Medizinische Universität Graz Austria/Österreich - Forschungsportal - Medical University of Graz
Gewählte Publikation:
Bytyqi, K.
Feasibility to Use a Continuous Glucose Monitoring System to Manage Diabetes During Hospitalisation
Humanmedizin; [ Diplomarbeit ] Medizinische Universität Graz; 2021. pp. 62
[OPEN ACCESS]
FullText
- Autor*innen der Med Uni Graz:
- Betreuer*innen:
- Elsayed Hesham
- Mader Julia
- Altmetrics:
- Abstract:
- Introduction Continuous glucose monitoring (CGM) systems and clinical decision support systems (cDSS) are frequently used in diabetes management nowadays. As CGM offers detailed information on glycaemic control, it can be an important alternative to capillary glucose measurements taken with point-of-care (POCT) devices, which are current best practice in inpatient care. The aim of our study was to assess the feasibility of using a cDSS steered with CGM vs. POCT glucose values. Methods: A retrospective analysis of inpatients with type 2 diabetes treated using a cDSS was performed. The cDSS was fed with POCT glucose values measured four times per day (pre-meal, bedtime). CGM was used in blinded mode. POCT measurements were matched to corresponding CGM values within 15min of each other. Conventional insulin dosing suggestions using POCT data were compared to simulated insulin dose calculations based on paired CGM data plus any CGMbased hypoglycaemic event. Results: Data from 30 patients (age 74.1±10.9 years, 60% female, BMI 28.6±5.6 kg/m², diabetes duration 13.2±11.6 years, HbA1c 72±22 mmol/mol, creatinine 1.5±1.2 mg/dl) were used. Median (Q1; Q3) POCT glucose was 135 (106; 181) mg/dl (n=885), median CGM glucose was 110 (69; 160) mg/dl (n=1075). Median (Q1; Q3) bolus insulin dose suggestion (only correction doses at night) was 3 (2; 3) IU (n=73) based on POCT-measurements and 2 (1; 3) IU (n=73) based on CGM-measurements. Median (Q1; Q3) for all other bolus insulin dose suggestions was 6 (3; 10) IU (n=643) based on POCT-measurements and 6 (2; 10) IU (n=643) based on CGM-measurements. Median (Q1; Q3) basal insulin dose suggestion was 14 (10; 20) IU (n=215) based on POCT-measurements and 13 (10; 19) IU (n=215) based on CGM-measurements. Finally, median (Q1; Q3) daily insulin dose suggestion was 28 (20; 42) IU (n=213) based on POCT-measurements and 25 (19; 39) IU (n=207) based on CGM-measurements. Conclusion: Combining CGM with cDSS provides similar insulin dose suggestions compared to POCT measurements in a simulation study. As CGM detects more hypoglycaemic events, the insulin dose suggestions by cDSS are slightly lower, thus less hypoglycaemia can be expected. Using CGM data to steer a cDSS will be evaluated within a clinical study.