ATLANTA — Using a weight-based method for calculating insulin correction dosing resulted in superior blood glucose control, compared with a traditional sliding-scale method in a pilot study of hyperglycemic inpatients.
In recent years, it has become clear that nonindividualized sliding-scale regimens often result in out-of-target blood glucose levels. “Sliding scales look backwards to correct the past, and fail to anticipate the future,” Diane M. Thompson, R.N., said at the annual meeting of the American Association of Diabetes Educators.
In cooperation with a local endocrinologist and critical care physician, Ms. Thompson and Heatherann Cundiff, R.N., who both work for Nashville, Tenn.–based Diabetes Healthways, which contracts with the hospital for diabetes management, developed a method for using a patient's weight to calculate the insulin dose needed to correct an elevated blood sugar. In a 2006 pilot study at Monroe Regional Medical Center, Ocala, Fla., the weight-based method resulted in 70% of 681 blood sugar values being less than 180 mg/dL in 54 patients, versus 46% of 719 values in 64 patients treated based on the sliding-scale method.
The weight-based method uses this formula to derive a sensitivity factor (SF): 3,000/kg body weight. The SF expresses how many mg/dL the blood glucose is lowered by 1 U of fast-acting insulin. The SF is calculated on admission, and documented on the patient's order sheet after verification by two nurses.
The correction dose is designed to bring the blood glucose to a target of 110 mg/dL, assuming that basal and prandial needs are met, either with medication or naturally by the body. This formula is used to calculate the required number of units of insulin: subtract 110 from the blood glucose in mg/dL, divide the result by the SF, then round the result to the nearest whole number.
For example, a 124-kg patient would have a SF of 24 (3,000/124). If her blood glucose was 250 mg/dL before breakfast, her correction dose would be 5.83 (250 − 110 = 140; 140/24 = 5.83), which would be rounded up to 6 U.
In the pilot study, the weight-based method was used in 64 patients during 1 month, yielding 31 different SFs (range 19–86). The number of blood glucose readings greater than 300 mg/dL dropped from 85 to 38 in those 64 patients, and the number of readings between 180 mg/dL and 300 mg/dL dropped from 270 to 168. The average blood glucose was 186 mg/dL with the sliding-scale method and 149 with the weight-based method, Ms. Cundiff reported.
'Sliding scales look backwards to correct the past, and fail to anticipate the future.'
Source MS. THOMPSON