Table 2.
Quantile regression analyses assessing the interaction of body mass/adiposity with cardiorespiratory fitness on inflammatory markers.
| hsCRP | ||||||
|---|---|---|---|---|---|---|
| B | SE | 95% CI | p | |||
| BMI | 0.209 | 0.063 | 0.082 | , | 0.336 | 0.002 |
| CRF | 0.01 | 0.005 | -0.001 | , | 0.02 | 0.065 |
| BMI×CRF | NS* | |||||
| WHtR | 4.661 | 5.193 | -5.811 | , | 15.133 | 0.374 |
| CRF | 0.005 | 0.005 | -0.006 | , | 0.016 | 0.371 |
| WHtR×CRF | NS* | |||||
| BF% | 0.048 | 0.038 | -0.028 | , | 0.124 | 0.213 |
| CRF | 0.006 | 0.005 | -0.048 | , | 0.017 | 0.268 |
| BF%×CRF | NS* | |||||
| IL-6 | ||||||
| B | SE | 95% CI | p | |||
| BMI | 5.504 | 1.478 | 2.509 | , | 8.5 | 0.001 |
| CRF | 0.229 | 0.071 | 0.085 | , | 0.372 | 0.003 |
| BMI×CRF | -0.009 | 0.003 | -0.015 | , | -0.004 | 0.002 |
| WHtR | 142.8 | 71.72 | -2.522 | , | 288.1 | 0.054 |
| CRF | 0.122 | 0.069 | -0.019 | , | 0.263 | 0.088 |
| WHtR×CRF | -0.245 | 0.135 | -0.519 | , | 0.029 | 0.078 |
| BF% | 1.351 | 0.537 | 0.264 | , | 2.438 | 0.016 |
| CRF | 0.076 | 0.035 | 0.004 | , | 0.148 | 0.039 |
| BF%×CRF | -0.002 | 0.001 | -0.004 | , | 0 | 0.026 |
| Leptin | ||||||
| B | SE | 95% CI | p | |||
| BMI | 3.188 | 0.568 | 2.038 | , | 4.338 | <0.001 |
| CRF | 0.016 | 0.047 | -0.079 | , | 0.111 | 0.741 |
| BMI×CRF | NS* | |||||
| WHtR | -356.93 | 295.76 | -956.21 | , | 242.33 | 0.235 |
| CRF | -0.587 | 0.283 | -1.16 | , | -0.013 | 0.045 |
| WHtR×CRF | NS* | |||||
| BF% | 2.442 | 0.346 | 1.742 | , | 3.142 | <0.001 |
| CRF | 0 | 0.048 | -0.097 | , | 0.097 | 0.997 |
| BF%×CRF | NS* | |||||
The markers of inflammation were high sensitivity C-reactive protein (hsCRP), interleukin 6 (IL-6) and leptin. The markers of body mass/adiposity were body mass index (BMI), waist-to-height ratio (WHtR) and body fat percentage (BF%); CRF, cardiorespiratory fitness.
B, unstandardized regression coefficient indicating the expected unit change in the dependent variable for one-unit change in the independent variable; SE, standard error NS, non-significant.
Quantile regression models were built including each inflammatory marker as dependent variable in separate models and the body mass or adiposity indicator, CRF, and the body mass/adiposity×CRF interaction as independent variables. All the analyses were adjusted for age, SLEDAI, and accumulated corticosteroid intake. When the interaction was not significant, the interaction term was removed from the regression model and the results are presented without interaction (i.e., the independent association of body mass/adiposity and CRF with the inflammatory marker).