Several stroke scales have been implemented to enhance early recognition of large vessel occlusion (LVO) in the field. These scales necessitate a tiered approach requiring emergency medical services (EMS) to utilize two scales, one for identifying stroke and another for differentiating LVO from non-LVO. Ideally, a single stroke scale should be utilized by EMS for triage.
This is a prospective analysis of 150 consecutive patients presenting with stroke symptoms from the field. The stroke scale modified Gaze-Face-Arm-Speech-Time (mG-FAST) was used to simultaneously identify stroke and detect LVO in the pre-hospital setting. Imaging was used to confirm the presence of a LVO and determine the sensitivity and specificity of mG-FAST. The receiver operating curve (ROC) was plotted to calculate the area under the curve (AUC). Youden's index was used to determine the optimal cutoff score. Inter-rater reliability was obtained by comparing the EMS and stroke provider mG-FAST scores. EMS dispatch-to-thrombectomy-capable stroke center (mothership, MS) arrival time and groin puncture time were compared before and after the implementation of mG-FAST.
33/150 patients had a confirmed LVO by imaging. 32/33 patients had an mG-FAST score ≥3. The AUC of mG-FAST was 0.899. An mG-FAST cut-off point of ≥3 yielded a sensitivity of 0.97 and specificity of 0.55 for LVO. The accuracy of this cut-off point was 64%. The EMS dispatch-to-MS time and groin puncture time decreased by 22 and 40 min after implementation of mG-FAST, respectively. With admission to the MS, the EMS dispatch-to-MS time decreased by 174.7 min compared to admission to a drip-and-ship (DS) hospital.
Utilizing a single stroke scale in the field improves EMS dispatch-to-MS time, EMS dispatch-to-groin puncture time, and EMS door-to-intervention time. Implementation of mG-FAST as a pre-hospital screening tool is an effective method of triaging patients to the MS or DS hospitals.