To the Editor I read with great interest the article by Nakanishi et al. assessing the relationship between HbA1c levels and the phase angle (PhA) obtained by a bioelectrical impedance analysis (BIA) in non-elderly patients with type 2 diabetes mellitus (DM) (1). They concluded that the PhA, as a reflection of muscle quality, serves as an independent indicator of glycemic control and may contribute to individualized treatment strategies. However, this conclusion is somewhat premature.
First, the authors stratified patients into tertiles according to PhA; however, the mean HbA1c values in all three groups exceeded 7%, suggesting that they did not identify a subgroup with good glycemic control, a so-called trouble with tertiles (2). A finer stratification (e.g., quartiles) might be necessary to isolate patients with HbA1c ＜7%; however, in that case, it would be more appropriate to present a scatter plot using continuous values of PhA and HbA1c and perform correlation analysis. Furthermore, if the authors wish to propose PhA as an indicator of good glycemic control, they should determine the optimal cutoff value for identifying HbA1c ＜7% through a receiver operating characteristic analysis - something the readership of Internal Medicine would certainly appreciate.
An attempt to explain the association between PhA and glycemic control in terms of “muscle quality” also seems somewhat hasty and insufficiently supported. Although a BIA allows for a detailed assessment of body composition, the authors discussed only PhA and BMI, parameters with limited resolution. Given that a BIA can estimate muscle mass and fat mass separately, it would have been reasonable to include the skeletal muscle mass index, which reflects “muscle quantity,” in their analysis (3).
Finally, I would like to highlight a key parameter that should not be overlooked when examining the relationship between BIA and HbA1c: Hb. In our recent study of hospitalized cardiovascular patients, we demonstrated that the extracellular water to total body water ratio (ECW/TBW) derived from a BIA was strongly correlated with Hb levels, suggesting that a BIA may be useful in predicting anemia (4). Since PhA is known to have a strong negative correlation with ECW/TBW and HbA1c is naturally related to Hb (5), hemoglobin may serve as the missing link in this study's observed relationship.
We hope that further clinically meaningful insights can be extracted from the authors' valuable data.
The author states that he has no Conflict of Interest (COI).
References
- 1.Nakanishi S, Shimoda M, Kimura T, et al. Association between phase angle and HbA1c in non-elderly patients with type 2 diabetes mellitus: a cross-sectional study using outpatient clinical data. Intern Med 65: 1458-1463, 2026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lamb KE, White SR. Categorisation of built environment characteristics: the trouble with tertiles. Int J Behav Nutr Phys Act 12: 19, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Taniguchi M. Clinical application of skeletal muscle quantity and quality assessment using bioelectrical impedance and ultrasound images. Phys Ther Res 27: 49-57, 2024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Nishikawa T, Higaki A, Okada Y, et al. Role of a bioelectrical impedance analysis in predicting anemia among cardiovascular disease patients. Intern Med 64: 2534-2540, 2025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Sakamoto N, Hu H, Nanri A, et al. Associations of anemia and hemoglobin with hemoglobin A1c among non-diabetic workers in Japan. J Diabetes Investig 11: 719-725, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]