Table 4.
Diagnostic performance and characteristics of BMI in predicting poor collaterals.
| Characteristics | All Cases | ICA | M1 | Proximal M2 | ||||
|---|---|---|---|---|---|---|---|---|
| Value | 95% CI | Value | 95% CI | Value | 95% CI | Value | 95% CI | |
| Poor Collaterals from Good Collateral | ||||||||
| AUC | 0.560 | 0.401–0.720 | 0.813 | 0.465–1.000 | 0.550 | 0.320–0.781 | 0.465 | 0.199–0.730 |
| p-value | 0.446 | 0.149 | 0.663 | 0.790 | ||||
| Cut point | ≤35.0 | ≤35.0 | ≤35.0 | ≤35.0 | ||||
| Sensitivity | 96.4% | 81.7–99.9% | 100% | 39.8–100% | 100% | 75.3–100% | 90.9% | 58.7–99.8% |
| Specificity | 30.8% | 14.3–51.8% | 75.0% | 19.4–99.4% | 23.1% | 5.0–53.8% | 22.2% | 2.8–60.0% |
| DA | 64.8% | 50.6–77.3% | 87.5% | 47.3–99.7% | 61.5% | 40.6–79.8% | 60.0% | 36.1–80.9% |
| YI | 27.2% | 8.2–46.2% | 75.0% | 32.6–100% | 23.1% | 0.2–46.0% | 13.1% | 18.9–45.2% |
| PPV | 60.0% | 44.3–74.3% | 80.0% | 28.4–99.5% | 56.5% | 34.5–76.8% | 58.8% | 32.9–81.6% |
| NPV | 88.9% | 51.8–99.7% | 100% | 29.2–100% | 100% | 29.2–100% | 66.7% | 9.4–99.2% |
AUC: Area under the curve; CI: Confidence interval. DA: Diagnostic accuracy. YI: Youden’s Index. PPV: Positive Predictive value. NPV: Negative Predictive value.