Changes in ischemic stroke presentations and associated workflow during the first wave of the COVID-19 pandemic: A population study in Alberta, Canada

ABSTRACT Background Pandemics may promote hospital avoidance among patients with emergencies, and added precautions may exacerbate treatment delays. There is a paucity of population-based data on these phenomena for stroke. We examined the effect of the COVID-19 pandemic on the presentation and treatment of ischemic stroke in an entire population. Methods We used linked provincial administrative data and data from the Quality Improvement and Clinical Research Alberta Stroke Program – a registry capturing stroke-related data on the entire population of Alberta(4.3 million)– to identify all patients presenting with stroke in the pre-pandemic(1-January-2016 to 27-February-2020, n=19,531) and pandemic(28-February-2020 to 30-August-2020, n=2,255) periods. We examined changes in thrombolysis and endovascular therapy(EVT) rates, workflow, and in-hospital outcomes. Results Hospitalizations/presentations for ischemic stroke dropped (weekly adjusted-incidence-rate-ratio[aIRR]:0.48, 95%CI:0.46-0.50, adjusted for age, sex, comorbidities, pre-admission care needs), as did population-level incidence of thrombolysis(aIRR:0.49,0.44-0.56) or EVT(aIRR:0.59,0.49-0.69). However, the proportions of presenting patients receiving acute therapies did not decline (e.g. thrombolysis:11.7% pre-pandemic vs 13.1% during-pandemic, aOR:1.02,0.75-1.38). Onset-to-door times were prolonged; EVT recipients experienced longer door-to-reperfusion times (median door-to-reperfusion:110-minutes, IQR:77-156 pre-pandemic vs 132.5-minutes, 99-179 during-pandemic; adjusted-coefficient:18.7-minutes, 95%CI:1.45-36.0). Hospitalizations were shorter but stroke severity and in-hospital mortality did not differ. Interpretation The first COVID-19 wave was associated with a halving of presentations and acute therapy utilization for ischemic stroke at a population level, and greater pre-hospital and in-hospital treatment delays. Our data can inform public health messaging and stroke care in current and future waves. Messaging should encourage attendance for emergencies and stroke systems should re-examine “code stroke” protocols to mitigate inefficiencies.