Randomized clinical trial of the impact of insulin therapy on liver function in patients undergoing major liver resection

• Published in: British journal of surgery Vol. 100; no. 5; pp. 610 - 618
• Main Authors: Hassanain, M., Metrakos, P., Fisette, A., Doi, S. A. R., Schricker, T., Lattermann, R., Carvalho, G., Wykes, L., Molla, H., Cianflone, K.
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 01.04.2013; Oxford University Press
• Subjects: Administration, Cutaneous; Adult; Aged; Blood Glucose; Glucose - administration & dosage; Hepatectomy - adverse effects; Hepatectomy - methods; Humans; Hypoglycemic Agents - administration & dosage; Infusions, Intravenous; Insulin, Regular, Human - administration & dosage; Liver Diseases - metabolism; Liver Diseases - prevention & control; Liver Glycogen - metabolism; Middle Aged; Perioperative Care - methods; Postoperative Complications - prevention & control; Preoperative Care - methods; Young Adult
• ISSN: 0007-1323; 1365-2168; 1365-2168
• DOI: 10.1002/bjs.9034

Background Postoperative liver dysfunction is the major source of morbidity and mortality in patients undergoing partial hepatectomy. This study tested the benefits of a metabolic support protocol based on insulin infusion, for reducing liver dysfunction following hepatic resection. Methods Consecutive consenting patients scheduled for liver resection were randomized to receive preoperative dextrose infusion followed by insulin therapy using the hyperinsulinaemic normoglycaemic clamp protocol (n = 29) or standard therapy (control group, n = 27). Patients in the insulin therapy group followed a strict dietary regimen for 24 h before surgery. Intravenous dextrose was started at 2 mg per kg per min the night before and continued until surgery. Hyperinsulinaemic therapy for a total of 24 h was initiated at 2 munits per kg per min at induction of anaesthesia, and continued at 1 munit per kg per min after surgery. Normoglycaemia was maintained (3·5–6·0 mmol/l). Control subjects received no additional dietary supplement and a conventional insulin sliding scale during fasting. All patients were tested serially to evaluate liver function using the Schindl score. Liver tissue samples were collected at two time points during surgery to measure glycogen levels. Results Demographics were similar in the two groups. More liver dysfunction occurred in the control cohort (liver dysfunction score range 0–8 versus 0–4 with insulin therapy; P = 0·031). Median (interquartile range) liver glycogen content was 278 (153–312) and 431 (334–459) µmol/g respectively (P = 0·011). The number of complications rose with increasing severity of postoperative liver dysfunction (P = 0·032) Conclusion The glucose–insulin protocol reduced postoperative liver dysfunction and improved liver glycogen content. Registration number: NCT00774098 (http://www.clinicaltrials.gov). Maintaining normoglycaemia is better