The study by Baldwin et al. (1) had the laudable intention of training house staff in moving away from inpatient sliding-scale insulin in favor of a basal-bolus approach or active titration of oral agents. However, some key points in their report remain unexplained. The authors state that in insulin-treated patients, the premeal short-acting insulin (usually regular) in combination with a basal insulin (usually NPH) was used. However, no details are given as to how the NPH was titrated and what parameters, if any, were used for adjustment. Even more notable is the lack of information about adjustment of short acting insulin. How were the initial doses of regular insulin arrived at, and what end point was the basis for periodic alterations of the premeal dose of bolus insulin? Were 2-h postprandial blood glucoses measured, and was the insulin-to-carbohydrate ratio and sensitivity factor utilized? If a best-guess approach was used, how is that much different from, or superior to, a sliding scale? The authors state that premeal regular insulin was administered twice daily, whereas for true prandial coverage, usually three and sometimes more injections are necessary. Thus the claim that the basal-bolus insulin was used in the study patients is not entirely true. Simply vowing to eliminate sliding-scale insulin without replacing it with a rational, scientific, and target-based alternative insulin regimen does not solve the problem. It is entirely possible that the HbA1c improvement seen in the study was due to intensive escalation of insulin or oral-agent therapy secondary to the increased attention provided by the twice-daily watchful eye of an endocrinologist and had very little to do with refusing to write sliding-scale orders per se.
The endocrine service at our institution utilizes a basal-bolus insulin regimen designed to overcome the above drawbacks. Daily adjustments of basal insulin are made based on fasting, premeal, and periodic 3:00 a.m. blood glucose readings. Two-hour postprandial readings, insulin-to-carbohydrate ratios, and sensitivity factor calculations dictate initiation and fine tuning of short-acting premeal insulin (2). It has required a reeducation of nurses, trainees, other health care professionals, and patients. We have attempted to involve persons from various disciplines to try to change the mindset vis-à-vis inpatient diabetes control through a collaborative effort. Although we are still in the midst of evaluating the efficacy of our inpatient subcutaneous insulin orders, we have been pleased with the results so far and feel that this approach is on a more scientific footing.
The optimal method of using subcutaneous insulin in the hospital remains to be determined. However, we feel that the only enduring philosophy is one that tailors both basal and short-acting insulin to the needs of the patient by means of a rational approach that has inherent flexibility.
