0002-9149/14/$ - see front matter 2014 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.amjcard.2014.06.006
752
Table 1
Baseline characteristics in relation to observed cardiovascular mortality
Characteristic
Clinical and biochemistry
Age (years)
Women
Body mass index (kg/m2)
Systolic BP (mm Hg)
Diastolic BP (mm Hg)
LDL cholesterol (mg/dl)
LDL cholesterol (mmol/l)
HDL cholesterol (mg/dl)
HDL cholesterol (mmol/l)
eGFR (ml/min/1.73 m2)
12-Lead ECG
Baseline atrial brillation
T-wave inversion V4e6
Sum T-wave inversion (mm)
ST-depression V4e6
Sum ST-depression (mm)
Cornell product (mm ms)
SV1 RV5e6 (mm)
Q-wave precordial leads
Sum Q-wave precordial (mm)
Q-wave limb leads
Sum Q-wave limb (mm)
Heart rate (beats/min)
Echocardiography
LVIDD (cm)
LVIDS (cm)
LAS volume/BSA (ml/m2)
LAD volume/BSA (ml/m2)
Peak aortic jet velocity (m/s)
AVA/BSA (cm2/m2)
Mean aortic gradient (mm Hg)
Mitral regurgitation (grade 2)
LV stroke volume (ml)
LV mass index (g/m2)
LV ejection fraction (%)
Medicine
Simvastatin/ezetimibe
RAS inhibitor
Ca2-blocker
b-blocker
Diuretic
Aspirin
No CV Death (n 1,403)
CV Death (n 70)
p-Value
67.3 9.7
573 (39%)
27.0 4.4
145.1 20.1
82.1 10.4
137.3 34.1
3.6 0.9
58.4 16.8
1.5 0.4
68.2 12.1
67.0 9.6
546 (39%)
27.0 4.4
144.7 20.0
82.0 10.3
137.5 34.0
3.56 0.88
58.4 16.8
1.51 0.43
68.4 12.0
73.4 8.7
27 (39%)
26.7 4.5
152.8 20.8
84.2 12.1
133.4 35.6
3.45 0.92
58.8 16.8
1.52 0.45
64.2 14.8
<0.001
0.95
0.58
<0.001
0.13
0.33
0.80
0.03
50 (3%)
336 (24%)
0 (3.5 to 0)
241 (17%)
0 (0e0)
1600 (1160e2070)
26.6 9.5
351 (24%)
0 (0e0)
438 (30%)
0 (0e2)
65.1 11.2
45 (3%)
312 (23%)
0 (3.5 to 0)
224 (16%)
0 (0e0)
1575 (1160e2040)
26.5 9.4
330 (24%)
0 (0e0)
413 (29%)
0 (0e2)
64.9 11.0
5 (7%)
24 (36%)
1 (5.5 to 0)
17 (26%)
0 (1.5 to 0)
1710 (1320e2320)
27.5 10.4
21 (30%)
0 (0e2)
25 (36%)
0 (0e3)
70.3 12.7
0.09
0.01
0.01
0.04
0.04
0.01
0.40
0.23
0.18
0.26
0.09
<0.001
5.0 0.6
3.20 0.57
34.0 (26.2e42.8)
17.0 (11.6e23.8)
3.1 0.5
0.6 0.2
22.8 8.8
146 (11%)
75.8 17.4
99.5 30.0
65.8 8.3
5.03 0.62
3.19 0.56
33.9 (26.0e42.4)
16.8 (11.6e23.7)
3.08 0.54
0.61 0.19
22.7 8.8
135 (10%)
76.1 17.5
98.9 29.3
66.0 8.2
5.12 0.79
3.35 0.68
34.8 (29.5e54.7)
19.6 (13.1e31.3)
3.19 0.55
0.56 0.19
24.9 9.2
11 (17%)
70.9 15.0
111.1 40.2
63.2 8.9
0.40
0.06
0.04
0.02
0.12
0.04
0.05
0.08
0.02
0.02
0.01
733
600
418
738
674
386
(50%)
(41%)
(28%)
(50%)
(46%)
(26%)
701
570
394
704
630
365
(50%)
(41%)
(28%)
(50%)
(45%)
(26%)
32
30
24
34
44
22
(46%)
(43%)
(34%)
(49%)
(63%)
(31%)
0.49
0.71
0.26
0.79
0.003
0.32
BP blood pressure; GFR glomerular ltration rate; HDL high-density lipoprotein; LAS-D left atrial volume in systole and diastole; LDL low-density
lipoprotein; LV left ventricular; LVIDD left ventricular internal end-diastolic diameter; LVIDS left ventricular internal end-systolic diameter; NA not
applicable.
Values reect medians with twenty-fth to seventy-fth percentiles and the corresponding p-values results from the Wilcoxon test.
investigating whether intensive lipid lowering with simvastatin and ezetimibe combination versus placebo could
reduce the need for aortic valve replacement (AVR) and risk
of CV morbidity and mortality in 1,873 patients, aged 45 to
85 years, with asymptomatic mild-to-moderate AS (dened
as echocardiographic aortic valve thickening accompanied
by Doppler-measured aortic peak ow velocity 2.5 and
4.0 m/sec and normal systolic LV function). The primary
outcome including study design, organization, clinical
measurements, exclusion criteria, and baseline characteristics and the main outcome have been published previously.4,5 This study uses post hoc analysis of SEAS data to
753
Table 2
Multivariate predictors of ECG ndings independently associated with cardiovascular death
ECG Finding
Q-wave sum in limb leads (mm)
Stroke volume (per 10 mL)
High density lipoprotein (per mmol/ml)
Cornell product (per 100 mm ms)
Cornell product (100 mm ms)
Female gender (yes/no)
LV mass index (per g/m2)
SV1 RV5e6 voltage (per mm)
T-wave inversion V4e6 (yes/no)
LAS volume/BSA (per ml/m2)
Body mass index (per kg/m2)
Heart rate (min1)
Stroke volume (per ml)
Mean aortic gradient (per mm Hg)
AVA/BSA (per cm/m2)
Atrial brillation (yes/no)
Body mass index (kg/m2)
Female gender (yes/no)
LAD volume/BSA (ml/m2)
Regression Coefcients
0.130 (0.068e0.192)
0.412 (0.657 to 0.166)
0.017 (0.003e0.031)
416 mm ms
0.007
0.126
203 mm ms
0.030
0.121
0.298
0.517
29.8
10.7 min1
0.424
2.8 min1
0.089
(334e498)
(0.054e0.083)
(0.082e0.171)
(108e298)
(0.008e0.052)
(0.029e0.215)
(0.352 to 0.244)
(0.412e0.622)
(23.7e36.0)
(7.2e14.1)
(0.283e0.566)
(1.6e4.1)
(0.141 to 0.037)
t Value*
p-Value
4.15
3.29
2.44
<0.001
0.001
0.01
9.93
9.30
5.53
4.20
2.65
2.58
<0.001
<0.001
<0.001
<0.001
0.01
0.01
10.84
9.70
9.50
6.09
5.87
4.44
3.37
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
AVA aortic valve area; BSA body surface area; LAD left atrial volume in diastole; LAS left atrial volume in systole.
* Variables are sorted by descending t values, which is a numerical ratio of the distance from no association. The t value does not depend on the sample
parameters and therefore allows for comparison of relative signal strength.
Because discrete variables do not have a regression slope, estimated marginal means were used to depict the multivariate relations for gender, T-wave
inversion (yes/no), and atrial brillation (yes/no) with the respective ECG nding, that is, using the arithmetic means from the regression model including all
other variables listed under each ECG nding.
leads I, II, III, aVF, aVR, and aVL; Supplementary Table 1).
Q waves were not coded in the presence of left bundle branch
block (n 43), which resulted in these patients being
excluded from the present analyses. Missing data were due to
100% paced rhythm and lead switch and/or noise in an ECG
lead needed for coding. Because of suspected colinearity, an
alternative analysis was performed substituting ECG LV
hypertrophy with ECG repolarization abnormalities.
Echocardiographic study protocol, reading procedures,
and reproducibility have been published.6 In short, transthoracic echocardiograms were recorded at the local study
centers, after which videotapes were read blinded to the
randomization and study visit at the SEAS echocardiography
core laboratory. Aortic valve area was calculated applying
the continuity equation7,8 and averaged over 10 consecutive
beats in patients with atrial brillation. LV dimensions and
wall thicknesses were measured on 2-dimensional images
following the American Society of Echocardiography
guidelines, using an anatomically validated formula.9 Left
atrial volume was measured in LV end-systole and enddiastole by the modied Simpson monoplane method in the
apical 4-chamber view and indexed by body surface area.10
Mitral regurgitation was assessed by color Doppler using
previously described 4-point grading scale (grade 0 to 3);
grade 2 corresponding to a moderate-to-severe mitral
regurgitation.11
An end point committee blinded to randomized treatment
classied end points according to a prespecied end point
manual.4 The primary end point in this substudy was CV
death (dened as death from complications of myocardial
infarction, progressive heart failure, cerebrovascular disease,
complications of cardiac surgery or intervention, or other
754
established pretest probabilities for CV death in asymptomatic mild-to-moderate AS, we used annual risks of 1% to 3%
for CV death to estimate improvements in the net reclassication index; the groups correspond to that proposed by the
American College of Cardiology/American Heart Association guidelines for the management of chronic stable
angina.13 Differences in area under the receiver operating
characteristic curve between the models with and without
ECG data were compared by the method of DeLong.14 To
assess if the predictive value of ECG ndings of CV mortality
was different in operated versus nonoperated patients, a test of
interaction was performed with respect to time-varying AVR
status. To avoid overtting, a p >0.01 resulted in elimination
from the multivariate models. For all other hypothesis testing,
a 2-tailed p <0.05 was required for statistical signicance.
Figure 1. Combining prognostically relevant ECG ndings resulted in
optimal prediction of cardiovascular mortality during 4.3 years of followup. Combining the prognostic information contained in each ECG variable independently associated with cardiovascular death resulted in optimal
discrimination of observed cardiovascular mortality (area under the curve
0.69 with 95% CI 0.63 to 0.75). AUC area under the curve.
Figure 2. Including ECG ndings improved area under the curve for
prediction of cardiovascular mortality during 4.3 years of follow-up. Area
under the curve for prediction of cardiovascular mortality increased by 0.06
(95% CI 0.02 to 0.09, p <0.001) when adding ECG data (heart rate,
Q-wave amplitude, and Cornell voltage-duration product) to the non-ECG
variables (age and LV ejection fraction) that were independently associated with cardiovascular mortality. AUC area under the curve.
Results
Baseline electrocardiograms were available in 1,563 patients (83.4%), among which the ECG abnormalities under
study could be assessed in 94% (n 1,473). Compared with
patients without ECG data for this study (n 400), subjects
with ECG data did not differ in peak aortic jet velocity
(p 0.95) but were likely to be younger subjects (mean
difference 1.2 years) with slightly higher ejection fraction
(mean difference 2%, both p 0.04). The 1,473 patients with
ECG data for this study included 900 men (61%) and 573
women (39%), followed for a mean of 4.3 years (6,362
patient-years of follow-up). Among all screened ECG variables (Supplementary Figure 1) displaying univariate association with CV mortality, the ECG variables that were
independently associated with CV mortality, at the 99%
condence level, were heart rate; sum of Q-wave amplitudes
in leads I, II, III, aVL, aVR, and aVF; and the Cornell
voltage-duration product (Supplementary Table 1).
Baseline characteristics independently associated with
each of the 3 ECG multivariate predictors of CV death, at
the 99% condence level, are listed in Table 2. There was
little or no colinearity in pairwise comparisons of the 3 ECG
variables under study (all Pearson correlation coefcients
0.08). During the course of the study, 139 participants
(9%) died, including 70 CV deaths (5%) and 26 categorized
as sudden cardiac death. Unadjusted heart rate (HR 1.5 per
11.2 minute1 [1 SD], 95% CI 1.2 to 1.8), summed Q-wave
amplitude (HR 1.2 per 2.0 mm [1 SD], 95% CI 1.0 to 1.4),
and Cornell voltage-duration product (HR 1.4 per
763 mm ms [1 SD], 95% CI 1.2 to 1.7, all p 0.01) were
associated with clear excess in the risk of CV death
(Supplementary Table 1). Adjusted by each other in addition
to age and LV ejection fraction (non-ECG variables
remaining in backward elimination with a p >0.01 resulting
in removal from the model), heart rate (HR 1.5 per
11.2 minute1 [1 SD], 95% CI 1.2 to 1.8), summed Q-wave
amplitude (HR 1.3 per 2.0 mm [1 SD], 95% CI 1.1 to
1.6), and Cornell voltage-duration product (HR 1.4 per
763 mm ms [1 SD], 95% CI 1.2 to 1.7, all p <0.001)
remained signicantly associated with CV mortality. The
combined prognostic information contained in each of the 3
ECG variables improved prediction of CV mortality beyond
the individual ECG ndings (Figure 1). Adding ECG strain
to heart rate and summed Q-wave amplitude in lieu of the
No ECG
Data
ECG
Data
No ECG
vs. ECG
p-Value
42.9
0.73
0.079
0.046
NA
68.3
0.79
0.102
0.045
NA
25.4
0.06
0.024
0.001
0.0249
<0.001
<0.001
0.013
0.183
0.008
NA
NA
NA
NA
NA
NA
NA
NA
15
46
9
0.086
0.221
NA
NA
NA
NA
NA
NA
NA
NA
114
1095
194
0.057
0.143
<0.001
0.044
Reects the addition of ECG ndings to a basic model with age and
LV ejection fraction (non-ECG variables that remained as independent
predictors of cardiovascular mortality).
IDI integrated discrimination improvement; NA not applicable;
NRI net reclassication index.
Cornell product provided a nearly identical amount of predictive information (overall model chi-square 35.8 vs 37.9,
respectively). However, when adjusting by non-ECG variables that were also associated with CV mortality, ECG
strain was, unlike the Cornell product, removed by backward elimination. In a model with CV death as the outcome,
the addition of ECG data to other important baseline
covariates signicantly increased area under the curve
(Figure 2), improved integrated discrimination by 2.5%, and
the overall net reclassication index by 14.3% (all p 0.04;
Table 3). Of note, the net reclassication index improvement was largely due to reclassication of subjects with a
priori intermediate risk of CV death (Supplementary
Table 3). There was no detectable risk dependency between the ECG ndings and time-varying AVR on risk of
CV mortality (all p 0.09 for interaction). The latter may be
reective of the fact that the ECG variables under study
were not included in the clinical evaluation of the need for
AVR. This hypothesis was supported in 2 separate explorative analyses. First, when restricting the multivariate analyses to the 52 CV deaths before AVR (74%), heart rate (HR
1.5 per 11.2 minute1 [1 SD], 95% CI 1.2 to 1.9), summed
Q-wave amplitude (HR 1.4 per 2.0 mm [1 SD], 95% CI 1.2
to 1.7), and Cornell voltage-duration product (HR 1.5 per
763 mm ms [1 SD], 95% CI 1.2 to 1.8, all p 0.001)
remained highly associated with CV death. Second, in
univariate comparisons of the 18 CV deaths occurring after
AVR, the heart rate was the only ECG variable displaying
an association with risk of postoperative CV death (HR 1.6
per 11.2 minute1 [1 SD], 95% CI 1.1 to 2.4, p 0.01).
755
Discussion
In this study of asymptomatic patients with mild-tomoderate AS, ECG ndings added incremental information of CV mortality beyond established risk factors. This
argues that AS severity alone does not adequately estimate
risk of CV mortality in asymptomatic patients with AS.
Furthermore, the 3 ECG abnormalities under study did not
display signicant colinearity and related differently to CV
risk markers at baseline. This suggests that co-morbidities
are common in AS and may be part of explaining unexpectedly high event rates in some patients with perceived
low-risk asymptomatic mild-to-moderate AS.
Annual ECG examination is already a guideline recommendation in asymptomatic adolescents and young adults
with congenital AS.15 However, before this study, there was
insufcient evidence to determine whether this strategy
would be useful for predicting CV mortality in the rapidly
growing population of older adults with degenerative AS.
The ndings of the present study indicate that the role of
ECG ndings in adults with AS may be to identify a group
of patients in need of more elaborate testing to determine the
cause of the ECG abnormality, such as hypertension and/or
the presence of clinically silent coronary artery disease, and
the proper therapeutic measures (coronary revascularization,
antihypertensive treatment, AVR, and/or other available
therapy on an individual basis), as determined by conventional measures, including echocardiography and coronary
angiogram.
Other studies trying to establish risk models for asymptomatic AS have identied poor renal function, peak aortic
jet velocity, LV hypertrophy, and LV wall stress as independent predictors of poor prognosis.16e18 However, most
previous studies have used AVR as an end point when
testing their risk score. Though, the latter outcome may be
based on guideline recommendations, that is, echocardiographic degree of AS severity for prediction of AVR as a
surrogate for progression may provide partly tautologic
results.
Concordant with data from the general and other specic
populations, we observed the ability of ECG ndings to
reclassify subjects to lesser risk.19e21 Importantly, we also
noted that the largest reclassication into correct greater and
lesser risk groups by adding ECG data occurred in the patients with an a priori intermediate risk of CV death. Future
studies should therefore examine if altered clinical observation or treatment strategy, in response to ECG ndings,
could improve prognosis in adult patients with AS.
This work has several limitations. The event rates in the
SEAS study population might not be representative of AS in
the general population because of several exclusion criteria in
the present study. The SEAS study might have been underpowered to detect the predictive value of combined ECG
strain and ECG LV hypertrophy.22 Data on some important
prognostic variables, such as presence of coronary artery
disease, degree of aortic valve calcication, and diastolic
function, were not obtained as part of the SEAS study protocol. Furthermore, it was not possible to ascertain whether
therapy was guided versus not guided by the ECG ndings,
which limits the causal inference of prognostic ndings in
this study. The observed discrepancy between various
756
11.
Disclosures
Drs. Boman, Devereux, Gohlke-Baerwolf, Nienaber,
Rosseb, and Wachtell received honoraria from Merck & Co,
Inc. Whitehouse Station, New Jersey, the funding sponsor of
the SEAS study. Dr. Okin serves on a medical advisory board
for GE Medical Systems and serves as a consultant and receives grant support from Novartis. Dr. Devereux has served
as a consultant to Novartis and Sano. Dr. Wachtell is supported by the Novo Nordisk Fonden, Denmark.
12.
13.
Supplementary Data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.
amjcard.2014.06.006.
1. Greve AM, Boman K, Gohlke-Baerwolf C, Kesniemi YA, Nienaber
C, Ray S, Egstrup K, Rosseb AB, Devereux RB, Kber L,
Willenheimer R, Wachtell K. Clinical implications of electrocardiographic left ventricular strain and hypertrophy in asymptomatic patients
with aortic stenosis: the Simvastatin and Ezetimibe in Aortic Stenosis
study. Circulation 2012;125:346e353.
2. Siegel RJ, Roberts WC. Electrocardiographic observations in severe
aortic valve stenosis: correlative necropsy study to clinical, hemodynamic, and ECG variables demonstrating relation of 12-lead QRS
amplitude to peak systolic transaortic pressure gradient. Am Heart J
1982;103:210e221.
3. Greve AM, Gerdts E, Boman K, Gohlke-Baerwolf C, Rosseb AB,
Hammer-Hansen S, Kber L, Willenheimer R, Wachtell K. Differences
in cardiovascular risk prole between electrocardiographic hypertrophy
versus strain in asymptomatic patients with aortic stenosis (from SEAS
data). Am J Cardiol 2011;108:541e547.
4. Rosseb AB, Pedersen TR, Allen C, Boman K, Chambers J, Egstrup K,
Gerdts E, Gohlke-Brwolf C, Holme I, Kesniemi VA, Malbecq W,
Nienaber C, Ray S, Skjaerpe T, Wachtell K, Willenheimer R. Design
and baseline characteristics of the simvastatin and ezetimibe in aortic
stenosis (SEAS) study. Am J Cardiol 2007;99:970e973.
5. Rosseb AB, Pedersen TR, Boman K, Brudi P, Chambers JB, Egstrup
K, Gerdts E, Gohlke-Brwolf C, Holme I, Kesniemi YA, Malbecq W,
Nienaber CA, Ray S, Skjaerpe T, Wachtell K, Willenheimer R.
Intensive lipid lowering with simvastatin and ezetimibe in aortic
stenosis. N Engl J Med 2008;359:1343e1356.
6. Cramariuc D, Ciof G, Rieck AE, Devereux RB, Staal EM, Ray S,
Wachtell K, Gerdts E. Low-ow aortic stenosis in asymptomatic
patients: valvular-arterial impedance and systolic function from the
SEAS substudy. JACC Cardiovasc Imaging 2009;2:390e399.
7. Carabello BA. What is new in the 2006 ACC/AHA guidelines on
valvular heart disease? Curr Cardiol Rep 2008;10:85e90.
8. Baumgartner H, Hung J, Bermejo J, Chambers JB, Evangelista A,
Grifn BP, Iung B, Otto CM, Pellikka PA, Quiones M. Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for
clinical practice. Eur J Echocardiogr 2009;10:1e25.
9. Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I,
Reichek N. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy ndings. Am J Cardiol 1986;57:450e458.
10. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E,
Pellikka PA, Picard MH, Roman MJ, Seward J, Shanewise JS, Solomon SD, Spencer KT, Sutton MS, Stewart WJ. Recommendations for
chamber quantication: a report from the American Society of Echocardiographys Guidelines and Standards Committee and the Chamber
14.
15.
16.
17.
18.
19.
20.
21.
22.