Reducing in-hospital delay to 20 minutes in stroke thrombolysis

Atte Meretoja, Daniel Strbian, Satu Mustanoja, et al.

Neurology 2012;79;306; Published online before print May 23, 2012;
DOI 10.1212/WNL.0b013e31825d6011

This information is current as of August 15, 2012

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://www.neurology.org/content/79/4/306.full.html

Neurology ® is the official journal of the American Academy of Neurology. Published continuously
since 1951, it is now a weekly with 48 issues per year. Copyright © 2012 by AAN Enterprises, Inc. All
rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
ARTICLES

Reducing in-hospital delay to 20 minutes

in stroke thrombolysis

Atte Meretoja, MD, PhD, ABSTRACT

MSc (StrokeMed) Objectives: Efficacy of thrombolytic therapy for ischemic stroke decreases with time elapsed from
Daniel Strbian, MD, PhD symptom onset. We analyzed the effect of interventions aimed to reduce treatment delays in our
Satu Mustanoja, MD, single-center observational series.
PhD, MSc
Methods: All consecutive ischemic stroke patients treated with IV alteplase (tissue plasminogen
(StrokeMed)
activator [tPA]) were prospectively registered in the Helsinki Stroke Thrombolysis Registry. A
Turgut Tatlisumak, MD,
series of interventions to reduce treatment delays were implemented over the years 1998 to
PhD
2011. In-hospital delays were analyzed as annual median door-to-needle time (DNT) in minutes,
Perttu J. Lindsberg, MD,
with interquartile range.
PhD
Markku Kaste, MD, PhD Results: A total of 1,860 patients were treated between June 1995 and June 2011, which in-
cluded 174 patients with basilar artery occlusion (BAO) treated mostly beyond 4.5 hours from
symptom onset. In the non-BAO patients, the DNT was reduced annually, from median 105 min-
Correspondence & reprint utes (65–120) in 1998, to 60 minutes (48–80) in 2003, further on to 20 minutes (14–32) in
requests to Dr. Meretoja: 2011. In 2011, we treated with tPA 31% of ischemic stroke patients admitted to our hospital. Of
atte.meretoja@fimnet.fi
these, 94% were treated within 60 minutes from arrival. Performing angiography or perfusion
imaging doubled the in-hospital delays. Patients with in-hospital stroke or arriving very soon from
symptom onset had longer delays because there was no time to prepare for their arrival.
Conclusions: With multiple concurrent strategies it is possible to cut the median in-hospital delay
to 20 minutes. The key is to do as little as possible after the patient has arrived at the emergency
room and as much as possible before that, while the patient is being transported. Neurology®
2012;79:306–313

GLOSSARY
BAO ⫽ basilar artery occlusion; DNT ⫽ door-to-needle time; EMS ⫽ emergency medical service; ER ⫽ emergency room;
INR ⫽ international normalized ratio; IQR ⫽ interquartile range; mRS ⫽ modified Rankin Scale; NIHSS ⫽ NIH Stroke Scale;
NNT ⫽ number needed to treat; OTT ⫽ onset-to-treatment time; POC ⫽ point-of-care; RCT ⫽ randomized controlled trials;
tPA ⫽ tissue plasminogen activator.

The phrase “time is brain” has been quoted over and over in the stroke literature, and remains
true.1 Recanalization therapies restoring blood flow are more effective the sooner they are
administered. The latest pooled analysis of IV alteplase (tissue plasminogen activator [tPA]) in
stroke revealed that a number needed to treat (NNT) is 4.5 to achieve little or no symptoms
Editorial, page 296
after stroke, i.e., modified Rankin Scale (mRS) score of 0 to 1, when the treatment was initiated
Supplemental data at
within 11⁄2 hours. The NNT doubled to 9 when treatment was initiated in the 11⁄2- to 3-hour
www.neurology.org interval, and further on to 14.1 in the 3- to 41⁄2-hour interval.2 Accordingly, one could deduce
Supplemental Data
that the NNT of IV thrombolysis increases by 1 for every 20 minutes of extra delay.
We have previously published the results of 100 IV thrombolytic therapies between 1995
and 2004, and the effect of our emergency room reorganization on the door-to-needle times
(DNT).3 With currently almost 2,000 patients and significantly shortened DNT over the last
few years, we present our experience on how to reduce the DNT, down to 20 minutes.
Podcast

From the Department of Neurology (A.M., D.S., S.M., T.T., P.J.L., M.K.), Helsinki University Central Hospital; and Research Programs Unit,
Molecular Neurology, Biomedicum Helsinki (P.J.L.), and Department of Clinical Neuroscience, University of Helsinki, Helsinki, Finland.
Study funding: Supported by the Helsinki University Central Hospital EVO Fund, Yrjö Jahnsson and Biomedicum Helsinki Foundations (A.M.),
Sigrid Jusélius Foundation (A.M., P.J.L., T.T.), and the Finnish Academy (P.J.L., T.T.).
Go to Neurology.org for full disclosures. Disclosures deemed relevant by the authors, if any, are provided at the end of this article.

306 Copyright © 2012 by AAN Enterprises, Inc.

 The aim of the present study is to describe
in a single-center series of IV tPA for acute
ischemic stroke what measures led to a me-
dian DNT of 20 minutes, and how these af-
fected the number of annual patients treated.
METHODS Registry setup. This report is based on the
Helsinki Stroke Thrombolysis Registry. The registration meth-
ods and treatment protocols have been previously described in
detail.4 Briefly, this is a prospective registry of consecutive isch-
emic stroke patients treated with IV alteplase at the neurologic
emergency room (ER) of the Helsinki University Central Hospi-
tal since June 1995. Delays were registered prospectively, with
no missing data. Patients were followed up at 3 months, includ-
ing mRS score evaluation, as part of routine care.
Standard protocol approvals, registrations, and patient
consents. The Helsinki Stroke Thrombolysis Registry has been
approved by institutional authorities. As a routine observational
quality registry, no patient consent for registration was required
by Finnish legislation.
Measures to reduce treatment delays. Over the years, 12
measures have been systematically implemented to reduce all
possible delays in treatment (table 1). Many of these have been
previously described.3,5,6 Constant cooperation with the emer-
gency medical service (EMS) was maintained throughout the
study. Our stroke team convened periodically to consider new
ideas for more rapid in-hospital management of patients.
Nationwide stroke awareness campaigning was performed
periodically and supported by Finnish laypeople organizations to
educate the general public. The emergency call center and EMS
personnel were systematically educated. A stroke priority dis-
Table 1 Twelve measures to reduce treatment delays
Measure Description Year
EMS involvement Education of dispatchers and EMS personnel, stroke 1998
high-priority dispatch
Hospital EMS contacts stroke physician directly via mobile 2001
prenotification phone
Alarm and preorder Laboratory and CT computer-ordered and alarmed 2001
of tests at prenotification
No-delay CT Stroke physician interprets the CT scan, not waiting 2001
interpretation for formal radiology report
Premixing of tPA With highly suspect thrombolysis candidates, tPA 2002
premixed prior to patient arrival
Delivery of tPA on Bolus administered on CT table 2002
CT table
CT relocated to ER Patient transfers of several hundred meters, 2003
including elevators, were no longer needed
CT priority and CT CT emptied prior to patient arrival, and patient 2004
transfer transferred straight onto CT table, not ER bed
Rapid neurologic Patient is examined upon arrival, on CT table 2004
evaluation
Preacquisition of Statewide electronic patient records and 2005
history eyewitness interview before/during transportation
Point-of-care INR Laboratory personnel draw blood while patient on CT 2005
table, and perform instant POC INR
Reduced imaging While all patients have a CT, advanced imaging 2005
reserved for unclear cases only
Abbreviations: EMS ⫽ emergency medical service; ER ⫽ emergency room; INR ⫽ interna-
tional normalized ratio; POC ⫽ point-of-care; tPA ⫽ tissue plasminogen activator.
patch code was instituted in 1998, with equally high priority as
myocardial infarction, surpassed only by cardiac arrest. In the
Province of Uusimaa, with a 9,000 km2 surface area and a 1.5
million population, all suspected thrombolysis candidates were
transported to our ER, the only 24/7 neurologic service in the
province. Despite being in a university teaching hospital, it is the
primary, not tertiary, center for all acute neurologic emergencies
serving roughly one-fourth of the Finnish population.
We concluded that for a correct diagnosis and the decision
on thrombolysis, the treating physician needs a valid history,
physical examination, CT scan, and some laboratory tests.
Taking the history usually took the most time, especially with
aphasic patients, and information was lost or distorted at any
in-betweens. Therefore, the hospital prenotification call was
made directly to the mobile phone of the ER stroke physician
on-call, whose phone number was widely distributed to the EMS
units. Subsequently, the stroke team ER nurse was alarmed
about an incoming thrombolysis candidate. The ER nurse va-
cated the CT table and called the laboratory nurse to join the
stroke physician at the CT to wait for the patient’s arrival.
However, the preliminary history relayed by the EMS, espe-
cially on symptom onset, was often not accurate, and reaching
next of kin or eyewitnesses afterwards for treatment decisions
was difficult. The treating stroke physician therefore requested
communication with the primary informants over a mobile phone
already at prenotification, during EMS contact on-scene, and pref-
erably to have the next of kin cotransported by EMS to allow for
rapid additional history taking when need arose. Such history taking
during transport was complemented by accessing the province-
wide electronic patient records prior to patient arrival.
Computer systems of modern hospitals sometimes also cause
hindrance. It takes time to write a patient into a system and
order laboratory tests or CT scans. We did all of this during
patient transportation. The only laboratory tests really affecting
treatment decisions were blood glucose and the international
normalized ratio (INR). Glucose was measured already by the
EMS and point-of-care (POC) INR sample taken immediately
on admission and reported within a minute. After relocating the
CT from a more remote location to our ER in 2003 it was
possible to achieve CT priority and to clear the CT table prior to
patient arrival.
Upon arrival, the patient was moved from the ambulance
stretchers straight onto the CT table, instead of an ER bed. Here,
while the stroke physician performed a rapid neurologic evalua-
tion using the NIH Stroke Scale (NIHSS), the laboratory nurse
sampled blood for the POC INR. Delay from ER arrival to start
of CT scanning was generally about 5–10 minutes, and the non-
contrast CT took only a few minutes to perform. The CT was
interpreted by the treating stroke physician, and the oral radiol-
ogy report utilized only when immediately available.
Unless the treating stroke physician was concerned about
diagnostic uncertainty (suspected stroke mimic) or questioned
the treatment risk/benefit ratio (mild or rapidly improving
symptoms), no further imaging was needed, and a treatment
decision was made. In uncertain cases, CT angiography or perfu-
sion imaging was performed immediately, through a large-bore
cannula the EMS always inserted at the antecubital vein of sus-
pected stroke victims during transport. The radiologist reported
the angiography and perfusion images immediately.
Prior to treatment, all patients and next of kin when present
were informed of the risks and benefits involved, which usually
took less than a minute. Initially, consent was sought from next
of kin if the patient was unable to communicate or comprehend.
Following the approval of stroke thrombolysis by European reg-
Neurology 79 July 24, 2012 307
 Figure 1 Number of annually treated patients and median door-to-needle times
Annual patients, with those treated beyond 3 hours in red (bars, left axis) and median door-to-needle time in minutes with
interquartile range (line, right axis). Total n ⫽ 1,686. The projected number of patients for 2011 is based on the observed
numbers of the first 6 months.
ulatory authorities in 2002, also noncommunicating patients
without next of kin were treated when considered to be in their
best interest. The alteplase bolus was preferably administered on
the CT table, followed by infusion in an adjacent room.
Statistical analysis. Due to non-normal distribution of age,
NIHSS, and all treatment delays, data are presented as median
and interquartile range, grouped by year for demonstrating
trends in time. Mean and SD are given for the most recent
data to allow comparisons to other series. Groups were com-
pared with the independent samples Mann-Whitney U or
Kruskal-Wallis tests, with statistical significance set at 0.05
(2-sided). Analyses were performed on IBM SPSS 18 (IBM
Corp, Armonk, NY).
RESULTS Between June 1995 and June 2011, a to-
tal of 1,860 IV thrombolytic therapies were adminis-
tered with a current rate of 1 per day over the first
half of 2011.
A total of 174 basilar artery occlusions (BAO)
were included, all treated with an institutional proto-
col involving concomitant full dose IV heparin and
extended time window up to 48 hours in patients
with progressing symptoms.7 Therefore the data for
BAO patients are given separately (table e-1 on the
Neurology® Web site at www.neurology.org). From
1995 to 1997, only BAO patients received tPA out-
side of randomized controlled trials (RCT), while
non-BAO patients were recruited in RCTs evaluat-
ing thrombolysis. Only 34% of the BAO patients
were treated within 4.5 hours from symptom onset
(onset-to-treatment time [OTT]). The median
DNT in BAO patients was several hours up to 2005,
and then reduced to 28 minutes, interquartile range
(IQR) 25–37, in 2011.
308 Neurology 79 July 24, 2012
The rest of the results are based on the non-BAO
patients only (n ⫽ 1,686). The proportion of pa-
tients treated with OTT beyond 3 hours has in-
creased since before 2008 from roughly 7% to 22%
in 2011 (figure 1). The median age was 69 years
(IQR 59 –77). Treatment of milder strokes has be-
come more common (table 2).
The in-hospital delays have steadily shortened
( p ⬍ 0.001), with currently half of the patients being
treated within 20 minutes, 72% within 30 minutes,
and 94% within 60 minutes of arrival (figure 1).
CT angiography or CT perfusion imaging prior
to tPA administration became routinely available at
our institution in 2002, and was used in almost a
third of the patients over the next 3 years. At this
point, we were concerned whether the imaging-
associated delays were paid off by clinical benefit and
decided to reduce additional imaging. In 2009, a
new ER CT machine and a rapid protocol for ad-
vanced CT imaging became available, leading again
to wider use of CT angiography and perfusion imag-
ing (table 2). Advanced imaging prior to treatment
still produced an extra delay of about 20 minutes,
practically doubling the in-hospital delays in these
patients (figure 2, p ⬍ 0.001).
The OTT declined up to 2008 ( p ⬍ 0.001), but
hinted a slight increase since ( p ⫽ 0.26) (figure 3).
The prehospital delays have not been reduced over
the years, on the contrary ( p ⬍ 0.001). These 2
trends presumably reflect the time window extension
up to 4.5 hours in late 2008 after the ECASS III
report.8 The prehospital delays for patients treated
 Table 2 Baseline characteristics, imaging, and outcome of patients treated with thrombolytic therapy for
acute ischemic stroke at the Helsinki University Central Hospital, by yeara

Patients with CT
NIHSS at angiography or Good
treatment perfusion imaging outcome Mortality
Year No. Age, y decision prior to therapy at 3 mob at 3 mo

1998 7 73 (59–75) 11 (8–14) 0 2 (40) 0

1999 17 69 (56–73) 12 (8–14) 0 6 (38) 1 (6)

2000 33 64 (52–74) 11 (8–14) 0 8 (28) 1 (3)

2001 23 64 (50–74) 10 (7–14) 0 9 (41) 2 (9)

2002 34 66 (57–73) 11 (6–17) 2 (6) 14 (42) 2 (6)

2003 61 71 (60–75) 12 (8–17) 18 (30) 15 (25) 8 (13)

2004 83 72 (64–77) 10 (7–18) 28 (34) 32 (39) 10 (12)

2005 171 71 (61–79) 10 (6–15) 42 (25) 70 (41) 22 (13)

2006 170 70 (60–77) 9 (5–15) 15 (9) 62 (37) 20 (12)

2007 180 70 (60–78) 9 (5–15) 15 (8) 67 (38) 18 (10)

2008 206 70 (60–76) 8 (5–13) 20 (10) 68 (34) 17 (9)

2009 249 69 (59–77) 7 (4–12) 78 (31) 98 (40) 21 (9)

2010 285 67 (58–76) 7 (4–13) 89 (31) 112 (40) 19 (7)

2011c 167 69 (60–77) 7 (4–11) 62 (37) 63 (38) 10 (6)

Total 1,686 69 (59–77) 8 (5–14) 369 (22) 626 (38) 151 (9)

Abbreviation: NIHSS ⫽ NIH Stroke Scale.

a
All values are n (%) or median (interquartile range).
b
Defined as modified Rankin Scale score 0 or 1. Thirty-one patients (1.8%) were lost to 3-month follow-up.
c
First 6 months.

within 3 hours have remained fairly stable, at median
71 minutes (IQR 54 –97) (figure 3). In 2011, the
mean ⫾ SD (median; IQR) treatment delays in min-
utes were 106 ⫾ 58 (89; 62–138) from symptom to
door, 28 ⫾ 23 (20; 14 –32) from door to needle, and
133 ⫾ 62 (119; 80 –176) total OTT.
DNT was longer in patients arriving within 30
minutes of symptom onset (n ⫽ 32) and patients
with in-hospital stroke (n ⫽ 59) compared to pa-
tients arriving 30 –150 minutes from symptom onset
(table e-2). These longer times may have reflected a
lack of time to prepare the ER for patient arrival. By
contrast, the DNT was consistent for patients arriving
between 30 and 150 minutes from symptom onset
(p ⫽ 0.173), and somewhat shorter for those arriving
just before the 3-hour time window (table e-2).
DISCUSSION Our results show that DNT can be
effectively shortened with multiple simple interven-
tions and when the DNT reduction is achieved,

Figure 2 Advanced imaging and in-hospital delays

Annual median door-to-needle time in minutes for patients treated with (upper red line, n ⫽ 369) and without (lower blue
line, n ⫽ 1,317) prior CT angiography or perfusion imaging.

Neurology 79 July 24, 2012 309

 Figure 3 Trends in onset-to-treatment and symptom-to-door times
Median onset-to-treatment times in minutes, with interquartile range (A), and median symptom-to-door time separately for
patients treated within (B, blue line, n ⫽ 1,465) and beyond (B, red line, n ⫽ 221) 3 hours from symptom onset.
more ischemic stroke patients can be treated with
thrombolytic therapy. The annual number of treated
patients has increased 10-fold during a decade, with a
current rate of 1 patient per day. We have reduced
the OTT for a median patient by 45 minutes, from
160 minutes to 115 minutes. This reduction has
stemmed from EMS–ER cooperation and stream-
lined in-hospital processes, and enables a higher like-
lihood of good outcome to a larger number of stroke
patients.2 It has been achieved with little or no extra
investment, but rather by reorganizing existing pro-
cedures. Reducing delays to thrombolytic therapy is
probably highly cost-effective.
Of 1,200 ischemic stroke patients admitted annu-
ally to our ER, we currently treat 31% with tPA.
However, there are altogether 2,300 annual ischemic
stroke patients in the whole province for which we
are the only thrombolysis service. Many patients who
are not candidates for tPA are admitted to other hos-
pitals.9,10 Therefore the population-based thrombol-
ysis rate for our province is currently 16%.
Early reports of tPA in routine clinical practice
have demonstrated average DNTs from 100 to 48
minutes, with improvement over time suggesting a
learning curve.11–13 Despite this, only 26.6% of pa-
tients treated with tPA in the Get With The
310 Neurology 79 July 24, 2012
Guidelines–Stroke program have a DNT below 60
minutes,14 compared with 94% in our center.
Several interventions reported to reduce throm-
bolysis delays have been previously reviewed.15,16 The
involvement and education of the emergency call
center and EMS on stroke allows rapid patient
identification and transport.17,18 We have imple-
mented these since 1998,19 and have previously
reported that faster treatment associates with bet-
ter patient outcome.5,6
Alarming a stroke team on admission may be ben-
eficial,20 but does not allow for preparation prior to a
patient’s arrival. The effect of hospital prenotifica-
tion on in-hospital delays has been reported in sev-
eral studies.15,21 Computerized order entry has
reduced in-hospital treatment delays in a Korean
hospital from 79 to 56 minutes.22 Since 2001, we
have computer-ordered the blood tests and CT, but
contrary to most centers, do this already before the
patient arrives. It would take us at least 30 minutes to
receive standard INR results, but with POC INR,
they are available before starting the CT. POC INR
also has proved reasonably reliable and useful in re-
ducing treatment delays elsewhere.23
Significant results in reducing delays are never
achievable by single interventions, but rather result
from continuous analysis and improvement of the
whole system,15,24,25 a fact acknowledged during
the pivotal National Institute of Neurological Dis-
orders and Stroke trial.26 A large American Heart
Association/American Stroke Association initia-
tive is currently rolling out a set of interventions to
reduce DNT.15 We have been implementing them
at our stroke service since 1998 and our results
verify that they can make a difference: shorten the
DNT, and increase the number of patients eligible
for thrombolysis.
Previous studies have suggested that having more
time translates to more delay due to lower staff-
perceived urgency, so that patients arriving 0 to 60
minutes from stroke onset have longer in-hospital
delays.27,28 This may be true, but in-hospital delays in
this patient group were longer because very early ar-
rivals did not provide time to activate the stroke
team, vacate the CT table, and check the history of
the patient prior to their arrival. In a study of the Get
With The Guidelines program, 12,545 patients
treated with tPA within 3 hours of symptom onset
were analyzed, excluding in-hospital strokes, to show
that patients arriving early had longer in-hospital de-
lays.29 However, the mean DNT in that study was 86
minutes, 91 minutes in those arriving within 1 hour
of symptom onset, and 77 minutes in those arriving
later. With such delays, it is obvious that patients
arriving for example after 100 minutes from symp-
tom onset and having a 91-minute in-hospital delay
were simply not treated. Hence, they were excluded
from the analysis, which can explain the apparently
faster action in those arriving late, a selection bias
acknowledged by the authors. In our data such a se-
lection could be detected for the patients arriving just
prior to the 3-hour time window in effect up to late
2008. With symptom-to-door times between 30 and
150 minutes the in-hospital delays were consistent.
In our series, having more time did not translate
to more delay, confirming similar data from
Germany.25
Our study has several limitations. First, we do not
have comprehensive data on how many of our stroke
patients were not treated with tPA due to in-hospital
or prehospital delays, as they are not included in the
Helsinki Stroke Thrombolysis Registry. The subject
has been studied extensively, and a recent review re-
vealed that about half of stroke patients in reports
over the last 10 years have arrived within 4 hours,
which is also our general experience.30 Second, with
several concomitant and consecutive measures to re-
duce our treatment delays, together with the con-
stantly accumulating experience and the resulting
confidence and routine of our stroke team including
EMS personnel, it is not possible to say for certain
how much each step contributed to the overall im-
provement. While we earlier thought that relocating
the CT scanner inside the ER was the most impor-
tant step,3 there was a period in 2010 when the orig-
inal CT in another part of the hospital had to be used
due to renovation, but this caused little or no effect
on delays (figure 1). Direct transport from ambu-
lance stretchers onto the CT table, instead of an ER
bed, and the stroke team and drug infusion waiting
for the patient at the CT seems to be more crucial
than the location of the CT. Lack of reliable history,
mostly on the stroke onset time, is a major cause of
delay in the evaluation of thrombolysis candidates.
Taking the history during patient transport has been
extremely helpful,5 and is likely to be a major con-
tributor to our reduced delays. Third, we do not
know if our interventions, while successful in reduc-
ing delays, actually produced better patient out-
comes. Lacking a randomized setting or even a
control group, our data do not provide answers to
this question. As only 1.4% of our treated patients
eventually proved to be stroke mimics, a rate low in
comparison to other series, fast-tracked evaluation
and treatment do not seem to promote misdiagno-
sis.31 As faster treatment has been shown to produce
better outcomes in clinical trials,2 it is likely that also
our patients have benefited. In addition, faster initial
evaluation is likely to benefit also those stroke pa-
tients who will not receive IV tPA, such as patients
with intra-arterial interventions,32 or primary intra-
cerebral hemorrhage. Fourth, the set of interventions
described in this article led to significant DNT re-
duction in our setting, but might not be directly ap-
plicable in other settings, for example due to lack of
electronic patient records or neurologist on-site 24/7.
However, most of the interventions are likely to be
reproducible elsewhere, although confirmatory stud-
ies might be warranted.
The target of initiating treatment within 60 min-
utes of arrival, set in the year 1996 by the National
Institute of Neurological Disorders and Stroke,33 is
no longer ambitious enough. The present study
shows that cutting this target to 20 minutes is feasi-
ble, and reports how this could be achieved. Treat-
ment delays have to be analyzed, evaluated, and
improved constantly. The numbers of patients
treated and treatment delays are linked both ways;
with more routine and experience, it is possible to
treat patients faster; and with faster treatment, it is
possible to treat more patients while staying within
evidence-based time windows. The key to success in
reducing the delays is to do only the basics while the
patient has arrived, and to do as much as possible
before, during transport.
Neurology 79 July 24, 2012 311
 AUTHOR CONTRIBUTIONS
A. Meretoja conceived and designed the study, drafted the manuscript,
and performed the statistical analyses. M. Kaste and T. Tatlisumak super-
vised and coordinated the study and obtained study funding. All authors
acquired the data, analyzed and interpreted the data, and edited the man-
uscript for intellectual content.
DISCLOSURE
A. Meretoja has received honoraria for speaking in educational symposia
and compensation for expert consultancy with Boehringer Ingelheim, and
travel expenses from H. Lundbeck A/S. D. Strbian and S. Mustanoja
report no disclosures. T. Tatlisumak has research contracts with
Boehringer-Ingelheim, Sanofi Aventis, H. Lundbeck A/S, Mitsubishi
Pharma, Schering Plough, Concentric Medical, PhotoThera, and Brains-
Gate. He has received a grant from Boehringer Ingelheim and has served
on advisory boards or as a consultant for Boehringer-Ingelheim, Mitsubi-
shi Pharma, and BrainsGate. P. Lindsberg reports no disclosures. M.
Kaste has received honoraria and his travel expenses have been covered for
participating in the Steering Committee meetings of the DIAS-4 trial, and
for serving as a consultant for Boehringer-Ingelheim and H. Lundbeck
A/S. Go to Neurology.org for full disclosures.
Received October 6, 2011. Accepted in final form December 1, 2011.
REFERENCES
1. Saver JL. Time is brain— quantified. Stroke 2006;37:263–
266.
2. Lees KR, Bluhmki E, von Kummer R, et al. Time to treat-
ment with intravenous alteplase and outcome in stroke: an
updated pooled analysis of ECASS, ATLANTIS, NINDS,
and EPITHET trials. Lancet 2010;375:1695–1703.
3. Lindsberg PJ, Happola O, Kallela M, Valanne L, Kuisma
M, Kaste M. Door to thrombolysis: ER reorganization and
reduced delays to acute stroke treatment. Neurology 2006;
67:334 –336.
4. Meretoja A, Putaala J, Tatlisumak T, et al. Off-label
thrombolysis is not associated with poor outcome in pa-
tients with stroke. Stroke 2010;41:1450 –1458.
5. Strbian D, Soinne L, Sairanen T, et al. Ultraearly throm-
bolysis in acute ischemic stroke is associated with better
outcome and lower mortality. Stroke 2010;41:712–716.
6. Puolakka T, Vayrynen T, Happola O, Soinne L, Kuisma
M, Lindsberg PJ. Sequential analysis of pretreatment de-
lays in stroke thrombolysis. Acad Emerg Med 2010;17:
965–969.
7. Sairanen T, Strbian D, Soinne L, et al. Intravenous throm-
bolysis of basilar artery occlusion: predictors of recanaliza-
tion and outcome. Stroke 2011;42:2175–2179.
8. Hacke W, Kaste M, Bluhmki E, et al. Thrombolysis with
alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl
J Med 2008;359:1317–1329.
9. Meretoja A, Roine RO, Kaste M, et al. Stroke monitoring
on a national level: PERFECT stroke, a comprehensive,
registry-linkage stroke database in Finland. Stroke 2010;
41:2239 –2246.
10. Meretoja A, Roine RO, Kaste M, et al. PERFECT Stroke
annual reports. Available at: http://www.thl.fi/fi_FI/web/
fi/tutkimus/hankkeet/perfect/aivohalvaus/perusraportit.
Accessed October 6, 2011.
11. Koennecke HC, Nohr R, Leistner S, Marx P. Intravenous
tPA for ischemic stroke team performance over time,
safety, and efficacy in a single-center, 2-year experience.
Stroke 2001;32:1074 –1078.
12. Grond M, Stenzel C, Schmulling S, et al. Early intrave-
nous thrombolysis for acute ischemic stroke in a
312 Neurology 79 July 24, 2012
community-based approach. Stroke 1998;29:1544 –
1549.
13. Chiu D, Krieger D, Villar-Cordova C, et al. Intravenous
tissue plasminogen activator for acute ischemic stroke: fea-
sibility, safety, and efficacy in the first year of clinical prac-
tice. Stroke 1998;29:18 –22.
14. Fonarow GC, Smith EE, Saver JL, et al. Timeliness of
tissue-type plasminogen activator therapy in acute isch-
emic stroke: patient characteristics, hospital factors, and
outcomes associated with door-to-needle times within 60
minutes. Circulation 2011;123:750 –758.
15. Fonarow GC, Smith EE, Saver JL, et al. Improving door-
to-needle times in acute ischemic stroke: the design and
rationale for the American Heart Association/American
Stroke Association’s Target: Stroke Initiative. Stroke 2011;
42:2983–2989.
16. Price CI, Clement F, Gray J, Donaldson C, Ford GA.
Systematic review of stroke thrombolysis service configura-
tion. Expert Rev Neurother 2009;9:211–233.
17. Zweifler RM, Drinkard R, Cunningham S, Brody ML,
Rothrock JF. Implementation of a stroke code system in
Mobile, Alabama: diagnostic and therapeutic yield. Stroke
1997;28:981–983.
18. Wein TH, Staub L, Felberg R, et al. Activation of emer-
gency medical services for acute stroke in a nonurban pop-
ulation: the T.L.L. Temple Foundation Stroke Project.
Stroke 2000;31:1925–1928.
19. Lindsberg PJ, Soinne L, Roine RO, et al. Community-
based thrombolytic therapy of acute ischemic stroke in
Helsinki. Stroke 2003;34:1443–1449.
20. Gomez CR, Malkoff MD, Sauer CM, Tulyapronchote R,
Burch CM, Banet GA. Code stroke: an attempt to shorten
inhospital therapeutic delays. Stroke 1994;25:1920 –1923.
21. Belvis R, Cocho D, Marti-Fabregas J, et al. Benefits of a
prehospital stroke code system: feasibility and efficacy in
the first year of clinical practice in Barcelona, Spain. Cere-
brovasc Dis 2005;19:96 –101.
22. Nam HS, Han SW, Ahn SH, et al. Improved time inter-
vals by implementation of computerized physician order
entry-based stroke team approach. Cerebrovasc Dis 2007;
23:289 –293.
23. Rizos T, Herweh C, Jenetzky E, et al. Point-of-care inter-
national normalized ratio testing accelerates thrombolysis
in patients with acute ischemic stroke using oral anticoag-
ulants. Stroke 2009;40:3547–3551.
24. Sattin JA, Olson SE, Liu L, Raman R, Lyden PD. An
expedited code stroke protocol is feasible and safe. Stroke
2006;37:2935–2939.
25. Kohrmann M, Schellinger PD, Breuer L, et al. Avoiding in
hospital delays and eliminating the three-hour effect in
thrombolysis for stroke. Int J Stroke 2011;6:493– 497.
26. Tilley BC, Lyden PD, Brott TG, Lu M, Levine SR, Welch
KM. Total quality improvement method for reduction of
delays between emergency department admission and
treatment of acute ischemic stroke: The National Institute
of Neurological Disorders and Stroke rt-PA Stroke Study
Group. Arch Neurol 1997;54:1466 –1474.
27. Albers GW, Bates VE, Clark WM, Bell R, Verro P, Ham-
ilton SA. Intravenous tissue-type plasminogen activator for
treatment of acute stroke: the Standard Treatment with
Alteplase to Reverse Stroke (STARS) study. JAMA 2000;
283:1145–1150.
28. Ferrari J, Knoflach M, Kiechl S, et al. Stroke thrombolysis:
having more time translates into delayed therapy: data
 from the Austrian Stroke Unit Registry. Stroke 2010;41:
2001–2004.
29. Saver JL, Smith EE, Fonarow GC, et al. The “golden
hour” and acute brain ischemia: presenting features and
lytic therapy in ⬎30,000 patients arriving within 60 min-
utes of stroke onset. Stroke 2010;41:1431–1439.
30. Evenson KR, Foraker RE, Morris DL, Rosamond WD. A
comprehensive review of prehospital and in-hospital delay
times in acute stroke care. Int J Stroke 2009;4:187–199.
31. Artto V, Putaala J, Strbian D, et al. Stroke mimics and intra-
venous thrombolysis. Ann Emerg Med 2012;59:27–32.
32. Khatri P, Abruzzo T, Yeatts SD, Nichols C, Broderick JP,
Tomsick TA. Good clinical outcome after ischemic stroke
with successful revascularization is time-dependent. Neu-
rology 2009;73:1066 –1072.
33. Marler JR, Winters Jones P, Emr M, eds. The National
Institute of Neurological Disorders and Stroke: proceed-
ings of National Symposium on Rapid Identification and
Treatment of Acute Stroke: NIH Publication No. 97-
4239. Available at: http://www.ninds.nih.gov/news_and_
events/proceedings/strokeworkshop.htm. Accessed Octo-
ber 6, 2011.

Neurology Podcast ®

Reducing in-hospital delay to 20 minutes in stroke thrombolysis

At www.neurology.org, click on the “Download Latest Issue” link or “Subscribe Now” to
subscribe to the RSS Feed.
This week’s podcast features an interview by Dr. Andy Southerland with Dr. Atte Meretoja about
his paper on reducing in-hospital delay.
NEW CME Opportunity: Listen to this week’s Neurology® Podcast and earn 0.5 AMA PRA
Category 1 CME Credits™ by answering the multiple-choice questions in the online Podcast quiz.

Scan this code with your smart phone to directly access the Podcast

Neurology Online CME Program

®

Earn CME while reading Neurology. This program is available only to online Neurology
subscribers. Simply read the articles marked CME, go to www.neurology.org, and click on CME.
This will provide all of the information necessary to get started. The American Academy of
Neurology (AAN) is accredited by the Accreditation Council for Continuing Medical Education
(ACCME) to sponsor continuing medical education for physicians. Neurology is planned and
produced in accordance with the ACCME Essentials. For more information, contact AAN Member
Services at 800-879-1960.

Neurology 79 July 24, 2012 313

 Reducing in-hospital delay to 20 minutes in stroke thrombolysis
Atte Meretoja, Daniel Strbian, Satu Mustanoja, et al.
Neurology 2012;79;306; Published online before print May 23, 2012;
DOI 10.1212/WNL.0b013e31825d6011
This information is current as of August 15, 2012

Updated Information & including high resolution figures, can be found at:
Services http://www.neurology.org/content/79/4/306.full.html

Supplementary Material Supplementary material can be found at:

http://www.neurology.org/content/suppl/2012/05/23/WNL.0b01
3e31825d6011.DC1.html
http://www.neurology.org/content/suppl/2012/05/23/WNL.0b01
3e31825d6011.DC2.html
http://www.neurology.org/content/suppl/2012/07/22/WNL.0b01
3e31825d6011.DC3.html
References This article cites 31 articles, 20 of which can be accessed free
at:
http://www.neurology.org/content/79/4/306.full.html#ref-list-1
Citations This article has been cited by 1 HighWire-hosted articles:
http://www.neurology.org/content/79/4/306.full.html#related-ur
ls
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
Electronic medical records
http://www.neurology.org/cgi/collection/electronic_medical_rec
ords
Infarction
http://www.neurology.org/cgi/collection/infarction
Models of care
http://www.neurology.org/cgi/collection/models_of_care
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
http://www.neurology.org/misc/about.xhtml#permissions
Reprints Information about ordering reprints can be found online:
http://www.neurology.org/misc/addir.xhtml#reprintsus
Door-to-needle times in acute ischemic stroke : How low can we
go?
Eric E. Smith and Rudiger von Kummer
Neurology 2012;79;296; Published online before print May 23, 2012;
DOI 10.1212/WNL.0b013e31825d602e

This information is current as of August 15, 2012

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://www.neurology.org/content/79/4/296.full.html

Neurology ® is the official journal of the American Academy of Neurology. Published continuously
since 1951, it is now a weekly with 48 issues per year. Copyright © 2012 by AAN Enterprises, Inc. All
rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
EDITORIAL
Door-to-needle times in acute
ischemic stroke
How low can we go?
Eric E. Smith, MD, IV human recombinant tissue plasminogen activator
MPH, FAHA (rt-PA) is the best proven and most effective treat-
Rudiger von Kummer, ment for acute ischemic stroke. The benefits of rt-PA
MD, FAHA are highly dependent on the time elapsed since stroke
symptoms.1 Accordingly, the Brain Attack Coalition
recommends that the time from emergency room ar-
Correspondence & reprint rival to initiation of rt-PA (that is, the “door-to-
requests to Dr. Smith:
eesmith@ucalgary.ca needle” time) should be 60 minutes or less.2
Are patients actually receiving rt-PA within 60
Neurology® 2012;79:296–297 minutes in clinical practice? Unfortunately, recent
studies clarify that they are not. Data from the large
American Heart Association/American Stroke Asso-
ciation Get With The Guidelines–Stroke program
and the multinational Safe Implementation of Treat-
ment in Stroke–International Stroke Thrombolysis
Register registry showed that median door-to-needle
times were 75 minutes and 65 minutes, respec-
tively.3,4 Among 641 US hospitals reporting ⱖ10 rt-
PA–treated patients, only 6.7% treated more than
half of their patients within 60 minutes.3
Clearly, neither US nor European hospitals are
meeting the proposed 60-minute door-to-needle
benchmark. Either the Brain Attack Coalition’s
benchmark is unrealistic and unfeasible— except
possibly in very experienced stroke centers or in the
context of prospective clinical trials— or a large
number of hospitals are missing out on processes of
care that could reduce door-to-needle times. So,
which is it?
This issue of Neurology® provides a convincing
answer by Meretoja et al.5 In this single center retro-
spective cohort study from Helsinki University Cen-
tral Hospital, a succession of improvements to their
rt-PA protocols resulted in a substantial reduction in
median door-to-needle times over a 10-year period.
A strikingly low median door-to-needle time of 20
minutes was achieved, with 94% of patients treated
within 60 minutes. Innovative strategies to reduce
door-to-needle time included obtaining history and
provisional consent for rt-PA by telephone in the
field, and bypassing the emergency department to
See page 306
From the Calgary Stroke Program (E.E.S.), Hotchkiss Brain Institute, University of Calgary, Calgary, Canada; and Department of Neuroradiology
(R.v.K.), Dresden University Stroke Center, Dresden, Germany.
Go to Neurology.org for full disclosures. Disclosures deemed relevant by the authors, if any, are provided at the end of this editorial.
296 Copyright © 2012 by AAN Enterprises, Inc.
take patients directly from emergency medical trans-
port to the CT scanner.
This remarkable improvement in door-to-needle
time contributed to a 45-minute shortening of the
time from stroke onset to treatment between 1999
and 2011. According to the pooled rt-PA trial data,
reducing onset-to-treatment time by 45 minutes
should result in an approximate 0.16 absolute in-
crease in the odds ratio for good recovery for rt-PA
compared to placebo.1 For a hypothetical “average”
patient in the 1995 National Institute of Neurologi-
cal Disorders and Stroke trial with a 26% chance of
good outcome on placebo but a 1.70-fold increase in
the odds of good outcome with rt-PA, increasing that
odds ratio from 1.70 to 1.86 would increase the
probability of a good outcome with tPA treatment
from 37% to 40%. This is a modest shift in an indi-
vidual’s expected outcome, but similar improve-
ments averaged over the tens of thousands of patients
treated annually with rt-PA6 could save hundreds of
patients per year from a lifetime of disability.
Meretoja et al. have demonstrated that door-to-
needle times of 60 minutes—indeed, much less than
60 minutes— can be consistently achieved. Given
this, it is an ethical imperative for us to work toward
similar results in our own hospitals. However, chang-
ing practices may not be easy. For many hospitals,
emulating the protocols used by Helsinki University
Central Hospital will require substantial changes to
systems of care involving not only neurologists, but
also others involved in emergency medical services
and in emergency medicine and radiology depart-
ments. Additionally, it is important that enough time
be reserved for accurate clinical evaluation, including
identification of contraindications to rt-PA. Patient
safety must not be sacrificed for the sake of speed.
Fortunately, there is some help for hospitals seek-
ing advice on how to reduce their door-to-needle
times. In 2010 the American Heart Association/
American Stroke Association launched the Target:
Stroke Initiative—accessible at www.targetstroke.
org— designed to help hospitals enact strategies to
reduce door-to-needle times with the ultimate goal of
treating more than half of their patients in 60 min-
utes or less. Registered hospitals can access informa-
tion on strategies and care pathways for acute stroke,
and share examples of best practices.7
By showing us how rapidly rt-PA can be given,
Meretoja et al. also show us the limits of what rt-PA
can achieve. Despite a remarkable reduction in treat-
ment times, 60% or more of their patients still had
poor outcomes throughout the years of their study.
Unfortunately, rt-PA usually fails to recanalize oc-
cluded arteries8 and fails to help patients whose arter-
ies have spontaneously recanalized, with the
consequent risk of hemorrhage. Clearly, rt-PA is not
a panacea. Better ways are needed to quickly identify
patients with an occluded artery but who still have
viable brain, and better drugs or devices are needed
to recanalize the occluded artery.9 Randomized con-
trolled trials to test new approaches must remain a
priority.
Are the door-to-needle times achieved by Mere-
toja et al. the limit, or can we reduce times further? It
is hard to imagine a system better optimized for
timely administration of rt-PA. Nonetheless, one
should not discount the possibility of further innova-
tions to reduce door-to-needle times. Ruling out
hemorrhage in the field during transport, for exam-
ple, by as-yet undiscovered biomarkers, might be one
way to further shorten times. Additionally, a renewed
emphasis on prehospital care is warranted, including
public education to activate emergency response sys-
tems immediately when stroke symptoms occur, cou-
pled with triage and rapid transport to the nearest
stroke center by emergency medical services. Im-
provements in the speed of prehospital care will fur-
ther shorten the time from stroke onset to rt-PA
administration, even after hospital door-to-needle
times are minimized.
For current and future treatments, door-to-needle
times less than 60 minutes should be the rule, not the
exception—now it is up to us to make it so.
DISCLOSURE
Dr. Smith reports grant funding from the NINDS, Canadian Institutes of
Health Research, Alberta Innovates–Heath Solutions, Canadian Stroke
Network, and Heart and Stroke Foundation of Canada; is an unpaid
volunteer for the American Heart Association Get With The Guidelines
program; and is an Assistant Editor of Stroke. Prof. von Kummer reports
personal compensation for serving on the Advisory Board of Lundbeck
AC, serving as Co-Chair on the Steering Committee of the DIAS-3 and -4
trials, serving on the image adjudication committee for these trials, and
consulting Synarc; and is Section Editor for Interventional Neuroradiol-
ogy of Neuroradiology. Go to Neurology.org for full disclosures.
REFERENCES
1. Lees KR, Bluhmki E, von Kummer R, et al. Time to treat-
ment with intravenous alteplase and outcome in stroke: an
updated pooled analysis of ECASS, ATLANTIS, NINDS,
and EPITHET trials. Lancet 2010;375:1695–1703.
2. Alberts MJ, Hademenos G, Latchaw RE, et al. Recom-
mendations for the establishment of primary stroke cen-
ters: Brain Attack Coalition. JAMA 2000;283:3102–3109.
3. Fonarow GC, Smith EE, Saver JL, et al. Timeliness of
tissue-type plasminogen activator therapy in acute isch-
emic stroke: patient characteristics, hospital factors, and
outcomes associated with door-to-needle times within 60
minutes. Circulation 2011;123:750 –758.
4. Ahmed N, Wahlgren N, Grond M, et al. Implementation
and outcome of thrombolysis with alteplase 3– 4.5 h after
an acute stroke: an updated analysis from SITS-ISTR.
Lancet Neurol 2010;9:866 – 874.
5. Meretoja A, Strbian D, Mustanoja S, Tatlisumak T,
Lindsberg PJ, Kaste M. Reducing in-hospital delay to
20 minutes in stroke thrombolysis. Neurology 2012;79:
306 –313.
6. Adeoye O, Hornung R, Khatri P, Kleindorfer D. Recom-
binant tissue-type plasminogen activator use for ischemic
stroke in the United States: a doubling of treatment rates
over the course of 5 years. Stroke 2011;42:1952–1955.
7. Fonarow GC, Smith EE, Saver JL, et al. Improving door-
to-needle times in acute ischemic stroke: the design and
rationale for the American Heart Association/American
Stroke Association’s Target: stroke initiative. Stroke 2011;
42:2983–2989.
8. Alexandrov AV, Molina CA, Grotta JC, et al. Ultrasound-
enhanced systemic thrombolysis for acute ischemic stroke.
N Engl J Med 2004;351:2170 –2178.
9. von Kummer R. After European Cooperative Acute Stroke
Study 3: mission accomplished? Stroke 2009;40:2268 –
2270.
Neurology 79 July 24, 2012 297
 Door-to-needle times in acute ischemic stroke : How low can we go?
Eric E. Smith and Rudiger von Kummer
Neurology 2012;79;296; Published online before print May 23, 2012;
DOI 10.1212/WNL.0b013e31825d602e
This information is current as of August 15, 2012

Updated Information & including high resolution figures, can be found at:
Services http://www.neurology.org/content/79/4/296.full.html

References This article cites 9 articles, 5 of which can be accessed free at:
http://www.neurology.org/content/79/4/296.full.html#ref-list-1
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
http://www.neurology.org/misc/about.xhtml#permissions
Reprints Information about ordering reprints can be found online:
http://www.neurology.org/misc/addir.xhtml#reprintsus