A REAl-life Study on Short-term DAPT in Patients With Ischemic Stroke or TIA

The READAPT will depict the benefit/risk profile of DAPT in a clinical setting, and address subgroups of patients such as those with small cerebral vessel diseases or those treated with revascularization procedures.

Randomized clinical trials (RCTs) proved that short-term DAPT is superior over single antiplatelet treatment in reducing the ischemic recurrence risk, without a remarkable increased hemorrhagic risk thanks to the short treatment course. However, RCTs excluded patients treated with revascularization procedures (i.e. intravenous thrombolysis and thrombectomy), did not provide data on neuroimaging, and had slightly different treatment procedures such as time-to-DAPT start and antiplatelets loading dose.

The study comprises a baseline coinciding with the index event, when the investigators will collect demographics and characteristics of the event, and a 90±5 day follow-up from the index event, when patients will be screened for treatment compliance, tolerability and ischemic or hemorrhagic events. Follow-up visit can be performed remotely.

The investigators did not establish strict NIHSS or ABCD2 score cutoff for patients' inclusion, as treatment decision has to be taken independently from the study, and highly recommend physicians to adhere to guidelines. Each participating center will include all consecutive patients (hospitalized or non-hospitalized) who will meet inclusion criteria.

Data were entered in an electronic anonymized database created on the Research Electronic Data Capture (REDCap) software for the analyses hosted at University of L'Aquila. The local PI or the co-investigators will be able to upload patients' data through a single form specifically created for the study, which will include a user-friendly drop-down menu. Anonymized data will be stored on a secured server under the responsibility of University of L'Aquila. The data will be automatically backed-up once a week. Data will not be shared with unauthorized persons. Plausibility of the entered data will be checked by the study manager and the statistical data manger and data queries will be resolved by Local PI. Cases with missing data or unresolved queries will be rejected to retain only the highest quality data in the registry. Data from centers not ensuring consecutive recruitment of patients or adequate follow-up will not be included in the final database.

All analyses will be performed according to the intention-to-treat principle in all included patients completing the 90-day follow-up or having a fatal outcome event within 90 days. Descriptive statistics will be used to report baseline information. The investigators will analyze the time from index event to the first occurrence of primary and secondary outcome events with the use of a Cox proportional hazards model. Two statistical models will be used: Model 1 unadjusted and Model 2 adjusted for demographics and characteristics of the index event. P values for interaction will be calculated according to the following subgroups: type of event (ischemic stroke vs TIA), time to DAPT (≤24 hours vs >24 hours from symptom onset), type of DAPT (aspirin+clopidogrel vs aspirin+ticagrelor), DAPT duration (≤21 vs >21 days and ≤30 vs >30 days), NIHSS score at onset (≤3 vs >3 and ≤5 vs >5), revascularization procedure (i.v. thrombolysis and/or mechanical thrombectomy vs no interventions). Hazard ratios with 95% confidence intervals will be reported. Should multiple events of the same type occur, the time to the first event will be used in the model. Data from patients who had no events during the study will be censored at the time of study termination or death. Assuming a 95% confidence interval, an estimated sample size of 1067 subjects would be required to detect a conservative 50% proportion of primary outcome occurrence with a two-sided 2.5% margin of error.