低血糖
医学
连续血糖监测
糖尿病
血糖自我监测
1型糖尿病
观察研究
内科学
重症监护医学
内分泌学
作者
Cecilie H. Svensson,Marie Moth Henriksen,Birger Thorsteinsson,Ulrik Pedersen‐Bjergaard
标识
DOI:10.1089/dia.2021.0216
摘要
Aim: Continuous glucose monitoring (CGM) is widely used in clinical practice and research to detect hypoglycemia. A consensus definition of CGM-recorded hypoglycemia is made by a group of international experts under the auspice of the Advanced Technologies and Treatments for Diabetes (ATTD). The purpose of this study is to compare the definition with patient-reported hypoglycemia. Methods: In a prospective, observational study of 186 patients with type 1 diabetes using blinded Medtronic iPro 2 CGM for 6 days, every patient-reported symptomatic hypoglycemic event and interstitial glucose (IG) values at the registration time were classified according to the ATTD definition of CGM-recorded hypoglycemia. For comparison between CGM and self-monitored blood glucose (SMBG) values, the International Hypoglycemia Study Group (IHSG) classification of hypoglycemia and chi-square test were used. Results: A total of 321 events of symptomatic hypoglycemia were reported by 68% of the patients, corresponding to 2.0 ± 2.3 events (mean ± standard deviation) per patient-week. A total of 206 (64%) events met the CGM consensus definition. In the remaining 115 (36%) not-confirmed events, 5 events had an IG <3.9 mmol/L, which lasted <15 min. The overall mean IG value was 3.6 ± 1.1 mmol/L (median 3.1, range 2.2-10.4). In symptomatic hypoglycemic events with both CGM and SMBG data, SMBG confirmed significantly more symptomatic hypoglycemic events than CGM (P < 0.001). Conclusion: CGM-recorded hypoglycemia according to the consensus definition is present at two thirds of all patient-reported events when recorded by the Medtronic iPro 2 system. The recommended minimum duration of a hypoglycemic event of 15 min is supported by the study. SMBG measurements detect significantly more symptomatic hypoglycemic events than Medtronic iPro 2 CGM.
科研通智能强力驱动
Strongly Powered by AbleSci AI