Table 2:
Performance measures of NCCT, CTP, final ASPECTS, and 24-hour NIHSS change for clinical outcome
| Thresholda | Good Outcome, mRS ≤2 (% Patients) | RR (95% CI) | P* | Sensitivity (95% CI) | Specificity (95% CI) | Accuracy |
|---|---|---|---|---|---|---|
| NCCT >7 vs ≤7 | 44 (6) | 8.0 (1.1–57.6) | .018 | 0.63 (0.4–0.8) | 0.89 (0.47–0.99) | 0.69 |
| CBV ≥8 vs <8 | 45 (0) | 15.4 (0.96–246) | .002 | 0.60 (0.4–0.8) | 100 (0.66–100) | 0.69 |
| Final NCCT >6 vs ≤6 | 47 (0) | 17 (1.1–27.3) | .001 | 1.00 (0.7–1.0) | 0.63 (0.44–0.78) | 0.72 |
| CBF/MTT >5 vs ≤5 | 46 (13) | 3.54 (1.1–11.8) | .046 | 0.74 (0.5–0.9) | 0.67 (0.35–0.88) | 0.69 |
| Change NIHSS 24 hours ≥8 | 86 (10) | 8 (2.7–25.2) | .0001 | 0.96 (0.8–1.0) | 0.67 (0.3–0.9) | 0.89 |
Note:—RR indicates relative risk; CBV, cerebral blood volume; CBF, cerebral blood flow; MTT, mean transit time.
a
Dichotomized thresholds were determined by ROC analysis. Percentage of patients presenting with good outcome above and below the selected threshold is illustrated. The best early predictors of clinical outcome were 24-hour NIHSS and presentation CBV.
*
P value was calculated using Fisher exact test.