Abstract 2497: Achieving Door-to-Needle Times of 37 minutes using a Streamlined Resident-Based Acute Stroke Protocol

Abstract

Background: While tPA efficacy is enhanced by shorter delivery times, a recent analysis of the National Get-With-the-Guidelines stroke data demonstrated that only 29% of patients receive tPA < 60 min from hospital arrival. To further improve door-to-needle (DTN) times (time from patient arrival to time of tPA administration) within our resident-driven stroke protocol, we carefully reviewed our process to identify inefficiencies, leading to a revised protocol. We tested the efficiency and safety of the new, streamlined protocol comparing protocol metrics, symptomatic hemorrhage rates, and discharge outcomes before and after the changes. Methods: In early 2011, a value stream analysis (VSA) was conducted, identifying several inefficiencies within our tPA protocol which included: (1) routing of patients to room, then CT, then back to room; (2) serial processing of work flow; and (3) delays in waiting for labs. On 3/1/11, a new resident-driven protocol was adopted incorporating changes to minimize delays: (1) routing patients directly to head CT prior to patient room, (2) utilizing a parallel approach to process work flow, and (3) implementing point-of-care (POC) labs. We directly compared the “Pre-VSA” epoch (1/1/09-2/28/11) to the “Post-VSA” epoch (3/1/11-7/31/11) with regard to baseline variables, protocol metrics, and discharge outcomes. Wilcoxon Rank Sum test was used for continuous data and Fisher’s Exact test was used for binary data with p<.05 required for significance. Results: In the Pre-VSA epoch, 132 patients were treated with IV tPA (0-4.5hr window). In the Post-VSA epoch, 37 patients were treated with IV tPA. Baseline variables such as age, NIH Stroke Scale, and past medical history were similar between the 2 epochs except for greater proportion of African-Americans and hypertension in the Post-VSA epoch ( Table ). Median DTN and onset-to-needle times were lower in the Post-VSA epoch: 60 min pre-VSA vs. 37 min post-VSA (p<0.0001) and 131 vs. 102 min. (p=0.056, strong trend), respectively. Of the remaining metrics, door-to-CT time differed between the 2 epochs (16 min vs. 1 min, p<0.0001). To ensure that efficiency was not gained at the cost of patient safety, “Favorable” discharge location (discharge to home or inpatient rehabilitation), symptomatic hemorrhage rate, and percentage of stroke mimics (patients with non-stroke diagnosis at discharge) were compared and did not differ between the two groups ( Table ). Conclusions: A resident-based stroke protocol using efficient choreography, parallel processing, and POC labs can accelerate tPA delivery without compromising patient safety.