Article Text

Download PDFPDF

Original research
Picture-to-puncture time in acute stroke endovascular intervention: are we getting faster?
  1. J van Heerden1,
  2. B Yan2,
  3. L Churilov3,
  4. R J Dowling4,
  5. P J Mitchell4
  1. 1Department of Radiology, The Royal Melbourne Hospital, Melbourne, Victoria, Australia
  2. 2Melbourne Brain Centre, The Royal Melbourne Hospital, University of Melbourne, Melbourne, Australia
  3. 3Florey Institute of Neuroscience and Mental Health, Melbourne, Australia
  4. 4Neurointervention Service, Department of Radiology, The Royal Melbourne Hospital, Melbourne, Australia
  1. Correspondence to Dr J van Heerden, Department of Radiology, The Royal Melbourne Hospital, Grattan Street, Parkville, Melbourne, VIC 3050, Australia; jolandivh{at}


Background In acute stroke management, time efficiency in the continuum of patient management is critical. We aimed to determine if implementation of system improvements at our institution translated to reduced picture-to-puncture (P2P) times over a 6-year period.

Methods We conducted a single-center retrospective analysis using a prospective acute stroke database of patients treated with intra-arterial therapy from October 2007 to October 2013. Patient demographics, stroke severity, neuroimaging and treatment time points were collected. Annual P2P times, defined as the interval between pretreatment neuroimaging (picture) and commencement of intra-arterial therapy (puncture), were assessed and compared.

Results From 2007 to 2013 a total of 189 patients were identified, of which 181 met the study criteria. At initial presentation, median baseline NIH Stroke Severity score was 17.00 (IQR 11.00–22.00). Annual median P2P times decreased from 171 to 123.5 min, showing a median decrease of 11.5 min per annum (95% CI −23.9 to 0.9) and trending towards statistical significance (p=0.069). Plotted data revealed longer P2P times in instances where stroke onset or CT acquisition times were out-of-hours. Using median regression modeling, the annual decrease in P2P median time reached statistical significance when independently adjusting for CT acquisition time (13.5 min P2P median time reduction, 95% CI −27.0 to −0.1, p=0.048) and for stroke onset time (14.5 min annual P2P median time reduction, 95% CI −26.1 to −2.8, p=0.015).

Conclusions As a consequence of systems improvement at our institution, we were able to demonstrate improved annual median P2P times from 2007 to 2013.

  • Angiography
  • CT
  • Stroke
  • Thrombolysis
  • Brain

Statistics from

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.


In the context of acute ischemic stroke management, each treatment step is time-critical.1 It has been well established that, within the 4.5 h time window, prompt administration of intravenous tissue plasminogen activator (IV-tPA) results in improved patient outcome.2 Subsequently, there has been much focus on systems improvement to facilitate faster door-to-needle times for the administration of IV-tPA, with strategies based on a rapid access ‘code-stroke’ approach (including pre-hospital dispatch center and ambulance personnel involvement, hospital pre-notification, prompt ambulance to CT transfer, prompt CT interpretation, premixing of IV-tPA and on-CT table administration of IV-tPA) implemented with good effect.3–5

Intra-arterial therapy (IAT) can be offered following IV-tPA6 or as an alternative treatment strategy when IV-tPA is contraindicated.7 However, this technique is associated with additional treatment steps and potential time delays. Inter-hospital transport, angiography theatre preparation and staffing (particularly after hours), in-hospital patient transfer to the angiography suite, in-hospital procedure consent and administration of preprocedural general anesthesia all impact on overall treatment time.8

To this end, picture-to-puncture (P2P) time (defined as the time interval between initial CT brain scan to groin puncture for IAT) has been formulated as a novel new metric that assesses a broader continuum of acute stroke care, particularly in the context of IAT.9 It has been shown that faster P2P times translate to better patient outcome,9 but there is a paucity of published work aimed at improving this time metric.

Over a 7-year period, our 24 h acute stroke center incrementally endeavored to shorten P2P times. This was achieved by making use of time-critical ambulance transfers, instituting hospital pre-notification of the stroke and neurointervention teams, expediting direct patient transfer from ambulance to angiography suite, doing rapid in-hospital clinical assessment at time of patient arrival and ensuring that patient consent was obtained en route to the angiography suite. We also moved towards conscious sedation (CS) instead of general anesthesia to achieve faster puncture times. We subsequently aimed to investigate our P2P times achieved over a 6-year period, with the hypothesis that improved systems implementation would translate to faster P2P times at The Royal Melbourne Hospital.


Study population

This was a single-center retrospective study of a prospectively collected acute stroke database of all consecutive patients with acute stroke treated with IAT at The Royal Melbourne Hospital from October 2007 to October 2013.

A total of 189 patients were recorded (2007: n=6; 2008: n=17; 2009: n=23; 2010: n=27; 2011 n=39; 2012 n=41; 2013 n=36), of which 181 met the inclusion criteria (table 1). Inclusion criteria were: all acute stroke patients treated at The Royal Melbourne Hospital with IAT (primarily or as an adjunct to intravenous thrombolysis) with recorded initial CT brain times and recorded intra-arterial groin puncture times. All anterior and posterior circulation infarcts were included to get an overall impression of systems improvement at our institution. There were no wake-up stroke patients treated with IAT. Other data collected included age, sex, stroke onset time, and stroke severity at presentation measured by National Institutes of Health Stroke Severity (NIHSS) scores. In 171 patients, 90-day modified Rankin Scale (mRS) scores were available to be assessed and compared as a measure of clinical outcome.

Table 1

Distribution of patients who received intra-arterial therapy and met study criteria over a 7-year period

Imaging and intervention

P2P time was defined as the time interval between the first CT brain study and the time of groin puncture for IAT. Patients either had their initial CT brain at our institution or at remote referral hospitals prior to transfer. None of the patients had MRI studies prior to IAT. Patients had various combinations of CT brain±CT angiogram head±CT perfusion, the full combination taking less than 8 min in total.

From 2007 to 2013 the following strategies were implemented to ensure faster P2P times: activation of time-critical ambulance transfer codes for patients referred from remote hospitals; pre-notification of the stroke and neurointervention teams (to enable prompt clinical assessment, time-efficient procedural consent as well as early staffing and preparation of the angiography suite); direct ambulance-to-angiography suite transfer; and use of CS instead of general anesthesia. Not all systems improvements were implemented simultaneously; however, there was a marked overlap in the time frames of the strategy roll-outs, particularly from 2010 to 2012.

Patients with acute stroke arriving directly at our institution requiring IAT were managed according to the quick-access ‘code stroke’ pathway.3 ,5 Following the CT brain study, direct communication between the stroke team and the neurointervention team facilitated direct CT-to-angiography suite transfer. After arrival at the angiography suite and administration of CS, patients had a diagnostic cerebral catheter angiogram performed via the common femoral artery approach followed by IAT. Differing IATs were employed including Solitaire AB Device (ev3 Neurovascular, Irvine, California, USA), mechanical thrombectomy with the Merci Retrieval System (Concentric Medical, Hertogenbosch, The Netherlands), percutaneous transluminal angioplasty with Gateway Balloon Catheter and Ultrafast (Boston Scientific, Natick, Massachusetts, USA) and IA thrombolysis (urokinase, dose range 500 000–1 000 000 units).

Statistical analysis

Data analysis was performed using STATA V.12 (StataCorp, College Station, Texas, USA). The value of p=0.05 was selected as a threshold for statistical significance.

Spearman's correlation coefficient and median regression modeling were used to investigate the association between independent variables (year, CT time and time of stroke onset) and P2P times. Logistic regression modeling was used to investigate the association between P2P time and a favorable clinical outcome (90-day mRS ≤2).


From 2007 to 2013, 181 patients met the inclusion criteria for the study (table 1). The patients had a median age of 67 years (IQR 56–76) with a 65% male prevalence. Annual age ranges were widely distributed each year (table 2). At initial presentation, the median baseline NIHSS was 17 (IQR 11–22); however, annual presenting NIHSS scores ranged from mild to severe each year (table 2). The percentage of inter-hospital transfers gradually increased from 2007 to 2013 (table 2). The number of acute anterior circulation strokes treated with IAT gradually increased annually resulting in a relative reduction in the percentage of posterior circulation infarcts treated with IAT (table 2).

Table 2

Annual patient demographics

There was an overall increase in the percentage of patients receiving CS from 2007 to 2013; however, comparative annual percentages varied (2007: 16.7% (n=1/6); 2008: 17.6% (n=3/17); 2009: 13% (n=3/23); 2010: 18.5% (n=5/27); 2011: 5% (n=2/39); 2012: 24.4% (n=10/41); 2013: 42.9% (n=15/35)).

Spearman's correlation coefficient showed a significant negative association between P2P times and calendar year (r = −0.16, p=0.03).

Annual median P2P times decreased from 171 min to 123.5 min (table 3). Using median regression modeling, this translated to an annual P2P median time reduction of 11.5 min (95% CI −23.9 to 0.9), trending towards statistical significance (p=0.069, figure 1).

Table 3

Picture-to-puncture times, 2007–2013

Figure 1

Box plot of picture-to-puncture times from 2007 to 2013. This figure illustrates overall constriction of annual picture-to-puncture time extremes over a 7-year period as well as a reduction in overall median picture-to-puncture times.

When adjusting for stroke onset time, a statistically significant reduction in annual median P2P time of 14.5 min (95% CI −26.1 to −2.8, p=0.015) was observed and, when adjusted for CT acquisition time, an annual reduction in median P2P time of 13.5 min was observed, also reaching statistical significance (95% CI −27.0 to −0.1, p=0.048). Scatter plots show that the improved adjusted P2P times can be explained by prolonged P2P times observed when stroke onset or CT acquisition occurred out-of-hours (figures 2 and 3).

Figure 2

Scatter plot demonstrating that prolonged picture-to-puncture times continue to be mostly observed when stroke onset time occurred out-of-hours (00:00–08:00 and 17:00–24:00).

Figure 3

Scatter plot showing that prolonged picture-to-puncture times are observed mostly when initial CT is performed out-of-hours (00:00–08:00 and 17:00–24:00).

Although the improved P2P times translated to statistically significant improvement in 90-day mRS scores (OR=1.002, per minute 95% CI of 1 to 1.004, p=0.032), in the analysis adjusted for age and initial NIHSS scores the difference in good mRS outcome did not reach statistical significance. Using logistic regression modeling, it was shown that the odds of a favorable 90-day mRS score, defined as mRS ≤2, was reduced by approximately 6% for every 60 min of prolonged P2P time (OR=0.998, 95% CI 0.996 to 1.000, p=0.069).


In the context of acute stroke, the truth of ‘time is brain’ is well established, quantified as 1.9 million neurons destroyed per minute without treatment.1

There are several large randomized controlled trials currently in progress assessing the efficacy of endovascular therapy in the setting of acute stroke. THERAPY (The Randomized, Concurrent Controlled Trial to Assess the Penumbra System's Safety and Effectiveness in the Treatment of Acute Stroke) is a trial across four centers in the USA investigating the superiority of endovascular therapy using the Penumbra aspiration device over IV-tPA in severe strokes, aiming to enroll 692 patients with large anterior circulation strokes (clot burden >8 mm).10 The EXTEND-IA (Extending the Time for Thrombolysis in Emergency Neurological Deficits-IA) randomized controlled trial is furthering the hypothesis that more prudent imaging-based selection in patients with internal carotid artery or middle cerebral artery occlusions with CT/MR evidence of perfusion mismatch within 4.5 h of stroke onset may maximize the efficacy of endovascular stroke therapy.11 The aim is to randomize 100 patients across eight centers to receive mechanical thrombectomy with the Solitaire device after IV-tPA versus tPA alone.11

The REVASCAT (Endovascular Revascularization With Solitaire Device vs Best Medical Therapy in Anterior Circulation Stroke Within 8 h) trial addresses the utility of endovascular therapy using the Solitaire device as a rescue option for patients who are either unable to receive; or have failed IV-tPA therapy after an internal carotid or proximal middle cerebral artery stroke within 8 h.12 A predicted 690 patients across four hospitals in Spain are expected to be randomized in a 1:1 ratio between the treatment and control groups.12

SWIFT-PRIME (Solitaire FR as Primary Treatment for Acute Ischemic Stroke) is a study following on the promising results of the SWIFT trial,13 aiming to randomize a maximum of 833 patients to receive either combined IV-tPA and endovascular therapy with the Solitaire device within 6 h of anterior circulation stroke onset versus IV-tPA alone.13

The MR Clean (Multicenter Randomized Clinical trial of Endovascular treatment for acute ischemic stroke in the Netherlands) recently randomized its final patient across 17 centers in the Netherlands, looking at the effect of chemical and/or mechanical endovascular therapy within 6 h of anterior circulation acute stroke onset.14

Although the efficacy time window for the administration of IV-tPA has been expanded to 4.5 h,15 it has been well shown that earlier administration of IV-tPA translates to better clinical outcomes.2 Subsequently, a drive to reduce door-to-needle times to less than 60 min was initiated by the American Heart Association, proposing the implementation of 10 key strategies.16 Despite this, a review of patient data from hospitals participating in the “Get with the Guidelines” Stroke Program (n=25 504 acute stroke patients treated)17 as well as sizeable North American registries (n=6867 acute stroke patients recorded)18 subsequently reported that only a small minority of patients (less than 26.6%17 and less than 19.6%18) achieved door-to-needle times within the recommended 1 h.

This prompted many acute stroke centers, including our own, to implement system strategies to improve door-to-needle times.3–5 At The Royal Melbourne Hospital we were able to achieve faster door-to-needle times by implementing a quick-access ‘code stroke’ pathway3 that was later further improved by adopting key components of the Helsinki acute stroke model which included pre-hospital ambulance and emergency staff involvement, hospital pre-notification, pre-ordering of tests, pre-acquisition of history if possible, direct ambulance to CT transfer, no-delay CT interpretation and IV-tPA administration on the CT bed.4 ,5

In the context of IAT in acute stroke management for patients who do not qualify for IV-tPA or in whom IV-tPA did not achieve recanalization, the mantra of ‘time is brain’ is continued with emphasis on timely recanalization.19 Limited data currently exist regarding the time metrics surrounding IAT.9 ,20 Procedure time as an appropriate metric has been suggested, with low rates of poor outcomes in procedure times of <30 min and overall target goals of <60 min suggested.21 Although procedure time with reperfusion as the endpoint is probably more physiological, difficulties remain in quantifying successful reperfusion as an endpoint.20 Recently, a novel new IAT time metric, P2P time, was proposed which demonstrated that every 10 min delay in P2P time correlated with a 6% relative lower probability of achieving a good outcome.9 P2P time expands the continuum of stroke care to incorporate the transfer time in those patients receiving IAT after arrival from remote centers,9 highlighting the importance of achieving systems improvement related to this time metric, a topic not yet addressed in the literature.

In Melbourne, Australia, the impact of ambulance practice on acute stroke care has been investigated.22 Recognition of acute stroke signs by ambulance and dispatch personnel as well as hospital pre-notification were identified as the main factors associated with shorter ‘call-to-doctor’ pre-hospital times and ‘door-to-doctor’ in-hospital times, highlighting the importance of pre-hospital care in the continuum of effective stroke management.22 Subsequently, public and emergency services awareness campaigns of acute stroke signs were launched in Australia (Face, Arm, Speech, Time, Emergency Response) that significantly reduced in-hospital assessment time and time to IV-tPA, thus allowing more patients access to time-appropriate treatment.23

At our center, acute stroke systems improvements implemented both pre-hospital and in-hospital over the last 7 years translated to a median P2P time improvement of 11.5 min per annum. Patients with acute stroke requiring IAT who present primarily at The Royal Melbourne Hospital benefit from time efficacy achieved by the time-critical ‘code stroke’ pathway followed by direct CT to angiography transfer facilitated by prompt communication between the stroke team and the neurointervention team. For patients transferred from remote hospital sites requiring IAT, improved time-critical ambulance transfer protocols, hospital pre-notification of the stroke team and neurointervention team, direct transfer from ambulance to CT (for patients who had not yet had a CT brain) or ambulance to angiography suite (for patients who had a CT brain remotely), as well as prompt in-hospital clinical assessment and consent obtained en route to the angiography suite have contributed to improved P2P times. In cases of severe stroke/dysarthria, every effort is made to expedite the process of consent, sometimes necessitating telephone consent from a next-of-kin.

In the angiography suite, a move toward CS instead of general anesthetic helped to achieve faster puncture times. CS has been shown to correlate with better outcomes in the context of acute anterior circulation strokes in other studies,24 ,25 unlikely to be completely explained by patient selection.

For out-of-hours transfers, pre-notification of the radiology registrar on call who then activates the stroke team and the neurointervention team expedited the IAT process. Scatter graphs (figures 2 and 3) indicate that prolonged P2P times continue to be observed during the out-of-hours period, highlighting the fact that future efforts should be aimed at improving this aspect of our stroke service. The delay in out-of-hours P2P times were mostly due to a shift in staffing demographics after hours, often with more junior staff members (emergency department staff, on-call neurology staff and on-call radiology staff) being on-site for initial stroke assessment followed by key senior staff members being called on-site in case of a patient qualifying for IA therapy.


P2P time is a novel new metric in acute stroke care that broadens time efficacy considerations, particularly in the context of patients transferred for IAT following a CT brain obtained remotely. Through acute stroke management systems improvements, The Royal Melbourne Hospital achieved an annual improvement in median P2P time of 11.5 min over a 7-year period.



  • Contributors All authors made substantial contributions to the conception and design of the work; contributed to drafting and revision of the work; approved the final submitted version of the manuscript and agree to be accountable for all aspects of the content of the manuscript.

  • Competing interests None.

  • Ethics approval Institutional ethics committee approval was obtained from the Royal Melbourne Hospital.

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

  • Data sharing statement All data relevant to this study have been included in the manuscript. There are no additional unpublished data relevant to this retrospective review.