Download Electrocardiography and Doppler echocardiography for risk

Journal of the American College of Cardiology
© 2005 by the American College of Cardiology Foundation
Published by Elsevier Inc.
Vol. 45, No. 7, 2005
ISSN 0735-1097/05/$30.00
doi:10.1016/j.jacc.2004.12.064
Electrocardiography and Doppler
Echocardiography for Risk Stratification
in Patients With Chronic Heart Failure
Incremental Prognostic Value of QRS
Duration and a Restrictive Mitral Filling Pattern
Christian Bruch, MD,* Michael Gotzmann,* Jörg Stypmann, MD,* Frauke Wenzelburger, MD,†
Markus Rothenburger, MD,† Matthias Grude, MD,* Hans H. Scheld, MD, FESC, FETCS,†
Lars Eckardt, MD,‡ Günter Breithardt, MD, FESC, FACC,* Thomas Wichter, MD, FESC*
Münster, Germany
This prospective study tested whether Doppler echocardiographic variables add incremental
value to QRS duration in determining the prognosis of patients with chronic heart failure
(CHF) and systolic dysfunction.
BACKGROUND Diastolic dysfunction frequently is observed in patients with CHF, but its prognostic impact
relative to that of QRS duration is unknown.
METHODS
A total of 193 patients with CHF and an ejection fraction ⬍45% were enrolled prospectively.
Echo measurements included left ventricular dimensions/volumes, ejection fraction, mitral
early/late diastolic velocity ratio, deceleration time, and tissue Doppler mitral annular
velocities. The mitral filling pattern was classified as either restrictive (RFP) or nonrestrictive.
A cardiac event (cardiac death or urgent cardiac transplantation) was defined as combined
study end point.
RESULTS
During a follow-up of 385 ⫾ 270 days, 24 patients suffered an event (cardiac death, n ⫽ 21;
urgent transplantation, n ⫽ 3). The RFP, QRS duration, left ventricular systolic diameter,
and mitral annular early diastolic velocity were independent predictors of an event. In patients
with QRS duration ⬎144 ms, the outcome was markedly poorer in the presence of RFPs as
compared with their absence. Similarly, despite a QRS duration ⱕ144 ms, the outcome was
worse in the presence of a RFP. A risk-stratification model based on the three strongest
independent predictors separated groups into those with good prognosis and those with high,
intermediate, and low event-free survival rates.
CONCLUSIONS In subjects with CHF and systolic dysfunction, transmitral flow patterns add incremental
value to QRS duration in determining the prognosis. (J Am Coll Cardiol 2005;45:1072–5)
© 2005 by the American College of Cardiology Foundation
OBJECTIVES
Despite advances in the medical management of patients
with chronic heart failure (CHF), morbidity and mortality
remain high (1). The prolongation of the QRS interval has
been identified as an independent predictor of adverse
outcome (2). However, in patients with CHF, echoDoppler indices of diastolic filling also provide prognostic
information, and a restrictive filling pattern (RFP) is indicative of a poor prognosis (3). This prospective study tested
whether Doppler echocardiographic variables add incremental value to QRS duration in determining the prognosis
of patients with CHF and systolic dysfunction.
METHODS
Study patients. Two hundred seventy-one patients were
recruited consecutively from our outpatient CHF clinic,
which is jointly run by the Departments of Cardiology/
Angiology and Cardiothoracic Surgery in our institution.
From the Departments of *Cardiology and Angiology, †Thoracic and Cardiovascular Surgery, and ‡Electrophysiology, University Hospital of Münster, Münster,
Germany
Manuscript received October 30, 2004; revised manuscript received December 8,
2004, accepted December 21, 2004.
Inclusion criteria were a history of CHF according to
Framingham criteria (4), left ventricular ejection fraction
(LVEF) ⬍45%, and clinical stability after at least two
months on standard medical therapy. Patients with congenital heart disease (n ⫽ 4), malignancy (n ⫽ 2), severe
valvular disease (n ⫽ 6), atrial fibrillation (n ⫽ 28), or under
permanent pacemaker stimulation (n ⫽ 27) were excluded.
Follow-up information was obtained during routine ambulatory visits but also by telephone contact with patients or
their physicians. All patients gave their informed consent
before study entry.
Electrocardiography (ECG) analysis. QRS duration was
measured in a 12-lead ECG using leads V3 to V6 (interobserver and intraobserver correlations for QRS duration: 0.97
and 0.98, respectively).
Echocardiography. All patients underwent a standard
echo examination, including the assessment of transmitral
peak early and late diastolic velocities (E, A) and deceleration time (DT). A RFP was defined by an E/A ratio ⬎2, a
DT ⬍150 ms, and a mitral annular E= velocity ⬍8 cm/s (5).
Tissue Doppler imaging (TDI)-derived peak systolic (S=),
early (E=), and late (A=) diastolic velocities were derived
Bruch et al.
ECG and Doppler Echo for Risk Stratification in Heart Failure
JACC Vol. 45, No. 7, 2005
April 5, 2005:1072–5
a prognostic index to classify patients into different risk
groups. A p value of ⬍0.05 was considered significant.
Abbreviations and Acronyms
A
⫽ peak late diastolic mitral filling velocity
A=
⫽ peak late diastolic mitral annular velocity
CHF ⫽ chronic heart failure
DT ⫽ deceleration time
E
⫽ peak early diastolic mitral filling velocity
E=
⫽ peak early diastolic mitral annular velocity
ECG ⫽ electrocardiography
LVEF ⫽ left ventricular ejection fraction
RFP ⫽ restrictive filling pattern
S=
⫽ peak systolic mitral annular velocity
TDI ⫽ tissue Doppler imaging
RESULTS
During a follow-up of 385 ⫾ 270 days, 24 patients suffered
an event (cardiac death, n ⫽ 21; urgent cardiac transplantation, n ⫽ 3) and thus reached the study end point. Eleven
patients were censored (elective cardiac transplantation, n ⫽
9; death from noncardiac cause, n ⫽ 2).
Patients with or without event did not differ significantly
with respect to the etiology of CHF, but QRS duration was
significantly longer in patients with an event (Table 1). In such
patients, LVEF and DT were reduced, and RFP was more
frequent (Table 2). The independent predictors of an event
identified by the multivariate Cox analysis are listed in Table 3.
Survival analysis. In patients with a QRS duration ⬎144
ms, the event-free survival was significantly lower than in
patients with a QRS duration ⱕ144 ms (event-free survival
rate of 69% vs. 90%, p ⫽ 0.0021) (Fig. 1). In patients with a
QRS duration ⬎144 ms, the presence of a RFP indicated a
poor prognosis (event-free survival rate 51% vs. 79% in those
with a non-RFP, p ⫽ 0.023) (Fig. 2A). Likewise, in patients
with a QRS duration ⱕ144 ms, a RFP indicated a lessfavorable outcome than that in patients without RFP (eventfree survival rate rate 59% vs. 97%, p ⬍ 0.0001) (Fig. 2B).
Construction of a noninvasive risk score. The noninvasive predictive model was based on the three strongest
independent predictors, i.e., presence of a RFP, QRS
duration ⬎144 ms, and left ventricular systolic diameter
index ⬎2.75 cm/m2. Very low-, low-, intermediate-, and
high-risk groups were identified by the absence of any risk
factor or the presence of one, two, or three risk factors with
an event-free survival of 100%, 91%, 64%, and 41%,
respectively (Fig. 3).
from the septal and lateral mitral annulus and averaged for
each patient (6). Interobserver and intraobserver correlations for conventional echo measurements and TDI variables reached 0.94 and 0.98, respectively.
Outcome measurements and statistical analysis. Death
from a cardiac cause or urgent cardiac transplantation was
considered as the combined study end point. Numerical
values are expressed as mean ⫾ SD. Continuous variables
were compared between groups using an unpaired t test (for
normally distributed variables) or Mann-Whitney U test
(for non-normally distributed variables). The chi-square test
was used to compare categoric variables. Clinical, ECG, and
echo variables were evaluated for the combined study end
point in a univariate Cox proportional hazard model. All
variables with a significant association were entered in a
multivariate Cox model to identify independent predictors
of outcome. Receiver operating characteristic curves were
generated to define cut-off values for independent predictors. Event-free survival was analyzed by the Kaplan-Meier
method, and survival curves were compared by the log-rank
test. Independent predictors identified by the multivariate
Cox proportional hazard survival model were used to derive
Table 1. Clinical Characteristics of Study Patients
Age (yrs)
Male/female (%)
BSA (m2)
ILVD/non-ILVD (%)
NYHA class
DM, n (%)
ICD, n (%)
QRS duration (ms)
LBBB (%)
Medication (%)
ACE-I or ARB
Diuretics
Digoxin
Beta-blockers
Nitrates
1073
Total
(n ⴝ 193)
Patients With Event
(n ⴝ 24)
Patients Without Event
(n ⴝ 169)
p Value
58 ⫾ 11
76/24
1.9 ⫾ 0.2
63/37
2.6 ⫾ 0.5
28 (15)
81 (42)
137 ⫾ 37
66 (34)
64 ⫾ 11
87/13
1.9 ⫾ 0.2
71/29
2.9 ⫾ 0.4
9 (38)
6 (25)
159 ⫾ 38
12 (50)
57 ⫾ 11
75/25
1.9 ⫾ 0.2
63/37
2.6 ⫾ 0.5
19 (11)
75 (46)
133 ⫾ 36
54 (32)
0.01
0.164
0.427
0.439
0.027
⬍0.001
0.111
0.002
⬍0.001
96
87
61
87
27
100
96
67
71
29
95
86
60
89
26
0.264
0.169
0.51
0.016
0.777
Mann-Whitney U test was used for comparison of continuous variables owing to their non-normal distribution.
ACE-I ⫽ angiotensin-converting enzyme inhibitor; ARB ⫽ angiotensin receptor blocker; BSA ⫽ body surface area; DM ⫽
diabetes mellitus; ICD ⫽ implantable cardioverter-defibrillator; ILVD ⫽ ischemic left ventricular dysfunction; LBBB ⫽ left
bundle branch block; NYHA ⫽ New York Heart Association.
1074
Bruch et al.
ECG and Doppler Echo for Risk Stratification in Heart Failure
JACC Vol. 45, No. 7, 2005
April 5, 2005:1072–5
Table 2. Echocardiographic Characteristics of Study Patients
LAD (cm)
LVDDI (cm/m2)
LVSDI (cm/m2)
LVDVI (ml/m2)
LVSVI (ml/m2)
LVEF (%)
FS (%)
PW Doppler
Mitral E/A ratio
DT
RFP, n (%)
Tissue Doppler
S= (cm/s)
E= (cm/s)
A= (cm/s)
E/E= ratio
Total
(n ⴝ 193)
Patients With Event
(n ⴝ 24)
Patients Without Event
(n ⴝ 169)
4.9 ⫾ 0.8
3.5 ⫾ 0.5
2.8 ⫾ 0.5
116 ⫾ 46
81 ⫾ 39
31 ⫾ 10
20 ⫾ 7
5.3 ⫾ 0.7
3.7 ⫾ 0.5
3.1 ⫾ 0.5
133 ⫾ 51
96 ⫾ 42
27 ⫾ 10
17 ⫾ 5
4.8 ⫾ 0.8
3.5 ⫾ 0.5
2.8 ⫾ 0.5
113 ⫾ 45
79 ⫾ 38
31 ⫾ 10
20 ⫾ 8
1.58 ⫾ 1.02
187 ⫾ 78
49 (25)
2.19 ⫾ 1.12
141 ⫾ 49
15 (63)
1.49 ⫾ 0.96
194 ⫾ 79
34 (20)
0.003
⬍0.001
⬍0.001
4.85 ⫾ 1.18
6.08 ⫾ 1.73
6.62 ⫾ 2.28
12.9 ⫾ 6.4
4.3 ⫾ 0.88
5.26 ⫾ 1.14
5.22 ⫾ 1.68
15.5 ⫾ 5.1
4.93 ⫾ 1.19
6.19 ⫾ 1.77
6.81 ⫾ 2.29
12.5 ⫾ 6.4
0.022
0.007
0.002
0.004
p Value
0.009
0.141
0.032
0.056
0.043
0.030
0.056
An unpaired t test was used for comparison of LAD and LVEF (normal distribution), a Mann-Whitney U test was used for
comparison of all other continuous variables (non-normal distribution).
A ⫽ peak late diastolic mitral filling velocity; A= ⫽ peak late diastolic mitral annular velocity; DT ⫽ deceleration time; E ⫽ peak
early diastolic mitral filling velocity; E= ⫽ peak early diastolic mitral annular velocity; FS ⫽ fractional shortening; LAD ⫽ left atrial
diameter; LV ⫽ left ventricular; LVDDI ⫽ LV end-diastolic diameter index; LVDVI ⫽ LV end-diastolic volume index; LVEF ⫽
LV ejection fraction; LVSDI ⫽ LV end-systolic volume index; LVSVI ⫽ LV systolic volume index; PW ⫽ pulsed-wave; RFP ⫽
restrictive filling pattern; S= ⫽ peak systolic mitral annular velocity.
DISCUSSION
This study is the first to combine the prognostic impact of
the ECG and the echocardiogram, two methods that are
routinely used in the follow-up of patients with CHF. The
main finding is that a RFP provides independent prognostic
information that is incremental to QRS duration in patients
with CHF.
Given the increasing number of individuals affected by
CHF, risk stratification is of tremendous importance. In such
patients, QRS prolongation is a known predictor of adverse
outcome. Shamim et al. (2) found a mortality of 50% in CHF
patients with a QRS duration ⬎140 ms as opposed to a
mortality of 23% in the remaining patients. However, our
study and others show that although QRS prolongation has
prognostic value, there is a substantial overlap of QRS duration
between patients with or without event (Table 2). In this
setting, the assessment of diastolic function added incremental
prognostic information. In CHF patients with or without
QRS prolongation, outcome was significantly worse in the
presence of a RFP (Figs. 2A and 2B). These findings are in line
with observations by Hansen et al. (7), who found an incremental value of a RFP to peak oxygen consumption in
determining the prognosis in patients with CHF.
Our findings indicate that a combination of predictive
factors more accurately predicts the individual risk than a
single parameter or cut-off value. In our risk model, in the
presence of two or three risk factors, outcome was significantly worse as compared with the presence of ⱕ1 risk
factor. In the absence of any risk factor, no patient suffered
a cardiac event during follow-up (Fig. 3). The variables
considered in our model are readily available in the daily
clinical setting and are cost-effective, enabling serial
follow-up examinations. Other markers, such as natriuretic
peptides or TDI analysis of left ventricular asynchrony (8),
may have added further prognostic information but were
not considered in the present analysis.
For the TDI-derived mitral annular E= velocity, Wang et al.
(9) recently reported an incremental predictive power for
cardiac mortality compared with standard clinical and echo
measurements. However, in their study, patients with various
Table 3. Multivariate Cox Proportional Hazard Analysis:
Predictors of Cardiac Events
Variable
Chi-Square
Relative Risk
(95% CI)
p Value
RFP
QRS duration ⬎144 ms
LVSDI ⬎2.75 cm/m2
E= ⬍5.5 cm/s
19.93
10.96
4.82
4.82
6.62 (2.7–16.4)
4.26 (1.7–10.6)
3.34 (1.1–10.3)
2.48 (1.0–6.0)
⬍0.0001
⬍0.0001
0.028
0.04
CI ⫽ confidence interval; E= ⫽ peak early diastolic mitral annular velocity; LVSDI ⫽
left ventricular systolic diameter index; RFP ⫽ restrictive filling pattern.
Figure 1. Kaplan-Meier survival curves in subgroups of patients according
to a QRS duration ⬍144 ms and ⬎144 ms. Comparison between groups
by log-rank test yielded a significant difference (p ⫽ 0.021).
JACC Vol. 45, No. 7, 2005
April 5, 2005:1072–5
Bruch et al.
ECG and Doppler Echo for Risk Stratification in Heart Failure
1075
Figure 3. Risk model based on a restrictive filling pattern, QRS duration
⬎144 ms, and left ventricular systolic diameter index ⬎2.75 cm/m2.
Notably, in subjects without any risk factor (RF) the event-free survival was
100%.
REFERENCES
Figure 2. Kaplan-Meier survival curves in patients with a QRS duration
⬎144 ms (A) or ⱕ144 ms (B). Outcome was significantly poorer in the
presence of restrictive filling patterns (RFPs) as compared with their
absence (p values derived from log-rank test).
cardiac diseases and a wide range of resting LVEFs were
analyzed. Thus, the prognostic impact of a RFP and E= may
vary in subsets of CHF patients with different etiologies and
resting LVEF.
Study limitations. Because patients with severe valvular disease, atrial fibrillation, permanent pacemaker stimulation, or
primary diastolic heart failure were excluded, our results should
not be extrapolated to these patient populations.
Reprint requests and correspondence: Dr. Christian Bruch, Universitätsklinikum Münster, Medizinische Klinik und Poliklinik C
(Kardiologie und Angiologie), Albert-Schweitzer-Str. 33, D-48129
Münster, Germany. E-mail: [email protected].
1. Packer M, Coats AJ, Fowler MD, et al. Effect of carvedilol on survival
in severe chronic heart failure. N Engl J Med 2001;344:1651– 8.
2. Shamim W, Francis DP, Yousufuddin M, et al. Intraventricular
conduction delay: a predictor of mortality in chronic heart failure? Int
J Cardiol 1999;70:171– 8.
3. Giannuzzi P, Temporelli PL, Bosimini E, et al. Independent and
incremental prognostic value of Doppler-derived mitral deceleration
time of early filling in both symptomatic and asymptomatic patients
with left ventricular dysfunction. J Am Coll Cardiol 1996;28:383–90.
4. Ho KK, Anderson KM, Kannel WB, Grossman W, Levy D. Survival
after the onset of congestive heart failure in Framingham Heart Study
subjects. Circulation 1993;88:107–15.
5. Garcia MJ, Thomas JD, Klein AL. New Doppler echocardiographic
applications for the study of diastolic function. J Am Coll Cardiol
1998;32:865–75.
6. Rivas-Gotz C, Manolios M, Thohan V, Nagueh SF. Impact of left
ventricular ejection fraction on estimation of left ventricular filling
pressures using tissue Doppler and flow propagation velocity. Am J
Cardiol 2003;91:780 – 4.
7. Hansen A, Haass M, Zugck C, et al. Prognostic value of Doppler
echocardiographic mitral inflow patterns: implications for risk stratification in patients with congestive heart failure. J Am Coll Cardiol
2001;37:1049 –55.
8. Bader H, Garrigue S, Lafitte S, et al. Intra-left ventricular electromechanical asynchrony. A new independent predictor of severe cardiac
events in heart failure patients. J Am Coll Cardiol 2004;43:248 –56.
9. Wang M, Yip GW, Wang AY, et al. Peak early diastolic mitral annulus
velocity by tissue Doppler imaging adds independent and incremental
prognostic value. J Am Coll Cardiol 2003;41:820 – 6.