Optimization of evening insulin dose in patients using the short-acting insulin analog lispro.
OBJECTIVE: A three-way, crossover, open-label, randomized study was designed to compare the evening and night (1800-0800) glycemic control when the evening premeal lispro dose was reduced by 20% and the bedtime basal NPH dose increased by 25%, or when the basal NPH dose was moved to before dinner at 1800, compared with the control arm on standard premeal human regular insulin and pre-bedtime NPH insulin. RESEARCH DESIGN AND METHODS: A total of 13 type 1 diabetic patients who use a premeal plus basal insulin regimen were studied on three separate days, with identical meals and snacks at the same times on each study day. On the control study day, patients received human regular insulin before dinner and NPH at bedtime in their usual doses. On another day, lispro was given before dinner with a dose reduction of 20%, and NPH at bedtime at 125% of usual dose. In the third regimen, the lispro and NPH were administered together in their usual dose before the evening meal by separate injections. The three regimens were tested in random order. RESULTS: Postprandial (1800-2200) blood glucose concentrations were lower after reduced-dose lispro compared with human regular insulin (6.0 +/- 0.3 [SEM] vs. 7.4 +/- 0.3 mmol/l, P < 0.05). Nighttime (2400-0400) blood glucose concentrations were not different (8.6 +/- 0.3 vs. 9.2 +/- 0.3 mmol/l, NS), and prebreakfast concentrations were also unchanged (7.7 +/- 0.9 vs. 8.7 +/- 0.8 mmol/l) after lispro with increased-dose NPH compared with standard insulin. By contrast, both nighttime (10.0 +/- 0.3 mmol/l, P < 0.05) and fasting glucose concentrations (10.8 +/- 0.6 mmol/l, P < 0.05) were significantly higher with dinnertime usual-dose lispro plus dinnertime usual-dose NPH compared with standard human insulin. Hypoglycemia at night (blood glucose < 3.0 mmol/l) did not differ between study days, but it was more frequent postprandially after dinner usual-dose lispro plus early NPH (2 vs. 7 patients, P = 0.062). CONCLUSIONS: With lower mealtime and higher basal bedtime insulin doses, patients using insulin lispro may be able to gain an overall improvement in evening blood glucose control without deteriorated nighttime glucose levels. Earlier basal NPH dosage alone does not ameliorate the nighttime hyperglycemia of short-acting insulin analog regimens.