Abstract WP330: Process Measures and Outcomes in Defuse 3 Trial by Day and Week Time Patterns
Introduction: Several studies identified temporal variations in stroke care and suggested that quality of care declines during off-hours and on weekends. Methods: We specified 2 time definitions: a) 8-hour blocks: night (midnight – 7:59), day (8:00 – 15:59), evening (16:00-23:59); b) weekday vs. weekend. We compared process measures and outcomes of the EVT-treated DEFUSE 3 patients based on these definitions. To assign patients to time-blocks, we used date and time of admission to hospital that performed EVT. Results: 92 patients were randomized to EVT treatment: 30% arrived to the treating hospital at night, 49% at day, 21% at evening. Mean age by arrival time 71±14, 70±12, 63±15 (p=0.09); NIHSS 17±5, 15±6, 15±8 (p=0.40); female 39%, 51%, 63% (p=0.27). Weekend admission occurred in 21%. Weekday vs weekend mean age 68±14 vs 72±11 (p=0.19); female 53% vs 37% (p=0.20), higher weekend NIHSS 15±6 vs 18±6 (p=0.04). Onset to arrival at the EVT center varied by time of day: (hrs:mins) 7:04±2:43 night, 8:05±3:01 day, 4:51±2:30 evening, p<0.001. However, day admissions tended to be wake-up/not witnessed strokes more often: 76% vs. 47% evening and 64% night, p=0.09. Transfer times for 57 transfer patients were similar: mean 3:06, 3:09, 3:38. Time from arrival at the treating hospital to groin puncture varied by presenting times: 2:28±1:11 night, 1:45±0:46 day, 2:36±2:32 evening (p=0.02). Time-metrics for weekday vs weekend were similar. Rates of successful reperfusion, 90-day mRS and mRS=0-2 did not differ by time of day or week. There was higher 90-day mortality (32% vs 10%, p=0.02) and in-hospital mortality (21% vs. 4%, p=0.03) on weekend. Symptomatic ICH also occurred more commonly in weekend admits (21% vs 3%, p=0.01). However, after adjustment for age and NIHSS, presenting on weekend was not independently associated with mortality, p=0.13. Conclusions: DEFUSE 3 patients admitted during the day had the longest time from last known well to arrival at the study site due to the high percentage of wake-up strokes admitted during this time period, however, these patients had the shortest arrival to groin puncture times. Although mortality rates were higher for patients who presented on weekends, this may be explained by the fact that these patients were older and had higher NIHSS.