Background: The benefit of thrombolysis is highly time dependent. Strategies and system changes to reduce door-to-needle time (DNT) have been proposed but the effectiveness of such strategies has not been fully evaluated.
Hypothesis: Specific change strategies to the system of delivering thrombolysis significantly improve DNT. Stroke severity also affects DNT significantly.
Methods: The Hurry Acute Stroke Treatment and Evaluation (HASTE) project was implemented in 3 phases at a single academic medical center to reduce DNT using four strategies. In HASTE-I (Jun 6 2012 - Jun 5 2013), baseline performance was analyzed with no changes made to stroke treatment. In HASTE-II (Jun 6 2013 - Jan 24 2015), three changes were implemented: 1) a STAT! stroke protocol to pre-notify the stroke team of the severity of incoming stroke patients; 2) administering tPA in the CT scanner; and 3) registering the patient as unknown prior to exact identification, to allow immediate order entry in our electronic health system. In HASTE-III (Jan 25 2015 - Jun 29 2015), we implemented a process to bring the patient directly to CT on the EMS stretcher. Decrease in DNT was analyzed using Wilcoxon rank sum and Kruskal Wallis tests, and multivariable linear regression. Log transformed DNT was modeled using a backward selection approach.
Results: There were 350 patients treated with tPA during the project. The results of the univariable and multivariable analyses are shown in the table. In univariable analyses, the following strategies improved DNT: STAT! stroke, patient registered as unknown, and stretcher to CT. Additionally, DNT was lower if tPA was administered in the CT. The stroke severity also affected DNT. Multivariable regression showed the following factors to be significant: giving tPA in the CT, stretcher to CT, patient registered as unknown, and stroke severity.
Conclusions: Stretcher to CT, patient registered as unknown, and administering tPA in CT were most efficacious in reducing DNT.