ABSTRACT
Background. Few studies have analysed the natural course of cerebral ischaemia for predicting outcome. We aimed to determine the early clinical findings and the thresholds for deficit severity and symptom duration that make it possible to stratify outcome. Methods. We included 154 patients with transient ischaemic attack or ischaemic stroke. Stroke profiles and neurological status were assessed from onset to 24 hrs, on admission, at 48 hrs, and at discharge. Outcomes were evaluated using the modified Rankin Scale. Positive and negative predictive values were calculated for the different thresholds. The model was subsequently evaluated on a new prospective cohort of 157 patients. Results. Initial National Institute of Health Stroke Scale (NIHSS) score <5 and symptoms regressing within 135 min were predictive of good outcome. Initial NIHSS score >22 and symptom stability after 1,230 min were predictive of physical dependency or death. Conclusions. Low and high NIHSS cut-off points are effective positive predictive values for good and poor outcomes. Thresholds for symptom duration are less conclusive.
ABSTRACT
OBJECTIVES: We tested the hypothesis that calling emergency medical services ("15", French equivalent of 911 or 999 calls) and response by a mobile intensive care unit staffed by emergency physicians (MICU) reduces the time to treatment to within the 3-hour time window required for administration of recombinant tissue plasminogen activator. METHODS: This study compared the time from symptom onset to admission (prehospital time), from admission to treatment (imaging and treatment delays, hospital time), and total time from symptom onset to treatment in an observational cohort of 53 consecutive patients, according to how they reached the hospital (Group 1: MICU and group 2: standard emergency ambulance dispatched by EMS center [2a] or direct admission [2b]). RESULTS: The study included 52 patients (1 was excluded because hospitalized at the time of the stroke): 27 (51.9%) in group 1, 16 (30.8%) in group 2a, and 9 (17.3%) in group 2b. Calling "15" shortened total home-to-needle time by 24 minutes (p=0.034). The mean total time was not significantly shorter in group 1 (152 versus 162 min; p=0.27) but MICUs were used for patients farther away (mean distance 25 versus 11 km; p=0.02). The average prehospital time was thus higher in group 1 (86 versus 69 min; p=0.044), but was compensated by a reduction in the average hospital time (66 versus 93 min; p=0.0001), due mainly to shorter waits for imaging (22 versus 45 min; p=0.0001). CONCLUSION: Calling the emergency services number reduces mean total time. MICUs were associated with a longer prehospital time, mainly due to greater distances, but they facilitated in-hospital management.
