Total parenteral nutrition and alternate energy substrates in treatment of severe acute pancreatitis

The use of total parenteral nutrition (TPN) in the treatment of 73 patients with acute severe pancreatitis was prospectively studied during a two year period. Patients were divided into three groups on the basis of calorie substrate used. Glucose and twice weekly lipid infusion (glucose based) were used in 60 per cent; 27 per cent required daily lipid infusion (lipid based), and 13 per cent received no lipid because of pre-existing hyperlipemia or thrombocytopenia (no lipid). Nutritional indices (albumin, transferrin and total lymphocyte count) were initially abnormal in more than 80 per cent of patients, and 50 per cent had three or more of Ranson's criteria. After TPN, 81 per cent had improved nutritional indices, and none had hypertriglyceridemia or aggravation of pancreatitis develop. Patients who received lipid based or no lipid had higher insulin requirements (p < 0.01) than those receiving mainly glucose. Mortality was increased tenfold (2.5 versus 21.4 per cent, p < 0.01) in patients who did not achieve positive nitrogen balance. We conclude that TPN, either lipid or glucose based, is a safe and effective therapy to reverse the malnutrition of acute pancreatitis and that failure to achieve positive nitrogen balance is associated with increased mortality.