Appropriate insulin regimes for type 2 diabetes: a multicenter randomized crossover study

OBJECTIVE: To directly compare the rate of hypoglycemia and metabolic control achieved on once-daily ultralente insulin administration with twice-daily NPH insulin administration in patients with type 2 diabetes. Patient treatment satisfaction and quality of life were also examined before and during each treatment. RESEARCH DESIGN AND METHODS: A crossover study was performed involving five centers and 79 patients with type 2 diabetes (fasting blood glucose > 8 mmol/l) with a 2-month run-in followed by two 6-month periods of either NPH or ultralente insulin administration. Patients were managed by a specialist nurse using a dosage adjustment protocol. RESULTS: HbA1c was lower with NPH insulin therapy during each of the 6-month periods (9.7 +/- 0.2 vs. 9.1 +/- 0.3 and 9.8 +/- 0.2 vs. 9.0 +/- 0.3 mmol/l; both P < 0.01). The difference was accounted for by higher evening glucose levels with ultralente insulin (fasting 8.2 +/- 0.3 vs. 8.2 +/- 0.3 mmol/l, 6:00 P.M. 11.5 +/- 0.4 vs. 10.6 +/- 0.4 mmol/l). Despite worse control, the total number of hypoglycemic episodes was greater with ultralente insulin (220 vs. 171), and hypoglycemic episodes requiring third-party assistance occurred almost entirely with ultralente (14 vs. 1). Treatment satisfaction scores increased more with NPH insulin compared with ultralente and rose further upon changing to NPH insulin, but fell upon changing to ultralente insulin. These changes were highly significant (P < 0.001). Diabetes quality of life improved on both regimens. CONCLUSIONS: These data clearly demonstrate the lower hypoglycemia rate, better glucose control, and greater treatment satisfaction accompanying therapy for type 2 diabetes with twice daily NPH compared with once daily ultralente insulin.