Diabetes Metab J. 2022;46(3):439-450.

Comparison of Prevailing Insulin Regimens at Different Time Periods in Hospitalized Patients: A Real-World Experience from a Tertiary Hospital.

, , , , ,

제1저자 문선준, 최훈지 · 교신저자 조영민

PMID 34666425 PMC9171166 DOI 10.4093/dmj.2021.0065 IF 8.2 내분비대사내과

소속 (3)

Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Korea.
Division of Endocrinology and Metabolism, Department of Internal Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea.
Division of Endocrinology and Metabolism, Department of Internal Medicine, Seoul National University Hospital, Seoul, Korea.

Abstract

BACKGROUND Prevailing insulin regimens for glycemic control in hospitalized patients have changed over time. We aimed to determine whether the current basal-bolus insulin (BBI) regimen is superior to the previous insulin regimen, mainly comprising split-mixed insulin therapy.

METHODS This was a single tertiary center, retrospective observational study that included non-critically ill patients with type 2 diabetes mellitus who were treated with split-mixed insulin regimens from 2004 to 2007 (period 1) and with BBI from 2008 to 2018 (period 2). Patients from each period were analyzed after propensity score matching. The mean difference in glucose levels and the achievement of fasting and preprandial glycemic targets by day 6 of admission were assessed. The total daily insulin dose, incidence of hypoglycemia, and length of hospital stay were also evaluated.

RESULTS Among 244 patients from each period, both fasting glucose (estimated mean±standard error, 147.4±3.1 mg/dL vs. 129.4±3.2 mg/dL, P<0.001, day 6) and preprandial glucose (177.7±2.8 mg/dL vs. 152.8±2.8 mg/dL, P<0.001, day 6) were lower in period 2 than in period 1. By day 6 of hospital admission, 42.6% and 67.2% of patients achieved a preprandial glycemic target of <140 mg/dL in periods 1 and 2, respectively (relative risk, 2.00; 95% confidence interval, 1.54 to 2.59), without an increased incidence of hypoglycemia. Length of stay was shorter in period 2 (10.23±0.26 days vs. 8.70±0.26 days, P<0.001).

CONCLUSION BBI improved glycemic control in a more efficacious manner than a split-mixed insulin regimen without increasing the risk of hypoglycemia in a hospital setting.

Keywords

Biphasic insulinsDiabetes mellitus, type 2Glycemic controlInsulin, long-actingInsulin, short-acting

Figures

**Fig. 1.** Glycemic control during hospital stay. Data shown are the glucose values (estimated mean±standard error) obtained from each time point with 95% confidence intervals. Analysis of covariance (ANCOVA) analyses were performed, adjusted for age, sex, body mass index, weight, estimated glomerular filtration rate, diabetes duration, glycated hemoglobin, number of concurrent antidiabetic medications, the usage frequency of each antidiabetic medication, and the number of plasma glucose measurements by point-of-care testing. (A) Mean glucose from day 1 to 6 according to time of the day. (B) Mean glucose from day 1 to 2 vs. day 5 to 6. (C) Preprandial glucose from day 1 to 6. (D) Fasting glucose from day 2 to 6. (E) Kaplan-Meier analysis of target achievement of the mean preprandial glucose on each day. (F) Kaplan-Meier analysis of target achievement of all preprandial (i.e., pre-breakfast, pre-lunch, and pre-dinner) glucose on each day. ^aP<0.05 for period 1 day 1 to 2 vs. period 2 day 1 to 2, ^bP<0.05 for period 1 day 5 to 6 vs. period 2 day 5 to 6.

**Fig. 2.** Cox regression and subgroup analyses of the mean preprandial glycemic target achievement. Shown are the forest plots for patient subgroups with respect to the target glycemic achievements. Age, sex, body mass index, weight, estimated glomerular filtration rate (eGFR), diabetes duration, glycated hemoglobin (HbA1c), number of concurrent antidiabetic medications, the usage frequency of each antidiabetic medication, and the number of plasma glucose measurements by point-of-care testing were adjusted. P values describe the interaction between the target glycemic achievement and subgroup variables with no adjustment for multiple testing. CI, confidence interval; CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration.

**Fig. 3.** Comparison of conventional vs. structured basal-bolus insulin (BBI). Data shown are the glucose values (estimated mean±standard error) obtained from each time point with 95% confidence intervals. Analysis of covariance (ANCOVA) analyses were performed, adjusted for age, sex, body mass index, weight, estimated glomerular filtration rate, diabetes duration, glycated hemoglobin, number of concurrent antidiabetic medications, the usage frequency of each antidiabetic medication, and the number of plasma glucose measurements by point-of-care testing. (A) Proportion of conventional and structured BBI from 2009 to 2018. (B) Mean glucose from 2009 to 2018. (C) Mean glucose from day 1 to 6 according to time of the day. (D) Mean glucose from day 1 to 2 vs. day 5 to 6. (E) Kaplan-Meier analysis of target achievement of the mean preprandial glucose on each day. (F) Kaplan-Meier analysis of target achievement of all preprandial glucose on each day.

그림 출처: PubMed Central 공개 논문 (PMC9171166)

Conflict of interest statement

CONFLICTS OF INTEREST . No potential conflict of interest relevant to this article was reported.

그 밖의 정보