Assessing insulin effectiveness at the end of the day: Once-daily versus twice-daily insulin glargine injection ()
1. INTRODUCTION
The first long-acting insulin analog, insulin glargine, was introduced in the United States in 2001. It has since become a basal insulin of choice due to its once-daily dosing and its reduced peak activity, resulting in fewer episodes of hypoglycemia [1,2]. Nevertheless, it is a commonly held clinical notion that once-daily insulin glargine administration provides inadequate basal coverage for some patients with type 1 diabetes. Pharmacodynamic studies are consistent with the possibility that relative insulinopenia may occur with once-daily dosing. Lepore et al. showed that the “end of action” of insulin glargine was 22 ± 4 hrs in subjects with type 1 diabetes [2]. Similarly, a euglycemic glucose clamp study in non-diabetic subjects showed that glucose infusion rates began to decline approximately 22 hours after insulin glargine injection [3]. Clement et al. published a case report in 2002 which reported a patient with elevated bedtime glucose values on once-daily insulin glargine therapy administered in the evening, and this patient noted significant improvement in capillary blood glucose measurements after administering insulin glargine as a split dose every 12 hours [4]. In order to examine the hypothesis that twice-daily insulin glargine administration provides superior insulinization during the waning hours of a 24-hour period, this study was designed to compare insulin and glucose concentrations with administration of equivalent daily doses of insulin glargine injected once or twice daily during a 38 hour fast.
2. RESEARCH DESIGN AND METHODS
2.1. Study Subjects
Ten subjects with type 1 diabetes mellitus (9 female) were studied on two separate occasions in a prospective, randomized, cross-over study. Inclusion criteria included age 18 - 50 years, diagnosis of type 1 diabetes at least one year prior to study enrollment, an A1c less than 9%, and basal-bolus therapy using insulin glargine. Exclusion criteria included complications requiring hospitalization within the last 12 months, allergy to insulin glargine, clinically significant chronic medical disease (AST greater than 2.5 timers upper limit of normal, creatinine greater than 1.8 mg/dl, hemoglobin less than 11 for males and less than 10 for females, or NYHA class III or IV heart disease), systolic and diastolic blood pressure above 180 or 110 mmHg, respectively, or active alcohol or drug abuse. The study was approved by the University of New Mexico Human Research Protections Office and all participants provided written informed consent.
2.2. Study Protocol
After an initial screening visit to determine eligibility and to obtain informed consent, study subjects were randomized to receive insulin glargine as either one full-dose injection (0.2 units/kg) at 2000 hours or two half-dose injections (0.1 units/kg) at 0800 and 2000 hours. Subjects subsequently recorded fasting capillary blood glucose (CBG) levels prior to breakfast and in the evenings before dinner, and insulin glargine doses were titrated until fasting CBG’s were less than 150 mg/dl (8.33 mmol/l). Once the dose of study insulin remained unchanged for one week with fasting CBG’s less than 150 mg/dl (8.33 mmol/l), subjects were admitted for an overnight study.
Subjects were admitted to the University of New Mexico Inpatient Clinical Research Unit at 1500 on the afternoon prior to study. In order to avoid carryover of long-acting insulin from the previous day, subjects were instructed not to take any insulin glargine at home on the day of admission. All subjects received a 7 kcal/kg American Diabetic Association meal at 1800, and then fasted for the remainder of the study [5]. A continuous intravenous infusion of human regular insulin was administered between 1900 on day 1 and 0700 on day 2 to achieve a target CBG of 90 - 120 mg/dl (5.0 - 6.66 mmol/l). CBG’s were monitored on an hourly basis and the insulin infusion was adjusted as needed to maintain CBG’s in the target range. At 0800 on day 2, subjects received either fullor half-dose insulin glargine injected subcutaneously into the abdomen, corresponding to their assigned regimen. Subjects randomized to the twicedaily insulin glargine regimen received their second injection at 2000 on day 2. Blood samples were obtained for plasma glucose and PEG-treated serum insulin every 60 minutes between 0800 on day 2 and 0400 on day 3, and every 30 minutes between 0400 and 0800 (i.e., the final 4 hours of the 24 hour period since initial insulin glargine administration) on day 3. In addition, bedside CBG’s were performed every 60 minutes. If hypoglycemia occurred, defined as a CBG value less than 50 mg/dl (2.78 mmol/l), subjects received 25 ml 50% dextrose intravenously and the study was continued. The inpatient study concluded at 0800 on day 3. At the conclusion of the first overnight stay, subjects switched to the alternative study insulin regimen (either once or twice-daily insulin glargine) and repeated the protocol as described above, followed by a second inpatient admission. Upon conclusion of the second hospital admission, subjects returned to their pre-study insulin regimen.
2.3. Sample Analysis
CBG concentrations were determined in duplicate using mixed venous blood with a standard, commercially available meter (Accucheck Advantage, Roach Diagnostics, Indianapolis, IN). Plasma glucose was assessed using the ACE Glucose Reagent (Alfa Wassermann, Caldwell, NJ). Free insulin concentrations were determined using an Ultra Sensitive Human Insulin Radioimmunoassay (Linco, St Charles, Missouri), which employs the double antibody/PEG technique to achieve a sensitivity of 0.2 mcU/ml when using a 100 microliter sample.
2.4. Validation of the Insulin Glargine Assay
Validation of the use of the Linco ultra-sensitive human insulin radioimmunoassay for insulin glargine was performed using the methods of Owen et al. [6]. The Human Insulin RIA Kit from Linco Research (St. Charles, Missouri) uses a standard curve from 0.2 to 20 mcU/ml. To validate the assay, 100 U/ml of Glargine was diluted volumetrically with 6 g/dl of Bovine Serum Albumin to final concentrations of 40, 20, 10, and 5 mcU/ml. These dilutions were prepared in triplicate and treated with 25% PEG, and every dilution was then analyzed in duplicate using the Linco Research Free Insulin Extraction Procedure (PEG protocol). The percentage of cross-reactivity was calculated from a ratio of measured and nominal concentrations. The cross-reactivity was determined to be 94% at a standard concentration of 5 mcU/ml (35 pmol/l); levels that correspond well with those observed in our study.
2.5. Statistical Analysis
The primary outcome variable of the study was the serum insulin concentration during the final 4 hours of the 24-hour study period. It was determined that ten subjects would be adequate to provide 80% power to detect a 49 pmol/l difference in serum insulin during the last 4 hours of the 24-hour study period between the once-daily and twice-daily insulin glargine regimens using a paired student’s t-test with alpha equal to 0.05. Secondary outcome variables include all insulin and glucose concentrations over the 24-hour period and the results of home CBG testing during the week prior to admission. Results from the once-daily and twice-daily insulin dosing regimens were compared using ANOVA and the paired Student’s t test, when appropriate. Area under the curve (AUC) for insulin and glucose during the last four hours of study was calculated using the trapezoidal rule. Results are reported as mean ± SD, except in the figures, which depict mean ± SEM.
3. RESULTS
3.1. Patient Characteristics
Subjects had a mean age of 40 ± 8 years, a mean duration of diabetes of 16 ± 10 years, and a mean BMI of 31 ± 8 m/kg2. All enrolled subjects were C-peptide negative (less than 1 pmol/ml) during hyperglycemia 60 minutes after ingestion of 8 ounces of Boost® nutritional beverage (Nestle HealthCare Nutrition, Inc., Minnetonka).
3.2. Serum Free Insulin Concentrations
Figure 1 shows the mean free insulin concentrations obtained during the 24 hour study period. At baseline (0800), there was no statistical difference in serum free insulin levels between the once-daily and the twice-daily insulin glargine regimens (137 ± 97 vs. 146 ± 128 pmol/l, respectively, p = 0.86). Twelve hours after the first injection of insulin glargine, serum insulin levels in the once-daily treatment regimen were significantly higher than during the twice-daily treatment regimen (150 ± 99 vs. 63 ± 39 pmol/l, p = 0.02). At the end of the 24-hour study period, there was no significant difference between the serum insulin levels in the once-daily and the twicedaily insulin glargine regimens (70 ± 56 vs. 84 ± 63 pmol/l, p = 0.60). Table 1 summarizes the AUC for insulin during the last four hours of the 24-hour study period.