Background: classification of Transient Ischemic attacks (TIA) and minor stroke is challenging, as there is no classification systems developed specifically for the TIA and minor stroke patient population. Hypothesis: We hypothesize that the newly developed Causative Classification System (CCS) and the Atherosclerosis Small Vessel Disease Cardiac Source Other Source (ASCO) classification would reduce the proportion of patients classified as cause undetermined compared with The Trial of ORG 10172 in Acute Stroke Treatment (TOAST) classification in a large prospectively evaluated TIA and Minor stroke population. Methods: Using published algorithms for TOAST, CCS, and ASCO, a single rater classified the etiology in patients presenting with a high-risk TIA (weakness or speech disturbance lasting ≥ 5minutes) or minor ischemic stroke (National Institute of Health Stroke Scale score ≤ 3) who underwent CT/CTA and subsequent MRI as part of the CATCH study. Results: 419 patients with TIA or Minor stroke were classified using TOAST, CCS, and ASCO. The proportion of patients with an undetermined etiology was 51.3% (215/419) with TOAST. This was significantly reduced by both CCS 36% (151/419, p< 0.001) and ASCO 41% (172/419, p< 0.001). CCS was also less likely to have an undetermined etiology as compared to ASCO (36% versus 41%, p = 0.024). When compared with TOAST, there was a 23.9% (95%CI:18.1- 29.7, P< 0.001) and 17.4% (10.1- 24.7, P< 0.001) reduction in the proportion of patients assigned to the undetermined group using CCS and ASCO respectively. The 8.5 % reduction in the undetermined group between CCS and ASCO was also statistically different P=0.031). Compared with ASCO1, CCS increased the assignment of patients to large artery disease (relative increase 7.4% {4.3-10.4}, P< 0.001) and Cardio-embolism/cardio-aortic categories (relative increase 8.1% {4.6-11.5}, P< 0.001). Conclusions: Both CCS and ASCO were superior to TOAST in assigning fewer patients to an undetermined etiology category. CCS was superior to ASCO at reducing the proportion of patients with undetermined etiology. This was largely driven by increased assignment in the large artery and Cardio-aorto embolic categories.
Read full abstract