Contribution of Ischemic Mitral Regurgitation to Congestive Heart Failure After Myocardial Infarction

Transcription

1 Journal of the American College of Cardiology Vol. 45, No. 2, by the American College of Cardiology Foundation ISSN /05/$30.00 Published by Elsevier Inc. doi: /j.jacc Contribution of Ischemic Mitral Regurgitation to Congestive Heart Failure After Myocardial Infarction Francesco Grigioni, MD,* Delphine Detaint, MD,* Jean-François Avierinos, MD,* Christopher Scott, MS, Jamil Tajik, MD, FACC,* Maurice Enriquez-Sarano, MD, FACC* Rochester, Minnesota OBJECTIVES The purpose of this study was to define the contribution of ischemic mitral regurgitation (IMR) to the occurrence of congestive heart failure (CHF) after myocardial infarction (MI). BACKGROUND After MI, CHF is a frequent and serious complication, but its determinants and, particularly, the role of IMR are poorly defined. METHODS We analyzed 173 asymptomatic patients with previous Q-wave MI ( 16 days) with echocardiographic quantitation of IMR (measuring effective regurgitant orifice [ERO] and regurgitant volume). The 102 patients with IMR were matched to 71 patients without IMR for age (71 11 years vs years; p 0.11), gender (76% vs. 82% males; p 0.41), and left ventricular ejection fraction (EF) (37 14% vs %; p 0.92). RESULTS Five-year rates of CHF and of CHF or cardiac death (CD) were 36 5% and 52 5%, respectively. Independent determinants of CHF were EF, sodium plasma level, and presence and degree of IMR (p ). Five-year CHF rates were 18 5% without mitral regurgitation (MR), 53 7% with IMR, 46 9% with ERO 1 to 19 mm 2 and 68 12% with ERO 20 mm 2 (all p ). The adjusted relative risk of CHF was 3.65 (95% confidence interval [CI] 1.86 to 7.75) for IMR presence and 4.42 (95% CI 1.9 to 10.5) for ERO 20 mm 2. The adjusted relative risk of CHF/CD was 2.97 (95% CI 1.77 to 5.16) for IMR presence and 4.4 (95% CI 2.4 to 8.2) for ERO 20 mm 2. CONCLUSIONS After MI, incidence of CHF and of CHF/CD are high even in patients with no or minimal symptoms at baseline and are higher in patients with IMR. Congestive heart failure is independently determined by larger ERO of IMR. These data suggest that detecting and quantifying IMR is essential for risk stratification after MI. Value of IMR treatment in improving post-mi outcome should be investigated. (J Am Coll Cardiol 2005;45:260 7) 2005 by the American College of Cardiology Foundation Survivors of acute myocardial infarction (MI) are at high risk for congestive heart failure (CHF) (1), which adversely affects subsequent outcome, particularly with poor survival (2,3). Therefore, identification of predictors of CHF in patients presenting with no or minimal symptoms after MI is essential in potentially altering post-mi management and minimizing CHF rate (4). Ischemic mitral regurgitation (IMR) is a mitral incompetence due to coronary artery disease (CAD) in the absence of intrinsic valve lesions and is frequent after MI (5,6). Previous data from our institution (7) and others (5,6,8) were consistent in suggesting that IMR presence after MI affects survival. However, with regard to morbid cardiac complications such as occurrence of post-mi CHF, few data are available. The Survival And Ventricular Enlargement (SAVE) study provided seminal information that IMR may lead to higher CHF incidence (5). However, inclusion limited to the acute post-mi phase ( 16 days), more advanced symptoms, and left ventricular (LV) alterations in patients with IMR and lack of multivariate confirmation left doubt on the intrinsic IMR influence on From the *Division of Cardiovascular Diseases and Internal Medicine and the Section of Biostatistics, Mayo Clinic and Mayo Foundation, Rochester, Minnesota. The study was supported in part by the grant HL of the National Institutes of Health. Dr. Grigioni is currently located at the Istituto di Cardiologia dell Universita di Bologna, Ospedale S. Orsola Malpighi, Bologna, Italy. Manuscript received March 23, 2004; revised manuscript received September 27, 2004, accepted October 4, subsequent CHF. Moreover, the SAVE study was intriguing as it almost exclusively included patients with mild mitral regurgitation (MR). Therefore, the influence of IMR and, in particular, the influence of increasing degree of MR on subsequent occurrence of CHF in patients with MI presenting with no or minimal symptoms is uncertain. To address this issue, we enrolled patients who were in New York Heart Association (NYHA) functional class I and II in the chronic post-mi phase ( 16 days), simultaneously to extensive clinical and Doppler echocardiographic assessment. We hypothesized that IMR is a major determinant of the occurrence of subsequent CHF and that its quantitation provides important prognosis information. Verifying these hypotheses would emphasize the importance of the assessment of IMR for risk stratification after MI. METHODS Population. Patients were consecutively enrolled at the Mayo Clinic, Rochester, Minnesota, between 1990 and 1997 if they: 1) were in NYHA functional class I or II at diagnosis and had no history of CHF; 2) had a Q-wave MI on electrocardiogram and a history of MI older than 16 days; 3) had transthoracic echocardiography during the same evaluation in routine clinical practice, showing either IMR, which was prospectively quantitatively assessed, or no MR. Exclusion criteria were: NYHA functional class III, recent MI ( 16 days), previous cardiac surgery, papillary

2 JACC Vol. 45, No. 2, 2005 January 18, 2005:260 7 Grigioni et al. Ischemic MR and CHF 261 Abbreviations and Acronyms CAD coronary artery disease CD cardiac death CHF congestive heart failure CI confidence interval EF ejection fraction ERO effective regurgitant orifice IMR ischemic mitral regurgitation LA left atrium LV left ventricle MI myocardial infarction MR mitral regurgitation NYHA New York Heart Association PCI percutaneous coronary intervention RR risk ratio RVol regurgitant volume SAVE Survival And Ventricular Enlargement study muscle rupture, primary leaflet or chordal pathology, associated aortic valve, or congenital heart disease. No exclusions were made on the basis of treatment received or degree of MR. A computerized matching process was conducted between patients with and without IMR who satisfied inclusion criteria. To avoid baseline differences among these major determinants of outcome, patients without IMR were matched to those with IMR for age, gender, and LV ejection fraction (EF). The matching process was computerized on the basis of multivariate matched sampling incorporating the propensity score (9) and was performed in a blinded manner before any outcome information was obtained. Congestive heart failure during follow-up was diagnosed using validated criteria (10). Cardiac deaths (CDs) were ascertained by reviewing death certificates, coroners reports, or autopsy records. Medications were recorded if prescribed for three months or more. Echocardiographic methods. Dimensions of the left ventricle (LV) and left atrium (LA) were obtained using two-dimensional guided M-mode echocardiography. The EF was visually estimated in all and combined with calculated values in 119 patients (69%) (7,11). Data were used unaltered from the original echocardiographic report by means of electronic transfer. Presence of MR was detected by color flow imaging. When present, the degree of IMR was assessed with at least one of the following two quantitative methods. Results were eventually averaged. Quantitative Doppler measured mitral and aortic stroke volumes, and the regurgitant volume (RVol) was calculated as the difference between these two stroke volumes (12,13). The effective regurgitant orifice (ERO) area was calculated as the ratio of RVol to regurgitant time-velocity integral (14). The proximal isovelocity surface area method (15) analyzed the proximal flow convergence region, and ERO was the ratio of regurgitant flow to regurgitant velocity (16,17). The RVol was calculated as the product of ERO by regurgitant time-velocity integral. Statistical analysis. Continuous variables are expressed as mean SD values. Group comparisons were performed with the standard t test or chi-square test, as appropriate. Event rates were estimated by the Kaplan-Meier method, censoring at the time of surgery if surgery was eventually performed. Comparisons between groups were evaluated using the two-way log-rank test. End points were CHF and the combined end point of CD or CHF. Impact of IMR on outcome was analyzed in two ways: 1) with presence of IMR at baseline used as the categorical determinant of adverse cardiac events, or 2) with quantitative degree of IMR (RVol and ERO). Among variables of MR (ERO and RVol), the strongest predictor (highest chi-square) of cardiac events was selected for the final model. Identification of other baseline predictors of events was accomplished by performing univariate Cox proportional hazards analysis on candidate clinical, laboratory, and echocardiographic variables. Risk ratio (RR) and 95% confidence interval (CI) associated with previously determined thresholds were defined (18). To confirm independent predictive value, variables with p 0.10 were tested in multivariate models, with IMR expressed as both a categorical and a quantitative variable. Final models were repeated with variables significantly different at baseline between patients with and without IMR. To assess the effects of events occurring after diagnosis (follow-up atrial fibrillation, MI, or percutaneous coronary intervention) on outcome, Cox proportional hazards analyses using time-dependent variables were used. Furthermore, because the role of NYHA functional class II versus I on outcome is unknown, analysis was repeated with NYHA functional class in the model. To address the concern of possible overestimation of EF with MR, we repeated the analysis using EF values decreased by 4% and 8% in patients with IMR. A value of p 0.05 was considered significant. RESULTS Baseline characteristics. We enrolled 173 patients with no or minimal symptoms (NYHA functional class I or II) in the chronic phase after MI in the study. The mean time after MI was years. The 102 patients with IMR (ERO cm 2 and RVol ml) were compared with 71 patients without IMR. Overall, age was years, 136 patients (79%) were male, and EF was 36 13%. The diagnosis of previous MI was confirmed by electrocardiogram (in all patients) and by echocardiography as presence of regional wall motion abnormalities in 172 patients. A nuclear perfusion study available for 72 patients detected a previous MI in 67 (93%). Coronary angiography was ultimately performed in 102 patients, showing obstructive coronary disease in all. Baseline characteristics of patients with and without MR are summarized in Table 1. Importantly, age, gender, comorbidity, most risk factors, prevalence of multivessel disease, and EF were similar in both groups. As expected,

3 262 Grigioni et al. JACC Vol. 45, No. 2, 2005 Ischemic MR and CHF January 18, 2005:260 7 Table 1. Baseline Characteristics of Patients With and Without Ischemic MR Baseline Characteristics Association to Time to IMR (n 102) No MR (n 71) p Value CHF p Value CHF/Death p Value Age (yrs) Male gender (%) Atrial fibrillation (%) Comorbidity score 2 (%) Diabetes (%) Hypertension (%) Smoking (%) Hypercholesterolemia (%) Anterior MI (%) CAD, multivessel disease (%) Sodium plasma level (mg/dl) EF (%) LVSi (mm/m 2 ) LVDi (mm/m 2 ) LA (mm) Mitral deceleration time (ms) ERO (cm 2 ) RVol (ml/beat) CAD coronary artery disease; CHF congestive heart failure; EF left ventricular ejection fraction; ERO effective regurgitant orifice; IMR ischemic mitral regurgitation; LA left atrial diameter; LVDi left ventricular end-diastolic diameter indexed to body surface area; LVSi left ventricular end-systolic diameter indexed to body surface area; MI myocardial infarction; MR mitral regurgitation; RVol regurgitant volume. limited clinical differences were noted regarding history of hypertension and location of the causal MI. Other differences between patients with and without IMR were linked to the IMR, with larger atrial dimension, higher prevalence of atrial fibrillation, and larger ventricular size. No differences were noted for treatment by beta-blockers (p 0.98), statins (p 0.98), and vasodilators (p 0.06), but IMR patients were more often treated with diuretics and digoxin (both p 0.003). IMR and CHF. During follow-up of years, 49 patients incurred CHF under conservative management. The overall incidence of CHF at 3 and 5 years was 23 4% and 36 5%, respectively. Patients with IMR experienced significantly higher incidence of CHF than those without IMR (at 5 years: 53 7% vs. 18 5%, respectively; p 0.001) (Fig. 1). In univariate analysis, presence of MR was strongly associated with occurrence of CHF (RR 4.3 [95% CI 2.2 to 8.9]; p ). The other univariate baseline predictors of time to CHF are listed in Table 1. In multivariate analysis, presence of IMR was a strong and independent determinant of occurrence of CHF (RR 3.65 [95% CI 1.86 to 7.75]; p ) (Table 2). The other independent determinants of time to CHF were EF and baseline sodium level. Of note, atrial and ventricular diastolic and systolic dimensions and mitral deceleration time (a measure of ventricular diastolic function), which were univariately associated with occurrence of CHF, were not independently predictive of CHF and did not affect (when forced into the model) the significance and RR attached to IMR (RR of IMR, respectively, 4.90, 2.65, 3.29, 3.41; all p 0.01). Furthermore, adjustment for differences between IMR and non-imr patients did not affect the independent predictive value of IMR for CHF (RR of IMR after adjustment for baseline atrial fibrillation, 3.7 [95% CI 1.9 to 7.8], p ; for hypertension, 3.7 [95% CI 1.9 to 8.0], p ; for MI location, 3.5 [95% CI 1.7 to 7.6], p ). The results were not altered after adjustment for NYHA (functional class I or II) distribution, which was not an independent predictor of CHF (p 0.90) and did not affect the adjusted RR attached to presence of IMR (3.7 [95% CI 1.8 to 8.0]; p 0.002). Accounting for events and procedures occurring after diagnosis, the independent predictive value of IMR for CHF was not affected (RR of IMR, adjusting for post-diagnosis atrial fibrillation, 2.8 [95% CI 1.3 to 5.9], p 0.01; or MI, 3.1 [95% CI 1.5 to 6.4], p 0.01; or PCI, 3.2 [95% CI 1.5 to 6.6], p 0.01). A higher degree of IMR was associated with a higher incidence of CHF. Patients with RVol 30 ml incurred higher CHF incidence (at 5 years, 54 9%) than those with no MR (p ), but only marginally higher than those with RVol 1 to 29 ml (49 10%). However, patients classified according to ERO showed marked difference in CHF rates (at 5 years, 18 5% without IMR, 46 9% with ERO 1 to 19 mm 2, and 68 12% with ERO 20 mm 2 ;p ) (Fig. 2). Univariately, in comparison to patients with no MR, ERO 1 to 19 mm 2 was associated with a high risk of CHF (RR 3.55 [95% CI 1.67 to 7.87]; p 0.001) and ERO 20 mm 2 with an even higher risk of CHF (RR 6.4 [95% CI 2.9 to 14.3]; p ). Moreover, ERO remained an independent predictor of CHF once adjusted for the other independent predictors of CHF with adjusted RR 3.45 (95% CI 1.58 to 7.9), p for ERO 1 to 19 mm 2 and 4.42 (95% CI 1.9 to 10.5), p for ERO 20 mm 2.

4 JACC Vol. 45, No. 2, 2005 January 18, 2005:260 7 Grigioni et al. Ischemic MR and CHF 263 Figure 1. Incidence of congestive heart failure (CHF) in asymptomatic patients after myocardial infarction (MI) according to the presence (continuous line) or absence (dotted line) of ischemic mitral regurgitation (IMR) at diagnosis. The event rates at five years are indicated the standard error. IMR and combined adverse events. During follow-up 62 patients died, 46 from cardiac cause (14 sudden, 15 progressive CHF, and 16 miscellaneous causes such as new MI or CAD complications, thromboembolic complications, and bleeding due to anticoagulation). Thus, a total of 80 patients incurred either CHF or CD during follow-up under conservative management. The incidence of this combined end point at 3 and 5 years was, overall, 33 4% and 52 5%, respectively. Patients with IMR experienced higher incidence of events compared to those without MR (at 5 years, 69 6% vs. 30 6%; p ) (Fig. 3). In univariate analysis, presence of MR was strongly associated with higher event rates (RR 3.3 [95% CI 2.0 to 5.6]; p ). The other univariate predictors of time to CHF/CD are listed in Table 1 (right column). In multivariate analysis, IMR was a strong independent predictor of the combined end point (adjusted RR 2.97 [95% CI 1.77 to 5.16]; p ). The other independent determinants of the combined end point were age, sodium level, and EF with a borderline significance (Table 3). Of note, ventricular diastolic and atrial dimension, hyperlipidemia, and presence of multivessel disease did not affect (when forced into the model) the significance magnitude of effect of IMR (RR attached to IMR, respectively, 2.2, 2.86, 2.94, 2.79; all p 0.01). Furthermore, adjustment for differences between IMR and non-mr patients did not affect the independent predictive value of IMR for the combined end point (RR of IMR after adjustment for baseline atrial fibrillation, 3.0 [95% CI 1.8 to 5.2], p ; for hypertension, 2.9 [95% CI 1.7 to 5.0], p ; for MI location, 2.8 [95% CI 1.6 to 5.0], p ). The results were not altered after adjustment for NYHA (functional class I or II) distribution, which was not an independent predictor of the combined end point (p 0.86) and did not affect the adjusted RR attached to presence of IMR (2.9 [95% CI 1.7 to 5.2]; p ). Accounting for events and procedures occurring after diagnosis, the independent predictive value of IMR for the combined end point was not affected (RR of IMR adjusting for post-diagnosis atrial fibrillation, 2.5 [95% CI 1.4 to 4.4], p 0.01; or MI, 2.5 [95% CI 1.5 to 4.5], p 0.01; or PCI, 2.8 [95% CI 1.6 to 4.9], p 0.01). A higher degree of IMR was associated with a higher incidence of the combined end point of CHF or CD. In comparison to patients with no MR (at 5 years, 31 6%), patients with RVol 1 to 29 ml and 30 ml incurred higher risk (at 5 years, 62 9% and 71 7%, respectively; p ). Patients with ERO 1 to 19 mm 2 and 20 mm 2 also were at high risk (68 9% and 78 9%, respectively; p ) (Fig. 4). Univariately, in comparison to patients with no MR, ERO 1 to 19 mm 2 was associated with a high risk of CHF (RR 2.8 [95% CI 1.6 to 5.1]; p 0.001) and ERO 20 mm 2 with an even higher risk of CHF (RR 4.4 [95% CI 2.4 to 8.2]; p ). The ERO area remained independently predictive of the combined end point after adjustment for the other independent predictors of CHF with adjusted RR 2.8 (95% CI 1.5 to 5.2), p for ERO1to19mm 2 and 3.42 [95% CI 1.7 to 6.7], p for ERO 20mm 2. DISCUSSION The present study showed that in patients with no or minimal symptoms in the chronic phase after MI, the incidence of CHF and that of CHF or CD are high but are unequally distributed. Compared with patients without IMR of similar age, gender, and EF, patients with IMR incur markedly higher event rates, unadjusted or adjusted for all possible differences between groups. The risk of event is approximately tripled with IMR even after adjustment for all other baseline predictors of CHF and for all differences in baseline characteristics in multivariate analysis. A higher degree of MR, particularly quantified by the ERO, is independently associated with higher risk of CHF and of the combined end point of CHF or CD. Therefore, identifying the presence and quantifying the degree of IMR in the chronic phase after MI is essential for post-mi risk stratification, even in patients presenting with no or minimal symptoms. IMR: challenges and opportunities. During the past decade, the large-scale trials have significantly contributed to improve prognosis of patients with acute MI (19). However, the reduced fatality rate from acute MI has resulted in Table 2. Baseline Independent Predictors of Subsequent Congestive Heart Failure After MI Variables RR 95% CI p Value Age Male gender Comorbidity index EF (per 1%) Sodium plasma level (per meq) IMR CI confidence interval; RR risk ratio; other abbreviations as in Table 1.

5 264 Grigioni et al. JACC Vol. 45, No. 2, 2005 Ischemic MR and CHF January 18, 2005:260 7 Figure 2. Incidence of congestive heart failure (CHF) in asymptomatic patients after myocardial infarction according to the degree of mitral regurgitation measured by effective regurgitant orifice (ERO) 20 mm 2 (continuous line), 1to19mm 2 (dotted line), and absent mitral regurgitation (ERO 0) (dashed line) at diagnosis. The event rates at five years are indicated the standard error. increased numbers of patients at risk to develop CHF (20,21). Furthermore, despite new therapeutic enhancement, the individual risk to develop CHF during long-term follow-up remains high (1) and may have contributed to the current CHF epidemic. The occurrence of CHF is a turning point for outcome, as it is associated with considerable excess mortality (2,3,10,20,22) and catastrophic social cost. Thus, identification of predictors of CHF in patients with no or minimal symptoms represents a core approach in the management of MI survivors (4). To detect a potential role of IMR as determinant of outcome after MI is challenging because IMR is rarely heard clinically, and color flow imaging, although sensitive, tends to overestimate the degree of regurgitation. Therefore, we designed our study on the basis of quantitation of IMR performed in routine practice without knowledge of any outcome data. The hypothesis that IMR may be a determinant of subsequent CHF stems from recent data suggesting that in both the acute ( 16 days) and chronic ( 16 days) phases after MI, IMR is associated with excess mortality (5,6,23 25). However, we (7) and the SAVE study data (5) noted that presence of IMR was associated with higher prevalence of symptoms at baseline, so no data are available on the potential impact of IMR on outcome in patients presenting with no or minimal symptoms who tend to be considered at low risk. Furthermore, there are no mechanistic data linking IMR to specific complications that may delineate the path to excess mortality and take steps to prevent it. To address theses issues, we analyzed the occurrence of CHF in patients detected in routine practice (to avoid selection bias) and presenting in NYHA functional class I or II in the chronic phase after MI, matching patients with and without IMR for age, gender, history of MI, and EF. IMR and subsequent CHF. The current study showed that presence of IMR in patients with no or minimal symptoms seen in the chronic phase after MI was independently associated with a higher long-term incidence of CHF. This high risk is not related to differences in baseline characteristics, as matching and adjustment in multivariate analysis showed that IMR remained an independent predictor of CHF after diagnosis. The outcome implications of IMR are additive to those of EF or hyponatremia, a marker of neurohormonal activation (26), and to adjustment for mitral deceleration time, a major marker of LV diastolic function (27), showing that IMR in and by itself is associated with subsequent high risk of CHF. The pioneering data from the SAVE study suggested that IMR in acute MI was associated with increased risk of CHF episodes, but could not confirm it independently (5). Therefore, to our knowledge the current study, using the strength of matching patients with and without IMR for age, gender, and EF (9), for the first time independently links IMR with occurrence of CHF in the chronic phase after MI. These findings extend and emphasize the importance of IMR as a determinant of poor outcome in ischemic heart disease, in the acute (5), the chronic (7), or both phases (6), or after percutaneous intervention (8) and for mortality or for morbid events. An important point is that the higher rate of CHF in asymptomatic patients with IMR is not due to longer time at risk because of lower cardiac mortality, as the combined end point of CD and CHF also occurred more frequently in patients with IMR. Degree of IMR and subsequent CHF. Another important finding is the direct degree-effect relationship between quantified IMR severity and CHF risk. Indeed, patients with ERO 20 mm 2 (0.20 cm 2 ) displayed CHF risk more than six times higher than patients without IMR in univariate analysis and more than four times higher after adjustment for all predictors of CHF. The analysis of the combined end point of CD and CHF again confirmed that this gradient of CHF occurrence paralleling the degree of IMR is not due to lower mortality. Figure 3. Survival free of congestive heart failure or cardiac death (eventfree survival) in asymptomatic patients after myocardial infarction (MI) according to the presence (continuous line) or absence (dotted line) of ischemic mitral regurgitation (IMR) at diagnosis. The event-free survival rates at five years are indicated the standard error.

6 JACC Vol. 45, No. 2, 2005 January 18, 2005:260 7 Grigioni et al. Ischemic MR and CHF 265 Table 3. Multivariate Predictors of the Combined End Point of Congestive Heart Failure or Cardiac Death in Asymptomatic Post-MI Patients Variables RR 95% CI p Value Age Male gender EF (per 1%) Sodium plasma level (per meq) IMR Abbreviations as in Tables 1 and 2. The quantitative analysis also provides important insights into potential mechanistic aspects of the link between CHF and IMR. The ERO exhibited stronger predictive power for subsequent CHF than the RVol. This result is in agreement with previous results obtained by our group (7,28) and probably reflects the mechanisms of LV-to-LA energy transfer. Indeed, the regurgitant LV energy can be transferred as kinetic (flow) or potential (pressure) to the LA (29). The RVol reflects purely kinetic energy, whereas the regurgitant orifice is linked to both, with increase in LA pressure leading to subsequent development of CHF. In that regard, the observation of higher filling pressures independently related to the ERO area (28) fits well with the clinical consequences of IMR and the higher rate of CHF with larger ERO. Hence, the current study emphasizes the relevance of IMR severity assessment through ERO determination. A threshold of 20 mm 2 appeared to identify patients at highest risk to develop adverse cardiac events. This cutoff is lower than the 40 mm 2 threshold used in patients with organic MR (18), as IMR volume is usually smaller (14). Such ERO threshold of 20 mm 2 is probably related to the profound LV alteration of post-mi patients (5,6), explaining the poor LV tolerance to IMR and the severe clinical consequences observed. This threshold fits well with previous observations of links between ERO and hemodynamic consequences (28,30), and between ERO and clinical outcome (7) in such patients. Therefore, IMR of a degree that may otherwise be qualified as moderate (31) imposes severe clinical implication in post-mi patients (5), even those presenting with no or minimal symptoms. Clinical implications. The present data expand the growing body of evidence that IMR is a major complication of MI with severe outcome consequences (5 8,24). Therefore, in clinical practice it is essential to detect the presence of IMR. Clinically, detection of IMR on the basis of a murmur is unreliable (5,32), but detection can routinely be done by Doppler echocardiography (31). As routine evaluation of LV function after MI is recommended (33,34) even in asymptomatic patients, detection of IMR does not represent an additional cost burden in the risk stratification of such patients. The risk of adverse cardiac events is directly related to the severity of regurgitation quantitatively assessed by Doppler echocardiography. The quantitative measurements are not always routinely done, but the recent recommendations of the American Society of Echocardiography (31) emphasize this approach in routine practice. Our data support this emphasis by demonstrating the link between quantitative measurements and outcome and support quantitative assessment of IMR progression in future studies. The severe consequences of IMR suggest that it should be the focus of targeted and vigorous treatment. In our series, 23 patients underwent cardiac surgery (with 22 coronary bypass and 6 MR corrections), but the benefit of these procedures cannot be defined, and the notable risk of surgical therapy (35,36) of IMR should lead to prudent clinical implications. Medical (37,38), pacing (39), surgical repair (36), and possibly future catheter-based procedures (40) may all be relevant options. The results of the present study suggest that clinical trials testing these approaches are warranted and that currently individualized therapeutic decisions for IMR are warranted to prevent the occurrence of CHF after MI. Study limitations. We matched patients with and without IMR for age, gender, and LVEF. However, it may be argued that with MR, EF may be higher than expected. Such a conceptual effect is, at most, of low magnitude (41) because of low volume of IMR. Nevertheless, other, less load-dependent indices of LV function, such as wall-stress adjusted EF or end-systolic dimension, did not affect the link between IMR and outcome. Furthermore, even when EF of IMR patients was decreased by four and even eight points, IMR remained independently determinant of subsequent CHF, with, respectively, RR of 3.35 (p ) and 3.0 (p 0.003), demonstrating that uncertainties about EF do not alter the independent effect of IMR on outcome. Ischemic MR may be considered as a surrogate for coronary disease and is the consequence of local LV alterations (42). Differences in coronary disease extent are unlikely in view of the observations in patients with coronary angiograms. Also, we adjusted for all baseline differences in LV size, and systolic Figure 4. Survival free of congestive heart failure or cardiac death (eventfree survival) in asymptomatic patients after myocardial infarction according to the degree of mitral regurgitation measured by effective regurgitant orifice (ERO) 20 mm 2 (continuous line), 1to19mm 2 (dashed line), and absent mitral regurgitation (ERO 0) (dotted line) at diagnosis. The event-free survival rates at five years are indicated the standard error.

7 266 Grigioni et al. JACC Vol. 45, No. 2, 2005 Ischemic MR and CHF January 18, 2005:260 7 and diastolic function, which did not reduce the independent effect of IMR on outcome. Furthermore, the physiologic link between IMR and filling pressures (28) and the outcome link between degree and clinical outcome of IMR support the fact that IMR is the direct determinant of the high CHF rate and justifies consideration of therapy directly aimed at reducing its degree. We could not adjust for all neurohormonal activation markers, but IMR is linked to outcome independently of hyponatremia, which reflects in part this activation (26). In view of our results, future studies should evaluate the link between IMR, hormonal activation, and outcome. Various therapies affect the outcome of post-mi patients, but their utilizations were similar in patients with and without IMR. However, to ensure that these therapies did not affect our results, we forced them into the model predictive of subsequent CHF. Statin, beta-blockade, aspirin, or angiotensin-converting enzyme inhibitor use did not reduce the impact of IMR on CHF occurrence (all RR 3.0; all p 0.001). Therefore, the impact of IMR on outcome does not appear related to concomitant treatment. CONCLUSIONS Patients with previous MI and no or minimal symptoms are at high risk for subsequent CHF. Patients with IMR, assessed by Doppler echocardiography in routine practice, are at markedly increased risk of CHF, independent of all other clinical and LV characteristics. Moreover, the risk of CHF and CD is directly related to IMR severity as it is quantitatively defined. Particularly, patients with ERO 20 mm 2 incur a risk of CHF more than four times higher than patients without IMR, with an absolute rate close to 70% at five years. Therefore, detecting and quantifying IMR is essential after MI, and prospective randomized trials are needed to analyze the effect of treatment in preventing IMR complications. Reprint requests and correspondence: Dr. Maurice Enriquez- Sarano, Division of Cardiovascular Diseases and Internal Medicine, Mayo Clinic, 200 First Street SW, Rochester, Minnesota sarano.maurice@mayo.edu. REFERENCES 1. Kannel WB. Epidemiology and prevention of cardiac failure: Framingham Study insights. Eur Heart J 1987;8 Suppl F: The SOLVD Investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med 1991;325: Pfeffer M, Braunwald E, Moye L, et al. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 1992;327: The SOLVD Investigators. Effect of enalapril on mortality and the development of heart failure in asymptomatic patients with reduced left ventricular ejection fractions. N Engl J Med 1992;327: Lamas G, Mitchell G, Flaker G, et al. Clinical significance of mitral regurgitation after acute myocardial infarction. Circulation 1997;96: Hickey M, Smith L, Muhlbaier L, et al. Current prognosis of ischemic mitral regurgitation. Implications for future management. Circulation 1988;78:I Grigioni F, Enriquez-Sarano M, Zehr K, Bailey K, Tajik A. Ischemic mitral regurgitation: long-term outcome and prognostic implications with quantitative Doppler assessment. Circulation 2001;103: Ellis SG, Whitlow PL, Raymond RE, Schneider JP. Impact of mitral regurgitation on long-term survival after percutaneous coronary intervention. Am J Cardiol 2002;89: Rosenbaum PR, Rubin D. Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. Am Statistician 1985;39: Ho K, Anderson K, Kannel W, Grossman W, Levy D. Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation 1993;88: Ling H, Enriquez-Sarano M, Seward J, et al. Clinical outcome of mitral regurgitation due to flail leaflets. N Engl J Med 1996;335: Rokey R, Sterling L, Zoghbi W, et al. Determination of regurgitant fraction in isolated mitral or aortic regurgitation by pulsed Doppler two-dimensional echocardiography. J Am Coll Cardiol 1986;7: Enriquez-Sarano M, Bailey K, Seward J, Tajik A, Krohn M, Mays J. Quantitative Doppler assessment of valvular regurgitation. Circulation 1993;87: Enriquez-Sarano M, Seward J, Bailey K, Tajik A. Effective regurgitant orifice area: a noninvasive Doppler development of an old hemodynamic concept. J Am Coll Cardiol 1994;23: Recusani F, Bargiggia G, Yoganathan A, et al. A new method for quantification of regurgitant flow rate using color Doppler flow imaging of the flow convergence region proximal to a discrete orifice. An in vitro study. Circulation 1991;83: Enriquez-Sarano M, Miller FJ, Hayes S, Bailey K, Tajik A, Seward J. Effective mitral regurgitant orifice area: clinical use and pitfalls of the proximal isovelocity surface area method. J Am Coll Cardiol 1995;25: Vandervoort P, Rivera J, Mele D, et al. Application of color Doppler flow mapping to calculate effective regurgitant orifice area. An in vitro study and initial clinical observations. Circulation 1993;88: Dujardin K, Enriquez-Sarano M, Bailey K, Nishimura R, Seward J, Tajik A. Grading of mitral regurgitation by quantitative Doppler echocardiography calibration by left ventricular angiography in routine clinical practice. Circulation 1997;96: Tavazzi L, Volpi A. Remarks about postinfarction prognosis in light of the experience with the Gruppo Italiano per lo Studio della Sopravvivenza nell Infarto Miocardico (GISSI) trials. Circulation 1997;95: Senni M, Tribouilloy CM, Rodeheffer RJ, et al. Congestive heart failure in the community: trends in incidence and survival in a 10-year period. Arch Intern Med 1999;159: Braunwald E. Shattuck lecture cardiovascular medicine at the turn of the millennium: triumphs, concerns, and opportunities. N Engl J Med 1997;337: Gillum R. Acute myocardial infarction in the United States, Am Heart J 1987;113: Lehman K, Francis C, Dodge H. Mitral regurgitation in early myocardial infarction. Incidence, clinical detection, and prognostic implications. Ann Intern Med 1992;117: Barzilai B, Davis V, Stone P, Jaffe A. Prognostic significance of mitral regurgitation in acute myocardial infarction. Am J Cardiol 1990;65: Barzilai B, Gessler C, Perez J, Schaab C, Jaffe A. Significance of Doppler-detected mitral regurgitation in acute myocardial infarction. Am J Cardiol 1988;61: Packer M, Lee WH, Kessler PD, Gottlieb SS, Bernstein JL, Kukin ML. Role of neurohormonal mechanisms in determining survival in patients with severe chronic heart failure. Circulation 1987;75:IV Oh J, Ding Z, Gersh B, Bailey K, Tajik A. Restrictive left ventricular diastolic filling identifies patients with heart failure after acute myocardial infarction. J Am Soc Echocardiogr 1992;5:

8 JACC Vol. 45, No. 2, 2005 January 18, 2005:260 7 Grigioni et al. Ischemic MR and CHF Enriquez-Sarano M, Rossi A, Seward J, Bailey K, Tajik A. Determinants of pulmonary hypertension in left ventricular dysfunction. J Am Coll Cardiol 1997;29: MacIsaac A, McDonald I, Kirsner R, Graham S, Tanzer D. Left ventricular energy in mitral regurgitation: a preliminary report. Aust NZ J Med 1992;22: Abramson S, Burke J, Kelly JJ, et al. Pulmonary hypertension predicts mortality and morbidity in patients with dilated cardiomyopathy. Ann Intern Med 1992;116: Zoghbi WA, Enriquez-Sarano M, Foster E, et al. Evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography: a report from the Task Force on Valvular Regurgitation of the American Society of Echocardiography. Am J Echocardiogr In press. 32. Desjardins V, Enriquez-Sarano M, Tajik A, Bailey D, Seward J. Intensity of murmurs correlates with severity of valvular regurgitation. Am J Med 1996;100: Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non ST-segment elevation myocardial infarction summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40: Ryan TJ, Antman EM, Brooks NH, et al update: ACC/AHA guidelines for the management of patients with acute myocardial infarction. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Acute Myocardial Infarction). J Am Coll Cardiol 1999;34: Grossi EA, Bizekis CS, LaPietra A, et al. Late results of isolated mitral annuloplasty for functional ischemic mitral insufficiency. J Card Surg 2001;16: Gillinov AM, Wierup PN, Blackstone EH, et al. Is repair preferable to replacement for ischemic mitral regurgitation? J Thorac Cardiovasc Surg 2001;122: Stevenson L, Bellil D, Grover-McKay M, et al. Effects of afterload reduction (diuretics and vasodilators) on left ventricular volume and mitral regurgitation in severe congestive heart failure secondary to ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 1987; 60: Evangelista-Masip A, Bruguera-Cortada J, Serrat-Serradell R, et al. Influence of mitral regurgitation on the response to captopril therapy for congestive heart failure caused by idiopathic dilated cardiomyopathy. Am J Cardiol 1992;69: St John Sutton MG, Plappert T, Abraham WT, et al. Effect of cardiac resynchronization therapy on left ventricular size and function in chronic heart failure. Circulation 2003;107: Liddicoat JR, Mac Neill BD, Gillinov AM, et al. Percutaneous mitral valve repair: a feasibility study in an ovine model of acute ischemic mitral regurgitation. Catheter Cardiovasc Interv 2003;60: Enriquez-Sarano M, Tajik A, Schaff H, et al. Echocardiographic prediction of left ventricular function after correction of mitral regurgitation: results and clinical implications. J Am Coll Cardiol 1994;24: Yiu S, Enriquez-Sarano M, Tribouilloy C, Seward J, Tajik A. Determinants of the degree of functional mitral regurgitation in patients with systolic left ventricular dysfunction: a quantitative clinical study. Circulation 2000;102: