Research Article: Basal Insulin Dose in Adults with Type 1 Diabetes Mellitus on Insulin Pumps in Real-Life Clinical Practice: A Single-Center Experience

Date Published: June 5, 2018

Publisher: Hindawi

Author(s): Bartłomiej Matejko, Aneta Kukułka, Beata Kieć-Wilk, Agnieszka Stąpór, Tomasz Klupa, Maciej T. Malecki.

http://doi.org/10.1155/2018/1473160

Abstract

Basal insulin (BI) infusion in pump therapy of type 1 diabetes (T1DM) mimics physiological secretion during the night and between meals. The recommended percentage of the total BI to daily insulin dose (termed the %BI) ranges between 30 and 50%. We analyzed whether this recommendation was followed in adults with T1DM from a university center, and whether BI doses were linked with glycemic control.

We included 260 consecutive patients with T1DM (159 women and 101 men) treated with continuous subcutaneous insulin infusion at the Department of Metabolic Diseases, Krakow, Poland. Data were downloaded from patients’ pumps and collected from medical records. We analyzed the settings of BI and the association of %BI with HbA1c level. Linear regression was performed.

The mean age of T1DM individuals was 26.6 ± 8.2 years, BMI was 23.1 ± 3.0 kg/m2, T1DM duration was 13.3 ± 6.4 years, and HbA1c level was 7.4%. There were 69.6% (n=181) of T1DM patients with %BI in the recommended range. The T1DM duration and HbA1c level of patients with a %BI <30% (n=23) was 9.5 years and 6.4%, respectively; for a %BI of 30–50%, it was 13.2 years and 7.4%; and for a %BI >50% (n=56), it was 15.8 years and 7.8% (p < 0.001 for both three-group comparisons). Multiple regression identified %BI among independent predictors of the HbA1c level. In this real-life analysis, the recommendations concerning %BI dosing were not followed by almost one-third of adult T1DM patients. Low %BI was associated with better glycemic control; however, this requires further confirmation.

Partial Text

Continuous subcutaneous insulin infusion (CSII) therapy by insulin pumps has become a widely used treatment in patients with type 1 diabetes mellitus (TIDM). CSII is currently the most physiological method of insulin delivery available [1]. This mode of insulin therapy includes two components: basal insulin (BI) infusion that mimics physiological hormone secretion during the night and between meals, and boluses of insulin substituting acute postprandial insulin secretion [2]. The total BI dose is recommended to constitute 30–50% of the total daily insulin dose (DID) in all age groups of individuals with T1DM (the percentage of the total BI to daily insulin dose is termed here the %BI) [3–5]. One algorithm that is commonly used in adults with T1DM specifically suggests programming of BI as a half of the DID [6]. However, whether these recommendations are followed in real-life clinical practice and if %BI in adults with T1DM treated with CSII affects glycemic control have not been assessed so far. This is particularly intriguing in light of the clinical observations that keeping the %BI below 40% and encouraging the patient to bolus more frequently might lead to improved glycemic control with less weight gain [7].

We included 260 consecutive adult patients with T1DM (159 women and 101 men) treated with CSII and remaining under diabetes care of the Department of Metabolic Diseases, a tertiary university reference center in Krakow, Poland. We excluded individuals with proliferative retinopathy, diabetic kidney disease stage III or higher, pregnant women, and those using other antidiabetic agents as adjunct diabetes therapy or steroids. BI profiles were determined and verified at each clinical visit by certified diabetologists. Basal profile from a day directly preceding the patient’s visit was used for the %BI analysis. Data regarding BI profiles, daily basal insulin, DID, average glycemia from blood glucose measurements, and average blood glucose measurements per day from the last two weeks were downloaded from dedicated software (CareLink Professional and Accu-Chek 360). Other collected variables were gender, age, type of insulin analog, weight, body mass index (BMI), T1DM duration, years on CSII, and the HbA1c level, which were all retrieved from the medical record. For the purpose of further analysis, we divided the study group according to their achievement of the recommended glycemic goal: HbA1c equal to or below 7.0%, or HbA1c above 7.0%.

The T1DM study group included 159 women (61.2%) and 101 men (38.8%). The patients were using rapid-acting insulin analogues in their personal insulin pumps including lyspro (n=75; 47.3%), aspart (n=68; 42.7%), and glulisine (n=16; 10%). The study individuals were on average 26.6 ±8.2 years old, with a BMI of 23.1 ± 3.0 kg/m2, and a T1DM duration of 13.3 ± 6.4 years. There were 69.6% (n=181) of T1DM patients with their %BI in the recommended range of 30–50% (mean proportion was 42.1%), while 8.8% (n=23) of patients had their %BI below and 21.6% (n=56) were above the recommended range. The T1DM duration and HbA1c level of patients with a %BI <30% was 9.5 years and 6.4%, respectively; for those with a %BI of 30–50%, it was 13.2 years and 7.4%; and for a %BI >50%, it was 15.8 years and 7.8% (p < 0.001 for both comparisons between the three groups). The HbA1c level and T1DM duration for patients with a %BI <30% were significantly lower than the other groups (post hoc for HbA1c: versus %BI 30–50% p=0.002, versus %BI >50% p < 0.001; for T1DM duration: versus %BI 30–50% p=0.009, versus %BI >50% p < 0.001). There was no statistical difference between the three abovementioned groups with respect to gender, age, and BMI. In this observational study, we report that a substantial proportion of adults with T1DM remaining under the care of the university clinic use a %BI that is outside the recommended range. We also provide some evidence that lower than widely advised %BI might be associated with better glycemic control. In the real-life analysis from a diabetes university center, the recommendations concerning the %BI dosing were not followed by almost one-third of adult patients with T1DM on insulin pumps. Moreover, we identified some evidence for the association of low %BI with better glycemic control; however, this observation requires further confirmation in a larger, randomized, controlled study.   Source: http://doi.org/10.1155/2018/1473160

 

Leave a Reply

Your email address will not be published.