Diabetes mellitus and insulin resistance associate with left ventricular shape and torsion by cardiovascular magnetic resonance imaging in asymptomatic individuals from the multi-ethnic study of atherosclerosis

Table 3 Association of LV remodeling and function by HOMA-IR among participants without diabetes mellitus (n = 1266)*

Dependent variables (LV indices)	Multivariable liner regression coefficients
Dependent variables (LV indices)	Log-HOMA-IR	R²
End-diastolic volume, ml
Model 1	2.66 (1.40)	0.36
Model 2	−2.53 (1.48)	0.41
Mass, g
Model 1	11.58 (1.58)‡	0.49
Model 2	3.92 (1.61)†	0.56
Mass-to-volume
Model 1	0.07 (0.01)‡	0.18
Model 2	0.05 (0.01)‡	0.20
Sphericity index, %
Model 1	−1.06 (0.33)‡	0.11
Model 2	−1.26 (0.35)‡	0.11
Ejection fraction, %
Model 1	0.15 (0.38)	0.18
Model 2	0.25 (0.41)	0.18
Longitudinal shortening, %
Model 1	−0.26 (0.13)†	0.16
Model 2	−0.22 (0.14)	0.17
Circumferential shortening, %
Model 1	−0.30 (0.14)†	0.13
Model 2	−0.08 (0.15)	0.14
Torsion, ⁰/cm
Model 1	0.13 (0.06)†	0.16
Model 2	0.11 (0.07)	0.17

Coefficients (standard error) represents the change in dependent variables corresponds to 1 unit increase in independent variables. Model 1 included age, gender, race, height, smoking status(never, former, or current), heart rate, hypertension, total cholesterol, use of medication for dyslipidemia, estimated glomerular filtration rate, walking in METs per week, and alcohol status (non-drinker, former, or current). Model 2 included systolic blood pressure, anti-hypertensive medication use, and obesity (BMI ≥ 30 kg/m²) in addition to Model 1
Homeostasis model assessment-estimated insulin resistance (HOMA-IR) was calculated as insulin (mU/l) × (glucose [mg/dl] × 0.055)/22.5. [16, 17]
*Participants with diabetes mellitus were excluded because participants with DM were excluded from the HOMA-IR calculation
†p < 0.05, ‡p < 0.01

ISSN: 1532-429X