Continuous glucose monitoring for detection of glycemic variability, hypoglycemia, and hyperglycemia in women with eating disorders

Nao Uotani; Shun'ichi Noma; Momoko Akamine; Takashi Miyawaki

doi:10.1186/s13030-022-00251-4

Continuous glucose monitoring for detection of glycemic variability, hypoglycemia, and hyperglycemia in women with eating disorders

Biopsychosoc Med. 2022 Oct 27;16(1):22. doi: 10.1186/s13030-022-00251-4.

Authors

Nao Uotani¹, Shun'ichi Noma², Momoko Akamine¹, Takashi Miyawaki³

Affiliations

¹ Graduate School of Home Economics, Department of Living Environment, Food and Nutrition, Kyoto Women's University, 35 Kitahiyoshi-Cho, Imakumano, Higashiyama, Kyoto, 605-8501, Japan.
² Noma-Kokoro Clinic, 5-322-1 Sujikaibashi, Fukakusa, Fushimi, Kyoto, 612-0889, Japan.
³ Graduate School of Home Economics, Department of Living Environment, Food and Nutrition, Kyoto Women's University, 35 Kitahiyoshi-Cho, Imakumano, Higashiyama, Kyoto, 605-8501, Japan. miyawakt@kyoto-wu.ac.jp.

Abstract

Background: The aim of this study was to investigate the relationships between hypoglycemia, hyperglycemia, glycemic variability (GV), and eating behavior by measuring daily glucose levels through an intermittently scanned continuous glucose monitoring (isCGM) system in outpatients classified according to eating disorder subtypes.

Methods: We analyzed data for 18 patients (four ANR, nine ANBP, and five BN cases). A FreeStyle Libre Pro® device was attached to the posterior aspect of the upper arm for glucose monitoring. This device conducted measurements every 15 min for five consecutive days. We estimated the mean amplitude of glycemic excursions (MAGE), hypoglycemia, and hyperglycemia.

Results: The mean glucose levels were 91.1 ± 2.2 mg/dL in the ANR group, 94.8 ± 7.5 mg/dL in the ANBP group, and 87.1 ± 8.0 mg/dL in the BN group (P = 0.174). The overall mean MAGE index was 52.8 ± 20.5 mg/dL. The mean MAGE values according to the subtypes were 42.2 ± 5.6 mg/dL in the ANR group, 57.4 ± 23.7 mg/dL in the ANBP group, and 53.0 ± 21.8 mg/dL in the BN group (P = 0.496). Over the course of five days, the frequency of hypoglycemia was as follows: three occurrences in the ANBP group, five occurrences in the BN group, and no occurrences in the ANR group (P = 0.016). Moreover, the occurrence of hypoglycemia was statistically significantly higher in the BN group than in the ANR group (P = 0.013). In the BN group, the frequency of hypoglycemia was highest between 2 and 6 AM, while hypoglycemia was observed throughout the day in the ANBP group. The frequency of hyperglycemia was one occurrence in the ANR group, one occurrence in the BN group, and zero occurrences in the ANBP group (P = 0.641).

Conclusions: Varying GV, hypoglycemia, and hyperglycemia were observed in all subtypes of eating disorders. Our findings suggest that eating behaviors such as binge eating and purging are associated with GV and hypoglycemia. We showed the importance of developing different nutritional approaches tailored to the subtype of eating disorder to prevent hypoglycemia. Additional studies are needed to explore the relationship between glucose levels and eating behaviors in patients with eating disorders.

Keywords: Anorexia nervosa; Bulimia nervosa; Eating disorder; Glycemic variability; Hyperglycemia; Hypoglycemia; Intermittently scanned continuous glucose monitoring system; Mean amplitude of glycemic excursions.