Frequency, Predictors, and Outcomes of Prehospital and Early Postarrival Neurological Deterioration in Acute Stroke: Exploratory Analysis of the FAST-MAG Randomized Clinical Trial.

Importance: Studies of neurological deterioration in stroke have focused on the subacute period, but stroke treatment is increasingly migrating to the prehospital setting, where the neurological course has not been well delineated. Objective: To describe the frequency, predictors, and outcomes of neurological deterioration among patients in the ultra-early period following ischemic stroke or intracranial hemorrhage. Design, Settings, and Participants: Exploratory analysis of the prehospital, randomized Field Administration of Stroke Therapy-Magnesium (FAST-MAG) Trial conducted from 2005 to 2013 within 315 ambulances and 60 stroke patient receiving hospitals in Southern California. Participants were consecutively enrolled patients with suspected acute stroke who were transported by ambulance within 2 hours of stroke onset. Main Outcomes and Measures: The main outcome was neurological deterioration, defined as a worsening of 2 or more points on the Glasgow Coma Scale (GCS), a level of consciousness scale ranging from 3 to 15, with higher scores indicating more alertness. Imaging outcomes were ischemic or hemorrhagic injury extent identified during the first brain imaging scan. Outcomes at 3 months included global disability level (assessed using the modified Rankin Scale [mRS]; range, 0-6, with higher numbers indicating greater disability) and mortality. Results: Among the 1690 patients (99.4%), the mean (SD) age was 69.4 (13.5) years, and 43% were female. Final diagnoses were acute cerebral ischemia in 1237 patients (73.2%), intracranial hemorrhage in 386 patients (22.8%), and neurovascular mimic in 67 patients (4.0%). The median (interquartile range [IQR]) minutes between the last well-known time and GCS assessments were 23 (14-42) minutes for prehospital, 58 (46-79) minutes for ED arrival, and 149 (120-180) minutes for early ED course assessments. From prehospital to early postarrival, ultra-early neurological deterioration (U-END) occurred in 200 of 1690 patients (11.8%), more often among patients with intracranial hemorrhage than among those with acute cerebral ischemia (119 of 386 [30.8%] vs 75 of 1237 [6.1%], P < .001). Patterns of U-END were prehospital U-END without early recovery in 30 of 965 patients (3.1%), stable prehospital course but early ED deterioration in 49 of 965 patients (5.1%), and continuous deterioration in both prehospital and early ED phases in 27 of 965 patients (2.8%). Ultra-early neurological deterioration was associated with worse 3-month outcomes, including increased global disability (mRS score, 4.6 vs 2.4; P < .001), reduced functional independence (mRS score 0-2, 32 of 200 [16.0%] vs 844 of 1490 [56.6%]; P < .001), and increased mortality (87 of 200 [43.5%] vs 176 of 1490 [11.8%]; P < .001). Conclusions and Relevance: Ultra-early neurological deterioration occurs in 1 in 8 ambulance-transported patients with acute cerebrovascular disease, including 1 in 3 patients with intracranial hemorrhage and 1 in 16 patients with acute cerebral ischemia, and is associated with markedly reduced functional independence and increased mortality. Averting U-END may be a target for future prehospital therapeutics. Trial Registration: ClinicalTrials.gov Identifier: NCT00059332.