We enthusiastically agree with most of the recommendations in “It’s time to abandon the sliding scale” (J Fam Pract. 2011;60:266-270), but we see an inconsistency.
Guthrie et al support the use of the postprandial fingerstick as a measure of glycemic control, and note its endorsement in the product inserts for 2 well-known rapid-acting analog insulins. However, neither the RABBIT-2 study cited by the authors nor the 2010 American Diabetes Association (ADA) standards of care take this position. In RABBIT-2, preprandial fingersticks were used exclusively,1 and the ADA standards recommend correction doses, or supplemental insulin, to correct pre-meal hyperglycemia.2 The 2004 ADA position statement on inpatient glycemic control3 cited by the authors does not differentiate between preprandial and postprandial fingersticks.
We believe that postprandial testing, while often useful in the outpatient setting, is rarely helpful in the hospital, where such testing can lead to too-frequent dosing of rapid-acting insulin. And, because the serum glucose is not known until after the meal, an appropriate pre-meal insulin dose cannot be prescribed until the following day. Since the relatively recent relaxation of standards in inpatient glycemic control,2 postprandial monitoring rarely leads to meaningful improvement in glycemic control.
Justin Moore, MD
Tracy Williams, MD
Wichita, Kan
References
1. Umpierrez GE, et al. Randomized study of basal-bolus insulin therapy in the inpatient management of patients with type 2 diabetes (RABBIT 2 trial). Diabetes Care. 2007;30:2181-2186.
2. Executive summary: standards of medical care in diabetes— 2011. Diabetes Care. 2011;34(suppl 1):S4-S10.
3. Clement S, et al. Management of diabetes and hyperglycemia in hospitals. Diabetes Care. 2004;27:553-591.
The authors respond:
The issue raised by Drs. Moore and Williams is vital to management with modern insulins. To address it fully, however, an understanding of the pharmacodynamics of rapid-acting insulins is required.
In brief, the pharmaco-kinetic value of insulin is the time at which amounts of insulin in the bloodstream are measurable. The pharmacodynamic value—the time at which there is enough insulin in the bloodstream to control glucose levels—is much shorter. For rapid-acting insulin at the usual doses, this value is about 3 to 4 hours.
The problem is that pre-meal glucose levels are measured 4 to 6 hours after the rapid-acting insulin is administered; thus, a blood glucose level measured at 6:00 pm reveals nothing about the rapid-acting insulin given at noon. Pre-meal measurements more accurately reflect the basal insulin than the bolus insulin. Corrections made at that time can cause problems at the peak time of the insulin, 1 to 3 hours later.
We correct this problem in the outpatient setting by changing the insulin dose by the amount of food to be eaten, not by the blood glucose level at the time, then looking at the pattern of values over several days and making corrections as needed. This is harder to do in the hospital because the patient is there for a shorter time. So we make rounds in the late afternoon, when the blood glucose values for the day (except for the evening) are available in a tabular form, and use them, along with values from the previous day(s), to look for a pattern.
Using a patterned approach with both pre- and post-meal blood glucose values will improve control of diabetes; if IV insulin is used, values are obtained hourly and the insulin drip titrated within a certain range to meet the needs of the individual patient.
Diana W. Guthrie, PhD, ARNP, BC-ADM, CDE
Richard A. Guthrie, MD, FAAP, FACE
Wichita, Kan