10.1161/CIRCULATIONAHA.116.023336
Treatment with TPA in the “Golden Hour” and the Shape of the
4.5 Hour Time-Benefit Curve in the National US Get With The GuidelinesStroke Population
Running title: Kim et al., Golden-Hour thrombolysis and Time-Benefit Curve
Joon-Tae Kim, MD1; Gregg C. Fonarow, MD2; Eric E. Smith, MD3; Mathew J. Reeves, PhD4;
Digvijaya D. Navalkele, MD5; James C. Grotta, MD6; Maria V. Grau-Sepulveda, MD7; Adrian F.
Hernandez, MD7; Eric D. Peterson, MD7; Lee H. Schwamm, MD8; Jeffrey L. Saver, MD9
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
1
2
Department of Neurology, Chonnam National University Hospital, Gwangju, Korea;
Department of Medicine, David Geffen School of Medicine, University of Calif
California,
for
o nia, Los
Angeles, CA; 3Hotchkiss Brain Institute, University of Calgary, Calgary, Albe
Alberta,
ert
r a, C
Canada;
anad
an
ada;
ad
a;
4
5
Department of Epidemiology and Biostatistics, Michigan State University, East Lansing, MI;
Department of Neurology, University of Texas Health Science Center, Houston, TX; 6Clinical
Innovation
Innnovati
In
tion
ti
on andd Research Institute, Memorial Hermann
H rmann Hospi
He
Hospital,
itaal, Ho
Houston, TX; 7Outcome
Rese
Research
seearch and Asse
Assessment
essme
meent
n Group,
Gro
roup, Duke
Du e Clinic
Clinical
cal Research
Reseear
Re
arch Ins
Institute,
nsti
ns
t tute
te, Du
Durham,
urham
am, NC;
am
C;; 8De
Department
Depa
partmeent oof
pa
Neurology,
Massachusetts
General
Hospital,
Medical
School,
Boston,
Neur
urol
ur
o ogy, M
assaachuuseetts G
as
e eral H
en
osppital, Harvard
Harva
vardd M
ediccall Sch
chool,, B
ostton, MA;
os
A 9D
A;
Department
epartme
m n
eurolo
logy
ogy, Da
avi
vidd Ge
Geff
ffen
en Sch
hoo
ooll of
of Med
ediccin
ed
ne,
e U
nniivers
rsit
ityy off Cal
it
alif
ifor
o ni
or
niaa, Los
oss A
nggel
e es
es, CA
A
of N
Neurology,
David
Geffen
School
Medicine,
University
California,
Angeles,
Address for Correspondence
Jeffrey L. Saver, MD, FAHA, FAAN, FANA
UCLA Stroke Center
710 Westwood Plaza
Los Angeles, CA 90095, USA
Tel 310-794-6379
Fax 310-267-2063
E-mail: [email protected]
Journal Subject Terms: Cerebrovascular Disease/Stroke; Ischemic Stroke
1
10.1161/CIRCULATIONAHA.116.023336
Abstract
Background—Earlier tissue plasminogen activator (tPA) treatment improves ischemic stroke
outcome, but aspects of the time-benefit relationship still not well delineated are: 1) the degree of
additional benefit accrued with treatment in the first 60 minutes after onset, and 2) the shape of
the time-benefit curve through 4.5 hours.
Methods—We analyzed acute ischemic stroke patients treated with intravenous tPA within 4.5
hours of onset from the Get With The Guidelines-Stroke US national program. Onset to
treatment (OTT) time was analyzed as a continuous, potentially nonlinear variable and as a
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
categorical variable comparing patients treated within 60 minutes of onset with later epochs.
Results—Among 65,384 tPA-treated patients, the median OTT was 141 minutes (IQR 110-173)
and 878 patients (1.3%) were treated within the first 60 minutes. Treatment within 660
0 mi
minu
nute
nu
tes,
te
s,
minutes,
compared with treatment within 61-270 minutes, was associated with increased od
dds ooff
odds
discharge to home (adjusted OR 1.25, 95% confidence intervals [CI] 1.07-1.45), independent
ambulation at discharge (adjusted OR 1.22, 95% CI 1.03-1.45), and freedom from disability
mod
oddif
ifiied Rankin
Rank
nkin Scale
nk
Scale 0-1) at discharge
g (adjusted
d OR 1.72, 95%
% CI 11.21-2.46),
.21-2.46),
2
) without
),
(modified
ncrreased hemorrha
rhagicc complications
comp
mpli
mp
licationns or in-hospital
li
in-hosppitaal mortality.
mortalityy. The
mo
Thhe pace off ddecline
eclinee iin
n be
enefit of
increased
hemorrhagic
benefit
PA
A from
fro OTT
T ttimes
imees of
im
of 20 tthrough
hrough 2270
hr
70 m
inuttes
e was
was mildly
mildly nonlinear
n nlin
no
inear for
fo discharge
disc
di
s harg
ge to
o hhome,
ome,
tPA
minutes
with more rapid benefi
f t loss in the first 17
1170
0 minutes than later, andd linear for independe
d nt
benefit
independent
ambulation and in-hospital mortality.
Conclusions—Thrombolysis started within the first 60 minutes after onset is associated with best
outcomes for patients with acute ischemic stroke, and benefit declined more rapidly early after
onset for ability to be discharged home. These findings support intensive efforts to organize
stroke systems of care to improve the timeliness of thrombolytic therapy in acute ischemic
stroke.
Keywords: Golden hour thrombolysis, time-benefit curve, acute ischemic stroke; thrombolysis;
tissue-type plasminogen activator
2
10.1161/CIRCULATIONAHA.116.023336
Clinical Perspective
What is new?
x
In this large registry of over 65,000 ischemic stroke patients treated with intravenous tPA,
878 patients were treated within the first 60m after onset, a ten-fold increase over
previously available data.
x
Thrombolytic treatment within the first 60m was associated with the highest rates of
favorable discharge outcomes.
x
Also, the shape of the time-benefit curve throughout the first 4.5h post-onset was
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
nonlinear for some outcomes.
x
Discharge to home and discharge free of disability decayed more rapidly in the first 100170m after onset than later, while independent ambulation at discharge and in
n-hospital
in-hospital
mortality
y declined in a steady fashion throughout the 4.5h window.
What are the clinical implications?
x
T
he su
supe
periior outcomes with thrombolysis inn the first 60m,
pe
60m
m, and more
more rapid decay of
The
superior
benefit forr ssome
ome
mee out
tco
comes ea
earlie
ier afterr oonset
nse
set su
upportt, intensive
inten
ensivee efforts
efffor
ortts to speed
spee
sp
e d patient
patien
nt
outcomes
earlier
support,
presen
nta
tattion and
nd tPA
A ttreatment
reatmen
ent start
en
sttart inn all
all patients.
patie
ieent
n s.
presentation
x
T e findings
Th
findi
fi
dings reinforce
rein
i force the im
iimportance
porttance off qu
alit
lity improvementt progra
ams
m tto
o accel
leratte
The
quality
programs
accelerate
‘door to needle’ time, support further implementation of telemedicine systems enabling
delivery of thrombolysis in rural and other frontline hospitals, and provide impetus for
further research on the use of mobile stroke unit ambulances equipped with CT scanners
that enable faster, prehospital delivery of thrombolysis.
3
10.1161/CIRCULATIONAHA.116.023336
Introduction
Thrombolytic therapy with intravenous tissue plasminogen activator (tPA) is beneficial for
patients with acute ischemic stroke within 4.5 hours of stroke onset.1-3 Pooled analysis of
randomized trials show the benefit of tPA is strongly time-dependent.3, 4 Worldwide, stroke
physicians have implemented continuous quality improvement programs to accelerate delivery of
tPA, steadily reducing door-to-needle (DTN) times.5-7
To guide further improvement of tPA delivery, it would be helpful to clarify two aspects
of the time-dependence of tPA benefit: 1) the additional benefit associated with hyperacute
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
treatment in the first 60 minutes after last known well time, and 2) the shape of the time-benefit
curve of tPA through 4.5 hours.
Until recently, tPA start in the first 60m post-onset, the “golden hour,” was not achievable
except in a tiny proportion of patients. As a result, analyses of the relation of onset to treatment
time
ime
me (OTT)
(OTT) and
nd out
outcome
utco
ut
ome bbased
ased
as
ed oon
n ra
rrandomized
nd
dom
o ized
d tri
trials
ialls have
have
v oonly
n y co
nl
considered
ons
n id
i er
ered ttimes
imes
im
es ffrom
r m 60
ro
8-10
minutes
incognita.
minu
nute
nu
t s onward,
onward
rd, with
rd
wi the
thhe golden
gol
olden hour treated
ol
trreated as terra
teerraa in
inco
c gni
co
n ta.3,, 8-10
However,
Howe
Ho
wever, rec
we
recent
centt
developments mak
ke hy
hhyperacute
peracute tPA
A treatment within
withi
hin 60 minutes of onset attainab
ble in clinical
make
attainable
practice. Continued improvements in systems to shorten DTN times, to as low as an average of
20 minutes,5 make it feasible to start tPA in Emergency Departments within 60 minutes postonset in very early arriving patients. Also, even faster tPA initiation can be performed with use of
mobile stroke unit ambulances equipped with CT scanners, permitting prehospital initiation of
thrombolysis.11-14 Both of these approaches, however, require substantial expenditures of labor
and capital, which makes their cost-benefit uncertain. Information is needed on whether the
degree of additional benefit gained with hyperacute treatment within the first hour is sufficient to
justify these special efforts.14
4
10.1161/CIRCULATIONAHA.116.023336
In addition, the shape of the time-benefit curve for tPA between 1 to 4.5 hours has not been
previously directly studied. Prior time analyses imposed a linear relationship, rather than making
the shape the subject of empiric investigation.3, 8, 10, 15 But clinical trial data are also compatible
with an exponential pace of benefit decline.4 Larger registry datasets can provide insights into
how quickly benefits of tPA decay with time and show whether the benefit of tPA declines with
time in an exponential, linear, or other pattern.
Methods
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
The Get With The Guidelines-Stroke Program (GWTG–Stroke) is a national registry launched by
he American Heart Association and American Stroke Association to support cont
tin
nuo
uous
u qquality
us
uali
ua
l ty
li
the
continuous
mprovement of hospital systems of care for patients with stroke and transient ischemic attack
improvement
TIA).16, 17 Details of the design and conduct of the GWTG-Stroke Program have previously
(TIA).
1
been
en ddescribed.
escribed
d.17
Briefly,
B
Br
ieefl
fly,
y, the
the GWTG
GWT
WTG
G us
uses a W
Web-based
e -basedd patient
eb
pati
pa
tien
ti
ent management
en
mana
ma
nageeme
m nt tool
too
ooll (Quintiles)
( uiinttil
(Q
iles
es)) to
es
t
collect
coll
lec
ect clinicall ddata
atta onn conse
consecutively
secutively
se
y aadmitted
dmittedd ppatients,
dm
attien
ents
ts,, to provide
ts
pro
rovidde decision
decis
isio
on suppor
support,
rt,, and
and
n to
enable real-time online
onlline reporting ffeatures.
eatures.17 The
h G
GWTG-Stroke
W G-Stroke Program was made available in
WT
i
April 2003 to any hospital in the United States.17 Each participating hospital received either
human research approval to enroll patients without individual patient consent under the common
rule or a waiver of authorization and exemption from subsequent review by their institutional
review board. The institutional review board of the data analysis center at Duke University
approved the study.
Data from hospitals that participated in the program any time between January 1, 2009,
and September 30, 2013, were included in this analysis. The study focused upon patients with
acute ischemic stroke treated with tPA within 270 minutes (4.5 hours) of last known well time.
5
10.1161/CIRCULATIONAHA.116.023336
Detailed study inclusion and exclusion criteria are shown in Supplemental Figure 1.
Analyzed outcomes included (1) in-hospital mortality, (2) discharge status (ordinal: home, acute
rehabilitation, skilled nursing facility, or dead; and binary: home vs other), (3) and ambulatory
status at discharge (ordinal: ambulatory without another person’s assistance, ambulatory only
with another person’s assistance, nonambulatory, or dead; and binary: ambulatory without
assistance vs other). Safety outcomes included symptomatic intracranial hemorrhage (sICH)
within 36 hours and severe systemic hemorrhage. In addition, for exploratory analysis, the
modified Rankin Scale (mRS) at discharge was analyzed, during the time period after it began
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
being collected in October 2012. The mRS was analyzed both over the full, 7 level (0-6) ordinal
cale and dichotomized at two cutpoints. The binary analyses were: mRS 0-1 (exc
cellle
lent
nt ooutcome;
utco
ut
com
co
me
scale
(excellent
nondisabled), and mRS 0-2 (good outcome; functionally independent).
Statistical analysis
Demo
mographic an
mo
and cl
cclinical
in
niccal vvariables,
arria
iabl
bles
bl
es,, ho
es
osp
s ital-le
leve
le
v l char
arac
ar
acte
ac
teri
te
rist
ri
s ic
icss, and
and
n clinical
clinica
call outcomes
ca
outc
ou
t om
tc
mess were
wer
eree
Demographic
hospital-level
characteristics,
comp
mpared amo
mp
ong
n ppatients
atieents tr
reated in thee 0-60, 661-90,
1-90
90, 91
91-1
-180
180
80,, aand
nd 1181-270
81-27
2700 m
inutess OTT
OTT wind
dow
compared
among
treated
91-180,
minutes
windows
and also between patients treated within andd beyond the golden hour (0-60 minutes). Percentages
were reported for categorical variables, and medians and interquartile ranges (IQRs) for
continuous variables. The Pearson-Ȥtest, Student-t test, and Wilcoxon rank-sum/Kruskal-Wallis
tests, as appropriate, were used to compare variables in the OTT time epochs.
Multivariable regression analysis was also performed to explore the relationship between
OTT and clinical outcome of interests. For dichotomous outcomes, logistic regression with GEEs
method with an exchangeable correlation structure was used to account for hospital clustering.
These multivariable analyses adjusted for 23 patient-level and 7 hospital-level variables (Table 3
footnote). Missing data for select key variables were imputed to the mode or median but patients
6
10.1161/CIRCULATIONAHA.116.023336
with missing NIHSS (<7% of the total) were excluded from all regression analyses. For ordinal
outcomes, adjusted rates were derived as an average of individual predicted probabilities
computed for each response category and each OTT time epoch among all possible combination
of covariates using a multivariate generalized logit model for non-ordered categorical outcomes.
All P values were 2-sided and statistical significance was defined as P value of less than 0.05.
The shape of the relationship between OTT and binary outcomes was assessed using
binary logistic regression and OTT restricted cubic splines with knots at 5th, 35th, 65th and 95th
percentiles to avoid any presumption of linearity.18, 19 To generate time benefit curves, outcome
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
specific predicted probabilities for each value of OTT within the observed range (18-270
minutes) were computed while setting all other variables in the model to mean values.
val
alue
ues.
ue
s To
s.
visually assess linearity, ORs for each OTT value were computed with the outcome rate at the
golden hour (60 minutes) as the reference. Wald Chi Square tests for non-linearity were also
perf
for
ormed. SAS
AS (ve
versio
ve
ionn 99.3;
io
.3;; S
AS IInstitute,
nsti
titu
t te, Ca
Cary,, NC)
NC) statistical
stat
st
atis
at
isti
is
t ca
call software
s ft
so
f wa
ware was
was
a uused
s d fo
se
forr al
alll
performed.
(version
SAS
tattis
isti
t cal analys
ysess.
ys
statistical
analyses.
Results
Among all stroke patients in the GWTG-Stroke registry during the study period, 1,065,875 were
patients with ischemic stroke admitted to hospitals with 30 or more cases and less than 25% of
missing data in medical history panels (Supplemental Figure 1). Among these, 65,384 ischemic
stroke patients from 1,456 sites were treated with tPA within 4.5 hours of symptom onset and
constitute the study population. The OTT time for tPA administration was median, 141 minutes,
IQR 110-173 (mean, 145±46). There were 878 (1.3%) patients with OTT time of 0 to 60
minutes, 6,490 (9.9%) with OTT time of 61 to 90 minutes, 46,457 (71.1%) with OTT time of 91
7
10.1161/CIRCULATIONAHA.116.023336
to 180 minutes, and 11,559 (17.7%) with OTT time of 181 to 270 minutes.
The general characteristics of patients treated with tPA in each of the 4 time epochs are
shown in Table 1, and unadjusted and adjusted ORs of potential factors associated with
hyperacute therapy within 60 minutes of onset are shown in Table 2. Patient-level factors
independently associated with receiving tPA within the first 60 minutes after adjustment
included: more severe presenting neurologic deficit (higher NIHSS score); arrival during regular,
weekday hours; and less frequent arrival by Emergency Medical Services (EMS) ambulance. The
relationship between age and hyperacute thrombolysis differed in younger versus older patients:
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
up to age 65, increasing age was associated with higher likelihood of golden hour thrombolysis;
but after age 65 increasing age was associated with lower likelihood of hyperacute
te tthrombolysis.
hrom
hr
om
mbo
boly
lysi
ly
s s.
Hospital-level factors independently associated with receiving tPA within the first 60 minutes
ncluded: higher annual volume of tPA-treated cases; not being a Primary Stroke Center; and
included:
bein
ng in the W
esst. In
In addition,
addi
ad
diti
di
tion
ti
onn, patients
pati
pa
tien
ti
e ts treated
en
tre
r ated iin
n th
the go
gold
lden
ld
en hhour
ouur ha
hhad
d sh
shorte
er in
inte
terv
te
r alss fr
rv
from
om llast
at
as
being
West.
golden
shorter
intervals
know
own well to ED
ow
E arrival
arriv
ival and
nd DTN times
tim
mes thann longer
long
ngerr OTT
OTT ggroups.
rooupps.
known
U ad
Un
djusted
d rates of efficacy and safety outcomes in the 4 OT
O
T groups are shown in
Unadjusted
OTT
Supplemental Table 1. Overall, at discharge, 42.5% of patients were sent home and 42.0% were
ambulating independently. In the cohort enrolled after the program initiated collection of
discharge mRS information, 28.5% were nondisabled (mRS 0-1), and 38.0% were functionally
independent (mRS 0-2). sICH occurred in 4.4%, severe systemic hemorrhage in 1.0%, and inhospital death in 7.5%.
The association of OTT time epoch with adjusted outcomes is shown in Table 3. In the
adjusted analysis, for patients treated in the first 60 minutes, compared with those treated with
more than 60 minutes of onset, all dichotomized efficacy outcomes were better, including
8
10.1161/CIRCULATIONAHA.116.023336
increased odds of discharge to home, independence in ambulation, freedom from disability (mRS
0-1), and functional independent (mRS 0-2). The largest effect was seen for excellent outcome
(mRS 0-1), with OR 1.72 (95% CI 1.21-2.46). In contrast, although sICH, systemic hemorrhage,
and mortality increased across all 4 analyzed OTT time epochs, these safety outcomes were not
distinctively further reduced among hyperacutely treated patients. Considering ordinal efficacy
outcomes, each earlier OTT time epoch was associated with an incremental beneficial shift,
including for ambulatory status at discharge, discharge destination, and disability level at
discharge (Figure 1).
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
The continuous time-benefit curves for 4 dichotomized efficacy and safety outcomes are
hown in Figure 2 (adjusted probability analysis), Supplemental Figure 2 (unadju
ust
sted
ed pprobability
roba
ro
babi
ba
bili
bi
l ty
shown
(unadjusted
analysis), and Supplemental Figure 3 (odds ratio analysis). A non-linear relation between rates of
discharge to home and OTT was noted (adjusted P for non-linearity=0.006); the probability of
disccha
harge to hom
omee declined
om
d cl
de
clin
ned ste
tead
te
adil
ad
ilyy from
il
om OTT
T of 2200 minutes
min
inut
utes
ut
es too 170
170 minutes
minu
mi
n tees (10.5
(100.55 fewer
(1
fewe
werr patients
we
pati
pa
tien
ti
e ts
discharge
home
steadily
th
h
tre
reated
re
ed ddischarged
isscharrged home with
with every
ev
ver
ery 15 minute
min
inut
ute
t delay)
delay)
y) with
witth a less
leess steep
ep decline
dec
ecline
per thousand
treated
1 270 minutes (2.6 fewer pati
ients dis
i ch
harged home per thousand treated with
wiith every
between 171171-270
patients
discharged
15 minute delay). There was a visually non-linear relation between discharge free of disability
(mRS 0-1) and OTT, with steeper decline from 20 minutes to 100 minutes and slower thereafter,
though formal testing for non-linearity did not reach statistical significance (adjusted P for nonlinearity=0.34). For independent ambulation at discharge, a linear relationship with OTT was
observed throughout 4.5 hours window, with 9.6 fewer per 1000 treated per 15 minute delay.
Similarly, for in-hospital mortality, a linear relation with OTT was noted, with 1.4 more per 1000
treated per 15 minute delay.
9
10.1161/CIRCULATIONAHA.116.023336
Discussion
In this study of more than 65,000 patients treated with intravenous tissue plasminogen activator
throughout the United States, treatment in the golden hour, within 60 minutes of onset, was
associated with higher rates of excellent early outcome compared to later treatment. After
adjustment for other outcome predictors, at the time of discharge, nearly one half of patients
treated in the golden hour were ambulating independently, nearly half could be sent directly
home, and more than one-third were free of disability (mRS 0-1). In addition, over the entire 4.5
hour window for treatment, the pace of the decline in benefit of tPA with longer onset to
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
treatment times was found to be non-linear for discharge to home, with more rapid benefit loss in
he first 170 minutes then later, potentially mildly non-linear for discharge fr
ffree
ee off di
ddisability
sabi
sa
bili
bi
lity
li
ty wi
with
the
he steepest loss of benefit over the first 100 minutes, and linear throughout for independent
the
ambulation and in-hospital mortality.
Our find
ndin
nd
ngs substantially
sub
ubst
ub
stan
st
anti
tiial
ally
ly enlarge
enl
n arrge
g the reported
rep
e or
orted ex
expe
peri
pe
rien
ri
ence
cee with
wit
i h hy
hhyperacute
peera
racu
cu
ute thrombolysis
thrrom
mbo
boly
lysi
ly
ss
findings
experience
with
th
hin
i 60 minutes
minuute
tess of
of onset.
onset. The
The onlyy randomized
ran
andomizzed tr
rial
al ooff tP
tPA
A tto
o spe
p cifiicallly report tr
reat
atment
within
trial
specifically
treatment
within 660
0 minutes
miinutes of onset were the
h two NIN
NDS
S-tP
PA Study trials; only 2 patie
i nts received study
d
NINDS-tPA
patients
infusion in the first 60 minutes and both received placebo.20, 21 The golden hour cohort size of
878 patients in the current study exceeds by an order of magnitude the largest prior report, from
the Prehospital Acute Neurological Treatment and Optimization of Medical care in Stroke
(PHANTOM-S) study.12 In our study, compared with treatment beyond the first 60 minutes,
hyperacute thrombolysis in the first 60 minutes was associated after adjustment with having 25%
greater odds of discharge to home, 22% greater odds of independent ambulation at discharge,
72% greater odds of being nondisabled (mRS 0-1) at discharge, and 58% greater odds of being
functionally independent (mRS 0-2) at discharge. The PHANTOM-S study showed similar
10
10.1161/CIRCULATIONAHA.116.023336
results for the one functional outcome it examined, that patients who received golden hour tPA
treatment were more likely to be discharged home compared with patients with a longer OTT
time.14 The results of our study provide reliable information on the magnitude of improved early
outcome associated with tPA treatment within the first 60 minutes after stroke onset.
Our findings additionally provide novel information regarding the shape of time-benefit
curve for tPA throughout the 0-4.5 hour time window for outcomes at discharge. Prior studies
have generally simply assumed a linear decline of benefit and have had insufficient sample sizes
to explore time-benefit variation at a more granular level. 3,4,8,10 With the larger GWTG-Stroke
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
cohort, we were able to delineate the time-benefit curve in a data-driven manner, making no
assumptions regarding linear, exponential, or other We found that, for excellent fun
unct
ctio
ct
io
ona
n l
functional
outcome (discharge to home), there was a mildly nonlinear relationship with OTT, with 2.5-fold
faster loss of benefit in the first 170 minutes than the next 100 minutes. Analysis of another index
of eexcellent
xcellent out
xc
utccome
ut
m , mR
mRS
S 0-1,
0-1
-1,, wa
wass co
ons
n traine
n d by low
ne
wer
e ssample
ampl
am
plee si
ssize
z (due
ze
(due to more
mor
ore re
eceent
outcome,
constrained
lower
recent
add
diti
diti
t on of thiss ffield
ieeld
d too the re
registry),
) but
),
ut also su
ssuggested
gggesteed a nnonlinear
onl
nlineearr time
me-bbenefit curve,
cur
u ve
ur
ve, with
addition
time-benefit
faster decline in th
he first 60-100 minutes than later, although thi
h s was not statistically significant.
the
this
In contrast, for good functional outcome (ambulating independently at discharge) and for
mortality, benefit declined linearly throughout the 4.5 hour treatment window. That superior
outcome is disproportionately affected by OTT in the early part of the treatment window is
consonant with dependence of excellent functional outcome upon the achievement of the
smallest final infarct volumes.22, 23
Despite the superior outcomes associated with treatment in the first 60 minutes of onset,
only 1.3% of tPA patients in the GWTG-Stroke cohort were treated in this early time frame.
Hyperacute treatment within 60 minutes of onset required rapid responses by both prehospital
11
10.1161/CIRCULATIONAHA.116.023336
and in-hospital systems of care. Compared with other patients, golden hour-treated patients
arrived much earlier at the Emergency Department (median 22 minutes) and had much shorter
DTN times (median 30 minutes). Even faster average DTN times have been attained at best
practice hospitals worldwide, often employing direct paramedic delivery of patients to the CT
suite rather than the Emergency Department.5, 24
The superior outcomes with thrombolysis in the golden hour in the current study further
emphasize the importance of system interventions to accelerate prehospital and Emergency
Department stroke care processes. Multiple interventions have been found beneficial in
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
shortening onset to tPA times, including pre-arrival notification by EMS providers, written
ke te
team
am
protocols for acute triage and patients flow, single call system to activate all strok
stroke
members, and direct to CT patient delivery.5, 6, 25, 26 Randomized trials have demonstrated that
prehospital delivery of thrombolysis in an ambulance equipped with a mobile CT scanner is
associated
asso
oci
ciated wit
with
th su
substantial
ubs
b taanttia
iall reduction
reedu
duct
ctio
ct
ionn off OTT ttime.
io
im
me. Thee fi
find
findings
nddin
ingss ffrom
room ou
our st
study
tud
udyy su
sugg
suggest
g es
e t th
that
at th
the
extra
extr
ra labor andd ccapital
api
pitall ccosts
ostss ooff mobile stroke
sttroke units
unitss have
hav
avee the
av
th potential
potent
po
ntial too bee justified
justifiedd byy improved
improoved
functional outcomes.
Our results identify several particular targets for system improvement to onset to door
and DTN times. Patients who arrived at Emergency Department outside beyond working hours
were less likely to be treated within the golden hour, indicating a need for improved staffing
solutions for off hour time periods. Arrival by EMS ambulance actually occurred less often in
patients treated within 60 minutes of onset than in patients treated 61-180 minutes after onset.
The alarm to door time (interval from 911 call to ED arrival) for EMS-transported stroke patients
in the US is typically about 45 minutes, reflecting travel of the ambulance to the scene,
paramedic activities on scene, and travel from scene to door.27 This overall alarm to door time
12
10.1161/CIRCULATIONAHA.116.023336
affords little opportunity for treatment within 60 minutes of onset. These findings support efforts
to shorten prehospital care intervals to meet national targets.28 Hospitals with higher annual
volumes of tPA treatment were more often able to provide thrombolysis within 60 minutes of
onset, suggesting potential benefits to avoiding duplicate services when not mandated by
geographic distribution.
This study also provides important reassurance regarding the safety of rapid thrombolysis
initiation within 60 minutes of onset. It is theoretically possible that the very fast diagnostic and
treatment algorithms required to achieve treatment within 60 minutes of onset would be
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
associated with imperfect patient selection and therapeutic management, off-setting gains from
earlier reperfusion when there is a greater volume of still-salvageable threatened ttissue.
isssu
sue.
e 29
e.
However, we found that hyperacute treatment within 60 minutes of onset was as safe as later
herapy, in avoiding sICH or severe systemic bleeding. These findings accord with previous
therapy,
GWTG-Stroke
GW
WTG-Strokee sstudies
tud
udiess showing
ud
shhow
o in
ng th
that
hat sh
shorter
hor
o ter OTT
OT
TT an
and DTN
DTN ti
time
times
mess are
a e associated
ar
asso
as
s ciiat
ated
ed with
with reduced
r du
re
duce
cedd of
ce
tPA
PA
A complications
complicatiion
o s iincluding
nccluudingg sICH.6, 15 A
Accordingly,
ccorddin
inglly, the
he aavailable
vail
ilaable eevidence
il
videencee suggests
t tha
ts
that
hat
hrombolysis within
withi
h n golden hour reduces compli
l cations, in add
dition to improving functional
thrombolysis
complications,
addition
outcome.
This study has several limitations. First, it has the inherent limitation of a registry-based
study. However, to optimize data quality, the GWTG-Stroke Program includes detailed training
of site chart abstractors, standardized case definitions and coding instructions, predefined logic
and range checks on data fields at data entry, audit trails, and regular data quality reports for all
sites.17 Second, longer term, post-discharge functional outcome, e.g. at 90 days, was not
investigated, as post-discharge data were not collected in the GWTG-Stroke registry during the
study period. Assessment at discharge requires some degree of rater surmise regarding the
13
10.1161/CIRCULATIONAHA.116.023336
patient’s degree of independence (mRS 2 vs 3), as the patient has not yet fully attempted
community integration. Also, this early assessment does not reflect further improvements in
function that may occur over subsequent months. However, studies have shown that functional
outcomes at discharge highly correlate with outcomes at 3 months.30, 31 Third, our results analyze
the effects of earlier time to lysis versus later time to lysis, not of earlier time to lysis versus
earlier time to placebo. Fourth, patients treated in the first 60 minutes represented only 1.3% of
the dataset. While this constrained study power, this population of 878 patients is an order of
magnitude larger than in prior reports..
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Conclusions
In
n broad national United States clinical practice, compared to later time epochs, hyperacute
treatment
reatment with tPA in the golden hour is associated with increased independent ambulation at
discharge,
disc
cha
harge, discharge
disch
charrge too ho
ch
home
home,
me,, and
and fr
freedo
freedom
dom from
o ooff disa
om
disability
sabi
sa
b liity orr dependence
bi
depe
de
pend
ndencee at
a discharge.
dis
ischar
arge
ar
ge. Over
ge
Ove
the
he entire
en
n
4.5 hhour
our gguideline-endorsed
uid
idelinee-endorsed tP
etPA
PA tim
time
me w
window,
ind
ndow
ow, rrates
ow
ates
at
es of bbeing
eingg ddisability-free
isa
sability-ffre
sa
free aatt
discharge and disch
harge to home decay
d cay more rapidly
de
rapid
dly
l in the first 100-170
1000-1700 minutes of stroke onset
discharge
to treatment time, while independent ambulation at discharge and in-hospital mortality decline in
a linear fashion throughout. These findings lend further support of intensive efforts to speed
patient presentation and tPA treatment start in all patients, including increasing the frequency of
treatment within stroke’s golden hour.
14
10.1161/CIRCULATIONAHA.116.023336
Sources of Funding: The GWTG-Stroke Program is provided by the AHA/American Stroke
Association. The GWTG-Stroke Program is currently supported in part by a charitable
contribution from Janssen Pharmaceutical Companies of Johnson & Johnson. The GWTGStroke Program has been funded in the past through support from Boehringer-Ingelheim, Merck,
Bristol-Myers Squibb/Sanofi Pharmaceutical Partnership, and the AHA Pharmaceutical
Roundtable.
Conflict of Interest Disclosures: J-TK reports no potential conflicts of interest. GCF reports
serving as a member of the GWTG Steering Committee; receipt of research support (to the
institution) from the Patient Centered Outcome Research Institute; and being an employee of the
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
University of California, which holds a patent on retriever devices for ischemic stroke. EES
reports serving as a member of the GWTG Steering Committee, and as a member of a data safety
trial
al. MJR
MJR reports
repo
re
porrts
po
and monitoring board for Massachusetts General Hospital for the MR Witness trial.
eceiving salary support from the Michigan Stroke Registry and serving as a member
memb
ber off th
he
receiving
the
American Heart Association (AHA) GWTG Quality Improvement Subcommittee. DDN reports
ten
enti
tial
al cconflicts
o flic
on
icts of interest. JCG is employedd by
ic
b Memorial Herman
ann Hospital and receives
no pot
potential
Hermann
connsuulting feess from
from Frazer
Fraaze
zerr Ltd,
Ltd, and
and ambulance
amb
m ulance
c manufacturer,
ce
manuf
ufac
uf
actu
ac
tu
ure
rer, aand
nd S
tryk
yker,, an
andd gr
grant su
upp
ppor
ortt
or
consulting
Stryker,
support
m the Ameri
rica
c n Heart
Heeart Association.
Asssociation. M
G-S rreport
epoortt bein
ingg mem
in
m
ember
ers off the
he D
uke Cl
Clin
inicaal
in
from
American
MG-S
being
members
Duke
Clinical
Reseear
arch
ch Ins
nsti
ns
titu
ti
tute ((DCRI),
tu
DC
CRI), w
hi h sserves
hich
e vees as
er
a tthe
he AHA
HA GW
G
TG ddata
a a co
at
ccoordinating
orrdinaati
ting
ng ccenter.
entterr. A
en
FH
Research
Institute
which
GWTG
AFH
eports being a recipient of an AHA Pharmaceutical Roundtable grant and having received
reports
research support from Johnson & Johnson and Amylin. EDP reports serving as principal
investigator of the Data Analytic Center for AHA GWTG; receipt of research grants from
Johnson & Johnson, Eli Lilly, and Janssen Pharmaceuticals; and serving as a consultant to
Boehringer Ingelheim, Johnson & Johnson, Medscape, Merck, Novartis, Ortho-McNeil-Janssen,
Pfizer, Westat, the Cardiovascular Research Foundation, WebMD, and United Healthcare. LHS
reports serving as chair of the AHA GWTG Steering Committee; as a consultant to the
Massachusetts Department of Public Health, Lundbeck, and to the Joint Commission; and
provision of alteplase to Massachusetts General Hospital for a NINDS-funded multicenter study
of extended window thrombolysis in magnetic resonance–guided patient selection from
Genentech. JLS reports serving as a member of the Get With the Guidelines (GWTG) Science
Subcommittee; as a scientific consultant regarding trial design and conduct to Medtronic,
15
10.1161/CIRCULATIONAHA.116.023336
Neuravia, Stryker, Grifols, Brainsgate, Lundbeck, Boehringer Ingelheim (prevention studies
only), and St Jude Medical; receipt of a grant (to the institution) from the National Institutes of
Health (NIH)-National Institute of Neurological Disorders and Stroke (NINDS); and being an
employee of the University of California, which holds a patent on retriever devices for ischemic
stroke.
References
1.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
2.
3.
4.
5.
6.
7.
8.
Tissue plasminogen activator for acute ischemic stroke. The national institute of
neurological disorders and stroke rt-pa stroke study group. New Engl J Med.
1995;333:1581-1587
Jauch EC, Saver JL, Adams HP, Jr., Bruno A, Connors JJ, Demaerschalk BM, Khatri P,
McMullan PW, Jr., Qureshi AI, Rosenfield K, Scott PA, Summers DR, Wang DZ,
Wintermark M, Yonas H, American Heart Association Stroke C, Council on
Cardiovascular N, Council on Peripheral Vascular D, Council on Clinical C.
C Guidelines
Gui
uide
deli
de
line
li
n s
ne
for the early management of patients with acute ischemic stroke: A guideline for
healthcare professionals from the american heart association/american stroke association.
Stroke. 2013;44:870-947
Embe
Em
bers
be
r on JJ,, Lees KR, Lyden P, Blackwelll L,
rs
L Albers G, B
luhm
lu
mki
k E, Brott T, Cohen G,
Emberson
Bluhmki
Davis S, Donnan G, Grotta
Gro
roottta J, Howard G, K
aste M
K
g M
ga
mmer
mm
e R, Lans
nsbe
beerg
Kaste
M,, Ko
Koga
M,, von Kumm
Kummer
Lansberg
M, Lindley
y RI
I, M
urra
rayy G, Olivot
ra
Oli
l vo
vot JM, Pa
Pars
sonss M
illeey B,
il
B, Toni
Tonni D,, Toyoda
Toyod
odaa K, Wahlgren
od
Wahlg
lg
gre
ren
RI,
Murray
Parsons
M,, Til
Tilley
N, War
rdllaw
w J, W
hitel
eley W, del Z
el
oppo G
J, B
aig
igen
ig
entt C
en
San
ndercoc
ock P
trokee
Wardlaw
Whiteley
Zoppo
GJ,
Baigent
C,, Sandercock
P,, Hackee W
W,, S
Stroke
Thromb
mbolys
mb
ysis T
rialissts
ts' Collaborative
Colllab
abor
oraative G.
or
G E
ffecct of tre
ff
eattmeent dela
ay, ag
age, and
nd str
roke
Thrombolysis
Trialists'
Effect
treatment
delay,
stroke
sseverity
se
veri
rity
ri
ty oon
n th
tthee ef
eeffects
fect
fe
ctss of iintravenous
ntra
nt
rave
vennouss tthrombolysis
h om
hr
mbo
boly
lysiis with
ly
with alteplase
alt
ltep
pla
lasee for
forr acute
te ischaemic
isc
scha
haem
ha
emic
ic
stroke: A meta-analysis
meta analysis of individual patient data from randomised trials.
trials Lancet
Lancet.
2014;384:1929-1935
Saver JL, Levine SR. Alteplase for ischaemic stroke--much sooner is much better. Lancet.
2010;375:1667-1668
Meretoja A, Strbian D, Mustanoja S, Tatlisumak T, Lindsberg PJ, Kaste M. Reducing inhospital delay to 20 minutes in stroke thrombolysis. Neurology. 2012;79:306-313
Fonarow GC, Zhao X, Smith EE, Saver JL, Reeves MJ, Bhatt DL, Xian Y, Hernandez AF,
Peterson ED, Schwamm LH. Door-to-needle times for tissue plasminogen activator
administration and clinical outcomes in acute ischemic stroke before and after a quality
improvement initiative. JAMA. 2014;311:1632-1640
Kim J, Hwang YH, Kim JT, Choi NC, Kang SY, Cha JK, Ha YS, Shin DI, Kim S, Lim
BH. Establishment of government-initiated comprehensive stroke centers for acute
ischemic stroke management in South Korea. Stroke. 2014;45:2391-2396
Hacke W, Donnan G, Fieschi C, Kaste M, von Kummer R, Broderick JP, Brott T, Frankel
M, Grotta JC, Haley EC, Jr., Kwiatkowski T, Levine SR, Lewandowski C, Lu M, Lyden
P, Marler JR, Patel S, Tilley BC, Albers G, Bluhmki E, Wilhelm M, Hamilton S,
Investigators AT, Investigators ET, Investigators Nr-PSG. Association of outcome with
early stroke treatment: Pooled analysis of ATLANTIS, ECASS, and NINDS rt-PA stroke
16
10.1161/CIRCULATIONAHA.116.023336
9.
10.
11.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
12.
13.
14.
15.
16.
17.
18.
19.
20.
trials. Lancet. 2004;363:768-774
Marler JR, Tilley BC, Lu M, Brott TG, Lyden PC, Grotta JC, Broderick JP, Levine SR,
Frankel MP, Horowitz SH, Haley EC, Jr., Lewandowski CA, Kwiatkowski TP. Early
stroke treatment associated with better outcome: The NINDS rt-PA stroke study.
Neurology. 2000;55:1649-1655.
Lees KR, Bluhmki E, von Kummer R, Brott TG, Toni D, Grotta JC, Albers GW, Kaste M,
Marler JR, Hamilton SA, Tilley BC, Davis SM, Donnan GA, Hacke W, Ecass AN, Group
Er-PS, Allen K, Mau J, Meier D, del Zoppo G, De Silva DA, Butcher KS, Parsons MW,
Barber PA, Levi C, Bladin C, Byrnes G. Time to treatment with intravenous alteplase and
outcome in stroke: An updated pooled analysis of ecass, atlantis, ninds, and epithet trials.
Lancet. 2010;375:1695-1703
Bowry R, Parker S, Rajan SS, Yamal JM, Wu TC, Richardson L, Noser E, Persse D,
Jackson K, Grotta JC. Benefits of stroke treatment using a mobile stroke unit compared
with standard management: The BEST-MSU study run-in phase. Stroke. 2015;46:33703374
Ebinger M, Winter B, Wendt M, Weber JE, Waldschmidt C, Rozanski M, Kunz A, Koch
P, Kellner PA, Gierhake D, Villringer K, Fiebach JB, Grittner U, Hartmann A, Mackert
d es M,, Audebert
udebe HJ,
J, Consortium
Co so u S. Effect
ec of
o thee use of
o ambulance-based
a bu a ce based
BM,, Endres
thrombolysis on time to thrombolysis in acute ischemic stroke: A randomi
izeed clinical
clin
cl
inic
in
ical
ic
al
randomized
trial. JAMA. 2014;311:1622-1631
Walter S, Kostopoulos P, Haass A, Keller I, Lesmeister M, Schlechtriemen T, Roth C,
Papanagiotou P, Grunwald I, Schumacher H, Helwig S, Viera J, Korner H, Alexandrou
M, Yilmaz U, Ziegler K, Schmidt K, Dabew R, Kubulus D, Liu Y, Volk T, Kronfeld K,
Ruck
Ru
ckes
ck
es C, Bertsch
Bertsch T, Reith W, Fassbenderr K.
K. Diagnosis and
an treatment
trea
eatment of patients with
Ruckes
stroke iin
n a mo
m
b le sstroke
bi
t ok
tr
okee un
unit
i versus
ver
ersus in
n hhospital:
osspitaal:
l A rrandomised
andoomi
an
mise
sed co
ccontrolled
ntro
roll
ro
lled
ll
ed ttrial.
rial. La
Lanc
ncet
nc
et
mobile
Lancet
Neurol. 20
012;1
11:
1:3977-4404
72012;11:397-404
Ebinger
er M
unz A,, W
endt M, R
ozansk
s i M,
sk
M Winter
Win
inte
ter B,
te
B, Waldschmidt
Wal
aldsch
hmiidt C, Weber
Weeberr J,
J,
M,, Ku
Kunz
Wendt
Rozanski
V
Vi
llri
ring
ri
nger K,
ng
K, Fiebach
Fieebachh JB,
JB, Audebert
Auddebe
Au
bert
be
r HJ.
HJ.
J. Effects
Eff
ffeects
ts ooff go
gold
l en
ld
e hhour
o r th
ou
thromb
mbolys
mb
ysis
ys
is: A
is
Villringer
golden
thrombolysis:
prehospitall acute neurologic
i al treatment andd optimizatio
i n of med
dical care in stroke
neurological
optimization
medical
(PHANTOM-S) substudy. JAMA Neurol. 2015;72:25-30
Saver JL, Fonarow GC, Smith EE, Reeves MJ, Grau-Sepulveda MV, Pan W, Olson DM,
Hernandez AF, Peterson ED, Schwamm LH. Time to treatment with intravenous tissue
plasminogen activator and outcome from acute ischemic stroke. JAMA. 2013;309:24802488
LaBresh KA, Reeves MJ, Frankel MR, Albright D, Schwamm LH. Hospital treatment of
patients with ischemic stroke or transient ischemic attack using the "Get With The
Guidelines" program. Arch Intern Med. 2008;168:411-417
Schwamm LH, Fonarow GC, Reeves MJ, Pan W, Frankel MR, Smith EE, Ellrodt G,
Cannon CP, Liang L, Peterson E, Labresh KA. Get With The Guidelines-Stroke is
associated with sustained improvement in care for patients hospitalized with acute stroke
or transient ischemic attack. Circulation. 2009;119:107-115
Desquilbet L, Mariotti F. Dose-response analyses using restricted cubic spline functions
in public health research. Stat Med. 2010;29:1037-1057
Harrell FE. Regression Modeling Strategies: with Applications to Linear Models,
Logistic Regression, and Survival Analysis. New York, NY: Springer; 2001.
Dachs RJ, Burton JH, Joslin J. A user's guide to the ninds rt-PA stroke trial database.
17
10.1161/CIRCULATIONAHA.116.023336
21.
22.
23.
24.
25.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
26.
27.
28.
29.
30.
31.
PLoS Med. 2008;5:e113
Grotta JC. tPA for stroke: Important progress in achieving faster treatment. JAMA.
2014;311:1615-1617
Zaidi SF, Aghaebrahim A, Urra X, Jumaa MA, Jankowitz B, Hammer M, Nogueira R,
Horowitz M, Reddy V, Jovin TG. Final infarct volume is a stronger predictor of outcome
than recanalization in patients with proximal middle cerebral artery occlusion treated
with endovascular therapy. Stroke. 2012;43:3238-3244
Yoo AJ, Chaudhry ZA, Nogueira RG, Lev MH, Schaefer PW, Schwamm LH, Hirsch JA,
Gonzalez RG. Infarct volume is a pivotal biomarker after intra-arterial stroke therapy.
Stroke. 2012;43:1323-1330
Kohrmann M, Schellinger PD, Breuer L, Dohrn M, Kuramatsu JB, Blinzler C, Schwab S,
Huttner HB. Avoiding in hospital delays and eliminating the three-hour effect in
thrombolysis for stroke. Int J Stroke. 2011;6:493-497
Tveiten A, Mygland A, Ljostad U, Thomassen L. Intravenous thrombolysis for ischaemic
stroke: Short delays and high community-based treatment rates after organisational
changes in a previously inexperienced centre. Emerg Med J. 2009;26:324-326
A systems approach to immediate evaluation and management of hyperacute stroke.
pe e ce aat eeight
g ce
e s aand
d implications
p ca o s for
o co
u y ppractice
ac ce aand
d pa
e ca
e. Thee
Experience
centers
community
patient
care.
ke sstudy
tu
udy group.
gro
rouup.
national institute of neurological disorders and stroke (NINDS) rt-PA stroke
Stroke. 1997;28:1530-1540
Wojner-Alexandrov AW, Alexandrov AV, Rodriguez D, Persse D, Grotta JC. Houston
paramedic and emergency stroke treatment and outcomes study (HoPSTO). Stroke.
2005;36:1512-1518
Ack
ker JJE,
E, 33rd,
rd, Pancioli AM, Crocco TJ, Ec
ckstein MK, Jauch
Jau
uch
c EC,
EC, Larrabee H, Meltzer
Acker
Eckstein
NM, Me
Merg
gen
endaahl W
C, M
unnn JW
un
W, Pren
ntiiss S
M,, S
andd C,
an
C Saver
Sav
ver
e JJL,
L Eig
L,
igel
ig
e B
el
l in B
lp
R,
Mergendahl
WC,
Munn
JW,
Prentiss
SM,
Sand
Eigel
B,, Gilp
Gilpin
BR,
Schoeberl M
o is P,
ol
P, Bailey
Bailey
y JR,
JR, Horton
Horto
on KB,
KB Stranne
Sttrannee S
K,, American
Ameeriican
an Heartt A
merican
an
M,, S
Solis
SK,
A,, Am
American
Stroke Association
Asssociaation Expert
Expert Panel
Paneel on Eme
m rg
me
gency
cy M
ed
edic
dical S
ervice
ces Sy
Systems S
C.
Emergency
Medical
Services
SC.
IImplementation
Im
plem
pl
ementa
em
tation
on strategies
stratteg
egie
i s for
for emergency
em
mer
e ge
genncy medical
m dica
me
dica
c l se
serv
r iccess w
rv
ithhinn st
stro
roke ssystems
ro
ystems
ys
m of ca
are:
services
within
stroke
care:
A policy statement from the
th
he american heart association/a
/ merican stroke
k associa
i tion
association/american
association
expert panel on emergency medical services systems and the stroke council. Stroke.
2007;38:3097-3115
Saver JL. Time is brain--quantified. Stroke. 2006;37:263-266
Ovbiagele B, Saver JL. Day-90 acute ischemic stroke outcomes can be derived from
early functional activity level. Cerebrovascular diseases. 2010;29:50-56
Zhang Q, Yang Y, Saver JL. Discharge destination after acute hospitalization strongly
predicts three month disability outcome in ischemic stroke. Restor Neurol Neurosci.
2015;33:771-775
18
10.1161/CIRCULATIONAHA.116.023336
Table 1. Characteristics of ischemic stroke patients in different IV tPA onset-to-treatment time windows.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
N
Age
Female, sex
Race/ethnicity
White, non-hispanic
Black
Hispanic
Asian
an
Other
her
Arrival
al on hours (vs off hours)*
Arrival
al by EMS
LKW to arrival, median (IQR), min
Door to tPA, median (IQR), min
NIHSS,
S, median (IQR)
0-9
10-14
14
15-20
20
21”
”
Nott documented
doocum
cu
umented
Medical
cal hhistory
istory
Atrial
ial ffibrillation/flutter
ibrrillation/flutter
ib
CAD/prior
D/prrio
i r MI
Carotid
rotid sten
stenosis
enos
en
osis
sis
Diabetes
abetes melli
mellitus
l tus
Dyslipidemia
slipidemia
Hypertension
Prosthetic heart valve
Peripheral vascular disease
Heart failure
Smoker
Previous stroke/TIA
Medication prior to admission
Anticoagulants
Antiplatelets
Antihypertensive
Overall
65,384
70.2 (14.9)
33022 (50.5)
0-60 min
878
68.4 (14.4)
407 (46.4)
61-90 min
6,490
69.8 (14.8)
3074 (47.4)
91-180 min
46,457
70.9 (14.8)
23774 (51.2)
181-270 min
11,559
67.6 (14.9)
5766 (49.9)
46735 (71.5)
9455 (14.5)
4787 (7.3)
1714 (2.6)
2693 (4.1)
33328 (51.0)
48557 (74.3)
59 (39, 89)
71 (54, 94)
10 (6, 17)
28729 (43.9)
12366 (18.9)
11350 (17.4)
8445
45 (12.9)
(12.9)
9)
4494
44
9 (6
94
(6.9)
6.9)
9)
609 (69.4)
128 (14.6)
72 (8.2)
27 (0.01)
42 (4.8)
535 (60.9)
613 (69.8)
22 (15, 30)
30 (22, 38)
12 (6, 18)
347 (39.5)
174 (19.8)
176 (20.1)
131 (14.9)
(114.9)
5 (5
50
(5.7)
5.7
7)
4606 (71.0)
893 (13.8)
503 (7.8)
200 (3.1)
288 (4.4)
3731 (57.5)
4858 (74.9)
32 (24, 40)
48 (39, 56)
12 (7, 18)
2465 (38.0)
1400 (21.6)
(21
21.6)
1371 (21.1)
(211.1)
884
8884 (13.6)
(113.6)
370 ((5.7)
5.7)
33470 (72.1)
6605 (14.2)
3337 (7.2)
1170 (2.5)
1875 (4.0)
23586 (50.8)
35433 (76.3)
57 (41, 78)
74 (58, 93)
10 (6, 17)
19879 (42.8)
8871 (19.1)
8193 (17.6)
6278
62278 (13.5)
(13.5)
5)
3236
366 ((7.0)
7.0)
0)
8050 (69.6)
1829 (15.8)
875 (7.6)
317 (2.7)
488 (4.2)
5476 (47.4)
7653 (66.2)
125 (87, 156)
88 (62, 124)
8 (5, 15)
6038 (52.2)
1921
21 (16.6)
1610
10 (13.9)
1152
1 52 (10.0)
11
83
838
38 (7.3)
3)
113997
3997 ((21.6)
21.6)
116378
6378 (2
(25.2)
25.2)
1749 (2
(2.7)
2.7)
7)
116887
16
887 (26.0)
26705 (41.1)
(41 1)
47562 (73.2)
738 (1.1)
2326 (3.6)
5939 (9.1)
11572 (17.8)
16197 (24.9)
167 (1
(19.2)
19..2)
1193
93 (2
(22.2)
22..2)
1177 (2.0)
(2
2.0)
0)
193 (22.2)
2
343 (39
(39.5)
5)
596 (68.6)
8 (0.9)
32 (3.7)
69 (7.9)
152 (17.5)
199 (22.9)
11390
390 (21.6)
(211.6)
1540 (24.0)
(244.0))
163 (2.5)
(2.5)
1446
14
4 ((22.5)
222.5)
2626 (40.8)
(40 8)
4617 (71.8)
75 (1.2)
216 (3.4)
503 (7.8)
1143 (17.8)
1357 (21.1)
1042
10
10421
421
21 (22.6)
(2
22.6)
119577 (25.9)
(2
25.9)
1297
12
97 (2.8)
(2
2.8)
8
1213
12134
1 4 (26.3)
( 6.3)
(2
19083 (41.3)
(41 3)
34068 (73.8)
535 (1.2)
1682 (3.6)
4432 (9.6)
7990 (17.3)
11810 (25.6)
201
2019
19 (17.6)
(177.6
6)
2688
268
88 (23.4)
(233.4)
4)
272
2772 (2.4))
3114 (27.1)
(27.1)
4653 (40.5)
(40 5)
8281 (72.1)
120 (1.0)
396 (3.5)
935 (8.1)
2287 (19.9)
2831 (24.6)
<0.001
1
<0.001
<0.0001
0.02
0.02
02
<0.001
0.28
0 28
<0.001
0.69
0.56
<0.001
<0.001
<0.001
13
13830
(21.6
(21.6)
6)
16185 (25.3)
16
(25.3
3)
1732 (2.7)
16694 (26.1)
16
1
26362 (41.1)
(41 1)
46966 (73.3)
730 (1.1)
2294 (3.6)
5870 (9.2)
11420 (17.8)
15998 (25.0)
0.09
0.04
0.18
0.01
0.32
0.002
0.55
0.87
0.22
0.80
0.16
4215 (6.5)
26860 (41.1)
39144 (60.0)
41 (4.7)
356 (40.6)
478 (54.4)
327 (5.0)
2649 (40.8)
3664 (56.5)
3243 (7.0)
19325 (41.6)
28402 (61.1)
604 (5.2)
4530 (39.2)
6600 (57.1)
<0.001
<0.001
<0.001
4174 (6.5)
26504 (41.1)
38666 (59.9)
0.03
0.59
0.0002
19
P1
<0.001
<0.001
<0.001
<0.001
<0.0
001
<0.00
<0.001
01
<0.001
<0.001
<0.001
<0.001
>60 min
64,506
70.2 (14.9)
32615 (50.6)
46126 (71.5)
9327 (14.5)
4715 (7.3)
1687 (2.6)
26
2651
651
1 ((4.1)
4 1)
4.
32793
32793 (50.8)
(50
50.8
. )
.8
4794
47
47944
944
94
4 (7
(74.
(74.3)
4.3)
4.
3)
60 (40, 90)
72 (55, 94)
10 (6, 17)
28382 (44.0)
12192 (18.9)
11174 (17.3)
8314
83
3144 (12.9)
(12
2.9
.9))
P2
0.0001
0.01
0.48
<0.001
<
<
<0.001
<0.001
<
<0.001
<
0.0006
0
00.0014
10.1161/CIRCULATIONAHA.116.023336
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Cholesterol reducer
26310 (40.2)
329 (37.5)
2473 (38.1)
19006 (40.9)
4502 (39.0)
<0.001 25981 (40.3)
0.09
Antidiabetic
11380 (17.4)
121 (13.8)
913 (14.1)
8263 (17.8)
2083 (18.0)
<0.001 11259 (17.5)
0.004
Hospital size, median No. of beds
390 (268, 576) 397 (266, 569) 380 (264, 572) 389 (268, 572) 400 (273, 599) <0.001 390 (268, 576)
0.89
(IQR)
Hospital region
<0.001
<0.001
West
13956 (21.3)
246 (28.0)
1666 (25.7)
9761 (21.0)
2283 (19.8)
13710 (21.3)
South
23415 (35.8)
279 (31.8)
2110 (32.5)
16862 (36.3)
4164 (36.0)
23136 (35.9)
Midwest
11982 (18.3)
152 (17.3)
1144 (17.6)
8255 (17.8)
2431 (21.0)
11830 (18.3)
Northeast
16031 (24.5)
201 (22.9)
1570 (24.2)
11579 (24.9)
2681 (23.2)
15830 (24.5)
Academic hospital
41283 (63.1)
555 (63.2)
4012 (61.8)
29176 (62.8)
7540 (65.2)
<0.001 40728 (63.1)
0.90
Primary Stroke Center
35206 (53.8)
444 (50.6)
3474 (53.5)
25157 (54.2)
6131 (53.0)
0.03
34762 (53.9)
0.05
Rural location
1869 (2.9)
16 (1.8)
133 (2.1)
1367 (2.9)
353 (3.1)
<0.001
18
1853
853
3 ((2.9)
2.9)
2.
9
9)
0.06
Annual
al volume of ischemic stroke
239 (163, 373) 242 (161, 390) 239 (163, 366) 239 (163, 368) 240 (161, 375)
0.99
0.9
99
99
239
23
39 (163,
(11633, 369)
36 )
0.83
admission,
sion, median (IQR)
Annual
al volume of tPA
22 (12, 34)
24 (14, 40)
24 (14, 37)
22 (12, 34)
22 (12, 34)
<0.001
22 (12, 34)
0.0004
0
administration,
nistration, median (IQR)
Abbreviations:
viations: EMS; Emergency Medical Service, LKW; last known well time, tPA; tissue plasminogen activator, CAD; coronary artery disease, MI; myo
myocardial
ion, TIA; transient ischemic attack
infarction,
mparison across all 4 time epochs, p-values computed with chi-squared tests for categorical variables and Kruskal-Wallis test for continuous variables
variables,
P1; comparison
mparison
on ooff 00-60min
-600mi
minn vs
v >60min,
>600min, p-values computed with chi-squared tests for
fo categorical variables an
and Wilc
Wilcoxon
coxon rank sum test for continuous variables
variab
P2; comparison
*
On hours
ours w
were
erre 7AM to 6PM,
6PM
PM,, Mon-Fri.
20
10.1161/CIRCULATIONAHA.116.023336
Table 2. Factors potentially associated with golden hour thrombolysis (onset to treatment time
within 60 minutes).
Unadjusted OR
(95% CI)
1.05 (0.99-1.11)
0.89 (0.85-0.93)
0.84 (0.73-0.97)
P
Adjusted OR
(95% CI)
1.06 (1.01-1.12)
0.87 (0.82-0.91)
0.88 (0.76-1.01)
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
$JH”SHU\U
0.08
Age >65, per 5yr
<0.001
Female (ref; male)
0.02
Race
0.87
White
Ref
Ref
Black
1.04 (0.82-1.32)
1.03 (0.80-1.31)
Hispanic
1.13 (0.86-1.49)
1.01 (0.78-1.31)
Asian
1.13 (0.71-1.80)
0.96 (0.61-1.50)
Other
1.09 (0.75-1.57)
1.00 (0.71-1.40)
Arrival OFF hours
0.63 (0.54-0.72)
<0.001
0.62 (0.54-0.71)
Arrival mode (ref; no EMS)
0.68 (0.57-0.81)
<0.001
0.63 (0.52-0.75)
NIHSS, per 1, up to 25
1.02 (1.01-1.03)
0.0001
1.04 (1.03-1.05)
Hypertension
0.83 (0.70-0.97)
0.02
0.93 (0.78-1.11)
Diabetes mellitus
0.79 (0.66-0.95)
0.01
0.86 (0.68-1.07)
Dyslipidemia
0.96 (0.83-1.10)
0.54
1.04 (0.88-1.23)
Atrial fibrillation
0.88 (0.74-1.05)
0.15
0.99 (0.82-1.19
(0.82-1.19)
19
9)
Carotid stenosis
0.79 (0.51-1.24)
0.30
0.82 (0.52-1.30)
(0.52-1.30
3 )
30
Prosthetic heart valve
0.79 (0.31-2.03)
0.62
0.86 (0.34-2.19
(0.34-2.19)
9)
Smoker
0.97 (0.80-1.17)
0.76
0.83 (0.68-1.01)
Stroke/TIA
0.86 (0.72-1.02)
0.09
0.88 (0.74-1.04)
CAD/prior MI
0.83 (0.70-0.99)
0.04
0.87 (0.72-1.04)
PVD
D
1.16 (0.81-1.67)
(0.81-1.6
67)
0.41
1.32 (0.92-1.89)
Prior
Prio
io
or anticoagulants
anticoagul
ulan
ul
an
nts
00.76
.76 (0.55
(0.55-1.03)
55
5-1.0
03)
0.088
00.82
.82 (0
(0.60-1.12)
0.60
6 -1.12)
Prior
Pr
rio
ior antiplatelets
0.98
0.98 (0.84-1.15)
(0.84
4-1.1
15)
0.82
0.82
1.144 (0.97-1.35)
(0.97-1
1.35
35)
35
Prior
Pr
rio
ior antihypertensive
antihyperten
en
nsive
ve
00.82
.82 (0.70
(0.70-0.95)
0-0
0.9
95)
0.008
0.008
0.922 (0.76-1.10)
(0.76-1.10
10
0)
Prior
Prio
io
or cholesterol
cholestero
ol rreducer
edu
duccer
00.90
.90 (0.
(0.78-1.04)
.78
8-1.0
04))
0.16
0.98
8 ((0.82-1.17)
0.82-1.17
17))
17
Priorr antidiabetes
anti
an
t di
ti
d ab
abet
etees med
et
medication
dicaatiion
00.76
.76
7 ((0.61-0.95)
76
0.61-0
0.9
95))
00.02
.02
0
0.85 ((0.64-1.13)
0.64
64--1.13
64
13)
13
Region
0.003
Northeast
Ref
Ref
Midwest
0.99 (0.72-1.35)
1.03 (0.75-1.41)
South
0.94 (0.70-1.25)
0.89 (0.67-1.18)
West
1.44 (1.08-1.93)
1.38 (1.03-1.85)
Hospital location; rural
0.56 (0.32-0.98)
0.04
0.59 (0.33-1.03)
Hospital academic
1.03 (0.84-1.26)
0.78
1.01 (0.82-1.25)
Hospital size >300 beds
1.08 (0.88-1.32)
0.47
1.08 (0.85-1.37)
Hospital annual volume ischemic stroke,
1.02 (0.95-1.09)
0.64
0.91 (0.80-1.03)
per 100 cases
Hospital annual volume tPA treatment,
1.04 (1.01-1.06)
0.002
1.08 (1.04-1.12)
per 5 cases
Hospital primary stroke center
0.85 (0.70-1.04)
0.11
0.79 (0.64-0.96)
Adjusted variables: All variables included in the model. P-values computed with Wald tests.
21
P
0.03
<0.001
0.07
<0.001
<0.001
<0.001
0.44
0.17
0.64
00.91
.91
0.41
0.41
4
00.75
.75
75
0.06
0.14
0.13
0.13
0.21
0.21
21
0.11
11
0.355
0.82
2
0.27
7
0.004
0.06
0.91
0.52
0.12
<0.001
0.02
10.1161/CIRCULATIONAHA.116.023336
Table 3. Association of OTT with dichotomized efficacy and safety outcomes, comparing
patients treated within golden hour (0-60 minutes) to those treated later.
Outcomes
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Discharge to home
0-60 min
61-90 min
91-180 min
181-270 min
0-60 min (vs >60min)
Independent ambulation at discharge
0-60 min
61-90 min
91-180 min
181-270 min
0-60 min (vs >60min)
mRS 0-1 at discharge*
0-60 min
61-90 min
91-180 min
181-270 min
0-60 min (vs >60min)
mRS 0-2 at discharge*
0-60 min
61-90 min
91-180
91-18
91
180
18
0 min
min
188
181-270
min
miin
0-60
0
-60 min (vs >60min)
>60m
0min
n)
Symptomatic
Sy
ymptomatic intr
intracranial
trac
a ranniall hhemorrhage
emo
orr
rrhhage
0-60
0
-60 min
61-90
61
90 min
min
91-180 min
181-270 min
0-60 min (vs >60min)
Severe systemic hemorrhage
0-60 min
61-90 min
91-180 min
181-270 min
0-60 min (vs >60min)
In hospital mortality
0-60 min
61-90 min
91-180 min
181-270min
0-60 min (vs >60min)
Unadjusted OR
(95% CI)
1.06 (0.92-1.22)
0.96 (0.90-1.03)
0.84 (0.80-0.88)
Ref
1.21 (1.06-1.37)
1.08 (0.93-1.26)
1.01 (0.94-1.08)
0.89 (0.85-0.94)
Ref
1.17 (1.01-1.36)
1.43 (1.06-1.93)
1.03 (0.87-1.20)
0.86 (0.76-0.98)
Ref
1.57 (1.17-2.10)
1.31 (0.98-1.74)
1.04 (0.89-1.22)
0.86 (0.76-0.96)
(0.76-0.9
96))
Reff
11.44
.4
44 (1.09
(1.09-1.91)
9-1.9
91))
0.89
0.89 (0.63-1.27
(0.63
3-1
1.277 )
0.90
0 90 (0.77-1.04)
0.
(0.77
77-1.0
04))
1.04 (0.94-1.15)
Ref
0.88 (0.63-1.24)
0.69 (0.31-1.56)
0.63 (0.44-0.89)
0.95 (0.77-1.17)
Ref
0.76 (0.35-1.65)
1.21 (0.91-1.60)
1.04 (0.92-1.19)
1.14 (1.04-1.24)
Ref
1.10 (0.84-1.44)
P+
<0.001
0.40
0.23
<0.001
0.005
<0.001
0.30
0.86
<0.001
0.03
0.0004
0.02
0.76
0.02
0.003
<0.001
0.06
0.58
0.007
0.
0
0.01
011
0.12
0.
12
0.53
53
0.16
0.1
0.
16
16
0.47
0.46
0.04
0.38
0.009
0.603
0.48
0.03*
0.18
0.51
0.005
0.49
Adjusted OR
(95% CI)
1.46 (1.24-1.72)
1.44 (1.33-1.56)
1.17 (1.11-1.23)
Ref
1.25 (1.07-1.45)
1.50 (1.25-1.80)
1.49 (1.38-1.62)
1.24 (1.17-1.32)
Ref
1.22 (1.03-1.45)
2.11 (1.46-3.05)
1.58 (1.31-1.90)
1.21 (1.06-1.39)
(1.06-1.399)
Ref
1.72 (1.21-2.46
(1.21-2.46)
6)
1.97 (1.41-2.74)
1.68 (1.40-2.02)
1.22
1.2
22 (1.07-1.40)
Ref
ef
11.58
.5
58 ((1.15-2.16)
1..15-2.16
16))
16
00.78
.7
78 ((0.53-1.14)
0.53-1.144)
0.75
0.7
75 (0.64-0.87)
(0.
0.64
64-0
64
0.87
87))
0.87 (0.78-0.97)
Ref
0.89 (0.61-1.29)
0.60 (0.26-1.38)
0.50 (0.35-0.72)
0.77 (0.62-0.96)
Ref
0.77 (0.34-1.73)
0.98 (0.73-1.31)
0.78 (0.69-0.89)
0.87 (0.80-0.95)
Ref
1.12 (0.84-1.48)
P+
<0.001
<0.001
<0.001
<0.001
0.005
<0.001
<0.001
<0.001
<0.001
0.02
<0.001
<0.001
<0.001
0.006
0.00
0.
006
00
6
00.003
.00
003
00
3
<0.001
<0.001
<0.001
0.004
0.00
0.005
005
00
5
0.0031
31
0.200
0.003
0.00
03
0.0118
0.53
0.0022
0.23
0.000
0.02
0.53
0.001
0.89
0.0002
0.002
0.44
Variables for adjustment: age, gender, race, atrial fibrillation, prosthetic valve, CAD-Prior MI, carotid stenosis, diabetes, peripheral vascular disease,
hypertension, smoking, dyslipidemia, prior Stroke/TIA, NIHSS as continuous, arrival and admission information (EMS/on-off hours), hospital size,
hospital region, rural location, academic hospital, primary stroke center, annual volume of ischemic stroke admission, annual volume of tPA
administration, interaction term age x NIHSS, and meds prior to admission (antiplatelets, anticoagulants, antihypertension, cholesterol-reducers,
and meds for diabetes).
*
mRS 0-1 and 0-2 at discharge: limited to Oct. 2012-Sep. 2013 when mRS started to be collected.
+
P-values are computed using Wald tests. Global p-values for the overall association of OTT groups with the outcomes are given, as well as pvalues for the comparison of each category with the reference group 181-270 minutes or OTT golden hour vs longer times.
22
10.1161/CIRCULATIONAHA.116.023336
Figure Legends:
Figure 1. Ordinal Functional Outcomes in Different Onset to Treatment Windows. Adjusted
rates for ordinal outcomes for onset-to-treatment time windows for (A) functional status at
discharge, (B) discharge destination, and (C) modified Rankin Scale at discharge (covariates for
adjustment same as models in table 3).
Figure 2. Changes in Dichotomous Functional Outcomes with Continuous Variation in Onset
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
to Treatment Time. The adjusted predicted probabilities scales of the association between OTT
and 4 discharge outcomes; (A) discharge to home, (B) free of disability (mRS 0-1)
0-1)) at
at discharge,
disc
di
scha
sc
harg
ha
r e
(C)
scale).
C) independent ambulation at discharge, and (D) in-hospital mortality (note different y axis scale)
23
,
,
,
,
, ,
,,
,
,
,
,
,
,
+ #%##(, +,(,'' '(#, +,#!, +,*% &
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
,
,
,
,
,
,
,
,
, ,
,
,,
,,
,
,
+ $", +,)(, ,
,
,
,
))(, , +,($'% ,
,
,
,
,
,,
,,
,
,
,
,
,
,
! % % %%% %%%
%%$ % %# %%"!%
8IO
8H8O
8FFO
8*GO
8F8O
8 EFO
8 E8O
8DO
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
8#:FO
8:8-4.
/O
-4.
/
8O
>O
8
E8O
IO
8:8O015+L
:8O015+LMN2O
8O
:9 :>O :E8O :I
KO :9O
:IO ;KO ?O =>
=>O =E8O AO BO
,0$1&%4
,0$1&%4
&3(
&3(
4!4
!4 '')44"4
! %
%
% %%
% %!
!
%%
%%$
% $ %
%
%#
# %%"!%
%
%"
"!
!
!
=8
=8O
E8
E8O
H8
H8O
J8O :88O :=8O <E8O <H8O <J8O =88O ==8O =E8O =H8O =J8O
%,.4-&4-*-$%.44%4$%2-,4
%,4 -&4-*-$%4 4%4$%2-,4
4%4$%2-,
,/ $/&%4
&3*4#4 ''*4!4
!
! %
% %
%
%%
%
$%%
%%$
%
$
%
%
%#
# %
%%"!%
%"
8 H8O
88)=8O
8)=
=8O
=8
8FFO
8#:JO
8F8O
8":HO
H
8%EFO
8!CFO
8":EO
8#:=O
8:8O
O 8C8O
8&=FO
8:FO
8%E8O
8C8O
8'=8O
86O
8%E8O
8CFO
8@O
8$EFO
87O
! % % %%%%%
%%$ % %# %%"!%
O 88JO
8(=FO
88HO
88EO
8":FO
88=O
8":8O 02,3O
8O
=8O
E8O
H8O
J8O :88O :=8O <E8O <H8O <J8O =88O ==8O =E8O =H8O =J8O
%,-4-&4-+.$%.44%4$%2-,4
,/$/&%4
&3+4!4 ''*4!4
888O 0,5+LMN2O
8O
=8O
E8O
H8O
J8O :88O :=8O <E8O <H8O <J8O =88O ==8O =E8O =H8O =J8O
%,.4-&4-*-$%.44%4$%2-,4
,/ $/&%4
&3*44!4 ''*44!44
Treatment with TPA in the "Golden Hour" and the Shape of the 4.5 Hour Time-Benefit Curve
in the National US Get With The Guidelines-Stroke Population
Joon-Tae Kim, Gregg C. Fonarow, Eric E. Smith, Mathew J. Reeves, Digvijaya D. Navalkele, James
C. Grotta, Maria V. Grau-Sepulveda, Adrian F. Hernandez, Eric D. Peterson, Lee H. Schwamm and
Jeffrey L. Saver
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Circulation. published online November 4, 2016;
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2016 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circ.ahajournals.org/content/early/2016/11/04/CIRCULATIONAHA.116.023336
Data Supplement (unedited) at:
http://circ.ahajournals.org/content/suppl/2016/11/04/CIRCULATIONAHA.116.023336.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click Request
Permissions in the middle column of the Web page under Services. Further information about this process is
available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/
SUPPLEMENTAL MATERIAL
TITLE: Treatment with TPA in the “Golden Hour” and the Shape of the 4.5 Hour TimeBenefit Curve in the National US Get With The Guidelines-Stroke Population
Joon-Tae Kim, MD,1 Gregg C. Fonarow, MD,2 Eric E. Smith, MD,3 Mathew J. Reeves,
PhD,4 Digvijaya D. Navalkele, MD,5 James C. Grotta, MD,6 Maria V. Grau-Sepulveda, MD,7
Adrian F. Hernandez, MD,7 Eric D. Peterson, MD,7 Lee H. Schwamm, MD,8 Jeffrey L. Saver,
MD9
1
Department of Neurology, Chonnam National University Hospital, Gwangju, Korea
Department of Medicine, David Geffen School of Medicine, University of California, Los
Angeles, California, USA
3
Hotchkiss Brain Institute, University of Calgary, Calgary, Alberta, Canada
4
Department of Epidemiology and Biostatistics, Michigan State University, East Lansing,
Michigan, USA
5
Department of Neurology, University of Texas Health Science Center, Houston, Texas, USA
6
Clinical Innovation and Research Institute, Memorial Hermann Hospital, Houston, Texas,
USA
7
Outcome Research and Assessment Group, Duke Clinical Research Institute, Durham, North
Carolina, USA
8
Department of Neurology, Massachusetts General Hospital, Harvard Medical School,
Boston, Massachusetts, USA
9
Department of Neurology, David Geffen School of Medicine, University of California, Los
Angeles, California, USA
2
Supplemental Figure 1. Selection of the study population
Supplemental Figure 2. The unadjusted predicted probabilities scales of the association
between OTT and 4 discharge outcomes; (A) discharge to home, (B) free of disability (mRS
0-1) at discharge, (C) independent ambulation at discharge, and (D) in-hospital mortality.
Note the substantial differences in shape compared with the adjusted analysis shown in main
paper Figure 2 and in Supplemental Figure 3. These differences reflect the impact that
prognostic variables, especially initial stroke severity have upon both OTT and final outcome,
and the importance of adjusting for these effects
Supplemental Figure 3. The adjusted log-odds scales of the association between OTT and 4
discharge outcomes; (A) discharge to home, (B) free of disability (mRS 0-1) at discharge, (C)
independent ambulation at discharge, and (D) in-hospital mortality
Supplemental Table 1. Unadjusted analysis of outcomes in onset-to-treatment (OTT) groups
Total N
Discharge to home
Ambulation independent at discharge
Intracranial hemorrhage
Severe systemic hemorrhage
In-hospital mortality
mRS 0-1 at discharge
mRS 0-2 at discharge
*
Overall
60,890
25,793/60,697
*
(42.5)
20,418/48,630
(42.0)
2,697/59,785
(4.5)
582/59,785
(1.0)
4,539/60,890
(7.5)
2,555/8,964
(28.5)
3,404/8,964
(38.0)
0-60 min
828
395/826
(47.8)
311/661
(47.0)
30/806
(3.7)
6/806
(0.7)
66/828
(8.0)
71/179
(39.7)
86/179
(48.0)
61-90 min
6,120
2749/6,095
(45.1)
2,162/4,828
(44.8)
232/5,990
(3.9)
39/5,990
(0.7)
429/6,120
(7.0)
364/1,170
(31.1)
487/1,170
(41.6)
91-180 min
43,221
17,743/43,08
1 (41.2)
14,167/34,59
2 (41.0)
1,964/42,459
(4.6)
425/42,459
(1.0)
3,315/43,221
(7.7)
1,632/5,994
(27.2)
2,182/5,994
(36.4)
181-270 min
10,721
4,906/10,695
(45.9)
3,778/8,549
(44.2)
471/10,530
(4.5)
112/10,530
(1.1)
729/10,721
(6.8)
488/1,621
(30.1)
649/1,621
(40.0)
n cases/n total non-missing (%)
P-values are based on Pearson chi-square tests for all categorical row variables.
P1; comparison across all 4 time epochs, P2; comparison of 0-60min vs >60min
mRS 0-1 and 0-2 at discharge: mRS is limited to patients with non-missing mRS from Oct. 2012-Sep. 2013
+
P1+
P2
<0.001
0.002
<0.001
0.008
0.043
0.27
0.044
0.50
0.009
0.57
<0.001
0.0008
<0.001
0.005