Faculty Bibliography

BACKGROUND: Patients with upper gastrointestinal (GI) tract malignancies are at risk for malnutrition and postoperative morbidity and mortality. We examined the protein kinetic effects of early enteral feeding in this population and compared it with results in patients receiving IV fluid. METHODS: Twenty-nine patients undergoing resection of an upper GI tract malignancy were prospectively randomized to either enteral feeding starting on postoperative day (POD) 1 via a jejunostomy tube (FEED, n = 12) or IV fluid (IVF, n = 17). On POD5, all patients underwent a protein metabolic study using [3H]phenylalanine to determine forearm skeletal muscle (nmol phenylalanine/100 g/min) protein net balance. Free fatty acids (FFA, mEq/dL) and insulin levels (mU/mL) were measured. RESULTS: Protein net balance was significantly less negative in the FEED group compared with the IVF group (-1.4 +/- 0.8 vs -5.0 +/- 1.4, p < .05). Respiratory quotient was significantly increased in patients receiving enteral feeding (0.85 +/- 0.02 vs 0.78 +/- 0.02 FEED vs IVF, p < .05). FFA levels were significantly decreased in the FEED group (0.36 +/- 0.04 vs 0.85 +/- 0.07, p < .05). Insulin levels were significantly elevated in the FEED group (19.8 +/- 4.5 vs 9.3 +/- 0.8, p < .05). Insulin levels correlated with amino acid fluxes. CONCLUSIONS: Postoperative enteral nutrition in upper GI cancer patients results in an improvement in protein kinetic net balance and amino acid flux across peripheral tissue. In addition, insulin levels are elevated, and this elevation correlates with amino fluxes across the forearm. By improving peripheral protein kinetics, early postoperative enteral nutrition may potentially contribute to a decrease in postoperative morbidity and mortality in upper gastrointestinal cancer patients

The role of human growth hormone (hGH) as a nutritional adjunct for cancer patients is controversial because of its potential mitogenic effects on tumor growth. No studies to date have examined the effect of hGH on human tumor response in vivo. In Vitro: Athymic mice were injected (s.c.) daily with hGH (GH, n=14) or saline (CTL, n=14). On Day 10, serum was collected and added to human pancreatic carcinoma cells in culture. In Vivo: Athymic mice were inoculated (s.c.) with human pancreatic carcinoma cells. On Day 14, mice were randomized to receive daily either hGH (GH, n=14) or saline (CTL, n=12). On Day 29, animals received [3H]phenylalanine for tissue protein fractional synthetic rate (FSR) measurement. Tumors were excised and cell cycle kinetics analyzed. Data are expressed as mean +/- SEM. Statistical analysis was performed by unpaired t test and/or ANOVA where appropriate. In Vitro: Serum from GH-treated animals had elevated IGF-1 levels (287 +/- 34 ng/ml vs 157 +/- 53 ng/ml, P<0.001) and significantly stimulated cell growth (No. cells x 10(3)/well) compared with CTL serum (925 +/- 31 vs 747 +/- 38, P<0.001). In Vivo: Serum for GH-treated animals had elevated IGF-1 levels (287 +/- 34 ng/ml vs 157 +/- 53 ng/ml, P<0.001) and significantly stimulated cell growth (No. cells x 10(3)/well) compared with CTL serum (925 +/- 31 vs 747 +/- 38, P<0.001). In Vivo: Growth hormone had no significant effect on tumor growth rate (mm3/day) (1.45 +/- 0.47 CTL vs 1.57 +/- 0.66 GH), final tumor weight (mg) (0.19 +/- 0.15 CTL vs 0.17+/- 0.06 GH), DNA Index (1.5 +/- 0.1 CTL vs 1.5 +/- 0.1 GH), percent S phase (20.3 +/- 3.3 CTL vs 22.1 +/- 3.0 GH), or tumor FSR (%/day) (51.1 +/- 17.8 CTL vs 70.2 +/- 61.1 GH). Growth hormone significantly elevated serum IGF-1 levels (ng/ml) (176 +/- 48 CTL vs 222 +/- 53 GH, P<0.005) and liver FSR (%/day) (62.8 +/- 17.8 CTL vs 79.7 +/- 12.7 GH, P<0.005). Serum of GH-treated mice increased human pancreatic cell growth in vitro. In vivo, GH administration raised serum IGF-1 levels and increased liver protein FSR, without tumor growth, cell cycle kinetics, or protein FSR

BACKGROUND: Cancer cachexia is a significant cause of postoperative morbidity and mortality in patients with tumors of the upper gastrointestinal tract. Standard parenteral nutrition (TPN) has failed to alter this. The anabolic effect of insulin has been well documented, and its positive effect on protein economy in cancer patients has been recently demonstrated. This study examines the effect of high-dose insulin and parenteral nutrition on protein kinetics in postoperative cancer patients. METHODS: Eleven patients underwent surgery for pancreatic, esophageal, or gastric carcinoma. Postoperatively, patients received standard TPN for 4 days (1 g/kg/day amino acids, 1,000 kcal/day dextrose, 100 g/day lipid), and hyperinsulinemic parenteral nutrition for 4 days (same as standard TPN plus 1.44 U/kg/day regular human insulin) in a crossover design. All patients received both treatments, and the order of treatment was determined randomly. Euglycemia was maintained during insulin infusion via a variable 30% dextrose infusion. Patients underwent protein metabolic studies after each treatment period and rates of whole body and skeletal muscle protein synthesis, breakdown, and net balance were determined by radioisotopic tracer methods using 14C-leucine and 3H-phenylalanine. RESULTS: Compared with standard TPN (STD), hyperinsulinemic TPN (INS) resulted in a significant increase in skeletal muscle protein synthesis (INS: 52.04 +/- 10.22 versus STD: 26.06 +/- 6.71 nmol phe/100 g/min, p < 0.05) and net balance of protein (INS: 7.75 +/- 4.61 versus STD: -15.10 +/- 6.44 nmol phe/100 g/min, p < 0.01), but no difference in skeletal muscle protein breakdown (INS: 44.29 +/- 11.54 versus STD: 41.17 +/- 5.89 nmol phe/100 g/min). Whole-body net balance of protein also significantly increased with insulin-based TPN, compared with standard TPN (INS: 0.04 +/- 0.05 versus STD: -0.08 +/- 0.07 mumol leu/kg/min, p < 0.05), but no difference in whole-body protein synthesis (INS: 2.52 +/- 0.15 versus STD: 2.49 +/- 0.15 mumol leu/kg/min) or whole-body protein breakdown (INS: 2.48 +/- 0.16 versus STD: 2.58 +/- 0.19 mumol leu/kg/min) was observed. Patients received significantly more calories during the hyperinsulinemic TPN period than during the standard TPN period. There was no difference in total, essential, or branched-chain amino acids, and no difference in serum free fatty acids, triglycerides, or cholesterol was observed between the two treatment periods. CONCLUSION: High-dose insulin in conjunction with hypercaloric parenteral nutrition causes improved skeletal muscle protein synthesis, skeletal muscle protein net balance, and whole-body protein net balance compared with standard TPN in postoperative cancer patients