Successful endovascular therapy for acute ischemic stroke (AIS) requires optimal patient selection and effective tools, and also the rapid delivery of treatment from the time of symptom onset. Time to treatment for patients with AIS remains, arguably, the most important modifiable factor in improving outcomes.

Although intravenous tissue plasminogen activator (tPA) administration has been extended to 4.5 h,1 ,2 fewer than one-third of eligible patients receive it within the target door to needle (DTN) time of 60 min.3 This failure to achieve benchmark times has led to ‘Target: Stroke’, a national quality improvement initiative of the American Heart Association/American Stroke Association which aims to increase the timeliness of intravenous tPA administration. Over the past 20 years, tremendous efforts in developing primary stroke centers and implementing strategies based on the development of these centers have led to a dramatic reduction in DTN times, but there is much room for improvement.

Studies in myocardial infarction (MI) have shown the critical importance of time in improving outcome for patients undergoing percutaneous coronary intervention (PCI). In a large meta-analysis of 27 MI trials,4 a strong relationship between PCI-related time delay and benefits from primary angioplasty was observed in high-risk patients (r=–0.50, β= –0.075 (95% CI –0.189 to 0.04)), with 131 min as the equipoint between PCI and medical therapies. Moreover, for every 10 min reduction in time to PCI, MI mortality was also reduced.4 In a prospective PCI trial investigating the relationship between time and outcome, patients randomized <2 h after symptom onset demonstrated a strong trend towards a lower 30-day death rate with PCI than with medical treatment (2.2% vs 5.7%, p=0.058), with no difference observed in those randomized after 2 h.5

A similar emphasis on time to treatment in AIS studies has also been demonstrated.6–9 In a collaborative pooled analysis of seven AIS endovascular therapy databases, the impact of onset-to-reperfusion time (in large artery occlusions) on outcomes was assessed. A total of 480 cases with successful reperfusion (median time 285 min) were included in the pooled analysis (120 with internal carotid artery occlusion and 360 with isolated middle cerebral artery occlusion). The adjusted OR for each 30-min time increase was 1.21 (95% CI 1.09 to 1.34; p<0.001) for mortality, 0.79 (95% CI 0.72 to 0.87) for favorable outcome, 0.78 (95% CI 0.71 to 0.86) for excellent outcome, and 1.21 (95% CI 1.10 to 1.33) for intracerebral hemorrhage.6

In the Solitaire With the Intention For Thrombectomy (SWIFT) prospective stent retriever trial, a total of 63 patients achieved reperfusion, with a good clinical outcome (modified Rankin Scale (mRS) score ≤2) in 57% of patients and a mean time to treatment of 310 min (compared with 348 min in those with mRS >2 at 90 days, p=0.08). Patients who died (19%) had a statistically longer time to treatment than those who survived (320 vs 270 min, p=0.04).7 In another study of 29 patients treated with endovascular therapy, 11 patients achieved reperfusion within <60 min of CT/CT angiography imaging and 82% had mRS ≤1 compared with 66.7% of patients in the ≥60 min group.10 In a retrospective study of 193 patients which studied the time from the initial triage non-contrast CT scan to puncture time (‘Picture to Puncture’, P2P), a shorter P2P was found to be independently associated with good outcome in AIS endovascular therapy.11 In this study, further analysis demonstrated an increased likelihood of a good outcome with P2P <90 min.11

Despite the staggering amount of brain tissue lost with each minute of ischemia12 and the repeated demonstration of time-linked outcomes in revascularization therapies in the Interventional Management of Stroke (IMS) I and III trials, the pooled analysis and SWIFT trial,6–9 efforts to improve time to treatment have been limited.

It is alarming that strategies to develop DTN benchmark times in intravenous tPA have not been applied to endovascular AIS therapy. Moreover, comprehensive studies, guidelines and recommendations are lacking for endovascular therapy time intervals.13 Hence, considerable effort is needed to streamline the complex organization of personnel and resources involved in delivering endovascular therapy, which poses an even greater challenge in identifying benchmark times.11

Insight to improving our processes can be borrowed from other industries. Within the airline industry, on-time performance correlates directly with operating profit. Similarly, on-time performance in endovascular stroke therapy correlates with brain tissue profit. Considerable resources are used by the airline industry including simulations, process monitoring and evaluation of key performance indicators to identify causes of delay. Similar to the airline industry, rigorous analysis of sub-interval times can lead to process improvement. Medicine—specifically a rapid neurologic examination, data evaluation and accurate interpretation of a clinical scenario—is arguably a more complex encounter than airline coordination; nevertheless, many methods of process improvement may be transferable. Within the automobile industry, lean manufacturing techniques pioneered by the Toyota motor company have led to substantial increases in efficiency. These methods have been adapted to refine processes for intravenous tPA administration in patients with stroke, showing a reduction of 21 min in DTN time (60 min vs 39 min) which translates to a 35% reduction in time.14

While an onset-to-treatment time has not been agreed upon for endovascular stroke therapy, the mean (SD) time from symptom onset to reperfusion in the IMS III trial was 325 (52) min. This included mean times of 81 (27) min for intravenous start to groin puncture, 42 (21) min from groin puncture to start of intra-arterial therapy and 81 (43) min from start of intra-arterial therapy to end.9 Every delay of 30 min to reperfusion was associated with a 14% relative reduction in the probability of a good clinical outcome. It therefore seems feasible that this mean 204 min interval could be reduced by 30 min or more (a 15% reduction in time) to achieve a time improvement.

The importance of identifying only the critical steps needed to efficiently select the proper patients for endovascular therapy and developing expedited processes to move the patient within the system has never been more urgently needed. These critical steps can be divided into three large categories:

1. Outside the hospital time delay (from last known well to presentation): public education, emergency medical service, telestroke and hospital readiness.

2. In-hospital triage and emergency department time delay: effective trauma like triage, stroke bay, imaging, referral to stroke and interventional team.

3. Angio-suite time delay: puncture to reperfusion.

In-hospital time delays are the main modifiable targets that can be influenced and improved. Times that one may consider as potential targets to monitor and address in process improvement are:

1. Door to Picture (D2P): identify processes for triage to CT with no delay, may consider similar target to the intravenous tPA guidelines: 25 min.

2. Picture to Puncture (P2P): identify processes for activation of interventional and stroke teams simultaneously. Proposed target: 90 min11 vs mirror the cardiology door to balloon time of 90 min and adapt Door to Device deployment (D2D) time of 90 min.

3. Puncture to catheter: time from groin puncture to placement of the guide catheter/sheath into the base target vessel (carotid, vertebral or subclavian) which would identify cases with difficult anatomy.

4. Catheter to Capture (C2C): time from base catheter placement to first deployment of the thrombectomy device. This time interval would address the device success in reaching the target clot.

5. Puncture to Reperfusion to Thrombolysis In Cerebral Infarction (TICI) ≥2b (P2R).

The monitoring of the above proposed stroke intervention time intervals would shed light on where/what the delay is and may provide a platform for process improvement. Some of the time delay noted by Sun et al11 was related to adding more complex imaging than non-contrast CT such as CT angiography, ED neuro-consult and then referral versus direct referral to the interventional team and transfer from hospitals located 60 miles outside the stroke interventional laboratory facility.11

Reducing time to treatment holds the substantial potential to reveal demonstrated efficacy now that higher reperfusion rates can be achieved. A comprehensive and ambitious effort is needed to explore the detailed and complex pathways by which a patient with stroke enters the system, is evaluated and is treated. Interval time periods need to be captured in high detail to allow precise evaluation, and motivated teams are needed to develop process improvement. This may be achieved by streamlining the pathway and reducing the time to reperfusion, which has been demonstrated with newer generation thrombectomy technology.

A concerted and dedicated effort into understanding and improving the triage pathway to endovascular therapy and reducing time intervals holds the potential to unlock the benefit of endovascular therapy for patients with AIS.

To perform effective endovascular stroke therapy we need to perform it efficiently. Time is essential for success, and the ‘time’ for success is now.