Effectiveness of human ultralente versus NPH insulin in providing basal insulin replacement for an insulin lispro multiple daily injection regimen. A double-blind randomized prospective trial. The Canadian Lispro Study Group

Abstract

Objective: To compare human ultralente (UL) insulin with human NPH insulin as basal insulin replacement in patients who use insulin lispro before meals.

Research design and methods: There were 178 patients with type 1 diabetes who were randomized to receive either human NPH or UL insulin once daily at bedtime in a 1-year double-blind clinical study. Eight-point blood glucose profiles were collected once monthly in the first 4 months, then every 2 months for the remainder of the study. Patients were also asked to perform premeal blood glucose measurements every day throughout the study. If before-supper blood glucose levels consistently exceeded 8 mmol/l despite optimal postprandial control with the lunch dose of insulin lispro, a second dose of basal insulin before breakfast was administered.

Results: For the group as a whole, insulin doses before meals and basal insulin doses were similar at baseline. At study's end, meal doses remained the same (30 +/- 1 U/day for UL., 29 +/- 1 U/day for NPH), while basal requirements were somewhat higher for the UL group than the NPH group: 30 +/- 1 U/day vs. 26 +/- 1 U/day, respectively (P < 0.05). The rates of severe hypoglycemia were similar for patients on NPH (0.05 +/- 0.03 per patient every 30 days) and for UL (0.07 +/- 0.04 per patient every 30 days) insulin. There was no significant difference for glycemic control between the NPH and UL groups overall (HbAlc at the end of the study: 7.6 +/- 0.1 vs. 7.7 +/- 0.1%, respectively), and by study's end a similar number of patients in the NPH and the UL groups needed to be switched to twice daily basal insulin (21 and 24%, respectively). Patients requiring twice-daily injections of basal insulin had a longer duration of diabetes (17.8 +/- 1.5 vs. 14.0 +/- 0.8 years, P < 0.05) and a highest baseline HbAlc (8.6 +/- 0.1 vs. 8.0 +/- 0.1%, P < 0.002) and were significantly older (38 +/- 2 vs. 34 +/- 1 years, P < 0.007). Patients who were switched to twice-daily NPH insulin had lower HbAlc levels at study's end compared with those switched to twice-daily UL insulin (7.7 +/- 0.2 vs. 8.2 +/- 0.3%), but this difference was not statistically significant. Distribution of hypoglycemia across the day was also similar in both groups.

Conclusions: UL or NPH insulin, when used as the basal insulin for multiple injection regimens, results in similar glycemic control in patients using insulin lispro before meals. However, in patients who require a second injection of basal insulin, NPH insulin appears to provide lower prebreakfast and prelunch glucose levels compared with UL insulin.