Leptin and insulin response to long-term total parenteral nutrition depends on body fat mass

Background & aims Circulating leptin and insulin concentrations are physiologically representing energy homeostasis. However, the artificial situation of long-term total parenteral nutrition (TPN) and its effects on serum leptin and insulin is not fully understood. Methods We studied 42 gastroenterological patients who received TPN for 19±11 days. Serum leptin and insulin levels as well as body composition assessed by bioelectrical impedance analysis were evaluated on days 0, 7 and 14. Insulin sensitivity was estimated by calculating the QUICKI. Results Before the start of TPN, leptin correlated positively with female gender (P<0.03), BMI (P<0.02), fat mass (P<0.02), insulin levels (P<0.001) and QUICKI (P<0.001). Within the first week of TPN, an increase of leptin levels was found only in patients with a body fat mass of >30% (P<0.02). As these were predominantly women, their leptin levels increased likewise (P<0.003). In regression analysis, fat mass (P<0.001), female gender (P<0.04), insulin levels (P<0.03), and i.v. glucose supply rates (P<0.05) were independently associated to leptin levels. Conclusions TPN—especially glucose—induces a neurohumoral response as shown here for leptin and insulin that is mainly depending on the fat mass. Better understanding of this regulatory mechanism during artificial nutrition could offer a new approach to improve its therapeutic effects.