The logistic regression models of optimal NPS thresholds
| Factors Included in Logistic Regression Model | OR (95% CI) | AUC (95% CI) |
|---|---|---|
| Model 1 | ||
| Volume of NPS 1.3–1.4, per 1-mm3 increase | 6.25 (1.90–20.57) | 0.884 (0.822–0.945) |
| Stenosis degree, per 10% increase | 1.99 (1.24–3.19) | |
| Remodeling ratio, per 0.1 increase | 1.50 (1.08–2.10) | |
| Fat-suppressed imaging | 2.88 (0.93–8.94) | |
| Model 2 | ||
| Volumetric summation of NPS no less than 1.2, per 1-mm3 increase | 1.30 (1.06–1.58) | 0.863 (0.791–0.935) |
| Stenosis degree, per 10% increase | 1.98 (1.26–3.10) | |
| Remodeling ratio, per 0.1 increase | 1.45 (1.05–2.00) | |
| Fat-suppressed imaging | 2.63 (0.89–7.81) | |
| Model 3a | ||
| Stenosis degree, per 10% increase | 2.43 (1.61–3.67) | 0.806 (0.725–0.887) |
| Remodeling ratio, per 0.1 increase | 1.51 (1.11–2.06) | |
| Fat-suppressed imaging | 2.03 (0.73–5.69) |
↵a The AUC of model 3 is significantly lower than those of model 1 (P = .008) and model 2 (P = .027).