Abstract 2933: Combined Clinical, CT, And CTA-SI Scale To Expect The Presence Of Large Infarct Volume On DWI In Acute Stroke Patients Within 8h Of Onset.
Background and Purpose: Large ischemic volume (>70-80ml) on diffusion-weighted imaging (DWI) has been associated with a lack of response to early recanalization therapy of acute ischemic stroke. We investigated whether a scale that combines clinical, conventional computed tomography (CT) and CT angiography source imaging (CTA-SI) information can estimate large infarct of >80ml seen on early DWI.
Methods: Acute anterior circulation stroke patients within 8h of onset and examined with CT and CTA were prospectively enrolled. Only imaging findings evaluated within 1.5h interval among CT, CTA and DWI were analyzed. NIH Stroke Scale (NIHSS) score was assessed on admission. The Alberta Stroke Programme Early CT Score (ASPECTS) was scored on CT and CTA-SI. To devise the early infarct volume estimation scale, NIHSS score, CT and CTA-SI ASPECTSs were scored as 1, if the NIHSS score was greater than or equal to, and, CT and CTA-SI ASPECTS was smaller than or equal to the cut-off value to distinguish the DWI volume of >80ml from others. Collateralization was also defined as score of 1 if collateral filling was limited <50% on CTA-SI. We compared the DWI volume based on the scale, score of 0, 1, 2, 3 and 4. After that, we calculated the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the score to estimate the DWI volume of >80ml were calculated.
Result: Fifty-seven subjects (median age [quartiles]; 74 [56-82], male; 28 [49%]) were enrolled from May 2010 to July 2011. DWI volume of >80ml was seen in 10 (18%) patients. Using ROC curve, the cut-off value of NIHSS score to predict DWI of >80ml was 20 (sensitivity of 0.70 and specificity of 0.70, area under curve [AUC] 0.761, p=0.010), that of CT ASPECTS was 6 (sensitivity of 0.98 and specificity of 0.68, AUC 0.887, p<0.001), that of CTA-SI ASPECTS was 6 (sensitivity of 0.87 and specificity of 0.80, AUC 0.952, p<0.001), respectively. Based on the scale, DWI volume was 14 (8-23) ml in patients with score of 0, 22 (11-36) ml with of 1, 67 (36-88) ml with of 2, 127 (71-147) ml with of 3, and 177 (131-191) ml with of 4 (p<0.001). All of those with score of 0 or 1 had DWI volume of ≤80ml (sensitivity of 1.00, specificity of 0.87 PPV of 1.00, and NPV of 0.62 (p<0.001). Seven (54%) of the 13 patients with score of 2 or 3 had DWI volume of >80ml (sensitivity of 0.70, specificity of 0.87, PPV of 0.54, and NPV of 0.93, p<0.001), and all of the 3 patients with score of 4 had DWI volume of >80ml (sensitivity of 0.30, specificity of 1.00, PPV of 1.00, and NPV of 0.87, p=0.004).
Conclusion: Clinical and CT based radiological examinations may estimate the large ischemic volume on early DWI.
- © 2012 by American Heart Association, Inc.