Proposition and operational characteristics of a new dispatch scale to specifically identify acute strokes

Prehospital recognition of an acute stroke improves the time from onset to thrombolysis and rates of reperfusion therapy. Studies conducted to evaluate paramedic and dispatcher performance in suspecting stroke are disappointing. This study addresses the specific issue of stroke recognition by dispatchers, taking into account delay in reporting the onset of symptoms (<5 h).

This is an observational analysis conducted over a 12-month period. Dispatchers used a modified Cincinnati Stroke Scale to specifically identify acute strokes in a criteria-based dispatch. Data were extracted from the State's dispatch and the State's stroke centre. All calls to the dispatch were included. Dispatcher's suspicion of acute stroke and the patient's final destination and diagnosis were collected. Simple descriptive statistics were used. Sensitivity and positive predictive value were calculated.

The dispatch received 27 719 calls resulting in ambulance dispatches; 427 calls [1.5%; 95% confidence interval (CI) 1.4-1.7] were classified as suspicion of acute stroke by dispatchers, and 40 of them (9.4%; 95% CI 6.6-12.1) fulfilled the criteria for thrombolysis (sensitivity 67.8%; 95% CI 54.3-79.4%). Dispatchers missed 19 of 59 strokes (32.2%; 95% CI 20.3-44.1) that received thrombolysis; 16 cases were missed because of unspecific acute symptoms (unconsciousness, dyspnoea), and three more because of unspecific nonacute symptoms (vertigo, dizziness).

The revised Cincinnati Stroke Scale for dispatch adds the notion of delay in the process of triage. It identifies 67.8% and misses 32.2% of the stroke patients treated by thrombolysis. Its performance is similar to previous results using the regular Cincinnati Stroke Scale, but allows for targeting acute strokes.