Table 2.
Univariate and ROC analyses for correlation between the variables and ROTA-related coronary intimal dissection.
| Variables | Univariate analysis | ROC analysis | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| r | p | AUC | 95%CI | Sensitivity | Specificity | Cutoff value | p | YI | ||
| Dcib | −0.407 | <0.001 | 0.942 | 0.920 | 0.964 | 86.4% | 95.4% | 0.435 | <0.001 | 0.828 |
| Dcio | 0.139 | 0.006 | 0.664 | 0.592 | 0.737 | 67.9% | 65.1% | 1.125 | <0.001 | 0.330 |
| Dcmb | −0.625 | <0.001 | 1.000 | 0.999 | 1.000 | 98.8% | 100.0% | 0.720 | <0.001 | 0.988 |
| Dcmo | 0.169 | 0.001 | 0.661 | 0.584 | 0.738 | 50.0% | 80.7% | 2.875 | <0.001 | 0.307 |
| Final burr size | −0.133 | 0.009 | 0.581 | 0.510 | 0.652 | – | – | – | 0.053 | – |
| Touch angle | 0.912 | <0.001 | 0.988 | 0.968 | 1.000 | 94.6% | 98.2% | 98.2 | <0.001 | 0.946 |
| MinLD | −0.104 | 0.040 | 0.540 | 0.459 | 0.621 | – | – | – | 0.343 | – |
| MaxLD | −0.119 | 0.019 | 0.566 | 0.495 | 0.637 | – | – | – | 0.112 | – |
Univariate and ROC analyses for dissection predictors. AUC: Area under ROC curve (95%CI). Dcib/Dcio/Dcmb/Dcmo: Distances defined in Figure 1. MaxLD/MinLD: Lumen diameters (mm). YI: Youden’s index (sensitivity + specificity − 1). Optimal cutoffs: Dcmb = 0.720 mm (sensitivity 98.8%, specificity 100%), touch angle = 98.2° (94.6%/98.2%). p-values from DeLong’s test for AUC comparisons.