From the Department of Cardiology, Hospital General Universitario
Gregorio Marañón, Madrid, Spain.
Correspondence to Héctor Bueno, MD, PhD, Department of Cardiology, Hospital General Universitario Gregorio Marañón, Dr Esquerdo, 46, 28007 Madrid, Spain. E-mail hecbueno{at}jet.es
Methods and Results—The in-hospital clinical outcome of 798
consecutive patients admitted to the coronary care unit within
48 hours of symptom onset with AIMI was analyzed according to
patient age and to the presence of RVI diagnosed by ECG and/or
echocardiographic criteria. The total incidence of RVI
was 37%, and it increased as age advanced. Patients with RVI had a
significantly higher incidence of major complications (45% versus
19%, P<0.0001) and a higher in-hospital mortality rate
(22% versus 6%, P<0.0001). The prognostic effect of
RVI was independent of sex, smoking, diabetes, shock on admission, left
ventricular ejection fraction, and reperfusion therapy, all
age-dependent predictors. A multivariate
analysis showed a significant (P=0.03)
interaction between age and RVI on AIMI mortality. RVI increased
mortality risk only in the oldest patients.
Conclusions—In patients with AIMI, RVI substantially increases
mortality risk in elderly patients, whereas it has a nonsignificant
effect in young subjects.
Definitions
Variables
Statistical Analysis
Prognostic Effect of RVI
Influence of Age
Interaction Between Age and RVI
Our findings may have some clinical implications. Table 4
The improvement in risk stratification when the patient's age is
considered compared with that when only the presence of RVI is
considered may help the selection of a particular therapeutic approach.
Zehender and colleagues4 demonstrated that
patients with AIMI and RVI had an in-hospital mortality rate of 31%,
compared with 6% in patients without RVI. It was found that the
presence of an ST-segment elevation in lead V4R
is one of the most powerful predictors of in-hospital death in
inferior-wall MIs. Our data suggest that the increase in
mortality risk associated with RVI observed in that study is the
average effect of a significant increase in mortality risk in the
elderly and a null increase in young patients. The authors pointed out
that most deaths occurred in patients who were not candidates for
thrombolytic therapy, a finding that led to the
recommendation of making maximal therapeutic efforts to procure the
opening of the occluded coronary artery in patients with
inferior-wall MIs and RVI.3 It has
even been suggested that higher-risk patients with AIMIs, such as those
with RVI, AV block, or precordial ST-segment depression, probably
benefit most from reperfusion therapy.1 2 19 20
This hypothesis was recently reviewed.20
Unfortunately, none of the placebo-controlled trials have
analyzed the benefit of thrombolytic therapy in
high-risk versus low-risk subjects with inferior MI.
Therefore, there is no convincing evidence that reperfusion therapy
results in improved outcomes in such patients. Our study shows that the
increase in mortality associated with RVI depends strongly on patient
age and, conversely, does not depend so much on the use of reperfusion
therapy. In the German study,4 the high
mortality observed in nonthrombolyzed patients may have been more
closely related to their advanced age, because those >75 years old
were not considered candidates for thrombolysis. Our
results are concordant with the observation that a high proportion of
low-risk patients receive reperfusion therapy and that this proportion
decreases as age—directly related to mortality
risk—increases.21 22 23 In other words, there is
an inverse relationship between the use of reperfusion therapies and
patient risk. Because primary coronary angioplasty is available
24 hours a day at our institution, this paradox cannot be explained
only by the age-related increase in the proportion of patients with
contraindications for
thrombolysis.24 This study
suggests that the use of a very aggressive treatment in all patients
with AIMI and RVI on the basis of selecting high-risk patients may not
be needed but that, on the contrary, maximal therapeutic efforts should
be focused on elderly patients and on the few young patients with a
complicated clinical course. Prospective studies evaluating the
efficacy of different therapeutic strategies in elderly patients with
AIMI and RVI are needed.
The reasons for our finding are unclear. Although there was an
age-related increase in the proportions of women and of diabetics,
factors that are associated with higher mortality
rates,25 26 27 28 and a decrease in the proportion of
smokers, a subgroup of patients in whom a better prognosis after acute
MI has been described when they are treated with reperfusion
therapies,29 30 there were no differences in the
prevalence of these predictors between patients with or without RVI.
LVEF was lower in patients with RVI and also decreased as age advanced.
However, the multivariate analysis, which
considered all variables available during hospitalization,
confirmed that the impact of RVI on acute inferior-wall
infarction mortality is independent of LVEF, as has previously been
reported.4 6 The age-related decrease in the
proportion of patients treated with reperfusion therapies may
contribute to the worse prognosis of elderly patients. However, no
differences in the use of thrombolysis or primary
coronary angioplasty were found according to the presence or
absence of RVI. Furthermore, reperfusion therapy was not identified as
an independent predictor of in-hospital death in any of the predictive
models analyzed. Therefore, a less aggressive treatment does
not seem to explain such poor prognosis of elderly patients with
RVI.
Most deaths were caused by cardiogenic shock or by mechanical
complications. As Figure 1
The results of this study may be limited to some extent because of
methodological limitations associated with its retrospective character.
There may be a bias in echocardiographic diagnosis of
RVI, because echocardiography was usually performed
later than the first 48 hours after admission, when features of RV
infarction may have improved, particularly after coronary
reperfusion. Patients with a complicated clinical course during the
acute phase may need an echocardiographic study at an
early stage and may have a greater probability of being diagnosed with
RVI by that method. However, the exclusion of patients diagnosed by
echocardiographic criteria may induce a greater bias
because those who have right bundle-branch block or pacemaker
stimulation due to complete AV block in the admission ECG, situations
frequently associated with RVI but that hamper its ECG diagnosis, would
be excluded, and it is known that such patients have a poor
prognosis.1 6 35 36 The use of reperfusion
therapy and the type of treatment used (thrombolysis or
primary coronary angioplasty) was based on individual clinical
decisions; therefore, conclusions about therapeutic management must be
interpreted cautiously.
In conclusion, the increase in mortality associated with RVI in
patients with AIMI depends on patient age. Young patients, with or
without RVI, have a good prognosis. Elderly patients with
inferior MI and RVI have a particularly high risk of dying
in hospital, whereas their prognosis is relatively benign if there is
no RVI.
Received March 30, 1998;
accepted June 23, 1998.
© 1998 American Heart Association, Inc.
Clinical Investigation and Reports
Combined Effect of Age and Right Ventricular Involvement on Acute Inferior Myocardial Infarction Prognosis
Abstract
Top
Abstract
Introduction
Methods
Results
Discussion
Appendix 1
References
Background—In patients with
acute inferior myocardial infarction (AIMI), right
ventricular involvement (RVI) is one of the strongest
predictors of in-hospital death. We hypothesized that the impact of RVI
on AIMI prognosis depends on the patient's age.
Key Words: myocardial infarction • aging • coronary disease • mortality • prognosis
Introduction
Top
Abstract
Introduction
Methods
Results
Discussion
Appendix 1
References
It has been suggested that among patients with
acute inferior myocardial infarction (AIMI), those with
right ventricular involvement (RVI) probably benefit most
from reperfusion therapy1 2 and that maximal
efforts should be made to obtain an early patency of the
infarct-related coronary artery.3 These
recommendations are based on the observation that the presence of RVI
is one of the strongest predictors of in-hospital death in patients
with AIMI,4 a finding that may help to stratify
patients with acute inferior-wall infarctions into high-
and low-risk groups.1 3 Nevertheless, the
exponential increase in mortality risk associated with increasing
age5 and the high short-term mortality observed
in elderly patients with RVI6 suggest that the
previously reported increase in mortality risk observed in patients
with acute MIs of inferior location and RVI compared with
those without RVI4 may reflect the average effect
of a large increase in mortality in elderly patients, with a
smaller increase in younger subjects. In other words, the deleterious
effect of RVI on inferior-wall MI prognosis may depend on
patient age. To test this hypothesis, we studied the in-hospital
outcome of 798 consecutive patients with acute inferior or
posterior MI according to patient age and to the presence of RVI.
Methods
Top
Abstract
Introduction
Methods
Results
Discussion
Appendix 1
References
Patients
The population studied consisted of 821 patients with a definite
diagnosis of acute inferior or posterior MI, not
transferred from other hospitals, consecutively admitted to the
coronary care unit of our institution from January 1, 1991, to
December 31, 1995. A definite diagnosis was established when at least 2
of the following criteria were present: (1) chest discomfort of
30 minutes compatible with myocardial ischemia; (2) presence
of an ST-segment elevation 0.1 mV in 2 of leads II, III, and aVF in
the admission ECG and/or appearance of an R wave in
V1 or V2 >0.04 second with
an R/S voltage ratio >1; and (3) elevation of serum creatine kinase
(CK) level to at least twice the upper normal limit (195 U/L in our
institution), with an MB fraction (CK-MB) >10% of the total CK level.
Three patients in whom the presence of RVI could not be determined by
the defined criteria were excluded. An MI was considered acute if the
time from symptom onset to admission was not longer than 48 hours.
Twenty patients admitted with a longer time delay were also excluded.
The definitive study group consisted of 798 patients.
RVI was diagnosed by the presence of an ST-segment elevation
0.1 mV in lead V3R or
V4R4 7 8 9 10 in the ECG
performed immediately after admission to the coronary care unit
and/or by the presence of RV free wall motion abnormalities or RV
dilatation detected in a 2-dimensional transthoracic
echocardiographic study.11 12 13 14 15
The determination of the presence of RVI was made by physicians who
were unaware of the clinical outcome of the patient. Shock was defined
as the concurrence of persistent hypotension and clinical signs of low
cardiac output16 (due to either left
ventricular [LV] or RV failure or both) and was
considered cardiogenic after the exclusion of hypovolemia,
arrhythmias, and mechanical complications. Rupture of free
ventricular wall, interventricular septum, or
papillary muscle was defined as mechanical complications. Major
complications included death, cardiogenic shock, mechanical
complications, primary ventricular fibrillation, sustained
ventricular tachycardia (lasting >30 seconds
or causing hemodynamic compromise), complete
atrioventricular (AV) block, and reinfarction.
The clinical records of all patients who met the inclusion
criteria were studied retrospectively. At the time of patient admission
to the coronary care unit, a complete clinical history,
physical examination, 16-lead ECG (12 standard leads plus 2 right
precordial leads, V3R and
V4R, and 2 posterior leads,
V7 and V8), complete blood
analysis, and chest radiograph were routinely performed. Serial
laboratory studies, including CK and CK-MB and 12-lead ECGs, were
obtained at 6- to 8-hour intervals during the first 24 to 48 hours of
evolution. The patient's treatment was determined on an individual
basis by the attending physician. Intravenous volume
loading was used systematically in patients with RV infarction and
systemic hypotension or signs of low cardiac output.
Echocardiographic studies were usually performed later
than the first 48 hours after admission unless a clinical indication
was present. Left ventricular ejection fraction (LVEF)
was evaluated by 2-dimensional echocardiography in
most patients, and in exceptional cases, only by contrast left
ventriculography during cardiac catheterization. The
following variables were analyzed according to the
patient's age and the presence of RVI: (1) baseline characteristics:
age, sex, cardiovascular risk factors (systemic
hypertension, diabetes mellitus, dyslipemia, and current cigarette
smoking), and history of previous MI; (2) infarct features: incidence
of RVI, time from symptom onset to admission, Killip class on
admission, and peak CK and CK-MB values; (3) diagnostic and
therapeutic procedures: echocardiography (LVEF,
RVI, mitral regurgitation, and mechanical
complications), coronary angiography (infarct-related
coronary artery, number of coronary arteries with a
luminal obstruction 70%), and reperfusion therapy
(intravenous thrombolysis or primary
coronary angioplasty); and (4) in-hospital incidence of major
complications. The cause of death was classified into 3 groups: (1)
cardiogenic shock, (2) mechanical complications (deaths in patients who
were operated on as treatment for any mechanical complication that
occurred in the perioperative period were included in
this group regardless of the immediate cause of death), and (3)
other causes.
Continuous variables are expressed as medians (25th to 75th
percentiles). The 
2 test was used to assess
the significance of the differences between proportions, and Student's
t test was used for comparisons between means. Comparisons
between ordinal variables and examination of age-related trends in
predictors and outcomes were studied with the Mantel-Haenszel test for
linear association. The independent contribution of age and RVI to
in-hospital mortality was assessed in 2 multivariate
analyses: the first included all prognostic factors available
at the time of admission (age, sex, diabetes, smoking habit, previous
MI, RVI, shock on admission, and reperfusion treatment), and the second
adjusted for other variables obtained during hospitalization (LVEF
and complete AV block). The presence of an interaction between age and
RVI was assessed by statistical evaluation of the interaction factor
agexRVI added to the predictive model that contained all predictors
available on admission (saturated model).17 A
backward system of variable exclusion was performed from the
saturated model, keeping fixed the interaction factor, age, and RVI.
The final model was the one that included less predictive variables
and had a nonsignificant loss of predictive value as evaluated by the
area under the receiver operating characteristic curves of the
models.18 The predictions obtained with the final
model were used to calculate the age-related adjusted odds ratios of
in-hospital mortality risk of patients with RVI compared with those
without RVI. All probability values were 2-tailed and were considered
significant at P<0.05.
Results
Top
Abstract
Introduction
Methods
Results
Discussion
Appendix 1
References
Clinical Features of AIMI
The clinical characteristics of the patients are shown in Table 1
. RVI was diagnosed
in 296 patients (37% incidence): in 223 patients by ECG criteria
(75%), and in 73 patients with negative or inconclusive ECG, by
echocardiography (25%). Major complications
appeared during hospitalization in 228 patients (29%). The most
frequent were complete AV block (14%) and cardiogenic shock (8%).
In-hospital mortality rate was 12% in the entire group. Of the 94
deaths, 49 (52%) were due to cardiogenic shock, 32 (34%) to
mechanical complications, and 13 (14%) to other causes.
View this table:
[in a new window]
Table 1. Clinical Characteristics of Patients With AIMI
According to the Presence or Absence of
RVI
Patients with RVI were slightly older and had a higher prevalence
of hypertension. They were more frequently admitted in shock and had
greater LV systolic dysfunction (Table 1
). Reperfusion
therapies (thrombolysis and primary coronary
angioplasty) were used in similar proportions of patients with and
without RVI. Coronary angiography showed that index MI was
related to the right coronary artery more frequently in
patients with RVI. The presence of RVI was associated with a
significantly higher incidence of major complications, cardiogenic
shock, mechanical complications, complete AV block, and primary
ventricular fibrillation as well as death during
hospitalization (Table 2).
View this table:
[in a new window]
Table 2. Hospital Course of Patients According to the
Presence or Absence of RVI
The incidence of RVI, prevalence of prognostic factors, and
incidence of most important outcomes are presented according to
age groups in Table 3. As
age increased, the incidence of RVI as well as the proportion of women,
diabetics, patients with shock on admission, and those with a
moderately to severely depressed LVEF increased. Conversely, the
proportion of smokers and of patients treated with reperfusion therapy
decreased. Increasing age was also directly associated with the
in-hospital incidences of major complications and death in the entire
group (Table 3) and in the patients with and without RVI (Figure 1). The causes of death did not change
significantly with age (Figure 2).
View this table:
[in a new window]
Table 3. Relationship Between Age, Incidence of RVI,
Proportion of Patients With Prognostic Factors, Reperfusion Treatment,
and Major Complications
View larger version (15K):
[in a new window]
Figure 1. Comparison of age-related in-hospital mortality
rate (top), incidence of cardiogenic shock (bottom left), and incidence
of mechanical complications (bottom right) in patients with AIMI with
( 
) and without (
) RV involvement.
View larger version (20K):
[in a new window]
Figure 2. Relationship between age, mortality rate, and
cause of death in 798 consecutive patients with AIMI.
Age and RVI were confirmed as independent predictors of
in-hospital death in patients with AIMI after adjustment for all other
predictors available at the time of admission and during
hospitalization (P<0.001 for both parameters).
A significant (P=0.03) interaction between the effects of
age and RVI on mortality was found in a new
multivariate analysis that considered all early
predictors (Appendix, saturated model). A final model with a predictive
value similar to that of the saturated model (Appendix, final model and
graph) was used to perform adjusted mortality predictions (Table 4). When the adjusted
impact of RVI on mortality was analyzed considering different
age groups, a gradient in mortality risk increase was found (Figure 3). Thus, the presence of RVI was found
to be a significant predictor of in-hospital death only in elderly
patients. This finding remained unchanged when patients with only ECG
criteria of RVI were analyzed.
View this table:
[in a new window]
Table 4. Adjusted In-Hospital Mortality Rate
Predictions
View larger version (17K):
[in a new window]
Figure 3. Adjusted age-related predictions of increase in
in-hospital mortality risk associated with presence of RVI compared
with absence of RVI in patients with AIMI. Results are expressed as
odds ratios and 95% CIs. Increase is statistically significant when
none of the ends of CIs cross equivalent risk line (odds
ratio=1).
Discussion
Top
Abstract
Introduction
Methods
Results
Discussion
Appendix 1
References
The results of the present study confirm previous
investigations showing that RVI is a strong independent predictor of a
poorer outcome in patients with AIMI.4 6 However,
the study shows that the influence of RVI on mortality changes
with age, from an insignificant effect in young patients to a severe
increase in mortality risk in the elderly. shows the
in-hospital mortality risk estimations for patients with AIMI. If the
patient's age and sex, the presence of RVI, and whether the patient is
currently smoking are known, this predictive model allows a simple risk
stratification of patients with AIMI that can be performed when they
are admitted to the coronary care unit. Assuming arbitrarily a
12% in-hospital mortality risk as the cutoff point to determine high-
and low-risk patients, one may notice that most high-risk patients are
elderly and have RVI (boldface numbers). Therefore, an accurate
diagnosis of RVI should be performed as soon as possible in all
patients with AIMI, but particularly in the elderly. Patients in whom
the ECG is inconclusive regarding the presence of RVI or who have
hemodynamic impairment should have an urgent
echocardiographic study. shows, the age-related increase in death
rate observed in patients with RVI is roughly parallel to the
age-related increase in the incidence of cardiogenic shock but not of
mechanical complications. In a previous study, we found that the high
mortality of patients 75 years old with AIMI and RVI was essentially
due to the high incidence of low-cardiac-output cardiogenic
shock.6 Mortality in those patients was not
linearly associated with LVEF, as opposed to patients without RVI. We
speculated that LV diastolic dysfunction may play a role in
the poor hemodynamic tolerance of acute RV dysfunction
observed in elderly patients.6 In the present
nonselected population, the mortality risk associated with RVI
increases progressively as age advances (Figure 3), as with LV
diastolic dysfunction.31 32 33
Therefore, the present study is in agreement with but does not
demonstrate the hypothesis that LV diastolic dysfunction
plays a role in the poor prognosis for elderly patients with AIMI and
RVI. Other age-related changes, such as the increase in
pulmonary vascular resistance, may contribute to this
behavior.34
View larger version (20K):
[in a new window]
Figure 4. Predictive models by multiple logistic regression
analyses. Shock indicates shock on admission; pMI, previous MI;
diab, diabetes; rep, reperfusion treatment; and ROC, receiver operating
characteristic.
Appendix 1
Top
Abstract
Introduction
Methods
Results
Discussion
Appendix 1
References
Predictive Models by Multiple Logistic Regression Analyses
The Appendix is shown in Figure 4.
Acknowledgments
We are indebted to Jesús Almendral, MD, Esteban G.
Torrecilla, MD, and Manuel Martínez-Sellés, MD, for
critically reviewing the manuscript.
References
Top
Abstract
Introduction
Methods
Results
Discussion
Appendix 1
References
This article has been cited by other articles:
 |
|
 |
|
  A. Kumar, H. Abdel-Aty, I. Kriedemann, J. Schulz-Menger, C. M. Gross, R. Dietz, and M. G. Friedrich Contrast-Enhanced Cardiovascular Magnetic Resonance Imaging of Right Ventricular Infarction J. Am. Coll. Cardiol., November 21, 2006; 48(10): 1969 - 1976. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Martinez-Selles, R. Lopez-Palop, E. Perez-David, and H. Bueno Influence of Age on Gender Differences in the Management of Acute Inferior or Posterior Myocardial Infarction Chest, August 1, 2005; 128(2): 792 - 797. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-K. Wong and H. D. White Risk stratification of patients with right ventricular infarction: is there a need for a specific risk score? Eur. Heart J., November 1, 2002; 23(21): 1642 - 1645. [Full Text] [PDF]
|
|
|
|
|
|
S. R. Mehta, J. W. Eikelboom, M. K. Natarajan, R. Diaz, C. Yi, R. J. Gibbons, and S. Yusuf Impact of right ventricular involvement on mortality and morbidity in patients with inferior myocardial infarction J. Am. Coll. Cardiol., January 1, 2001; 37(1): 37 - 43. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 RV Infarction Increases Mortality in the Elderly but Not in the Young. Journal Watch Cardiology, December 11, 1998; 1998(1211): 3 - 3. [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||