Segmental Bioimpedance Variables in Association With Mild Cognitive Impairment

Dieu Ni Thi Doan; Boncho Ku; Kahye Kim; Minho Jun; Kyu Yeong Choi; Kun Ho Lee; Jaeuk U Kim

doi:10.3389/fnut.2022.873623

Segmental Bioimpedance Variables in Association With Mild Cognitive Impairment

Front Nutr. 2022 Jun 2:9:873623. doi: 10.3389/fnut.2022.873623. eCollection 2022.

Authors

Dieu Ni Thi Doan^{1

2}, Boncho Ku¹, Kahye Kim¹, Minho Jun¹, Kyu Yeong Choi³, Kun Ho Lee^{3

4

5}, Jaeuk U Kim^{1

2}

Affiliations

¹ Department of Digital Health Research, Korea Institute of Oriental Medicine, Daejeon, South Korea.
² Korean Convergence Medicine, University of Science and Technology, Daejeon, South Korea.
³ Gwangju Alzheimer's Disease and Related Dementias (GARD) Cohort Research Center, Chosun University, Gwangju, South Korea.
⁴ Department of Biomedical Science, Chosun University, Gwangju, South Korea.
⁵ Dementia Research Group, Korea Brain Research Institute, Daegu, South Korea.

Abstract

Objective: To examine the changes in body composition, water compartment, and bioimpedance in mild cognitive impairment (MCI) individuals.

Methods: We obtained seven whole-body composition variables and seven pairs of segmental body composition, water compartment, and impedance variables for the upper and lower extremities from the segmental multi-frequency bioelectrical impedance analysis (BIA) of 939 elderly participants, including 673 cognitively normal (CN) people and 266 individuals with MCI. Participants' characteristics, anthropometric information, and the selected BIA variables were described and statistically compared between the CN participants and those with MCI. The correlations between the selected BIA variables and neuropsychological tests such as the Korean version of the Mini-Mental State Examination and Seoul Neuropsychological Screening Battery - Second Edition were also examined before and after controlling for age and sex. Univariate and multivariate logistic regression analyses with estimated odds ratios (ORs) were conducted to investigate the associations between these BIA variables and MCI prevalence for different sexes.

Results: Participants with MCI were slightly older, more depressive, and had significantly poorer cognitive abilities when compared with the CN individuals. The partial correlations between the selected BIA variables and neuropsychological tests upon controlling for age and sex were not greatly significant. However, after accounting for age, sex, and the significant comorbidities, segmental lean mass, water volume, resistance, and reactance in the lower extremities were positively associated with MCI, with ORs [95% confidence interval (CI)] of 1.33 (1.02-1.71), 1.33 (1.03-1.72), 0.76 (0.62-0.92), and 0.79 (0.67-0.93), respectively; with presumably a shift of water from the intracellular area to extracellular space. After stratifying by sex, resistance and reactance in lower extremities remained significant only in the women group.

Conclusion: An increase in segmental water along with segmental lean mass and a decrease in body cell strength due to an abnormal cellular water distribution demonstrated by reductions in resistance and reactance are associated with MCI prevalence, which are more pronounced in the lower extremities and in women. These characteristic changes in BIA variables may be considered as an early sign of cognitive impairment in the elderly population.

Keywords: Alzheimer’s disease (AD); bioelectrical impedance analysis (BIA); body composition; mild cognitive impairment (MCI); segmental analysis.