The relationship between continuous glucose monitoring metrics, circadian rhythms, and the presence of albuminuria in insulin-using patients with type 2 diabetes

Lee Da Young; Jung-Been Lee; Inha Jung; Park So Young; Yu Ji Hee; Seo Ji A; Heon-Jeong Lee; Kim Nan Hee

doi:10.1530/endoabs.99.EP28

EP28

Prev Next

Section Contents Cite

Endocrine Abstracts (2024) 99 EP28 | DOI: 10.1530/endoabs.99.EP28

ECE2024 Eposter Presentations Diabetes, Obesity, Metabolism and Nutrition (383 abstracts)

The relationship between continuous glucose monitoring metrics, circadian rhythms, and the presence of albuminuria in insulin-using patients with type 2 diabetes

Da Young Lee ¹ , Jung-Been Lee ² , Inha Jung ¹ , So Young Park ¹ , Ji Hee Yu ¹ , Ji A Seo ¹ , Heon-Jeong Lee ³ & Nan Hee Kim ¹

Views

Author affiliations

¹Korea University Ansan Hospital, Ansan-si, Korea, Rep. of South; ²Sun Moon University, Asan Campus, Asan-si, Korea, Rep. of South; ³Korea University Anam Hospital, Seoul, Korea, Rep. of South

Background: We aimed to explore the relationship between continuous glucose monitoring (CGM)-derived indicators and the presence of albuminuria taking circadian features into consideration concurrently in Insulin-using patients with type 2 diabetes.

Methods: We analyzed data from 129 insulin-treated type 2 diabetes patients in Korea University Ansan Hospital. Over 10 days, we collected time-series data using Dexcom G6, Fitbit, and food diaries, along with cardiometabolic parameters. For CGM metrics, we computed daily time percentages within 70–180 mg/dl (TIR^70-180) and above 250 mg/dl (TAR^>250), standard deviation (SD), and coefficient of variation (CV). Cosinor analysis on Fitbit heart rate data yielded circadian rhythm indices, including amplitude and midline statistics of rhythm (MESOR). The amplitude represents the strength of the rhythm, whereas MESOR reflects mean heart rate. Logistic regression, adjusting for various factors, assessed albuminuria presence (urine albumin-creatinine ratio ≥30 mg/g). Factors included age, sex, BMI, eGFR, HbA1c, diabetes duration, systolic BP, daily step count, carbohydrate percentage, amplitude, and MESOR.

Results: Individuals with albuminuria were found to be older, with higher BP and HbA1c levels, a longer duration of diabetes, and lower amplitude compared to those without albuminuria. In logistic analysis, high HbA1c, systolic BP, and MESOR, and a low amplitude of among covariates were associated with a higher risk of albuminuria. Among the CGM-derived metrics, a low TIR^70-180and a high TAR^>250 were significantly linked to the presence of albuminuria, even after adjustment for metabolic parameters and circadian features. In case of SD and CV, absence of significance was found.

Table 1. Features associated with the presence of albuminuria
Clinical and circadian features	Odds ratio (95% CI)^a
Age(years)	1(0.93–1.06)
Sex, men	0.61(0.24–1.54)
BMI(kg/m²)	0.98(0.87–1.11)
eGFR	0.98(0.96–1.0)
HbA1c(%)	1.86(1.24–2.78)
Duration of diabetes(years)	1.06(1–1.14)
Systolic BP(mmHg)	1.06(1.03–1.10)
Ln(Daily step count)	1.22(0.61–2.46)
Percentage of carbohydrate	1.02(0.96–1.08)
Amplitude	0.81(0.68–0.96)
MESOR	1.06(1.01–1.11)
CGM metrics	Odds ratio(95% CI)^b
TIR^70–180	0.97(0.94–0.99)
LnTAR^>250	1.49(1.01–1.07)
SD	1.03(0.99–1.07)
CV	0.98(0.90–1.07)
^aAll 11 variables were included as covariates in the logistic regression analysis.^b Age, sex, BMI, eGFR, Hemoglobin A1c, duration of diabetes, systolic BP, daily step count, percentage of carbohydrate, amplitude, and MESOR were adjusted for each logistic regression analysis.

Conclusions: We illustrated that a disrupted circadian feature, represented by a low amplitude or high MESOR, was associated with the presence of albuminuria. Considering this significance, TIR^70-180and TAR^>250 are linked to the presence of microalbuminuria.