Restrictive Filling Pattern is a Powerful Predictor of Heart Failure Events Postacute Myocardial Infarction and in Established Heart Failure: A Literature-Based Meta-Analysis

Journal of Cardiac Failure Vol. 13 No. 5 2007

Clinical Investigations

Restrictive Filling Pattern is a Powerful Predictor of Heart Failure Events Postacute Myocardial Infarction and in Established Heart Failure: A Literature-Based Meta-Analysis JITHENDRA B. SOMARATNE, BHB, MBCHB, GILLIAN A. WHALLEY, PhD, MHSc, DMU, GREG D. GAMBLE, MSc, AND ROBERT N. DOUGHTY, MD, FCP, FRACP, FCSANZ Auckland, New Zealand

ABSTRACT Background: Two recent literature-based meta-analyses revealed that restrictive filling pattern (RFP) was associated with a 4-fold increase in the risk of death in patients with heart failure (HF) and postacute myocardial infarction (AMI). This similar but unique analysis evaluated the link between RFP and morbidity. Methods and Results: Prospective echocardiographic studies of patients post-AMI and with HF that reported HF morbidity were identified. Events (post-AMI: development of HF; HF: HF readmission) were compared between patients with and without RFP in both patient groups. Review Manager version 4.2.7 software was used for the analysis. Twelve post-AMI studies (1286 patients, 271 events) and 5 HF studies (647 patients, 176 events) were identified. RFP was associated with HF readmission in the HF patients (OR 2.96 [2.02e4.33] and development of HF post-AMI (OR 10.10 [7.02e14.51]). The event rate in the RFP group was the same regardless of disease category (49% post-AMI, 42% HF); however, RFP was less prevalent in the post-AMI group (22% versus 39%). Conclusions: This literature-based meta-analysis confirms that RFP is a powerful predictor of HF hospitalization in patients with HF and especially the development of HF post-AMI. This is an important prognostic sign and should be incorporated into routine clinical practice. (J Cardiac Fail 2007;13:346e352) Key Words: Diastole, echocardiography, morbidity, prognosis.

estimated to be $29.6 billion in 2006.1 Despite advances in management, morbidity and mortality secondary to HF remain high. Doppler echocardiographic assessment of mitral diastolic filling is commonly applied using a variety of pulsed-wave Doppler techniques, including the assessment of blood flow velocity and pattern across the mitral valve.2 In a normal heart, passive diastolic ventricular filling occurs early in diastole in response to ventricular suction and an atrioventricular pressure gradient3,4 and comprises the majority of ventricular filling. The velocity is directly related to the pressure gradient between the 2 chambers, but the deceleration slope is more closely related to rate of pressure equalization and chamber stiffness. The time for pressure equalization (mitral deceleration time) is thus an important marker of diastolic ventricular function, and in the setting of high filling pressures and increased chamber stiffness early filling terminates abruptly, resulting in a short deceleration time (Fig. 1). Thus short deceleration time (!140

Heart failure (HF) is a significant and ever increasing public health problem globally. In the United States, according to National Hospital Discharge Survey figures, hospital admissions for a principal diagnosis of HF increased from 399,000 in 1979 to 1,093,000 in 2003.1 Furthermore, the direct and indirect cost of HF to the US health system is From the Department of Medicine, School of Medicine, Faculty of Medical and Health Sciences, The University of Auckland, Auckland, New Zealand. Manuscript received August 11, 2006; revised manuscript received January 23, 2007; revised manuscript accepted January 25, 2007. Reprint requests: Gillian Whalley, PhD, MHSc, DMU, Department of Medicine, Faculty of Medicine and Health Sciences, The University of Auckland, Private Bag 92 019, Auckland, New Zealand. Dr. Jithendra Somaratne is supported by the Greenlane Educational and Research Fund and the Douglas Goodfellow Medical Research Fellowship from the Auckland Medical Research Foundation. Dr. Gillian Whalley is currently the National Heart Foundation of New Zealand Senior Fellow. 1071-9164/$ - see front matter Ó 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.cardfail.2007.01.010

346

Restrictive Filling and Heart Failure Events

Grade MFP

0 Normal

1 Abnormal

2 Pseudonormal



Somaratne et al

3 Restrictive

MV E/A

Between 1 and 2

<1

Between 1 and 2

>1.5

MV DT

>140ms

>220ms

>140ms

<140ms

TDI PV PWD MV Valsalva

Ea/Aa >1 Amitral

ARpulmonary

Unchanged E/A

Ea/Aa <1

Ea/Aa <1 Amitral

ARpulmonary

Amitral < ARpulmonary E/A

Unchanged E/A

347

Ea/Aa <1 Amitral < ARpulmonary Reversible: E/A Non-reversible: unchanged E/A

Fig. 1. Diastolic filling grades. MFP, mitral filling pattern; MV, mitral valve; E, peak early transmitral flow velocity; A, peak transmitral flow velocity during atrial contraction; DT, mitral E wave deceleration time; TDI, tissue Doppler imaging; PV, pulmonary venous; PWD, pulmonary venous Doppler; Ea, peak early diastolic velocity of mitral annulus; Aa, peak diastolic velocity of mitral annulus during atrial contraction; Amitral, duration of mitral A wave; ARpulmonary, duration of pulmonary venous atrial reversal.

ms) represents the most severe form of diastolic filling abnormality, the so-called restrictive filling pattern.3,4 Restrictive filling pattern has been linked with adverse outcomes in patients with HF and postacute myocardial infarction (postAMI). Two recent literature-based meta-analyses demonstrated that the presence of the restrictive filling pattern (RFP) was associated with an odds ratio for death of 4.10 (95% CI 3.38e4.99) in post-AMI5 patients and 4.36 (95% CI 3.60e5.29) in HF6 patients. Although the association between mortality and RFP is clearly established, the relationship between HF morbidity and RFP in both patient groups is less so. Studies in post-AMI patients7e18 and in established HF19e23 have demonstrated a link between RFP and HF morbidity. However, the reported odds ratios for HF morbidity have been wide ranging, with variable event rates and statistical significance. Furthermore, many of these studies were powered for combined end points (death and morbidity) and thus were too small to clearly establish an independent link between morbidity and RFP. This literature-based meta-analysis aims to more precisely assess the impact of RFP on the risk of developing de novo HF post-AMI and acute decompensation in HF patients. Methods

asked to confirm the extracted data. If the reported data were insufficient, the authors were asked to provide the missing information via e-mail or facsimile. Additional data were requested from the investigators of 3 of the 12 post-AMI studies and 2 of the 5 HF studies. Definition of Restrictive Mitral Filling RFP was defined within each study according to site-specific criteria (Tables 1 and 2). Although there was some regional and institutional variation in definitions of RFP, all were deemed acceptable and within normal variation of clinical echocardiography. End Points The primary end point was HF events. HF events were defined as development of HF in the post-AMI group and HF readmissions in the HF group. In 7 of the post-AMI studies, the end point used was readmission to hospital for HF, in the remaining 5 studies the end points included: development of HF symptoms (2 studies); HF symptoms $ New York Heart Association Class II, Killip Class $ II, and Killip Class $ III (Table 1). In the HF cohort, the end point was hospitalization from HF, with some minor variations. One study19 reported a combined end point of hospital admission or worsening symptomatic status and another20 reported all hospitalizations, which included hemodynamic decompensation, sustained arrhythmias, infections, and peripheral ischemic events. In both cases, no further breakdown was possible and thus all events were included.

Identification of Studies

Statistical Methods

The methodologic approach to this literature-based meta-analysis is similar to that described in greater detail previously.5,6 Briefly, published studies were identified by searching several online databases. Prospective studies which enrolled post-AMI and HF patients and reported HF events according to mitral filling pattern were included. Most of the required data was extracted from the published articles, and each investigator was contacted and

The Cochrane Collaboration Program Review Manager 4.2.7 was used for analysis. For each study, patients were stratified according to the individual study criteria as restrictive or nonrestrictive. The number of patients and the number of events allocated to each group were recorded. The odds ratio using a fixed effects model is presented, but a random effects model was also evaluated. Because the latter was not different from the former, we

AMI, acute myocardial infarction; DT, deceleration time of passive mitral filling; E/A, ratio of early to late mitral filling; EF, ejection fraction; E/Vp, ratio of peak early mitral filling to flow propagation velocity; F/U, follow-up; HF, heart failure; N/A, not available; RFP, restrictive filling pattern; SD, standard deviation.

E/A $1 or 2 and E/Vp $1.5 DT !140 5.5 28 48 !35 10/91 (11) 15/225 (7) 2004 2004 Kinova Møller 2

Bulgaria Denmark

Numbers in article Author provided numbers

62 67

72 h At randomization

12 52 7/33 (21) 2004 Karvounis

Greece

60/125 (48) 14/104 (13) 23/88 (26) Numbers in article Numbers in article Numbers in abstract 2000 2001 2002 Møller 1 Cerisano Brzezninska

Denmark Italy Poland

(36) (7) (14) (14) 73/205 4/58 6/27 14/102 Author provided numbers Numbers in article Numbers in abstract Numbers in article 1997 1999 1999 2000 Sakata Poulsen Tsai Burgess

Japan Denmark Taiwan United Kingdom

Author provided numbers

55

24 h

In-hospital 32 12 67 61 54

Killip class $II HF readmission Development of HF symptoms Development of HF symptoms HF readmission HF readmission

6h 3d At discharge 53 40 N/A

62 61 N/A 66 HF readmission HF readmission NYHA Class $II HF readmission

49 47 N/A 44

On admission to CCU 24 h 7d 4d

60 12 30 11

E/A O1.6 and DT !180 E/A $2 or E/A 1e2 and DT !140 A velocity 2 SD below mean DT #140 DT #125 E/A O2 or E/A 1e2 with DT #140 DT!140 DT #130 E/A $2 or E/A $1.5 with DT !140 E/A O2 In-hospital 32 24 hrs 24 hrs 30/133 (23) 5/95 (5) 1997 1997 Garcia-Rubira Nijland

Spain Netherlands

Numbers in article Numbers in article

44 49 62 61 Killip Class $III HF readmission

Echo timing post-AMI EF % Age (y) Event Type HF Events/n (%) Data source Country Year First Author

Table 1. Included Postacute Myocardial Infarction Studies

F/U (mo)

Definition of RFP

348 Journal of Cardiac Failure Vol. 13 No. 5 June 2007 only present the fixed-effects model. The program automatically weights the individual studies according to sample size and size of effect. Standard tests for heterogeneity including, chi-square (presented), I2, and funnel plots were performed.

Results AMI Studies

A total of 1286 patients with 271 events from 12 studies7e18 are included (Table 1). The pooled average age was 63 years with a range of 54e67 years and pooled average ejection fraction (EF) was 45%: mean age was not available in 1 study11 and average EF was inaccessible in 2 studies.11,15 One study18 recruited patients with an EF !35%, but did not provide an average EF. Follow-up was restricted to the duration of hospital admission in 2 studies.7,13 In the remaining studies, it varied between 5.5 and 60 months, with a mean of 23.5 months. The timing of the initial echocardiogram post-AMI varied from ‘‘on admission to CCU’’ to ‘‘at discharge’’. In 7 studies the echocardiogram was done in less than 24 hours after the index AMI (Table 1). A total of 287 (22%) patients had an RFP at baseline. There were 140 HF events (49%) in the RFP group and 131 HF events (13%) in the non-RFP group, resulting in an odds ratio of 10.10 (95% CI 7.02e14.51) (Fig. 2). No significant heterogeneity was observed (chi-square 5 13.67, P 5 .25, I2 5 19.5%) (Fig. 2). HF Studies

A total of 647 patients with 221 events from 5 studies19e23 are included (Table 2). Patients were recruited from outpatient HF clinics,19,21 the echocardiography laboratory,22 and inpatient services.23 The mean age was 64 years, with a range of 52 to 73 years. The pooled average EF was 30%, with a range of 24% to 34% (Table 2). Follow-up ranged from 10 to 22 months, with a mean of 15.8 months. At baseline, 250 (39%) patients had the RFP. There were 104 HF events (42%) in the RFP group and 72 HF events (18%) in the non-RFP group, producing an odds ratio of 2.96 (95% CI 2.02e4.33) (Fig. 3). There was no evidence of heterogeneity between studies (chi-square 5 2.28, P 5 .68, I2 5 0%) (Fig. 3). Discussion The relationship between restrictive filling pattern and death is well established in post-AMI patients5 and patients with established HF.6 On this premise, we sought to clarify the relationship between RFP and HF morbidity, specifically the development or exacerbation of HF. These metaanalyses bring together data from nearly 2000 patients and have clearly demonstrated that RFP is a strong predictor of future HF in post-AMI patients and HF hospitalization in those with existing HF. This unique analysis of HF morbidity in these 2 high-risk patient groups highlights

Restrictive Filling and Heart Failure Events

Study Garcia-Rubira Nijland Sakata Poulsen Tsai Burgess Moller 1 Cerisano Brzezninska Karvounis Kinova Moller 2 Events/Total (95% CI)

Restrictive Filling n/N 11/26 3/12 42/52 3/14 5/10 6/19 24/26 11/34 8/9 5/6 8/35 14/44 140/287

Non-Restrictive Filling n/N

OR (fixed) 95% CI

Weight %

19/107 2/83 31/153 1/44 1/17 8/83 36/99 3/70 15/79 2/27 2/56 11/181

24.44 2.16 17.24 2.16 2.11 11.63 6.57 7.57 1.94 0.69 6.77 16.72

131/999

100.00



Somaratne et al OR (fixed) 95% CI

3.40 [1.35, 13.50 [1.98, 16.53 [7.47, 11.73 [1.11, 16.00 [1.50, 4.33 [1.29, 21.00 [4.69, 10.68 [2.74, 34.13 [3.96, 62.50 [4.71, 8.00 [1.59, 7.21 [2.99,

8.54] 91.82] 36.58] 123.96] 171.20] 14.53] 94.07] 41.68] 294.07] 829.26] 40.30] 17.39]

349 Year 1997 1997 1997 1999 1999 2000 2000 2001 2002 2004 2004 2004

10.10 [7.02, 14.51]

Test for heterogeneity: Chi2 = 13.67, df = 11 (P = 0.25), I2 = 19.5% Test for overall effect: Z = 12.49 (P < 0.00001) 0.001 0.01

0.1

Restrictive Filling Better

1

10

100 1000

Restrictive Filling Worse

Fig. 2. Postacute myocardial infarction forest plot. CI, confidence interval; df, degrees of freedom; HF, heart failure; OR, odds ratio.

the important role of Doppler echocardiography to further aid risk assessment and thus management. Although both were highly significant, there was a striking relationship between RFP and development of HF in post-AMI patients with an odds ratio of 10 associated with the presence of restrictive mitral filling. The development of early or late HF post-AMI, the natural consequence of significant myocardial damage, portends a poor prognosis.23 This meta-analysis has demonstrated that it may possible to identify a group of patients at particularly high risk of developing HF subsequent to the initial ischemic event. Importantly, the incidence of HF post-AMI is not rare. A recent literature review of 9 studies with consecutive AMI patients reported a mean incidence of in-hospital HF of 36% (range 19% to 51%).24 Early development of HF portends a higher in-hospital25e27 and 1-year25,28,29 mortality. In the CARE Study,30 6.3% of AMI patients with no previous HF developed late-onset ($3 months after the index AMI) HF and this late-onset HF was found to be associated with a 10-fold increase in 5-year mortality. Furthermore, patients with HF post-AMI are also at high risk of future HF readmissions.31e33 Thus the use of echocardiography to identify these patients (approximately 20% of all AMI patients) early in their clinical course may offer substantial management benefits. The diagnosis of HF continues to be associated with poor quality of life, high morbidity, and mortality despite contemporary HF management.34e40 After being admitted to the hospital, patients experience higher rates of unplanned HF hospitalization37,40 and mortality.35e40 A study of HF patients in Australia and Scotland found previous hospitalization for HF was associated with an OR of 2.3 (95% CI 1.11e4.8) for unplanned readmission or out-ofhospital death within 3 months of initial HF hospitalization.37 One-year mortality rates after initial hospital admission for HF have been reported between 25 and 35 %.39,40 This meta-analysis has identified a subgroup of HF patients (those with restrictive mitral filling) who are at high risk of

hospitalization. It should be noted that this is not a small percentage of patients: approximately 40% of this cohort had RFP at baseline and 42% of those with RFP were rehospitalized. These 2 meta-analyses have demonstrated that RFP is a predictor of HF disease progression and represents an important similarity between 2 different patient groups. Although the 2 cohorts were similar in terms of age (63 versus 64 years), the pooled average EF was much higher in the post-AMI group (45% versus 30%) and the prevalence of the RFP in the post-AMI cohort was almost half that of the HF cohort (22% versus 39%). But, most importantly, when RFP was identified, the event rate was similar regardless of disease category: 49% in the post-AMI cohort and 42% in the HF cohort. This similarity is especially remarkable, because it supports the concept that RFP per se may be an important predictor of morbidity, rather than underlying myocardial damage, systolic impairment, or LV dilation that might be present in some individual patients. The higher OR in the post-AMI cohort compared with the HF cohort (10.10 versus 2.96) reflects both the lower event rate in the non-RFP group of the post-AMI cohort (13%); and the slightly higher event rate in the RFP group of the post-AMI cohort which may be due to unrecognized HF in this group. This draws attention to the greater prognostic value of the RFP in post-AMI patients: 1 in 2 postAMI patients with RFP will develop HF. Thus identification of RFP in post-AMI patients is an ominous sign because half of this group will develop HF. RFP is therefore a very important prognostic indicator in this group of patients. Restrictive filling pattern is related to elevated LV filling pressure.41e44 Moreover, it is a surrogate measure of elevated filling pressure and may demonstrate hemodynamic instability. RFP is also associated with higher neurohormone levels.45e48 Because BNP level is linked with mortality, hospitalizations, and worsening HF, it is logical that RFP, in these patients, is also related to outcome.49e55

CM, cardiomyopathy; DT, deceleration time of passive mitral filling; E/A, ratio of early to late mitral filling; EF, ejection fraction; F/U, follow-up; HF, heart failure; IVRT, isovolumic relaxation time; N/A, not available; RFP, restrictive filling pattern.

52 70 70 73 1997 2000 2002 2002 Pozzoli Bettencourt Dini Whalley

Italy Portugal Italy NZ

Numbers in article Author provided numbers Author provided numbers Numbers in article

69/173 30/97 37/207 31/115

(40) (31) (18) (27)

HF HF HF HF

readmission readmission readmission readmission

24 34 32 32

N/A 53 68 54

17 18 22 10

IVRT #50, filling velocity O105, DT !150 E/A O1 and DT #130 DT !130 DT !140 E/A O2 and DT !140 12 67 27 55 HF readmission 1995 Belardinelli

Italy

Numbers in article

9/55 (16)

Age (y) HF Event Type HF Events/n (%) Data Source Country Year First Author

Table 2. Included Heart Failure Studies

EF %

Ischemic CM %

F/U (mo)

Definition of RFP

350 Journal of Cardiac Failure Vol. 13 No. 5 June 2007 Similarly, it is no surprise that RFP may predict further hemodynamic instability and development of HF. This is supported by the findings of this study that the presence of RFP is a strong predictor of future HF events, especially in postAMI patients. Many of the studies linking RFP with elevated filling pressure and outcome have been conducted in HF populations with impaired systolic function. The HF patients in the current analysis display depressed EF, but, conversely, mean EF in the post-AMI studies is higher. Nevertheless, RFP remains a strong predictor of outcome in the post-AMI cohort. The literature-based approach adopted in this analysis precluded an evaluation of the individual and multivariate contribution of all echocardiographic and clinical variables to overall risk. However, an individual patient meta-analysis, pooling individual patient data from each included study, would potentially have the power to determine the independence of the RFP from systolic function, other echo parameters and clinical variables. Identification of the RFP is an accessible marker of prognosis which can be obtained during routine clinical echocardiography. Reversal of the RFP by means of preload reduction could be a target for intensified therapy. Patients with the RFP who respond to therapy (reversible RFP) have been shown to have better outcomes compared with those with an irreversible RFP.56,57 Limitations

Meta-analyses have many inherent limitations, including publication bias. Unpublished data, which may or may not be in agreement with these results, may have been omitted. Unpublished studies are often negative and so the omission of such studies may lead to exaggeration of the risk estimate. To minimize publication bias, all studies that included prognosis and diastolic parameters were identified and the search was not limited to RFP. Authors were contacted and asked if it were possible to breakdown their data on the basis of RFP versus non-RFP. Furthermore, all published authors were contacted to request any unpublished data. A further possible bias is duplication of patients. Because our search strategy was wide and encompassing, several publications containing duplicate patients with different echocardiographic variables were identified. After consultation with the original authors, duplicate patient cohorts were identified and excluded. We believe this rigorous methodologic approach has minimized these potential sources of bias and error. The timing of echocardiography may be important. A study of post-AMI patients with serial Doppler echocardiographic measurements demonstrated the greater prognostic value of detecting the RFP at approximately 2 weeks after the index AMI in comparison to within 2 days of the index AMI.58 In this meta-analysis, echocardiographic examinations were performed on postAMI patients between admission and 7 days after the index AMI. Similarly, the HF patients were studied when they were deemed clinically stable. Unfortunately, given the

Restrictive Filling and Heart Failure Events

Study Belardinelli Pozzoli Bettencourt Dini Whalley

Restrictive Filling n/N

Non-Restrictive Filling n/N

7/25 52/98 20/47 14/53 11/27

2/30 17/75 10/50 23/154 20/88

104/250

Events/Total (95% CI) 2

OR (fixed) 95% CI

Weight % 4.34 29.99 18.47 28.75 18.46

72/397

100.00



Somaratne et al OR (fixed) 95% CI

5.44[1.02, 3.86[1.97, 2.96[1.20, 2.04[0.96, 2.34[0.94,

29.19] 7.54] 7.31] 4.35] 5.84]

351 Year 1995 1997 2000 2002 2002

2.96[2.02, 4.33]

2

Test for heterogeneity: Chi = 2.28, df = 4 (P = 0.68), I = 0% Test for overall effect: Z = 5.60 (P < 0.00001) 0.1 1 0.01 Restrictive Filling Better

10 100 Restrictive Filling Worse

Fig. 3. Heart failure forest plot. CI, confidence interval; df, degrees of freedom; HF, heart failure; OR, odds ratio.

limitations of literature-based meta-analyses discussed previously, the effect of timing of echocardiography on risk estimates could not be accounted for. The criteria used by individual investigators for classification of RFP varied slightly. Importantly, none was deemed to fall far outside what might be considered international standards. In many cases, this was predetermined by the investigators to be the best cutoff for detecting atrisk subjects. This may have influenced the results, but we do not think this was the case because the variation was only slight. Follow-up was restricted to the duration of the index hospital admission in 2 studies.7,13 This was markedly different to the other studies in this cohort, which had a mean followup of nearly 2 years. A separate analysis, excluding these 2 studies, revealed the following event rates: 45% (105/235) in the RFP group and 10% (76/793) in the non-RFP group. The prevalence of the RFP was unchanged (23%) and the odds ratio for HF events was similar (OR 11.43 [95% CI 7.52e17.52]). One study20 reported readmission from multiple causes and it was not possible to isolate true HF readmission from the data. A sensitivity analysis excluding this study yielded a similar OR for HF events (OR 2.57 [95% CI 1.62e4.10]). The size of the risk estimates and confidence intervals around the risk estimates in conjunction with the sample size, would suggest that these potential sources of error, although possible, are likely to have minimal effect on the overall risk estimates. Conclusion Restrictive mitral filling is associated with a 10-fold increase in the odds of developing HF post-AMI and a 3fold increase in the odds of hospitalization in HF patients. The current focus of echocardiographic diagnostic and risk stratification strategies, in both post-AMI and HF settings, is systolic function. However, these analyses suggest that the presence of restrictive filling may be used to identify patients at very high risk of experiencing new HF or worsening HF. Importantly, after RFP is identified, the group event rate is similar whether they have existing HF or have recently experienced an AMI. Interestingly, because of the lower event rate in the post-AMI patients

without restrictive filling, the odds ratio of the RFP is much higher than in the HF cohort. As such, this literature-based meta-analysis confirms that RFP is a powerful predictor of HF development, especially in post-AMI patients, and should be incorporated into routine clinical practice.

References 1. Thom T, et al. Heart disease and stroke statisticsd2006 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 2006;113:e85ee151. 2. Oh JK, et al. The noninvasive assessment of left ventricular diastolic function with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr 1997;10:246e70. 3. Garcia MJ, Thomas JD, Klein AL. New Doppler echocardiographic applications for the study of diastolic function. JACC 1998;32:865e75. 4. Nishimura RA, Tajik AJ. Evaluation of diastolic filling of left ventricle in health and disease: Doppler echocardiography is the clinician’s Rosetta stone. JACC 1997;30:8e18. 5. Whalley GA, Gamble GD, Doughty RN. Restrictive diastolic filling predicts death post acute myocardial infarction: a meta-analysis of prospective studies. Heart 2006;92:1588e94. 6. Whalley GA, Gamble GD, Doughty RN. The prognostic significance of restrictive diastolic filling associated with heart failure: a meta-analysis. Int J Cardiol 2007;116:70e7. 7. Garcia-Rubira JC, et al. Abnormal filling pattern of the left ventricle and outcome in acute myocardial infarction. Int J Cardiol 1997;61: 143e9. 8. Nijland F, et al. Prognostic implications of restrictive left ventricular filling in acute myocardial infarction: a serial Doppler echocardiographic study. JACC 1997;30:1618e24. 9. Sakata K, et al. Prognostic value of Doppler transmitral flow velocity patterns in acute myocardial infarction. Am J Cardiol 1997;79:1165e9. 10. Poulsen SH, Jensen SE, Egstrup K. Longitudinal changes and prognostic implications of left ventricular diastolic function in first acute myocardial infarction. Am Heart J 1999;137:910e8. 11. Tsai WC, et al. Prognostic value of Doppler-derived mitral deceleration time in postinfarction patients with left ventricular ejection fractions of 35% or more. J Formosan Med Assoc 1999;98:70e2. 12. Burgess MI, Atkinson P, Ray SG. Restrictive left ventricular filling pattern after myocardial infarction: significance of concomitant preserved systolic function. Echocardiography 2000;17:659e64. 13. Møller JE, et al. Pseudonormal and restrictive filling patterns predict left ventricular dilation and cardiac death after a first myocardial infarction: a serial color M-mode Doppler echocardiographic study. JACC 2000;36:1841e6. 14. Cerisano G, et al. Prognostic implications of restrictive left ventricular filling in reperfused anterior acute myocardial infarction. JACC 2001; 37:793e9.

352 Journal of Cardiac Failure Vol. 13 No. 5 June 2007 15. Brzezninska B, et al. [Prognostic value of a model using restricted left ventricular filling for prediction of developing heart failure after myocardial infarction]. Przeglad Lekarski 2002;59:583e6. 16. Karvounis HI, et al. Evaluation of a Doppler-derived index combining systolic and diastolic left ventricular function in acute myocardial infarction. Angiology 2004;55:21e8. 17. Kinova E, Kozhuharov H. Left ventricular diastolic filling patterns as predictors of heart failure after myocardial infarction: a colour M-mode Doppler study. Hellen J Cardiol 2004;45:23e31. 18. Møller JE, et al. Effects of losartan and captopril on left ventricular systolic and diastolic function after acute myocardial infarction: results of the Optimal Trial in Myocardial Infarction with Angiotensin II Antagonist Losartan (OPTIMAAL) echocardiographic substudy. Am Heart J 2004;147:494e501. 19. Belardinelli R, et al. Exercise training improves left ventricular diastolic filling in patients with dilated cardiomyopathy. Clinical and prognostic implications. Circulation 1995;91:2775e84. 20. Pozzoli M, et al. Loading manipulations improve the prognostic value of Doppler evaluation of mitral flow in patients with chronic heart failure. Circulation 1997;95:1222e30. 21. Bettencourt P, et al. Predictors of prognosis in patients with stable mild to moderate heart failure. J Cardiac Failure 2000;6:306e13. 22. Dini FL, et al. Prognostic value of left atrial enlargement in patients with idiopathic dilated cardiomyopathy and ischemic cardiomyopathy. Am J Cardiol 2002;89:518e23. 23. Whalley GA, et al. Pseudonormal mitral filling pattern predicts hospital re-admission in patients with congestive heart failure. JACC 2002; 39:1787e95. 24. Hellermann JP, et al. Heart failure after myocardial infarction: a review. Am J Med 2002;113 324e30. 25. Emanuelsson H, Karlson BW, Herlitz J. Characteristics and prognosis of patients with acute myocardial infarction in relation to occurrence of congestive heart failure. Eur Heart J 1994;15:761e8. 26. Ali AS, et al. Clinical predictors of heart failure in patients with first acute myocardial infarction. Am Heart J 1999;138 1133e9. 27. Altun A, et al. Heart failure during first inferior acute myocardial infarction. Coron Artery Dis 1999;10:455e8. 28. Vaur L, et al. Epidemiology of myocardial infarction in France: therapeutic and prognostic implications of heart failure during the acute phase. Am Heart J 1999;137:49e58. 29. Nicod P, et al. Influence on prognosis and morbidity of left ventricular ejection fraction with and without signs of left ventricular failure after acute myocardial infarction. Am J Cardiol 1988;61:1165e71. 30. Lewis EF, et al. Predictors of late development of heart failure in stable survivors of myocardial infarction: the CARE study. J Am Coll Cardiol 2003;42:1446e53. 31. The Capricorn I. Effect of carvedilol on outcome after myocardial infarction in patients with left-ventricular dysfunction: the CAPRICORN randomised trial. Lancet 2001;357:1385e90. 32. Pfeffer MA, et al. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. Results of the survival and ventricular enlargement trial. The SAVE Investigators. N Engl J Med 1992;27:669e77. 33. Pitt B, et al. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 2003;348:1309e21. 34. Stewart S, et al. Trends in hospitalization for heart failure in Scotland, 1990e1996. An epidemic that has reached its peak? Eur Heart J 2001; 22:209e17. 35. Cowie MR, et al. Survival of patients with a new diagnosis of heart failure: a population based study. Heart 2000;83:505e10. 36. Blackledge HM, Tomlinson J, Squire IB. Prognosis for patients newly admitted to hospital with heart failure: survival trends in 12 220 index admissions in Leicestershire 1993e2001. Heart 2003;89: 615e20. 37. Stewart S, et al. Poles apart, but are they the same? A comparative study of Australian and Scottish patients with chronic heart failure. Eur J Heart Failure 2001;3:249e55.

38. Stewart S, et al. More ‘malignant’ than cancer? Five-year survival following a first admission for heart failure. Eur J Heart Failure 2001;3: 315e22. 39. Schaufelberger M, et al. Decreasing one-year mortality and hospitalization rates for heart failure in Sweden: data from the Swedish Hospital Discharge Registry 1988 to 2000. Eur Heart J 2004;25:300e7. 40. Lee DS, et al. Trends in heart failure outcomes and pharmacotherapy: 1992 to 2000. Am J Med 2004;116:581e9. 41. Appleton CP, et al. Estimation of left ventricular filling pressures using two-dimensional and Doppler echocardiography in adult patients with cardiac disease. Additional value of analyzing left atrial size, left atrial ejection fraction and the difference in duration of pulmonary venous and mitral flow velocity at atrial contraction. JACC 1993;22:1972e82. 42. Nishimura RA, et al. Noninvasive Doppler echocardiographic evaluation of left ventricular filling pressures in patients with cardiomyopathies: a simultaneous Doppler echocardiographic and cardiac catheterization study. JACC 1996;28:1226e33. 43. Yamamoto K, et al. Determination of left ventricular filling pressure by Doppler echocardiography in patients with coronary artery disease: critical role of left ventricular systolic function. JACC 1997;30:1819e26. 44. Pozzoli M, et al. Doppler echocardiography reliably predicts pulmonary artery wedge pressure in patients with chronic heart failure with and without mitral regurgitation. JACC 1996;27:883e93. 45. Yu CM, et al. Diastolic dysfunction and natriuretic peptides in systolic heart failure. Higher ANP and BNP levels are associated with the restrictive filling pattern. Eur Heart J 1996;17 1694e702. 46. Margulies KB, et al. Physiological significance of early deceleration time prolongation in asymptomatic elderly subjects. J Card Failure 1999;5:92e9. 47. Akioka K, et al. Prognostic value of Doppler transmittal flow patterns and cardiac natriuretic peptides in patients with chronic congestive heart failure admitted for episodes of acute decompensation. Heart Vessels 2000;15:53e60. 48. Patrianakos AP, et al. Restrictive filling pattern is associated with increased humoral activation and impaired exercise capacity in dilated cardiomyopathy. Eur J Heart Failure 2004;6:735e43. 49. Fisher C, et al. N-terminal pro B type natriuretic peptide, but not the new putative cardiac hormone relaxin, predicts prognosis in patients with chronic heart failure. Heart 2003;89:879e81. 50. Isnard R, et al. Combination of B-type natriuretic peptide and peak oxygen consumption improves risk stratification in outpatients with chronic heart failure. Am Heart J 2003;146:729e35. 51. Anand ISM, et al. Changes in brain natriuretic peptide and norepinephrine over time and mortality and morbidity in the Valsartan Heart Failure Trial (Val-HeFT). Circulation 2003;107:1278e83. 52. Gwechenberger M, et al. Interleukin-6 and B-type natriuretic peptide are independent predictors for worsening of heart failure in patients with progressive congestive heart failure. J Heart Lung Transplant 2004;23:839e44. 53. Latini R, et al. The comparative prognostic value of plasma neurohormones at baseline in patients with heart failure enrolled in Val-HeFT. Eur Heart J 2004;25:292e9. 54. Logeart D, et al. Predischarge B-type natriuretic peptide assay for identifying patients at high risk of re-admission after decompensated heart failure. JACC 2004;43:635e41. 55. de Groote P, et al. B-type natriuretic peptide and peak exercise oxygen consumption provide independent information for risk stratification in patients with stable congestive heart failure. JACC 2004;43:1584e9. 56. Temporelli PL, et al. Reversible restrictive left ventricular diastolic filling with optimized oral therapy predicts a more favorable prognosis in patients with chronic heart failure. JACC 1998;31:1591e7. 57. Pinamonti B, et al. Persistence of restrictive left ventricular filling pattern in dilated cardiomyopathy: an ominous prognostic sign. JACC 1997;29:604e12. 58. Temporelli PL, et al. Doppler-derived mitral deceleration time as a strong prognostic marker of left ventricular remodeling and survival after acute myocardial infarction: results of the GISSI-3 echo substudy. JACC 2004;43:1646e53.