Article Text

Download PDFPDF

Identifying delays to mechanical thrombectomy for acute stroke: onset to door and door to clot times
  1. Kaiz S Asif1,
  2. Marc A Lazzaro1,2,
  3. Osama Zaidat1,2,3
  1. 1Department of Neurology, Medical College of Wisconsin and Froedtert Hospital, Milwaukee, Wisconsin, USA
  2. 2Department of Neurosurgery, Medical College of Wisconsin and Froedtert Hospital, Milwaukee, Wisconsin, USA
  3. 3Department of Radiology, Medical College of Wisconsin and Froedtert Hospital, Milwaukee, Wisconsin, USA
  1. Correspondence to Dr M A Lazzaro, Vascular and Interventional Neurology, Medical College of Wisconsin and Froedtert Hospital, 9200 W Wisconsin Avenue, Milwaukee, WI 53226, USA; mlazzaro{at}


Outcomes from endovascular therapy for acute stroke are time dependent. Delays in the administration of this therapy have not been extensively studied and no performance benchmarks have been established. There are limited data on the complex factors that can affect these delays. In this review, we discuss the existing literature on the delays involved in endovascular therapy and have presented them as prehospital and inhospital factors. Some of these factors are common to intravenous thrombolysis; in addition, there are some that are unique to endovascular therapy. These include the awareness of the first responders, emergency medical services, interhospital transfer and triage systems, activation of the endovascular team, complex imaging decisions, and intraprocedural delays. A thorough understanding of these delays can help identify areas of improvement which may affect clinical outcomes.

  • Stroke
  • Thrombectomy

Statistics from


The emergence of novel approaches to improve the speed and extent of endovascular recanalization in acute stroke has created a concurrent need to analyze the complex prehospital and inhospital delays involved in the implementation of this therapy.

Time, indeed, is brain. Prolonged ‘symptom onset to needle’ time for intravenous thrombolysis in acute stroke worsens clinical outcome.1–5 In a pooled analysis of randomized controlled trials for intravenous tissue plasminogen activator (IV tPA), the number needed to treat increased from 4.5 when treatment was given within 1.5 h, to 14.1 when time to treatment was between 3 and 4.5 h.3 In a transcranial Doppler study that continuously assessed recanalization through 6 h after IV tPA, rapid arterial recanalization was associated with better short term improvement.6 With intra-arterial therapy (IAT), recanalization achieved at later times may be associated with a poor risk–benefit ratio in unselected patients.7 Compared with the Interventional Management of Stroke (IMS) I trial, the time to endovascular treatment in the recently published IMS III trial was 32 min longer and has been cited as one of the important reasons for the lack of benefit observed with endovascular therapy, despite achieving better revascularization compared with IV tPA.8

To guide institutional response, benchmark interval times have been recommended by national guidelines for intravenous therapy, with the goal door to needle time to be within 60 min.9 The Brain Attack Coalition restated the need to have door to needle time of less than 60 min in their updated recommendations for establishment of primary stroke centers in 2011.10–12

Despite these guideline recommendations and widespread awareness over the past decade, there has been a very modest increase in patients treated within the door to needle time of 60 min.13 A recent analysis of the ‘Get with the Guidelines Stroke’ (GWTG-Stroke) registry showed that less than one-third of patients with acute ischemic stroke are treated within the guideline recommended door to needle times.14 To address this issue, the American Stroke Association (ASA) announced a national campaign called ‘Target:Stroke’ which was modeled after other successful performance improvement initiatives like ‘GWTG-Stroke’ and the ‘Door-to-Balloon Alliance’ with a goal to reduce door to needle time to <60 min in 50% of all treated patients.15

Compared with IV tPA, the analysis of delays and establishment of benchmark interval times for IAT presents some unique challenges. One of the main reasons for this is the limited amount of data available on these delays which have not been studied uniformly for IAT in randomized controlled trials or large registries (table 1). Recently published metrics for measuring quality of care at comprehensive stroke centers have classified the speed of endovascular intervention as a supplemental metric with the goal ‘arrival to start of treatment’ to be within 2 h.16 Due to the lack of enough evidence for quantitative time standards, no specific performance benchmarks have been set. The factors affecting interval times for IAT can be discussed under two groups:

  • Prehospital factors that affect interval times—prehospital factors strongly affect inhospital delays for IV tPA in acute stroke.17 ,18 It is imperative to focus on these prehospital factors (emergency medical services (EMS), transport and triage systems, emergency room (ER) activation) from an IAT perspective, while probing into inhospital delays for IAT (door to puncture, door to recanalization).

  • Inhospital factors—factors that affect door to groin interval times include initial ER assessment, complex imaging interpretation, evaluation for and administration of intravenous thrombolysis, re-evaluation, and further decision making. An important part in the chain of events preceding IAT might be assessment for IV tPA eligibility and its administration; although it has not been studied, systems that ensure quicker IV tPA administration might result in a reduction in overall time to the assessment for IAT eligibility. Factors affecting time intervals after angiosuite entry to revascularization would be completely exclusive to IAT.

Table 1

Variability in reporting stroke intervention time intervals

Prehospital factors that affect interval times

Community education

A fundamental aspect of improving acute stroke treatment time involves community education; however, this can be the most challenging element. Despite widespread awareness of EMS use being associated with reduced prehospital and inhospital delays,19–22 only approximately 50% of patients with acute stroke use EMS.23 Educational programs for the public have shown an increase in the number of patients arriving at hospitals for thrombolysis.24 These programs used media strategies including billboard advertising, radio and television public service announcements, news stories, brochures, and posters. In addition, volunteers were trained in stroke recognition and response, and they in turn trained coworkers. Community figures were used as role models to show calling 911 immediately for stroke symptoms. These media strategies can be used to increase awareness about patient related delays, such as symptom recognition and EMS activation, with strong emphasis on the importance of early hospital arrival to avail the benefit of approved treatment modalities. The public can be informed about the developments in ongoing research related to endovascular revascularization methods which would not only decrease the resistance to participate in research but would also reinforce the importance of the effect of time on outcomes.

Emergency medical services

EMS play a vital role in the management of stroke patients. Important aspects within the EMS phase of acute stroke care include dispatch, assessment and pre-notification, and triage and transport strategies. Dispatch priority is a key factor that decides EMS response,25 and a stroke high priority dispatch code has been shown to reduce the prehospital times for IV tPA in acute stroke.26 Education of dispatchers and establishment of stroke priority dispatch codes could improve arrival at the scene.

As EMS personnel serve as the first contact point between the patient and the healthcare system, swift evaluation and hospital pre-notification may contribute a critical link to expedited acute stroke care. EMS evaluation to predict the candidacy for thrombolytic therapy, including IAT, can enhance preparedness of inhospital teams, including the emergency department, neurology, radiology, and the neurointerventional team. Paramedic accuracy of stroke identification in the field has markedly improved by the use of two common prehospital stroke instruments, the Los Angeles Prehospital Stroke Screen and the Cincinnati Prehospital Stroke Scale, which have demonstrated sensitivities in stroke identification of greater than 90%.25 Prehospital notification by EMS reduces delays in stroke evaluation and reduces the transfer and intrahospital processing times for IV tPA.17 ,18 ,27 Reduction in door to balloon times of >30 min have been shown in cardiology with early activation of the interventional team.28–30 Accuracy of ER activation of a cath lab in cardiology has been shown to be around 75%.31 EMS assessment for eligibility for IV thrombolytic therapy using scales that correlate with National Institutes of Health Stroke Scale score have been tested32 but no literature exists on training EMS in identification of potential IAT eligible patients. The marked heterogeneity and complex presentation of ischemic stroke, when compared with myocardial infarction, likely would limit similar success and would not be cost and personnel effective. However, early activation of the inhouse stroke neurology team by EMS may expedite assessment for IAT eligibility (box 1).

Box 1

Prehospital checklist for emergency medical services aimed at endovascular revascularization therapy (intra-arterial therapy)

Time points: Dispatch time, time of emergency medical services (EMS) arrival on scene, EMS departure, EMS arrival at emergency room (ER)

Patient identifiers and basic evaluation: Age, sex, symptoms, vitals

Prehospital Stroke Scale: Once focal symptoms confirmed, apply Prehospital Stroke Scale

Assess eligibility for intravenous tissue plasminogen activator (IV tPA) AND/OR intra-arterial therapy (IAT): (may necessitate the development of a prediction score)

Assess distance variables:

 Transfer to IAT capable hospital v/s PSC with drip and ship or hub and spoke strategy?

 Does the patient need direct air transport or may need air transport from PSC to CSC?

Do's: Insert a large bore peripheral intravenous catheter

Don’ts: Do not treat hypertension, do not give glucose (unless glucose <50)

Pre-notification: EMS stroke thrombolytic treatment eligibility prediction score (indicating need for IV tPA /IAT) which would ER + stroke team or ER + stroke team + IAT team

Relay other details: Age, onset, vitals, clinical status, allergies, medications, PMH, contact number of POA and witness, need for an interpreter for non-English speaking patients

Relay distance and transfer variables: Need for potential transfer to IAT capable hospital later, potential need for air transport

For patients with acute stroke, the ASA recommends single call activation of the stroke neurology team either by the ER on patient arrival or prior to patient arrival based on EMS notification.15 For IAT, the current practice at most centers is an initial assessment by the stroke team which then activates the IAT team.33 Early activation of the IAT team from the stroke team may reduce door to groin times but it may cause unnecessary false activation. Development of an assessment scale/score for EMS to achieve quick and reliable prediction for thrombolytic therapy, including IAT, followed by prehospital notification of the stroke neurology team, may lead to pre-arrival activation of an IAT team and shorter door to groin times. Predictive accuracy would need to be tested across centers along with a cost–benefit analysis.

The limited availability of hospitals with IAT services warrants EMS assessment and identification of potential IAT eligible patients based on time of onset of symptoms and to transfer patients outside the IV tPA window directly to an IAT capable center. This would prevent delays associated with interhospital transfer. There continues to be strong interest in developing triage algorithms for EMS responders, highlighted by a recent proposal to direct patients to a primary stroke center or comprehensive stroke center based on the severity and duration of the stroke.34 Equally important is EMS assessment of the need for air transportation, either directly to a hospital that offers IAT or for interhospital transfer later. While air transport costs can seem substantial, judicious use of air transport measures have shown to increase availability of IV thrombolysis to rural communities in a potentially cost effective manner.35 ,36 Triage policies should be designed with emphasis on available capabilities at the hospitals in the region.25 If it is not possible to rapidly transfer patients who would potentially need IAT to an appropriate hospital, then regional strategies could be developed. This would include the ‘drip and ship’ or ‘hub and spoke’ models, in which a patient is initially taken to a primary stroke center with intent to stabilize, initiate IV thrombolysis if possible, and transfer the patient to a center with IAT capability.

Patient evaluation and treatment involves patient registration and order entry for imaging and laboratory studies. Pre-arrival electronic order entry for CT scan and laboratory investigations reduces delays in IV tPA administration.27 IAT protocols for laboratory and imaging orders and order sets could be completed prior to patient arrival.

Interhospital transfer

Transfer from a primary hospital to a center that can offer IAT can cause considerable delays in administering stroke therapy. In one study involving a metropolitan hub and spoke model, the median transfer time for acute stroke patients receiving IAT was 104 min.37 Although delays can happen at different points related to weather, geographic patterns, contacting with ambulances, and traffic patterns, there can also be delays related to administrative steps, such as bed procurement, hospital policies, etc. Emergency departments at centers that do not offer IAT could develop protocols for quick assessment, interhospital communication, and interhospital transmission of relevant data, and quick transport of patients to hospitals with advanced therapies. Relevant data could include a checklist with initial examination, vitals, laboratory results, relevant past history, allergies, medications, and imaging results. Special emphasis should be laid on getting the contact number of the family for procedural consent. and relaying the need for an interpreter for non-English speaking patients as finding an appropriate family member and arranging for an interpreter can waste a considerable amount of time. These protocols for secondary transfers to a hospital capable of IAT could be developed in conjunction with EMS,25 and hospital policies for transferring and accepting patients would need to be clearly outlined.

Inhospital factors that affect interval times

Inhospital factors would include those that occur after the patient arrives at a hospital providing endovascular revascularization therapy, either by direct EMS transport or from interhospital transfer. Limited data exist on inhospital delays for IAT. The inhospital interval times have not been studied uniformly in randomized trials or large single arm mechanical thrombectomy cohorts, and consistent definitions have not been used (box 2).7 ,38–44 Inhospital factors affecting time to IAT would be influenced by complex clinical evaluation, imaging acquisition and interpretation, procedure consent, patient preparation, and intraprocedural events.

Box 2

Proposed checklist for endovascular revascularization team (intra-arterial therapy) activation by emergency room

Standard protocol (to be followed 24/7 at intra-arterial therapy (IAT) capable hospitals)

  1. Once stroke is the primary diagnosis based on prehospital emergency medical services (EMS) assessment or inhospital emergency room (ER) assessment, the emergency medicine physician notifies the ER operator with instruction to page either (A) for intravenous tissue plasminogen activator or, if the protocol suggests possible need for IAT, then page (B).

    1. Stroke team (stroke resident/physician, stroke nurse)

    2. Stroke team + IAT team. IAT team comprises technician, nurse, and IAT fellow.

  2. The emergency department operator sends one group page.

  3. The stroke physician on-call responds to the page. If there is no response within 5 min, then the operator re-pages the stroke physician on-call and pages the designated backup.

  4. If 1 (B) was started, then the IAT team responds to the page as well. If there is no response within 5 min, then the operator re-pages the IAT team member and pages the designated backup (attending).

  5. If IAT is indicated and the patient's consent obtained, the emergency department physician verifies with the operator that the interventional team (nurse, technologists, anesthesia team) is getting ready for the procedure.

  6. The emergency medicine physician ensures initial therapy is given and that the patient is readied for transport as soon as a cath lab is available.

  7. After the procedure, feedback is provided to the EMS and ER to ensure reinforcement of positive behaviors and correction of flaws.

Door to procedure

Intrahospital transport and patient flow may play an important role in reducing time to IAT for acute stroke patients. This includes identifying the optimal destination for the arriving patient, which may be the emergency department or CT scanner. A large proportion of patients do not meet the 25 min goal for door to CT. In a large registry published in 2008, among all patients arriving at the hospital within 2 h from symptom onset, only 23% received a CT scan within 25 min.22 Reorganization of the ER with relocation of the CT scanner in the ER and first assessment of the patient in the scanner instead of an ER bed has been shown to reduce delays to assessment and imaging in patients receiving IV tPA.27 ,45 This could be enhanced by pre-arrival notification and preparedness of the CT personnel.

The emergence of imaging based patient selection has added new challenges to the already complex management of acute stroke. Guidelines for imaging in acute stroke by the ASA recommend that for patients in whom IAT is being considered, parenchymal assessment with MRI (diffusion weighted imaging) or CT angiogram source images along with vascular imaging and perfusion studies should be performed, but the type of vascular or perfusion imaging to be done are not specified.46 The role of penumbral imaging and the significance of mismatch in the selection for IAT remains unclear. Although it has been shown that reperfusion is significantly associated with favorable clinical outcomes in patients with a mismatch compared with those without a mismatch,47 recently published data argue against it.48 There is heterogeneity in the application of multimodal imaging across different centers.49 To avoid the delay in decision making, some centers have routinely included multimodal imaging in their acute stroke response protocol, even for patients presenting within 3 h.50 It is unclear if the routine performance of multimodal imaging causes significant delays to therapy. In a study looking at time from CT to microcatheter placement, multimodal imaging did not produce significant delay in this particular time interval.33 In a retrospective analysis, multimodal imaging with CT perfusion or MRI was associated with significant delays to endovascular treatment compared with non-contrast CT without reducing hemorrhage rates or improving clinical outcomes.51 Door to first image of the multimodal scan has been recommended as a core metric by the Brain Attack Coalition for all centers that perform multimodal imaging.16 It is difficult to establish benchmark times when the imaging is not being uniformly applied across centers, and availability between centers varies. Furthermore, each center has system level issues in implementation of policies and these issues would need to be studied and resolved so that uniformity in application of multimodal imaging can be achieved across centers. Development of such a uniform protocol based approach across a large number of centers with strict record of delays involved with multimodal imaging could provide useful data that could be used to establish benchmark interval times.


Limited data exist for intraprocedural interval times. They have been studied variably and inadequately in randomized trials and in mechanical thrombectomy registries, and there is a lack of consistent definitions for different interval times. Time intervals could include angiosuite to puncture time (patient preparation, anesthesia administration), time from puncture to clot visualization, clot visualization to microcatheter placement, and microcatheter placement to recanalization/termination of procedure. Factors that might affect these intraprocedure time intervals include protocol for angiography suite and patient preparation, anesthesia method, angiography techniques, and thrombectomy or thrombolysis device.

Endovascular therapy for acute ischemic stroke requires angiography suite and device preparation, which includes multiple steps where time delays can manifest. Preparation of a sterile ‘dry tray’ at the end of the day might expedite room readiness for emergent cases after hours but it may not be cost effective as the trays have to be discarded within a certain time frame to maintain sterility. Furthermore, established protocols to preselect necessary devices, including an arterial access system, sheath, guide catheter, microcatheter, and wires may reduce time for room preparation. After commencement of the procedure, a scrubbed assistant to facilitate sterile device preparation may also reduce procedure time.

Anesthesia used for IAT is variable and is dependent on operator/institutional protocols.52 The merits and limitations of each type of anesthesia (conscious sedation vs general anesthesia) have been debated in recent literature. General anesthesia can help solve the problems that can arise from patient movements (difficulties in microcatheter navigation, risks of vessel injury, etc), and whether it can lead to time delays is still debated. In one study it was associated with median time delays of 15 min53; however, in a multicenter retrospective cohort study of anterior circulation stroke patients, no significant differences were found in the time to treatment or time to recanalization between conscious sedation and general anesthesia.54 Institutional protocols for the type of anesthesia for IAT might streamline acute stroke treatment.

Time from microcatheter placement in the thrombus to recanalization can vary among studies and its significance is not entirely clear. In one study, prolonged ‘microcatheter placement to revascularization time’ was found to be associated with unfavorable outcome even after adjusting for age, National Institutes of Health Stroke Scale score, and time between onset and microcatheter placement.55 This emphasizes the importance of achieving rapid recanalization once the clot is visualized. Newer generation devices involve fewer passes and achieve significantly faster recanalization times compared with older generation devices. In the recently published Solitaire with the Intention for Thrombectomy (SWIFT) trial, median time from the initial placement of the guide catheter to achievement of recanalization or end of procedure was 36 min for the Solitaire FR compared with 52 min for the Merci device.56


In concurrence with extensive ongoing efforts to develop advanced imaging for patient selection and innovate in areas of drug and device development for IAT, there is an urgent need to study the factors producing delays to administration of this treatment. The first step would be to use a uniform definition for different interval times across centers participating in randomized trials and large registries, which would need to be followed by a systematic study of their impact on outcomes (table 2). It is only then that benchmark times can be established and strategies developed that can be feasibly and cost effectively executed across a large range of hospitals.

Table 2

Suggested stroke intervention time intervals

Substantial changes need to be made in the current systems for prehospital stroke care and for interhospital transfers of these patients. More awareness needs to be created among the public as well as medical community about the existing initiatives like ‘Target : Stroke’ and GWTG-Stroke. As prehospital notification by EMS becomes more established and EMS training is enhanced to improve the accuracy of predicting therapy, a significant improvement in both prehospital as well as inhospital times could occur.

And finally, at the organizational level, resilience in large volume centers and strong support at the national level are required. Strong internal efforts would need to be made, which would need market and regulatory pressures to sustain them. Incentives for achieving better treatment times, public reporting, and pay for performance might prove to be useful tools. A strong resolve is required to improve systems of care in acute stroke and to make approved therapies available to as many patients as possible.



  • Contributors All authors contributed towards conception and design, reviewing the literature, drafting the article, revising it critically for important intellectual content, and final approval of the version to be published.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.