The long-acting insulin used in the United States is Ultralente. As indicated above, it is an insulin-zinc crystalline preparation that was developed to be a basal, long-acting insulin because in animal form its duration of action when injected subcutaneously exceeds 24 hours and is relatively peakless. Because human Ultralente is more soluble, its duration of action is significantly shorter (Table 2). In a recent study (9), human Ultralente was shown to have an onset of action of 1.0 ± 0.2 hour and a duration of action of 19 ± 5.8 hours. Peak action was at 10-12 hours. Moreover, the intersubject variability was twice that seen with a subcutaneous insulin pump. Other sources show that intrasubject variability is very high with Ultralente. Disappointingly, when twice-daily Ultralente was compared to twice-daily Lente insulin in 66 patients in a double crossover study, the major clinical difference was more severe hypoglycemia with Ultralente treatment, mostly between 5 a.m. and breakfast (41).
Because of the above-indicated drawbacks of NPH, Lente, and Ultralente as basal insulins, long-acting insulin analogs have been developed to try to better mimic physiological basal insulin secretion. The only commercially available long-acting analog at the time of this writing is insulin glargine (14). It is produced by modifying insulin at position 21 of the A chain by substituting glycine for asparagine, and also at the amino-terminus of the B chain by adding two arginines. This shifts the isoelectric point, causing glargine to be virtually insoluble at a physiological pH. It is kept soluble within the bottle by using a buffer with a pH of 4. When injected into the neutral pH of the subcutaneous space, glargine precipitates, markedly delaying absorption. The result is a nearly peak-
less insulin analog, with onset of action in 2-4 hours and a duration of action more than 24 hours. In the previously mentioned study of Lepore at al. (9), glar-gine's onset of action was 1.5 ± 0.3 hours and duration 20.5 ± 3.7 hours. Plus, its effect was essentially peakless, and there was no fall-off in action at 24 hours. Intersubject variability was —60% that of NPH.
Three clinical trials with insulin glargine for type 1 diabetes have been published. A phase 3 multicenter trial that compared glargine and bolus lispro with NPH and bolus lispro showed lower fasting blood glucose levels and more patients reaching target fasting glucose in the glargine group. Further, the glargine group showed less variability in fasting blood glucose and less weight gain (42). A 28-week study by Ratner et al. (43) compared glargine and NPH for basal insulin coverage. It showed a greater reduction in fasting plasma glucose, and there were fewer episodes of symptomatic or nocturnal hypoglycemia in the glargine group. A 4-week trial by Rosenstock et al. (44) showed better control of fasting blood glucose with glargine than with NPH.
In type 2 diabetes, a 1-year study compared bedtime NPH and bedtime glargine in 426 patients failing oral agents, and showed similar improvements in HbAlc but less nocturnal hypoglycemia and lower post-dinner glucose levels in the glargine group (45). Rosenstock et al. (46) performed a 28-week open-label trial in patients with type 2 diabetes who were already taking NPH so they continued the NPH or were switched to glargine. Once-daily bedtime glargine was found to be as effective as once- or twice-daily NPH in terms of glycemic control, with a lower risk of nocturnal hypoglycemia and less weight gain.
Compared to NPH insulin, glargine causes more pain at the injection site (as many as 6.1% of patients versus 0.3% for NPH), which may reflect its mild acidity. Glargine is clear, whereas other long-acting insulins such as NPH and the Lentes are milky, requiring care in the labeling and in patients' handling of the glargine vials. It cannot be mixed with other insulins because of the need to keep the pH at 4 to preserve solubility for injection. It has 60% of the lipogenic potency of human insulin but 783% the mitogenic potency (29). The mitogenic potency has, thus far, not been a clinical problem but long-term follow-up is needed. A problem—not unique to glargine—is that its profile does not perfectly mimic the dawn phenomenon seen in some patients with type 1 diabetes. Regardless, insulin glargine appears to provide a more predictable, longer-duration, peakless profile compared with the other available basal insulins (9,47).
Was this article helpful?