Hepatic fat content and insulin action on free fatty acids and glucose metabolism rather than insulin absorption are associated with insulin requirements during insulin therapy in type 2 diabetic patients

To determine causes of iuterindividual variation in insulin requirements, w e recruited 20 type 2 diabetic patients with stable glucose control and ins ulin doses for >1 year on combination therapy with bedtime NPH insulin and metformin. Insulin absorption (increase in free and total insulin over 8 h after a subcutaneous dose of regular insulin) and actions of intravenous (6 -h 0.3 mU . kg(-1) . min(-1) euglycemic insulin clamp combined with [3-H-3] glucose) and subcutaneous (glucose infusion rate required to maintain isogl ycemia and suppression of free fatty acids [FFAs]) insulin, liver fat conte nt (proton spectroscopy), visceral fat (magnetic resonance imaging), weight , and body composition were determined. We found the following variation in parameters: insulin dose range 10-176 U (mean 42 U, fold variation 17.6x) or 0.13-1.39 U/kg (0.44 U/kg, 10.7x), absorbed insulin 10.6 x, action of su bcutaneous insulin to suppress FFAs 7.5x and to stimulate glucose metabolis m (M value) 11.5x, body weight 67-127 kg (91 kg, 1.9x), liver fat 2-28% (12 %, 14x), and visceral fat 179-2,053 ml (1,114 mi, 11.5x). The amount of ins ulin absorbed, measured as either free or total insulin, was significantly correlated with its ability to suppress FFAs and stimulate glucose metaboli sm but not with the insulin dose per se. The actions of absorbed insulin we re, on the other hand, significantly correlated with the daily insulin dose (r = 0.70 for action on FFAs, P < 0.001, and r = -0.61 for nil value, P < 0.005). Actions of subcutaneous and intravenous insulin to suppress FFAs we re significantly correlated (r = 0.82, P < 0.001, R-2 = 67%). Of the measur es of adiposity, the percent hepatic fat was the parameter best correlated with the daily insulin dose (r = 0.76, P < 0.001). The percent hepatic fat was also significantly correlated with the ability of intravenous insulin t o suppress endogenous glucose production (r = 0.72, P < 0.005). We conclude that the major reason for interindividual variation in insulin requirement s in type 2 diabetes is the variation in insulin action. Variation in hepat ic fat content may influence insulin requirements via an effect on the sens itivity of endogenous glucose production to insulin.