Is Sliding-Scale Insulin on the Slippery Slope?
Abstract & Commentary
Synopsis: Adding sliding scale insulin to an inpatient’s usual diabetic medications does not reduce rates of hyperglycemia or hypoglycemia or shorten length of stay.
Source: Dickerson LM, et al. Ann Fam Med. 2003;1:29-35.
To ascertain the benefits of instituting a sliding-scale insulin (SSI) regimen of regular insulin in hospitalized patients, Dickerson and associates conducted a multicenter, randomized control study. They enrolled 153 adults with a comorbid diagnosis and a concurrent diagnosis of diabetes mellitus type 2 from 10 family practice residency programs. They excluded patients with diabetic ketoacidosis, hyperosmolar nonketotic state, hypoglycemia, pregnancy, acute myocardial infarction, hemodymanic instability, or acute cerebrovascular accident and patients who could not take food or medicine orally. All patients continued their usual diabetic medications (oral drugs or intermediate-acting insulin alone or in combination with regular insulin). The study group received SSI in addition.
The study group (n = 75) and the control group (n = 78) were similar in age (62.5 vs 65.9 years), gender (37.3% female vs 32.1% male), race, and ethnicity. On admission they had similar blood glucose values (202.9 vs 186.4 mg/dL). They had similar comorbidities, which were primarily cardiovascular, pulmonary, infectious, and neurologic. Their usual diabetic medications were sulfonylureas, metformin, and thiazolidinediones. There were significantly more patients in the control group using intermediate-acting insulin (33.3 vs 50.0%).
All patients had fingerstick blood glucose determinations 4 times a day, and all were taking an American Diabetes Association diet. The study end points were blood sugars greater than 300 mg/dL (hyperglycemia) or less than 50 mg/dL (hypoglycemia), glycemic events (combined rates of hyper- and hypoglycemia), and length of hospitalization.
The addition of SSI resulted in no difference in any end point. The percentages of patients (study vs control) with hyperglycemia (33.3 vs 34.6%), hypoglycemia (8.0 vs 9.0%), and glycemic events (36.0 vs 35.9%) were not significantly different. In both groups the average number of glycemic events was 1.3. The length of hospitalization (5.0 vs 5.3 days) did not differ significantly.
On multivariate analysis 3 factors were associated with an increase in glycemic events: use of intermediate-acting insulin, blood glucose greater than 250 mg/dL on admission, and receipt of corticosteroids.
Comment by Allan J. Wilke, MD
SSI regimens are passed down from senior resident to intern like family recipes. Like families arguing about who has the best recipe for spaghetti sauce, residents argue the virtues of the SSI they learned. In theory (at least according to my senior resident), use of SSI corrected for the alteration of glucose metabolism that comorbid disease wrought. SSI is not without problems. Patients are subject to an increase in needle sticks, and it ties up nursing time, a precious commodity these days. This is not the first article to call into question the use of SSI. Queale demonstrated a higher rate of hyperglycemia with SSI.1 Gearhart reported shorter hospital stays when diabetes is treated proactively, rather than retroactively with SSI.2 (Interestingly, the earliest reference I could find that argues against SSI was published in 1970,3 early enough that my senior resident should have known better!)
Some cautions: This study examined only one SSI regimen. It was not aggressive in its regular insulin dosing (eg, 4 units for a blood glucose between 201 and 250 mg/dL), which is understandable since the SSI augmented rather than replaced usual therapy. It is possible that a more aggressive regimen may have produced different results.
Is there any reason to suspect that the patients in this study are different than the ones you treat? The demographics and comorbidities look very much like the adult patients in a general internal medicine or family practice to me. Although Dickerson et al do not discuss it, patients from residency practices tend to be poorer and more inclined to postpone admission until it is absolutely necessary. If that is the case here, then these patients may have been sicker than the ones you admit. If anything, that only strengthens the study’s conclusions for me.
It is curious that patients using intermediate-acting insulin had more glycemic events and that patients in the control group were more likely to be using acting insulin, yet there was no difference between the 2 groups in rates of glycemic events.
Taking into consideration the limitations of this study (only diabetes type 2 patients who were not critically ill and who could take medicine orally), it is time to for SSI to join bloodletting and trepanation in the Museum of Medical Anachronisms.
Dr. Wilke, Assistant Professor of Family Medicine, Medical College of Ohio, Toledo, OH, is Associate Editor of Internal Medicine Alert.
References
1. Queale WS, et al. Arch Intern Med. 1997;157:545-552.
2. Gearhart JG, et al. Fam Pract Res J. 1994;14:313-322.
3. MacMillan DR. J Ky Med Assoc. 1970;68:577-579.
Taking into consideration the limitations of this study, it is time to for sliding-scale insulin to join bloodletting and trepanation in the Museum of Medical Anachronisms.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.