Background In eligible patients with acute ischaemic stroke, rapid revascularisation is crucial for good outcome. At our treatment centre, we had achieved and sustained a median door-to-needle time of under 30 min. We hypothesised that further improvement could be achieved through implementing a revised treatment protocol and in situ simulation-based team training sessions. This report describes a quality improvement project aiming to reduce door-to-needle times in stroke thrombolysis.

Methods All members of the acute stroke treatment team were surveyed to tailor the interventions to local conditions. Through a review of responses and available literature, the improvement team suggested changes to streamline the protocol and designed in situ simulation-based team training sessions. Implementation of interventions started in February 2017. We completed 14 simulation sessions from February to June 2017 and an additional 12 sessions from November 2017 to March 2018. Applying Kirkpatrick’s four-level training evaluation model, participant reactions, clinical behaviour and patient outcomes were measured. Statistical process control charts were used to demonstrate changes in treatment times and patient outcomes.

Results A total of 650 consecutive patients, including a 3-year baseline, treated with intravenous thrombolysis were assessed. Median door to needle times were significantly reduced from 27 to 13 min and remained consistent after 13 months. Risk-adjusted cumulative sum charts indicate a reduced proportion of patients deceased or bedridden after 90 days. There was no significant change in balancing measures (stroke mimics, fatal intracranial haemorrhage and prehospital times).

Conclusions Implementing a revised treatment protocol in combination with in situ simulation-based team training sessions for stroke thrombolysis was followed by a considerable reduction in door-to-needle times and improved patient outcomes. Additional work is needed to assess sustainability and generalisability of the interventions.