclusions: Weight loss in lung cancer patients is accompanied by ele- vated glucose and alanine flux, and increased gluconeogenesis from ala- nine. In contrast, values observed in WS patients are not different from those in healthy subjects. Aberrant gluconeogenesis may contribute to loss of muscle protein and body weight in lung cancer. Session 6 - GASTROINTESTINAL SYSTEM 0.36 Combined growth hormone (GH) and epidermal growth factor (EGF) administration upregulates H*-cou- pied oligopeptide transporter PEPT1 gene expression in rabbit jejunum after small bowel resection T. R. Ziegler, R lannoli, L. H. Gu, J. H. Miller, C. K. Ryan and H. C. Sax Emory University School of Medicine, Atlanta, GA, USA and University of Rochester Medical Center, Rochester, New York, USA. Background and purpose: In a rabbit model of massive small bowel resection (SBR), post-operative administration of GH plus EGF signifi- cantly upregulated brush-border amino acid transport in residual small intestine. To investigate amino acid transport capacity ir this animal model at the molecular level, we determined jejunal and ileal mRNA expression of the H+-coupled oligopeptide transporter PEPTI. Methods: Adult rabbits underwent 70% mid-jejunoileal SBR. After recov- ery, either saline (control), GH (0.2mg/kg/day, i.m.), EGF (l .5 #g/kg/hr s.c.) or GH + EGF was given for 7 days. Animals without SBR or hor- monal therapy served as unresected controls. Total RNA extracted from full-thickness intestine was used for Northern blotting. PEPT1 mRNA lev- els were determined using a full-length rabbit PEPT1 cDNA probe, nor- malized to 18S mRNA expression. Results: Expression of the 2.9kb PEPT1 mRNAtranscript was similar in jejunum versus ileum. SBR significantly decreased PEPT1 mRNA in jejunum versus unresected controls (to 52 +__17% of unresected; P = 0.032). In contrast, SBR was associated with a non-significant increase in ileal PEPT1 expression (to 144 _+16% of unresected; NS). PEPT1 mRNA expression after SBR and hormonal therapy is shown below (mean +_ SEM, as a % of SBR controls). Treatmentgroup Jejunum Ileum Control-saline 100 _+ 14 100 -+ 18 GH 117-+16 71 -+4 EGF 147-+19 98+_19 GH + EGF 220 -+51 t 105 _+ 11 1"= P< 0.01 vs control. Conclusions: PEPT1 mRNA is abundantly expressed in rabbit small intestine. PEPT1 gene expression in residual rabbit jejunum and ileum is differentially regulated by SBR itself and by post-SBR hormonal treat- ment. Administration of GH or EGF alone have no effect on gut PEPT1 mRNA levels. However, combined treatment with GH + EGF synergisti- cally upregulates expression of this major peptide transporter mRNA in residual jejunum, but not in ileum. These data represent the first in vivo demonstration of growth factor-stimulated peptide transporter gene expression, and suggest a possible therapeutic role for combined GH + EGF as a method to enhance oligopeptide transport in proximal small intestine after massive SBR. 0.37 Preventive enteral administration of ornithine alpha-ketoglutarate improves intestinal mucosal repair after transient ischemia in rats F. Raul, B. Duranton, E Gosse, M. Galluser, C. Bergmann and R. Schleiffer CJF INSERM 95-09, IRCAD, 67000-Strasbourg, France. The polyamines are involved in repair processes after intestinal ischemia. The objective of the present work was to investigate whether the preven- tive enteral administration of a polyamine precursor like ornithine alpha- ketoglutarate (OKG) has a beneficial effect on the morphological and 10