Intensive Care Medicine vol:34 issue:8 pages:1525-1533
OBJECTIVE: Tight glycaemic control by intensive insulin therapy (IIT) reduces morbidity and mortality in critically ill patients. As potential mechanisms contributing to the clinical benefits we hypothesized that glycaemic control affects regional nitric oxide (NO) bioavailability by changing NO synthases (NOS) activity, NOS transcription, NOS substrate availability or the endogenous NOS inhibitor asymmetric dimethylarginine (ADMA) levels. DESIGN: Prospective, randomized experimental study. SETTING: University medical laboratory. INTERVENTIONS: In a rabbit model of prolonged critical illness we assessed the relative impact of maintaining normal insulin/normoglycaemia (n=8), high insulin/normoglycaemia (n=8), normal insulin/hyperglycaemia (n=9) and high insulin/hyperglycaemia (n=8) plasma levels over 7 days on activity and gene expression of endothelial and inducible NOS isoforms in muscle, liver and aorta biopsies, and on plasma levels of NO, arginine and ADMA. MEASUREMENTS AND RESULTS: Compared with normoglycaemic groups, both hyperglycaemic groups revealed 53% higher day-3 NO plasma levels (p<0.05), 40% lower NOS activity in muscle (p<0.01) and 35% lower endothelium-mediated relaxation of aortic rings (p<0.01), 515% higher gene expression of iNOS in muscle (p<0.01) and 99% higher eNOS gene expression in aorta (p<0.01). Only the hyperglycaemic/hyperinsulinaemic group showed lower arginine plasma levels (53% lower, p<0.0001). Compared with healthy controls, normoglycaemic animals revealed 33% lower ADMA levels (p<0.05). CONCLUSIONS: In this animal model of prolonged critical illness, maintaining normoglycaemia, and not glycaemia-independent actions of insulin, prevented excessive systemic NO release on day 3 and appeared to preserve local endothelial function. Factors contributing to this finding may comprise direct endothelial cell damage, direct effects on the enzyme activity, decreased substrate availability or less NO-induced inhibition.