Avoiding Nocturnal Hypoglycemia: Consideration of an Extra Injection at Bedtime

  1. Andrew W. Norris, MD, PhD; and
  2. Lori M.B. Laffel, MD, MPH
  1. From the Joslin Diabetes Center, Harvard Medical School, Boston, Massachusetts

    While sleeping, how do you replace insulin, a life-sustaining hormone with a narrow therapeutic range whose requirements change throughout the night? This challenge has faced patients with type 1 diabetes mellitus and their physicians since the first use of purified insulin for subcutaneous injection in 1922. The Diabetes Control and Complications Trial (1) demonstrated the magnitude of this challenge, showing that intensive insulin therapy not only delays microvascular complications but also increases the risk for hypoglycemia by roughly threefold, with more than half of the episodes occurring during sleep (2).

    It has long been assumed that the risk for nocturnal hypoglycemia is lower when long-acting insulin is given at bedtime (split-dose regimen) rather than at the evening meal (mixed-dose regimen) (3). The risk is presumed lower with bedtime insulin because, in theory, when the long-acting insulin is given at dinnertime, it will peak earlier in the night, closer to the nadir of insulin needs (3). Thus, it is of interest that the Diabetes Control and Complications Trial found increased nocturnal hypoglycemia risk in its intensive treatment group, which was receiving a split-dose regimen, compared with the control group, which was receiving a mixed-dose regimen. The increased risk for nocturnal hypoglycemia in the intensive treatment group was naturally ascribed to the intensive glycemic goals rather than to the split-dose regimen of the intervention group; however, this conclusion lacked proof.

    In this issue, Fanelli and colleagues (4) report on their study, which tested this assumption directly by comparing a split-dose regimen with a mixed-dose regimen; both groups had intensive glucose targets. This prospective, open-label study was small, following 22 adults with type 1 diabetes mellitus randomly assigned to the two 4-month crossover …

    « Previous | Next Article »Table of Contents