|
Model 1
| | | |
Model 2
| | |
|---|
|
Variable
|
Exp(B)
|
95% CI
|
p
|
Exp(B)
|
95% CI
|
p
|
|---|
|
Intercept
|
0.003
|
0.000; 0.022
|
<.001
|
0.005
|
0.000; 0.050
|
<.001
|
|
Male sex
|
4.218
|
1.403; 14.207
|
0.014
|
4.167
|
1.050; 20.178
|
0.05
|
|
Age (y) - 50
|
1.111
|
1.047; 1.194
|
0.002
|
1.083
|
1.011; 1.177
|
0.037
|
|
CKD Stage 4 vs. 3
|
3.290
|
1.068; 10.773
|
0.041
|
N/A
|
N/A
|
N/A
|
|
Carotid plaque
|
6.131
|
1.605; 27.983
|
0.011
|
17.387
|
2.750;175.88
|
0.006
|
|
Cutoff point
| | | | | | |
|
((FEP/FGF23) < 1/3.9)
|
3.915
|
1.346; 12.364
|
0.015
|
6.873
|
1.703; 35.999
|
0.011
|
- Exp (B) for the intercept measures the estimated odds of KI > 5 for the reference or zero values of the explanatory variables in the model. Exp (B) of the predictor measures the odds ratio (B) associated to the variable category or 1 unit change depending of the nature of the variable. The table provides with Exp (B) and 95% confidence intervals (CI) for variables with a statistically significant (p < 0.05) contribution to explain the magnitude of abdominal aortic calcification (AAC) in a multivariate logistic regression model comparing KI > 5 vs. KI = 0 for all patients (Model 1), or among patients with an estimated GFR below 30 ml/min (Model 2). The ratio FEP/FGF23 was introduced as a binary variable with a cutoff point of (FEP/FGF23) < (1/3.9) (or equivalently, log2(FEP/FGF23) < log2(1/3.9)). The ROC curve in Figure 2 shows the high sensitivity and specificity of the logistic regression analysis in Model 1 (Area under the ROC curve = 0.89 and good model calibration, as measured by Hosmer-Lemenshow goodness-of-fit test p = 0.95). For Model 2, the area under the ROC curve = 0.899 and HL test p = 0.55.
