Abstract
ABSTRACTBackgroundEarly detection of large vessel occlusion (LVO) facilitates triage to a comprehensive or thrombectomy-capable stroke center to reduce treatment times and improve outcomes. Prehospital stroke scales, however, are not sufficiently sensitive, so here we investigate the ability of the portable Openwater optical blood flow monitor to detect LVO in patients undergoing acute stroke evaluation.MethodsPatients were prospectively enrolled at two comprehensive stroke centers during acute stroke evaluation within 24 hours of symptom onset with NIHSS ≥2. Each patient underwent a 60-second bedside optical blood flow evaluation with the Openwater instrument. The Openwater instrument generates cerebral blood flow waveforms based on relative changes in speckle contrast. Anterior circulation LVO was deteremined based on CT angiographic imaging and defined as occlusion of the ICA, or first/second segment of the MCA. A deep learning model, based on a transformer architecture trained on all patient data using fivefold cross-validation and learned discriminative representations from the raw speckle contrast waveform data. ROC analysis compared the Openwater diagnostic performance (i.e., LVO detection) with performance of prehospital stroke scales.ResultsAmongst 135 patients, the median NIHSS was 8 (IQR: 4-14), and 52 (39%) had an anterior circulation LVO based on CT angiogram. The Openwater instrument had 79% sensitivity and 84% specificity for the detection of LVO. The RACE scale had 60% sensitivity and 81% specificity. LAMS had 50% sensitivity and 81% specificity. In the ROC analysis, the binary Openwater classification (high-likelihood vs low-likelihood) had an AUC of 0.82 (95%CI: 0.75-0.88), which outperformed RACE (AUC: 0.70; 95%CI: 0.62-0.78; p=0.04) and LAMS (AUC: 0.65; 95% CI: 0.57-0.73; p=0.002).ConclusionsThe Openwater optical blood flow monitor outperformed prehospital stroke scales for the detection of LVO in patients who presented to the Emergency Department for acute stroke evaluation. These encouraging findings need to be validated in the prehospital environment.
Publisher
Cold Spring Harbor Laboratory