CLINICAL RESEARCH STUDY

Secondary Prevention after Ischemic Stroke

or Transient Ischemic Attack
Sripal Bangalore, MD, MHA,a Lee Schwamm, MD,b Eric E. Smith, MD, MPH,c Inder M. Singh, MD, MS,d
Li Liang, PhD,e Gregg C. Fonarow, MD,f Deepak L. Bhatt, MD, MPH;g for the Get With the Guidelines-Stroke
Steering Committee and Investigators
a
New York University Medical Center, New York; bMassachusetts General Hospital, Boston; cHotchkiss Brain Institute, University of
Calgary, Alberta, Canada; dMercy Heart and Vascular Institute, Sacramento, Calif; eDuke Clinical Research Institute, Durham, NC;
f
Ahmanson-UCLA Cardiomyopathy Center, Los Angeles, Calif; gBrigham and Womens Hospital Heart & Vascular Center, and Harvard
Medical School, Boston, Mass.

ABSTRACT
BACKGROUND: Patients with stroke or transient ischemic attack are at increased risk of recurrent stroke.
Transient ischemic attack is a harbinger for stroke merely hours to days after the initial transient ischemic
attack. There is thus a narrow window of opportunity to initiate evidence-based therapies for secondary
prevention of stroke. Our objective was to assess hospital adherence at discharge to secondary prevention
measures after transient ischemic attack or ischemic stroke.
METHODS: Observational study of patients in the Get With The Guidelines-Stroke registry from 2007 to
2011. Patients were divided into 2 cohorts based on presentation: transient ischemic attack versus ischemic
stroke. Adherence to evidence-based secondary prevention and other quality measures were assessed.
RESULTS: Among the 858,835 patients with transient ischemic attack or ischemic stroke, 259,319 (30%)
patients presented with a transient ischemic attack and 599,516 (70%) patients presented with an ischemic
stroke. After adjusting for patient and hospital characteristics, adherence to secondary prevention measures
was consistently lower for the transient ischemic attack cohort (vs ischemic stroke cohort), who had lower
odds of being discharged on antithrombotics (odds ratio [OR] 0.63; 95% condence interval [CI],
0.59-0.66; P <.0001), anticoagulants for atrial brillation (OR 0.65; 95% CI, 0.61-0.68; P <.0001), lipidlowering medication for LDL >100 mg/dL (OR 0.52; 95% CI, 0.50-0.54; P <.0001), intensive statin
therapy (OR 0.74; 95% CI, 0.72-0.76; P <.0001), LDL cholesterol measurement (OR 0.66; 95% CI,
0.64-0.68; P <.0001), smoking cessation counseling (OR 0.83; 95% CI, 0.78-0.89; P <.0001), stroke
education (OR 0.71; 95% CI, 0.69-0.73; P <.0001), or weight loss recommendations (OR 0.88; 95% CI,
0.85-0.90; P <.0001). The adherence to evidence-based therapies increased signicantly (P <.0001) over
time (2007-2011) for both the cohorts, but the increasing trend was consistently lower for patients who
presented with transient ischemic attack.
CONCLUSIONS: In patients surviving an ischemic stroke or transient ischemic attack, adherence to evidencebased secondary prevention discharge measures were consistently less for patients with transient ischemic
attack, thus representing a missed opportunity at instituting preventive measures to reduce the risk of
recurrent stroke.
2014 Elsevier Inc. All rights reserved.
The American Journal of Medicine (2014) 127, 728-738
KEYWORDS: Secondary prevention; Stroke; Transient ischemic attacks

Funding: See last page of article.

Conicts of Interest: See last page of article.
Authorship: See last page of article.
Requests for reprints should be addressed to Sripal Bangalore, MD,
MHA, FACC, FAHA, FSCAI, Cardiovascular Outcomes Group, Cardiac
0002-9343/$ -see front matter 2014 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.amjmed.2014.03.011

Catheterization Laboratory, Cardiovascular Clinical Research Center, The

Leon H. Charney Division of Cardiology, New York University School of
Medicine, New York, NY 10016.
E-mail address: sripalbangalore@gmail.com

Bangalore et al

Secondary Prevention after Stroke/Transient Ischemic Attack

729

Transient ischemic attack is a harbinger for stroke, with a

collected using an Internet-based patient management tool
90-day stroke rate of w11% and with a recurrent transient
(Outcome; Quintiles, Durham, NC). A validation study
ischemic attack rate of w13%.1-3 In addition, approximately
showed excellent reliability of the data collected in GWTGone quarter of ischemic strokes are preceded by a transient
Stroke.13 In addition, hospital characteristics including acischemic attack, often merely hours or days before the
ademic/nonacademic status, annual stroke volume, and
ischemic stroke.2 There is thus a narrow time window for
geographic regions were obtained from the American Hospital Association database. The
prevention of ischemic stroke after
institutional review boards at the
a transient ischemic attack.2 In
CLINICAL SIGNIFICANCE
participating center determined
addition to the increased risk of
that the study was exempt and
ischemic stroke, patients with tran
Transient ischemic attack (TIA) is a
hence, no individual patient consient ischemic attack have an
harbinger for stroke, with an increased
sent was required.10
increased risk of major adverse
risk of stroke merely hours to days after
cardiovascular events (w22% at 1
the initial TIA, leaving a narrow window
year),4-7 with the risk of adverse
of opportunity to initiate evidencePatient Selection and
cardiovascular events similar to
based therapies for secondary prevenStudy Cohort
that of patients with documented
tion of stroke.
ischemic stroke.5
For the purpose of this analysis,
patients in the GWTG-Stroke
Randomized trials have shown

Among the 858,835 patients with TIA or
registry with a diagnosis of tranthat aggressive secondary prevenischemic stroke, we found that adhersient ischemic attack or ischemic
tion measures after an ischemic
ence to evidence-based secondary prestroke were included. Transient
stroke or a transient ischemic
vention discharge measures were
ischemic attack was dened as
attack can substantially reduce the
consistently less for patients with TIA,
symptoms that were present at the
risk of recurrent ischemic stroke.
thus representing a missed opportunity
time of hospital arrival, lasting
The American Heart Association/
at instituting preventive measures to
<24 hours, with no alternative
American Stroke Association
reduce the risk of recurrent stroke.
diagnosis. Patients with symptoms
(AHA/ASA) guidelines for the
lasting <24 hours but with radiosecondary prevention of stroke
graphic evidence of ischemic
emphasize aggressive secondary
injury, or those with symptoms lasting >24 hours, were
prevention measures after either a stroke or transient
classied as ischemic stroke.14 Patients were grouped into
ischemic attack, not only to reduce recurrent stroke, but also
8,9
to reduce related cardiovascular morbidity and mortality.
2 cohorts: transient ischemic attack and ischemic stroke. The
study period was from January 1, 2007 to October 3, 2011.
The narrow time window for preventive measures after a
Patients were excluded if: 1) they were enrolled at sites
transient ischemic attack dictates that aggressive secondary
with missing data on >25% of patients; 2) they were
prevention strategies must be employed at the time of
admitted to the hospital without at least one transient
discharge during index admission for transient ischemic
ischemic attack; 3) they were transfer-in and -outs; 4) they
attack. However, whether this is followed in clinical practice
had an in-hospital stroke; 5) their discharge destination was
is not well characterized. In this study, using data from the
missing or they left against medical advice, died in the
Get With The Guidelines-Stroke registry, we aimed to
hospital, or were discharged to hospice (Figure 1).
assess the adherence to secondary prevention measures at
hospital discharge after transient ischemic attack or ischemic
stroke; compare rates of adherence for patients with tranSecondary Prevention and Other Quality-of-care
sient ischemic attack and ischemic stroke; and evaluate
Measures
temporal trends during the study period from 2007 to 2011.
The outcome measure of interest was adherence to secondary prevention measures at hospital discharge. This
included performance measures and other quality meaMETHODS
sures. The discharge performance measures included: 1)
Get With The Guidelines-Stroke (GWTG-Stroke)
antithrombotic medications at discharge, including antiProgram
platelet agents (aspirin, aspirin/dipyridamole, ticlopidine,
clopidogrel) and anticoagulants (heparin and warfarin); 2)
The GWTG-Stroke program is a voluntary quality
anticoagulation (warfarin, heparin, or other anticoagulants)
improvement program across the US, which collects inforfor atrial brillation (paroxysmal or persistent); 3) lipidmation on stroke admission. The methods of the GWTGlowering agents for patients with low-density lipoprotein
Stroke program have been described previously.10-12
cholesterol (LDL) >100 mg/dL; 4) smoking cessation
Briey, for each hospitalization for stroke or transient
counseling provided at or before discharge for current
ischemic attack, information on patient demographics,
smokers; 5) stroke education; and 6) weight loss education
medical history, in-hospital diagnostic work-up, treatment,
for patients with a body mass index >25. The other quality
discharge medications, counseling, and disposition were

730

The American Journal of Medicine, Vol 127, No 8, August 2014

Figure 1 Patient selection. LAMA left against medical advice; TIA transient ischemic attack;
tPA tissue plasminogen activator.

measures included: 1) antihypertensive treatment with

angiotensin-converting enzyme inhibitor (ACEi) or
angiotensin receptor blockers (ARB), beta-blockers, calcium channel blockers, or diuretics; 2) LDL documentation;
3) intensive statin therapy; and 4) recommendations
regarding diabetes management (medications or diet) in
patients with diabetes.

Statistical Analysis
Multivariable logistic regression analyses were performed
to evaluate the adjusted relationship between the 2 patient
groups and the secondary prevention measures and outcomes. All models used a Generalized Estimating Equation
approach to account for within-hospital clustering and were
adjusted for patients baseline characteristics. In addition,
the hospital characteristics such as the geographic region,
number of beds in the hospital, annual number of stroke
discharge, annual intravenous tissue plasminogen activator
volume, and teaching status were included in the model.
A time-trend analysis was performed to evaluate the
change over time in the core discharge measures. An
interaction of time-by-patient group was included in the
model to assess whether the temporal trends were different

between ischemic stroke and transient ischemic attack patients after adjusting for baseline covariates described above.
All statistical analyses were performed using SAS
version 9.2 (SAS Institute, Cary, NC). All P values were
2-sided, with P <.05 considered statistically signicant.

RESULTS
Among the 1.4 million patients with stroke or transient
ischemic attack, 858,835 patients with ischemic stroke or
transient ischemic attack from 1545 sites fullled the inclusion criteria and were included in this analysis
(Figure 1). Among them, 259,319 (30%) patients presented
with a transient ischemic attack and 599,516 (70%) patients
presented with an ischemic stroke.

Baseline Characteristics
The baseline characteristics of the 2 groups are described
in Tables 1 and 2. Patients presenting with a transient
ischemic attack were more likely to be women, white, with
prior stroke/transient ischemic attack, dyslipidemia, more
likely to be on lipid-lowering therapy, were able to ambulate
independently at admission, and had lower National

Bangalore et al
Table 1

Secondary Prevention after Stroke/Transient Ischemic Attack

731

Baseline Characteristics of Patients with IS or TIA

Variable
Demographic
Age
Sex, male
Race
White
Black
Hispanic
Asian
Insurance
Self pay/no insurance
Medicare
Medicaid
Private/other insurance
Medical history
Diabetes mellitus
Hypertension
Dyslipidemia
Smoker
CAD/prior MI
Atrial brillation/utter
Prosthetic heart valve
Previous stroke/TIA
Carotid stenosis
PVD
Heart failure
Sickle cell
Current pregnancy
Medications prior to admission
Antihypertensive
Lipid lowering
Diabetic medications
Arrival & admission
Arrival
Private transport/other
EMS
Pre-notied (among EMS patients)
Ambulatory status on admission
Unable to ambulate
With assistance
Able to ambulate independently
Ambulatory status prior to current event
Unable to ambulate
With assistance
Able to ambulate independently
NIH Stroke Scale
NIH Stroke Scale (categorized)
>25
21-25
16-20
11-15
6-10
0-5
Arrived at off-hours*
Time variables
Time from symptom onset to arrival, minutes
Time from symptom onset to arrival < 120
minutes

Overall (n 858,835)

TIA (n 259,319)

IS (n 599,516)

P-value

69.98  14.42
47.15

69.94  14.63
42.90

70.00  14.63
48.98

.1399
<.0001

71.07
16.04
6.68
2.33

74.65
13.25
6.74
1.71

69.53
17.24
6.65
2.60

<.0001

4.81
45.12
6.82
34.98

3.36
44.55
6.21
37.55

5.44
45.37
7.09
33.87

<.0001

33.17
80.56
45.59
19.31
27.49
15.67
1.51
33.83
4.45
4.71
6.10
0.04
0.02

30.75
78.82
48.09
15.22
27.99
13.54
1.70
37.13
4.48
4.30
5.31
0.04
0.02

34.21
81.31
44.52
21.06
27.28
16.58
1.43
32.42
4.43
4.89
6.44
0.04
0.02

<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
.3095
<.0001
<.0001
.0678
.1076

66.97
42.15
24.07

68.15
45.84
22.91

66.47
40.56
24.58

<.0001
<.0001
<.0001

43.53
48.28
59.22

51.20
42.38
58.62

40.22
50.83
59.44

<.0001

9.83
12.78
26.62

3.56
9.79
36.66

12.55
14.07
22.27

<.0001

1.55
3.38
74.24
3 (1-7)

1.10
2.90
75.75
1 (0-3)

1.74
3.59
73.58
4 (2-8)

<.0001

0.67
1.16
2.21
3.69
8.09
34.73
49.00

0.09
0.15
0.33
0.96
3.20
35.88
49.07

0.92
1.59
3.02
4.86
10.20
34.23
48.96

151 (62-442)
43.42

115 (60-269)
51.31

187 (66-552)
39.49

<.0001

<.0001
<.0001

.9397
<.0001
<.0001

732
Table 1

The American Journal of Medicine, Vol 127, No 8, August 2014

Continued

Variable

Overall (n 858,835)

Vital signs
Heart rate
Systolic blood pressure
Diastolic blood pressure
BMI

79.68
156.92
82.86
28.14






16.81
29.74
18.03
6.78

TIA (n 259,319)
78.12
153.36
80.55
28.31






15.41
28.42
16.57
6.83

IS (n 599,516)
80.33
158.44
83.84
28.06






17.32
30.16
18.53
6.76

P-value
<.0001
<.0001
<.0001
<.0001

BMI body mass index; CAD coronary artery disease; EMS emergency medical service; IS ischemic stroke; MI myocardial infarction;
NIH National Institutes of Health; PVD peripheral vascular disease; TIA transient ischemic attack.
*Off-hours dened as holiday or before 7 AM or after 6 PM from Monday through Friday.

Institutes of Health stroke scale score (transient ischemic

attack patients with symptoms that had not yet resolved or
were still resolving at presentation had scores >0), but were
less likely to have atrial brillation, diabetes, hypertension,
or be smokers, when compared with patients presenting with
an ischemic stroke (Table 1). Patients with transient
ischemic attack had lower total cholesterol, LDL cholesterol, and higher high-density lipoprotein cholesterol when
compared with patients with an ischemic stroke.

In-hospital Treatment and Quality of Care

Patients presenting with transient ischemic attack had
shorter symptom onset to arrival time (115 minutes vs 187
minutes), with 51% of patients presenting within 2 hours of
symptoms (Table 2). In patients arriving within 3 hours of
symptom onset, transient ischemic attack patients were less
likely to receive computed tomography scan within 25
minutes of arrival when compared with the ischemic stroke
patients (Table 2). Transient ischemic attack patients were
more likely to be treated in smaller, nonacademic hospitals
and centers from the Northeast, more likely to be able to
ambulate independently or with assistance, and were more
likely to be discharged home than patients with an ischemic
stroke (Table 2).
Adherence at hospital discharge to secondary prevention
and other quality-of-care measures was lower in the transient ischemic attack cohort who were less likely to be
prescribed an antithrombotic (96.3% vs 97.7%), anticoagulation for atrial brillation (90.6% vs 93.7%), or a
statin for LDL >100 mg/dL (75.0% vs 82.8%), and less
likely to receive stroke education (73.5% vs 79.4%) and
weight loss recommendations (52.2% vs 55.1%) (Table 2).
In addition, transient ischemic attack patients were less
likely to be prescribed an ACEi or an ARB, beta-blocker, or
a calcium channel blocker or receive intensive statin therapy
at discharge when compared with the ischemic stroke patients (Table 2).

Quality of Care: Multivariable Analyses

In the multivariable model adjusted for patient/hospital
characteristics and accounting for the clustering of data
within hospitals, adherence to secondary prevention

measures was consistently lower for the transient ischemic

attack cohort who had lower odds of being discharged on
antithrombotics (odds ratio [OR] 0.63; 95% condence interval [CI], 0.59-0.66; P <.0001), anticoagulants for atrial
brillation (OR 0.65; 95% CI, 0.61-0.68; P <.0001), lipidlowering medication for LDL >100 mg/dL (OR 0.52; 95%
CI, 0.50-0.54; P <.0001), intensive statin therapy (OR 0.74;
95% CI, 0.72-0.76; P <.0001), LDL cholesterol measurement (OR 0.66; 95% CI, 0.64-0.68; P <.0001), smoking
cessation counseling (OR 0.83; 95% CI, 0.78-0.89;
P <.0001), stroke education (OR 0.71; 95% CI, 0.690.73; P <.001), or weight loss recommendations (OR 0.88;
95% CI, 0.85-0.90; P <.0001) (Table 3). Consequently,
there were lower adjusted odds of defect-free measure in the
transient ischemic attack cohort when compared with the
ischemic stroke cohort (OR 0.73; 95% CI, 0.71-0.76;
P <.0001).

Time Trend
Adherence to various discharge measures increased signicantly over time for both the transient ischemic attack and the
ischemic stroke cohort from 2007 to 2011 (Figures 2, 3), even
after adjusting for patient and hospital characteristics.
Moreover, there was a signicant interaction (calendar time
by patient cohort) inferring that the magnitude of increase in
adherence to core discharge measures was consistently lower
for the transient ischemic attack cohort when compared with
the ischemic stroke cohort for antithrombotic therapy at
discharge (adjusted OR per 1 year 1.17; [95% CI, 1.12-1.21]
vs 1.39 [95% CI, 1.33-1.45]; Pinteraction <.0001), anticoagulant for atrial brillation at discharge (adjusted OR per 1 year
1.11 [95% CI, 1.07-1.16] vs 1.23 [95% CI, 1.19-1.27];
Pinteraction <.0001), lipid-lowering therapy for LDL >100
mg/dL at discharge (adjusted OR per 1 year 1.30 [95% CI,
1.27-1.34] vs 1.44 [95% CI, 1.40-1.48]; Pinteraction <.0001),
and stroke education (adjusted OR per 1 year 1.61 [95% CI,
1.54-1.69] vs 1.71 [95% CI, 1.63-1.80]; Pinteraction
.0002). However, the trend for increasing adherence over
time was greater in magnitude for the transient ischemic
attack cohort for weight loss counseling (adjusted OR per 1
year 1.19 [95% CI, 1.14-1.25] vs 1.14 [95% CI, 1.09-1.19];
Pinteraction .001), door to computed tomography scan within
25 minutes in patients presenting within 3 hours of symptoms

Bangalore et al
Table 2

Secondary Prevention after Stroke/Transient Ischemic Attack

733

Hospital Characteristics, In-hospital Treatment, Secondary Prevention, and Other Quality-of-care Measures

Variable
Hospital characteristics
Number of beds*
Annual IS/TIA patients*
301
101-300
0-100
Annual IV tPA cases
>10
>6 &  10
6
Region
West
South
Midwest
Northeast
Hospital type
Academic
In-hospital treatments
Antithrombotic therapy administered by the end
of hospital day 2
Door to CT time within 25 minutes (in patients
arriving < 3 hours of symptom onset)
Secondary prevention measures
Antithrombotics
Anticoagulation for AF
Statin for LDL > 100
Smoking cessation
Stroke education (since 2008)
Reducing weight recommendation for BMI  25
Composite measure
Defect-free measure
Antihypertensive treatment
Antihypertensive treatment
ACEi
ARB
ACEi or ARB
Beta-blockers
Calcium channel blockers
Diuretic
Others
Other quality of care measures
LDL documented
Intensive statin therapy (since 2010)
Diabetes treatment (medications or ADA diet)
Outcomes
Length of stay
Length of stay 3 days
Discharge destination
Rehabilitation
Skilled nursing facility
Home
Ambulatory status (among nonmissing)
Able to ambulate independently or with
assistance
Able to ambulate independently

Overall (n 858,835)

TIA (n 259,319)

IS (n 599 516)

P-value

347 (240-503)

327 (227-470)

353 (248-522)

<.0001

42.62
46.82
10.56

40.32
47.76
11.92

43.62
46.41
9.98

<.0001

28.62
30.47
40.91

25.61
29.58
44.81

29.92
30.86
39.22

<.0001

16.92
36.55
18.24
28.29

15.50
34.06
16.78
33.66

17.54
37.62
18.87
25.97

<.0001

55.19

51.92

56.6

<.0001

93.05

94.32

92.49

<.0001

40.28

27.66

45.03

<.0001

97.26
92.96
80.46
95.83
77.01
54.21
82.28  21.31
49.89

96.29
90.58
75.05
95.31
73.47
52.25
79.58  22.44
44.32

97.69
93.73
82.78
95.99
79.42
55.12
83.46  20.69
52.31

<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001

80.11
38.18
9.64
46.55
41.57
21.13
20.58
10.53

80.02
35.34
11.05
45.18
40.34
20.07
21.39
10.00

80.14
39.40
9.03
47.14
42.10
21.58
20.23
10.75

.1977
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001

83.59
22.84
86.04

79.11
19.32
87.56

85.53
24.21
85.46

<.0001
<.0001
<.0001

3 (2-5)
53.94

2 (1-3)
77.48

4 (3-6)
43.75

<.0001
<.0001

17.74
15.52
66.74

2.39
6.62
90.99

24.38
19.37
56.26

<.0001

90.96

97.86

87.86

<.0001

64.66

85.22

55.42

<.0001

ACEi angiotensin-converting enzyme inhibitors; ADA American Diabetes Association; ARB angiotensin receptor blocker; BMI body mass index;
CT computed tomography; DVT deep vein thrombosis; IS ischemic stroke; IV intravenous; LDL low-density lipoproteins; TIA transient ischemic
attack; tPA tissue plasminogen activator.
* Since 2008.

734
Table 3

The American Journal of Medicine, Vol 127, No 8, August 2014

Secondary Prevention and Other Quality-of-care Measures: Multivariable Analyses (TIA vs IS)
Unadjusted

Secondary prevention measures

Antithrombotics at discharge
Anticoagulation for atrial brillation
Statin for LDL > 100 or ND
Smoking cessation counseling
Stroke education (since 2008)
Reducing weight recommendation for BMI  25
Defect-free measure
Other quality of care measures
Antihypertensive treatment
Antihypertensive treatment - ACEi
Antihypertensive treatment - ARB
Antihypertensive treatment - ACEi or ARB
Antihypertensive treatment - Beta-blocker
Antihypertensive treatment - calcium channel blockers
Antihypertensive treatment - Diuretic
Antihypertensive treatment - Other
Door to CT within 25 minutes (in patients arriving within
3 hours)
Door to CT within 25 minutes
Lipid lowering medications
Statin
LDL documented
Lipid-lowering medications for LDL > 100 or ND
Lipid-lowering medications for LDL > 100
Intensive statin therapy (since 2010)
Diabetes medications or ADA diet
Outcomes
Discharge home
Able to ambulate independently (among non-missing)
Able to ambulate independently or with assistance
(among non-missing)
LOS  3 days

Adjusted

OR

Lower
95% CI

Upper
95% CI

P-value

OR

Lower
95% CI

Upper
95% CI

P-value

0.65
0.67
0.66
0.86
0.72
0.87
0.69

0.62
0.64
0.65
0.81
0.70
0.85
0.68

0.69
0.70
0.68
0.91
0.74
0.90
0.71

<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001

0.63
0.65
0.53
0.83
0.71
0.88
0.73

0.59
0.61
0.52
0.78
0.69
0.85
0.71

0.66
0.68
0.55
0.89
0.73
0.90
0.76

<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001

0.89
0.84
1.18
0.90
0.88
0.88
1.08
0.91
0.48

0.86
0.82
1.15
0.89
0.87
0.86
1.06
0.88
0.46

0.91
0.85
1.21
0.92
0.90
0.89
1.10
0.93
0.50

<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001

0.96
0.85
1.14
0.91
0.87
0.88
1.09
0.92
0.54

0.90
0.84
1.11
0.89
0.85
0.87
1.07
0.90
0.52

1.03
0.86
1.16
0.92
0.88
0.89
1.10
0.94
0.56

.2297
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001

0.68
0.69
0.69
0.68
0.67
0.70
0.76
1.01

0.66
0.68
0.67
0.67
0.66
0.69
0.74
0.98

0.69
0.71
0.70
0.69
0.68
0.72
0.78
1.04

<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
.5946

0.73
0.50
0.51
0.66
0.46
0.52
0.74
0.97

0.71
0.48
0.49
0.64
0.44
0.50
0.72
0.92

0.75
0.51
0.53
0.68
0.48
0.54
0.76
1.01

<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
.1267

7.84
4.43
5.89

7.64
4.30
5.56

8.06
4.56
6.23

<.0001
<.0001
<.0001

8.23
4.66
5.23

8.03
4.53
5.01

8.44
4.80
5.46

<.0001
<.0001
<.0001

4.55

4.44

4.66

<.0001

4.48

4.38

4.58

<.0001

ACEi angiotensin-converting enzyme inhibitors; ADA American Diabetes Association; ARB angiotensin receptor blocker; BMI body mass index;
CI condence interval; CT computed tomography; DVT deep vein thrombosis; IS ischemic stroke; LDL low-density lipoprotein; LOS length of
hospital stay; ND not documented; OR odds ratio; Ref reference; TIA transient ischemic attack.

onset (adjusted OR per 1 year 1.21 [95% CI, 1.18-1.24]

vs 1.17 [95% CI, 1.14-1.19]; Pinteraction .004), and
length of stay 3 days (adjusted OR per 1 year 1.12;
[95% CI, 1.11-1.13] vs 1.07 [95% CI, 1.07-1.08];
Pinteraction <.0001).

DISCUSSION
In this analysis of close to a million admissions for
ischemic stroke or transient ischemic attack, adherence to
evidence-based secondary prevention and other qualityof-care discharge measures were consistently lower
(except for antihypertensives usage) for the transient
ischemic attack cohort when compared with ischemic
stroke cohort. In addition, although adherence to the secondary prevention and other quality-of-care measures
increased with time (from 2007-2011), the magnitude of

the increase was less for the transient ischemic attack

cohort for most measures.

Prevention of Recurrent Stroke

Stroke is one of the leading causes of disability and the third
leading cause of death in the US.15 The National Stroke
association estimates that 40% of stroke survivors will
experience moderate to severe impairments requiring special
care.15 In addition to the clinical consequences, the nancial
consequences of stroke are considerable, with a total cost of
about $54 billion/year in the US,16 and stroke is one of the
leading causes of hospital and related health care resource
utilization.17 A large focus of stroke remains on the acute
management of stroke. However, 1 in 4 strokes are recurrent
strokes. In addition, it is estimated that transient ischemic
attack is an important harbinger for stroke, with 25% of

Bangalore et al

Secondary Prevention after Stroke/Transient Ischemic Attack

735

Figure 2 (A) Adherence to antithrombotics at discharge 2007 to 2011. (B) Adherence to anticoagulation for atrial brillation
at discharge 2007 to 2011. AF atrial brillation. (C) Adherence to statins at discharge for low-density lipoprotein >100 mg/dL
2007 to 2011.

patients with transient ischemic attack having a stroke,15 and

many of them within hours to days following a transient
ischemic attack.2 There is thus a narrow time window for
instituting preventive measures including medication, stroke
education, and counseling in patients with transient
ischemic attack; and this is at the time of admission for the
index transient ischemic attack. In the Early use of eXisting
PREventive Strategies for Stroke (EXPRESS) study, a
strategy of urgent assessment and early initiation of existing
treatment in patients with transient ischemic attack reduced
the 90-day risk of recurrent stroke by about 80%.18 In
addition, urgent assessment and treatment of patients with
transient ischemic attack reduced subsequent hospital beddays, acute costs, and 6-month disability.19 The AHA/ASA
guideline for secondary prevention of stroke therefore places
special emphasis on secondary prevention to prevent recurrent strokes and cardiovascular events.8 These include (but
are not limited to) blood pressure reduction (Class I) even in
patients without a documented history of hypertension (Class
IIa); glycemic control in patients with diabetes (Class I);
statin therapy with intensive lipid lowering in patients who
have evidence of atherosclerosis, an LDL level 100 mg/dL
(Class I); smoking cessation (Class I); antiplatelet therapy
with aspirin (Class I); and anticoagulation in patients with
paroxysmal or permanent atrial brillation (Class I).8,9 These
measures have been shown to reduce the risk of recurrent
stroke and cardiovascular events. However, the performance
measures endorsed by the National Quality Forum is focused

on patients with ischemic stroke and has largely ignored

patients with transient ischemic attack; a critical gap that
needs to be urgently addressed.

A Missed Opportunity
The results of the present study suggest that in patients
presenting with a clinically less dramatic event such as those
with a transient ischemic attack, the hospital adherence to
secondary prevention measures was consistently inferior,
with lower odds of being discharged on antithrombotics,
anticoagulants for atrial brillation, statins, or to receive
dysphagia screening, LDL cholesterol measurement, smoking
cessation counseling, stroke education, or weight loss recommendations. This represents a missed opportunity at
secondary prevention, as the window for prevention of a
recurrent stroke or transient ischemic attack is narrow for
these patients. Since the inception of the GWTG-Stroke
program, there have been considerable improvements in
adherence to these core discharge measures. In the present
study, adherence to antithrombotics for patients with transient
ischemic attack at discharge increased from 94.7% in 2007 to
97.1% in 2011. Similarly, the adherence to anticoagulation for
atrial brillation at discharge increased from 88.1% to 90.8%
and for statins from 62.6% to 85.7%. Although these adherence rates for the transient ischemic attack cohort have
improved in recent years, they were consistently lower
compared with the ischemic stroke cohort. There is, thus,

736

The American Journal of Medicine, Vol 127, No 8, August 2014

Figure 3 (A) Adherence to smoking cessation counselling at discharge 2007 to 2011. (B) Adherence to stroke education at discharge
2007 to 2011. (C) Adherence to weight management recommendations at discharge 2007 to 2011. (D) Trends in defect-free care 2007
to 2011.

room for improvement, and efforts must be directed to

close this gap in adherence. Although the adherence to
discharge measures was consistently lower for most measures for a transient ischemic attack patient, they were no
different for antihypertensive medication usage when
compared with ischemic stroke patients. In addition, there
is controversy about the optimal antihypertensive agent for
stroke prevention with the AHA/ASA guidelines advocating diuretics or a combination of diuretics with an
ACEi. In our study, only 20% were on diuretics and 46%
on an ACEi or an ARB. Interestingly, the second most
common antihypertensive agents used (after ACEi/ARB)
were beta-blockers, where the data are rather weak for
stroke prevention,20-22 and beta-blockers have been
relegated to fourth-line agents by major national and
international hypertension guidelines.23,24

STUDY LIMITATIONS
Our data are from a prospective registry of patients from a
voluntary quality-reporting program; therefore, adherence to
guideline-recommended secondary prevention therapies
may be higher in these hospitals than in hospitals not
participating in GWTG-Stroke. If so, it is possible that the
difference in quality of care between transient ischemic
attack and ischemic stroke might be even greater in
nonparticipating hospitals. However, this is still the largest
series reporting secondary prevention after an ischemic stroke

or transient ischemic attack, with data from close to a million

patients from more than 1500 acute care hospitals. Our data
assessed secondary prevention measures at the time of hospital discharge and do not provide data on adherence/
compliance with these recommendations post discharge.
Reecting a focus of GWTG-Stroke on acute thrombolytic
therapy in the emergency department, only transient ischemic
attack patients with ongoing symptoms at the time of activation of emergency medical services or admission to the
emergency department were included in the registry; therefore, patients presenting with past resolved transient ischemic
attacks are not included in our analysis. However, we have no
reason to think that quality of care would be better in patients
presenting with past resolved transient ischemic attacks, and
we suspect that, if different, it would be worse rather than
better. Moreover, disparity in the use of evidence-based
therapies could merely be a reection of physician assessment of increased risk of adverse effect for a particular patient
subgroup or a reection of patient choice, which is not
captured in the database. Moreover, although the relative
difference between the patients cohorts for secondary prevention and quality-of-care measures were large, in some
cases the absolute differences were rather small.

CONCLUSIONS
Data from over close to a million patients with ischemic
stroke or transient ischemic attack suggest that the hospital

Bangalore et al

Secondary Prevention after Stroke/Transient Ischemic Attack

adherence to evidence-based secondary prevention and

other quality-of-care measures at the time of discharge is
consistently lower for patients with transient ischemic attack
when compared with patients with ischemic stroke. The
adherence to these discharge measures has increased in
GWTG-Stroke program from 2007 to 2011 but is still
consistently lower for the transient ischemic attack cohort.
Efforts must be made to close these treatment gaps and
make use of the narrow treatment window available for such
patients to prevent recurrent cerebrovascular and cardiovascular events.

References
1. Johnston SC, Gress DR, Browner WS, Sidney S. Short-term prognosis
after emergency department diagnosis of TIA. JAMA. 2000;284(22):
2901-2906.
2. Rothwell PM, Warlow CP. Timing of TIAs preceding stroke: time
window for prevention is very short. Neurology. 2005;64(5):817-820.
3. Bhatt DL, Eagle KA, Ohman EM, et al. Comparative determinants of
4-year cardiovascular event rates in stable outpatients at risk of or with
atherothrombosis. JAMA. 2010;304(12):1350-1357.
4. Alberts MJ, Bhatt DL, Mas JL, et al. Three-year follow-up and event
rates in the international REduction of Atherothrombosis for Continued
Health Registry. Eur Heart J. 2009;30(19):2318-2326.
5. Venketasubramanian N, Rother J, Bhatt DL, et al. Two-year vascular
event rates in patients with symptomatic cerebrovascular disease: the
REACH registry. Cerebrovasc Dis. 2011;32(3):254-260.
6. Hill MD, Yiannakoulias N, Jeerakathil T, et al. The high risk of stroke
immediately after transient ischemic attack: a population-based study.
Neurology. 2004;62(11):2015-2020.
7. Atanassova PA, Chalakova NT, Dimitrov BD. Major vascular events
after transient ischaemic attack and minor ischaemic stroke: post hoc
modelling of incidence dynamics. Cerebrovasc Dis. 2008;25(3):
225-233.
8. Furie KL, Kasner SE, Adams RJ, et al. Guidelines for the prevention of
stroke in patients with stroke or transient ischemic attack: a guideline
for healthcare professionals from the american heart association/
american stroke association. Stroke. 2011;42(1):227-276.
9. Jauch EC, Saver JL, Adams HP Jr, et al. Guidelines 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(3):870-947.
10. Schwamm LH, Fonarow GC, Reeves MJ, et al. 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(1):107-115.
11. Bhatt DL, Steg PG, Ohman EM, et al. International prevalence,
recognition, and treatment of cardiovascular risk factors in outpatients
with atherothrombosis. JAMA. 2006;295(2):180-189.
12. Steg PG, Bhatt DL, Wilson PW, et al. One-year cardiovascular event
rates in outpatients with atherothrombosis. JAMA. 2007;297(11):
1197-1206.
13. Xian Y, Fonarow GC, Reeves MJ, et al. Data quality in the American
Heart Association Get With The Guidelines-Stroke (GWTG-Stroke):
results from a national data validation audit. Am Heart J. 2012;163(3):
392-398, 8.e1.
14. Easton JD, Saver JL, Albers GW, et al. Denition and evaluation of
transient ischemic attack: a scientic statement for healthcare professionals from the American Heart Association/American Stroke Association Stroke Council; Council on Cardiovascular Surgery and
Anesthesia; Council on Cardiovascular Radiology and Intervention;
Council on Cardiovascular Nursing; and the Interdisciplinary Council
on Peripheral Vascular Disease. The American Academy of Neurology
afrms the value of this statement as an educational tool for neurologists. Stroke. 2009;40(6):2276-2293.

737

15. Go AS, Mozaffarian D, Roger VL, et al. Heart disease and stroke
statisticse2013 update: a report from the American Heart Association.
Circulation. 2013;127(1):e6-e245.
16. Heidenreich PA, Trogdon JG, Khavjou OA, et al. Forecasting the
future of cardiovascular disease in the United States: a policy statement
from the American Heart Association. Circulation. 2011;123(8):
933-944.
17. Wolfe CD. The impact of stroke. Br Med Bull. 2000;56(2):275-286.
18. Rothwell PM, Giles MF, Chandratheva A, et al. Effect of urgent
treatment of transient ischaemic attack and minor stroke on early
recurrent stroke (EXPRESS study): a prospective population-based
sequential comparison. Lancet. 2007;370(9596):1432-1442.
19. Luengo-Fernandez R, Gray AM, Rothwell PM. Effect of urgent
treatment for transient ischaemic attack and minor stroke on disability
and hospital costs (EXPRESS study): a prospective population-based
sequential comparison. Lancet Neurol. 2009;8(3):235-243.
20. Bangalore S, Parkar S, Grossman E, Messerli FH. A meta-analysis of
94,492 patients with hypertension treated with beta blockers to determine the risk of new-onset diabetes mellitus. Am J Cardiol.
2007;100(8):1254-1262.
21. Bangalore S, Messerli FH. Beta-blockers as fourth-line therapy for
hypertension: stay the course. Int J Clin Pract. 2008;62(11):
1643-1646.
22. Bangalore S, Steg G, Deedwania P, et al. beta-Blocker use and clinical
outcomes in stable outpatients with and without coronary artery disease. JAMA. 2012;308(13):1340-1349.
23. National Clinical Guideline Centre. The clinical management of primary hypertension in adults. Available at: http://www.nice.org.uk/
guidance/index.jsp?actiondownload&o53228. Accessed December
3, 2011.
24. Mancia G, De Backer G, Dominiczak A, et al. 2007 Guidelines for the
management of arterial hypertension: the Task Force for the Management of Arterial Hypertension of the European Society of Hypertension
(ESH) and of the European Society of Cardiology (ESC). Eur Heart J.
2007;28(12):1462-1536.

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. GWTG-Stroke has been funded in the
past through support from Boehringer Ingelheim, Merck, Bristol-Myers
Squib/Sano Pharmaceutical Partnership, and the AHA Pharmaceutical
Roundtable. The industry sponsors of GWTG-Stroke had no role in the
design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the
manuscript.
Conicts of Interest: SB, EES, IMS, and LL report no disclosures. LS
reports the following: Serves as the unpaid chair of the AHA GWTG
Stroke Clinical Working Group; serves as a Stroke Systems of Care Expert
Consultant and measure development expert to the Joint Commission,
Coverdell Registry/MA Dept of Public Health; is on the steering committee for the Medtronic Victory AF trial, and the DSMB for the NovoNordisk DEVOTE trial. He receives research funding from NINDS and
PCORI. GCF reports the following: EmploymentUCLA Employee,
which holds a patent on stroke retriever devices National Institutes of
HealthGrants and grants pending. DLB discloses the following relationships: Advisory Board, Elsevier; Practice Update Cardiology,
Medscape Cardiology, Regado Biosciences; Board of Directors: Boston
VA Research Institute, Society of Cardiovascular Patient Care; Chair:
American Heart Association Get With The Guidelines Steering Committee; Data Monitoring Committees: Duke Clinical Research Institute;
Harvard Clinical Research Institute; Mayo Clinic; Population Health
Research Institute; Honoraria: American College of Cardiology (Editor,
Clinical Trials, Cardiosource), Belvoir Publications (Editor in Chief,
Harvard Heart Letter), Duke Clinical Research Institute (clinical trial
steering committees), Harvard Clinical Research Institute (clinical trial
steering committee), HMP Communications (Editor in Chief, Journal of

738
Invasive Cardiology); Population Health Research Institute (clinical trial
steering committee), Slack Publications (Chief Medical Editor, Cardiology
Todays Intervention), WebMD (CME steering committees); Other:
Clinical Cardiology (Associate Editor); Journal of the American College
of Cardiology (Section Editor, Pharmacology); Research Grants: Amarin,

The American Journal of Medicine, Vol 127, No 8, August 2014

AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Medtronic, Roche,
Sano Aventis, The Medicines Company; Unfunded Research: FlowCo,
PLx Pharma, Takeda.
Authorship: All authors had access to the data and a role in writing the
manuscript.