TABLE 6.
Multivariable-Adjusted Linear Regression Models for Changes in Abdominal Fat Attenuation With Changes in Cardiovascular Disease Risk Factors
| Model | Overall Participants | ||||
|---|---|---|---|---|---|
|
| |||||
| Δ in SAT Attenuation | p Value | Δ in VAT Attenuation | p Value | ||
| Δ in Systolic blood pressure | Sample size | 988 | |||
| MV* | 2.89 (1.86, 3.93) | <0.0001 | 2.55 (1.85, 3.24) | <0.0001 | |
| MV + BMI* | 2.81 (1.77, 3.84) | <0.0001 | 2.57 (1.84, 3.30) | <0.0001 | |
| MV + WC* | 2.80 (1.77, 3.84) | <0.0001 | 2.62 (1.88, 3.36) | <0.0001 | |
| MV + fat volume* | 2.82 (1.78, 3.85) | <0.0001 | 2.60 (1.84, 3.36) | <0.0001 | |
| Sex interaction† | – | 0.23 | – | 0.01 | |
|
| |||||
| Δ in Diastolic blood pressure | Sample size | 988 | |||
| MV* | 2.24 (1.54, 2.94) | <0.0001 | 2.01 (1.53, 2.48) | <0.0001 | |
| MV + BMI* | 2.20 (1.50, 2.90) | <0.0001 | 2.01 (1.51, 2.51) | <0.0001 | |
| MV + WC* | 2.21 (1.50, 2.91) | <0.0001 | 2.05 (1.55, 2.56) | <0.0001 | |
| MV + fat volume* | 2.21 (1.50, 2.91) | <0.0001 | 2.06 (1.54, 2.58) | <0.0001 | |
| Sex interaction† | – | 0.26 | – | 0.07 | |
|
| |||||
| Δ in Fasting plasma glucose | Sample size | 1,084 | |||
| MV* | 1.29 (0.03, 2.55) | 0.04 | 1.57 (0.71, 2.43) | 0.0004 | |
| MV + BMI* | 1.23 (−0.02, 2.49) | 0.06 | 1.55 (0.65, 2.44) | 0.0007 | |
| MV + WC* | 1.20 (−0.06, 2.45) | 0.06 | 1.53 (0.62, 2.44) | 0.001 | |
| MV + fat volume* | 1.19 (−0.07, 2.45) | 0.06 | 1.61 (0.68, 2.54) | 0.0007 | |
| Sex interaction† | – | 0.11 | – | 0.13 | |
|
| |||||
| Δ in Total cholesterol | Sample size | 988 | |||
| MV* | 5.01 (2.40, 7.62) | 0.0002 | 2.87 (1.06, 4.68) | 0.002 | |
| MV + BMI* | 5.20 (2.63, 7.78) | <0.0001 | 4.40 (2.53, 6.27) | <0.0001 | |
| MV + WC* | 5.35 (2.76, 7.93) | <0.0001 | 4.47 (2.57, 6.37) | <0.0001 | |
| MV + fat volume* | 5.31 (2.72, 7.90) | <0.0001 | 4.97 (3.04, 6.90) | <0.0001 | |
| Sex interaction† | – | 0.58 | – | 0.18 | |
|
| |||||
| Δ in HDL cholesterol | Sample size | 988 | |||
| MV* | −3.25 (−4.21, −2.29) | <0.0001 | −3.64 (−4.29, −2.99) | <0.0001 | |
| MV + BMI* | −3.23 (−4.20, −2.27) | <0.0001 | −3.70 (−4.38, −3.02) | <0.0001 | |
| MV + WC* | −3.21 (−4.17, −2.25) | <0.0001 | −3.66 (−4.35, −2.96) | <0.0001 | |
| MV + fat volume* | −3.23 (−4.19, −2.27) | <0.0001 | −3.93 (−4.64, −3.23) | <0.0001 | |
| Sex interaction† | – | 0.0005 | – | 0.64 | |
|
| |||||
| Δ in Log triglycerides | Sample size | 988 | |||
| MV* | 0.11 (0.07, 0.15) | <0.0001 | 0.11 (0.09, 0.14) | <0.0001 | |
| MV + BMI* | 0.11 (0.07, 0.15) | <0.0001 | 0.13 (0.10, 0.15) | <0.0001 | |
| MV + WC* | 0.11 (0.07, 0.15) | <0.0001 | 0.13 (0.10, 0.16) | <0.0001 | |
| MV + fat volume* | 0.11 (0.07, 0.15) | <0.0001 | 0.14 (0.11, 0.16) | <0.0001 | |
| Sex interaction† | – | 0.03 | – | 0.15 | |
Values are estimated β (95% confidence interval). The results show the association of changes in cardiovascular disease risk factor for each additional 5 HU decrease in the change in abdominal fat attenuation.
MV model, adjusted for baseline fat attenuation, baseline cardiovascular disease risk factors, age, sex, current smoking, alcohol intake, physical activity, postmenopausal status (women only), and hormone replacement therapy (women only). MV + BMI model was additionally adjusted for BMI change. MV + WC model was additionally adjusted for waist circumference change. MV + fat volume model was additionally adjusted for respective fat volume (SAT volume for SAT attenuation and VAT volume for VAT attenuation).
Sex interaction was examined based on the multivariable-adjusted model.