Skip Navigation

European Journal of Heart Failure 2008 10(8):765-771; doi:10.1016/j.ejheart.2008.05.015
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Najafi, F.
Right arrow Articles by Jamrozik, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Najafi, F.
Right arrow Articles by Jamrozik, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2008 European Society of Cardiology

Late-onset heart failure after myocardial infarction: Trends in incidence and survival

Farid Najafia,*, Annette J. Dobsonb, Michael Hobbsc and Konrad Jamrozikd

a School of Population Health, Kermanshah University of Medical Sciences Kermanshah, Iran
b School of Population Health, University of Queensland Australia
c School of Population Health, University of Western Australia Australia
d School of Population Health and Clinical Practice, University of Adelaide Australia

* Corresponding author. Tel.: +98 831 8262005; fax: +98 831 8263048. E-mail address: farid_n32{at}yahoo.com (F. Najafi).


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Background: Limited data are available on the epidemiology of heart failure (HF) after acute myocardial infarction (AMI). We have investigated trends in the incidence and outcome of HF developing more than 28 days after first-ever AMI.

Methods and results: We identified all residents of Perth, Western Australia aged 25–64 years with no history of HF, who had experienced an AMI between 1984 and 1993, and followed them to 2005 (at which time survivors of the index events would have been aged up to 85 years). Of 3109 patients identified, 406 (13.1%) had at least one subsequent admission to hospital with a diagnosis of HF and 211 died. Following adjustment for age and sex, the hazard ratio for late-onset HF for the period 1989–1993 relative to 1984–1988 was 0.85 (95%CI: 0.69 to 1.04). After adjustment for age, history of diabetes and recurrent acute coronary syndrome, the hazard ratio for death in patients with late-onset HF did not change over the period of study (HR per year=1.02, 95%CI: 0.99 to 1.05).

Conclusion: Our findings contradict recent claims that there is an epidemic of HF driven in part by improved survival after AMI.

Key Words: Heart failure • Myocardial infarction • Trends • Incidence • Survival • Western Australia

Received December 7, 2007; Revised April 2, 2008; Accepted May 8, 2008


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
In addition to recent decreases in the incidence and severity of acute myocardial infarction (AMI) [1,2], growing use of coronary artery revascularization procedures (CARPs) in the 1990s and better medical treatment have resulted in improved survival of patients following an AMI. Although there is some evidence that the incidence of hospitalisations for heart failure (HF) may be declining [3,4], others claim that improving survival after AMI has made a major contribution to the reported epidemic of heart failure [5].

There are few population-based studies of trends in the epidemiology of HF after AMI on which to make a judgment regarding this hypothesis. In addition, such analyses of HF after AMI need to distinguish the time of its occurrence, because the mechanisms for early-onset HF (HF complicating an index AMI within the first 28 days) and late-onset HF (HF developing beyond 28 days after an index AMI) might be different [6,7]. Furthermore, the available data regarding trends in the incidence of early-and late-onset HF after AMI are not consistent [8-10]. While a decline in the severity of AMI events, improvement in in-hospital treatment and growing use of CARPs may all contribute to the observed decline in early-onset HF after AMI, trends in late-onset HF after AMI still need to be investigated.

There are also limited population-based data on survival after a diagnosis of HF, although there are some indications that survival of such patients has improved in recent years [4,11,12]. Nevertheless, HF is a pathophysiological state brought about by a variety of different underlying causes. It is therefore important to investigate trends in HF within the context of specific conditions such as AMI [13], especially as AMI is one of its most frequent precipitants.

The World Health Organization (WHO) MONItoring trends and determinants of CArdiovascular disease (MONICA) Project was a population-based study of trends in mortality, morbidity and risk factors for AMI in people aged 25-64 over a 10-year period from 1984 to 1993 [14]. Using the Western Australian Linked Database System (WALDS) [15] and MONICA data from Perth, Western Australia, we have investigated trends in the incidence and outcome of late-onset HF.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
The Perth MONICA Register covered all residents of the Perth Statistical Division, effectively the metropolitan area of Perth, aged 25-64 years. The Register included all major coronary events occurring between 1984 and 1993 and used the ‘cold pursuit’ method to identify non-fatal potential cases of AMI through surveillance of hospital discharge codes [14]. A research nurse abstracted data from medical records for each admission to hospital with an International Classification of Diseases (9th revision, clinical modification) (ICD-9-CM) code for AMI or subacute coronary heart disease (codes 410 and 411, respectively).

The present analyses concern all patients with events which met the following criteria: the patient had no history of previous AMI or HF and no evidence of early-onset HF (HF during the first 28-days after the index AMI event); the event met the MONICA criteria for ‘definite AMI’ [14]; and the patient was still alive 28 days after the onset of the symptoms of AMI.

2.1. Definition of heart failure
Using the WALDS [15], we followed up all patients included in our study sample for a subsequent admission to hospital with a diagnosis of HF. In order to capture all cases of HF (even those complicating a recurrent AMI), we regarded a patient as having HF when the electronic record for a new hospital admission included the ICD-9-CM code for HF (428) in either the first or second diagnostic position. We refer to such cases as ‘late-onset HF’, as opposed to ‘early-onset HF’ which denotes HF complicating the first-ever AMI within the first 28 days.

2.2. Definition of covariates
A history of angina was recorded if there was a good description of typical retrosternal discomfort or pain brought on by exercise and relieved by rest, before the index AMI. The first recorded systolic blood pressure after the onset of symptoms of AMI was categorized as <100 mmHg or ≥100 mmHg. Diabetes mellitus (DM) was defined by treatment with insulin or oral hypoglycaemic agents immediately prior to admission; patients managed by diet alone were considered as diabetic if their diagnosis was confirmed by the referring doctor's letter or hospital record. Similarly, hypertension was defined by treatment for this condition at the time the symptoms of the index AMI developed. To define disturbances of cardiac rhythm related to the index AMI, we classified maximum pulse rate within the first 24 h of the index AMI into two categories: 60-100 beats/min (‘normal’), and <60 or >100 (defining bradyarrhythmias or tachyarrhythmias). Patients were categorized as currently smokers or non-smokers. The length of hospital stay (LOS) during the index hospital admission was categorized as either ≤10 days or >10 days (75th centile). Peak creatine phosphokinase (CPK) was defined as the ratio of the maximum CPK value for the event to the relevant upper limit of normal. The MONICA database includes Minnesota codes for up to four electrocardiograms together with summary variables for any ST-elevation, new Q-wave or anterior wall infarction. A recurrent acute coronary syndrome (ACS) was defined as a patient having another admission to hospital with a relevant ICD-9-CM code (410, 411 or 413) in first diagnostic position. A CARP was recorded if a patient had any of following ICD-9-CM procedure codes either during the hospital admission for the index AMI or subsequently: 36.01-36.07, and 36.10-36.19. Angiotensin converting enzyme inhibitors (ACEi) were introduced during the period covered by the MONICA Project and were included in a category named ‘other antihypertensive agents’, which excluded diuretics, beta blockers and calcium channel blockers (CCBs).

Using the WALDS, we followed up all patients included in our study sample for deaths from any cause. Very few residents of Western Australia with a history of AMI die in other parts of the country [16]. For predicting late-onset HF, age at the first-ever hospital admission with AMI was added to the statistical models as a categorical variable (24-54, 55-59 and 60-64 years). Age at the time of developing incident HF was calculated and entered into relevant models as a continuous variable.

2.3. Statistical analysis
We analyzed the data using Stata 9 [17]. For categorical variables, we calculated crude relative risks and corresponding 95% confidence intervals (95%CIs) to compare the characteristics of patients who did or did not develop late-onset HF. The median of age and CPK ratio among two groups were compared using the Mann-Whitney U test. To check trends in severity of the index AMI, changes in systolic blood pressure, pulse, LOS, and ECG presentation were analyzed using the chi-squared test, as were changes in use of treatments.

All patients included in this sample were followed to the end of 2005 for either re-admission to hospital with HF or death. The cumulative incidence of late-onset HF was calculated using the Kaplan-Meier method, taking into account death as a competing risk [18]. To investigate the trends between 1984-1988 and 1989-1993 in the risk of occurrence of late-onset HF and ‘late-onset HF or death’, we used simple and multivariable Cox proportional hazard regression models. Because of collinearity between the indicators of ECG changes, we used development of a new Q-wave in conjunction with the index AMI as the summary ECG measure. We entered ACS and CARP before the occurrence of late-onset HF into the models as dichotomous time-dependent variables.

In addition, we standardized the proportions of men and women in each 5-year period who eventually developed HF during 10 years of follow-up using the direct method with the age distribution of individuals on the Perth MONICA register as the external reference.

We also used a Cox proportional hazards model to identify predictors of death among patients with late-onset HF, summarizing the findings as hazard ratios (HRs) and corresponding 95%CIs. Trends in all-cause mortality after a diagnosis of late-onset HF were assessed by two methods. First, we calculated changes in annual hazard of mortality after late-onset HF, using the Cox approach. Secondly, we calculated the proportions of patients with late-onset HF developing in two periods, 1985-1992 and 1993-2000, who died during 5 years of follow-up, for men and women separately. For models predicting survival after late-onset HF, ACS and CARP after the occurrence of HF were entered into the models as dichotomous time-dependent variable.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
Between 1984 and 1993, the Perth MONICA Register included 4006 patients with definite, first-ever non-fatal AMI, and without previous HF. Of these, 897 developed early-onset HF. Among the remaining 3109 cases, 406 (13.1%) were admitted to hospital with late-onset HF during a median follow-up of 14.4 years. The total number of first-ever non-fatal AMI events increased from 1451 in 1984-1988 to 1658 in 1989-1993.

Patients who developed late-onset HF were older, more likely to be women, to have diabetes and hypertension, and to be current smokers at the time of index AMI (Table 1). There was no difference between patients who did or did not develop late-onset HF in terms of history of angina before the index AMI.


View this table:
[in this window]
[in a new window]

  		Table 1 Characteristics of patients with first-ever non-fatal myocardial infarction, Perth MONICA, 1984-1993

 
Of the 2703 patients who did not develop late-onset HF, 504 (18.6%) had at least one admission for ACS after the index AMI compared with 140 (34.5%) of the 406 patients with late-onset HF (p<0.001) (Table 1). Of those 140 patients, 104 were re-admitted with ACS before the development of late-onset HF: 21 developed late-onset HF within 28 days after the ACS admission and 83 patients were admitted to hospital with a diagnosis of late-onset HF more than 28 days after the ACS admission. The remaining 36 patients were admitted to hospital for a recurrent ACS after their first admission for late-onset HF.

3.1. Severity of index myocardial infarction
Patients with late-onset HF had longer hospital admissions for the index AMI and marginally higher CPK ratios (Table 1). They were also more likely to have bradyarrhythmia or tachyarrhythmia during the index AMI. However, they did not show more ECG abnormalities and were no more likely to have a blood pressure of less than 100 mmHg than patients who did not develop HF. Among those with late-onset HF, there was no evidence of any relationship between the indices of severity and the time from index AMI to the occurrence of late-onset HF.

There was some evidence that the severity of the index AMIs declined over time. The percentage of patients with ST-elevation (61.3% in 1984-1988 vs. 55.6% in 1989-1993, p=0.001), Q-wave (53.3% vs. 47.7%, p=0.002), or ST-depression (47.6 vs. 43.4, p=0.02) and of those with a LOS>10 days (27.5% vs. 16.6%, p<0.001) declined, but there was no change in maximum pulse rate or median CPK ratio and a small increase in the percentage of patients with SBP less than 100 mmHg (4.8% vs. 6.7%, p=0.03).

3.2. Management of myocardial infarction
There were differences in drug treatment before admission for the index AMI in patients who did and those who did not develop late-onset HF, which probably reflect the prevalence of other cardiovascular conditions (Table 2). Patients with late-onset HF were also more likely to have been treated with diuretics, ‘other antihypertensive agents’ and CCBs either during or at discharge from their index hospital admission (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2 Treatment of patients with first-ever non-fatal myocardial infarction during index hospital admission, Perth MONICA, 1984-1993

 
Over the study period, there were marked changes in medications. From 1984-1988 to 1989-1993 there were significant increases in the prescription at discharge of beta blockers (64.0% vs. 80.0%), ‘other antihypertensive agents’ (2.3% vs. 12.6%), and anti-platelet agents (46.7% vs. 87.4%). In addition, there was a significant increase in the use of streptokinase (17.0% vs. 45.9%).

Over a median follow-up of 14.4 years after the index AMI, 784 patients (25.2%) underwent a CARP. The proportion of patients without late-onset HF who underwent a CARP (26.0%) was higher than those with late-onset HF (20.0%, p=0.009). From the total of 784 patients, 354 had procedures performed within 3 months of the index AMI. The increase in the proportion of patients who had a CARP performed was statistically significant (17.7% in 1984-1988 vs. 31.8% in 1989-1993, p<0.001).

3.3. Late-onset heart failure: secular trends and predictors
The cumulative incidence of late-onset HF during follow-up was 17.3% (95%CI: 15.3%-19.4%) and it increased at a rate of 0.87% per year. However, there was some evidence of a decrease in the hazard of late-onset HF over the period of study; the age- and sex-adjusted hazard of late-onset HF in 1989-1993 compared with 1984-1988 was 0.85 (95%CI: 0.69-1.04) (Table 3). For the same period, the hazard ratio of ‘late-onset HF or death’, adjusted for age and sex, was 0.82 (95%CI: 0.72-0.93). When the analysis was limited to ten years of follow-up for each patient (232 cases of late-onset HF), the age-adjusted cumulative incidence of late-onset HF showed a statistically significant decline only among men (ratio of proportions for 1989-1993 vs. 1984-1988: 0.73; 95%CI: 0.59-0.91) (Table 4). In addition, the cumulative number of cases of late-onset HF declined from 122 in 1984-1988 to 110 in 1989-1993.


View this table:
[in this window]
[in a new window]

  		Table 3 Predictors of late-onset heart failure and ‘heart failure or death’ after first non-fatal definite myocardial infarction, Perth MONICA 1984-1993: results from Cox proportional hazard models

 


View this table:
[in this window]
[in a new window]

  		Table 4 Temporal trends in percentage of patients who developed late-onset heart failure after first-ever myocardial infarction during 10 years of follow-up, Perth MONICA, 1984-1993

 
While age, history of DM and hypertension, smoking, and recurrent admission for ACS were positively and significantly associated with occurrence of late-onset HF in univariate and multivariable models, undergoing a CARP after the index AMI was negatively and significantly associated with late-onset HF (Table 3). The significant univariate associations of late-onset HF with female sex and LOS (>10 days) were not apparent in the multivariable model. The HRs for predictors of ‘late-onset HF or death’ were similar to those for late-onset HF, but somewhat attenuated.

3.4. Survival after late-onset HF: predictors and secular trends
After a median follow-up of 3.20 years (first and third quartiles were 0.92 and 7.23 years, respectively), 211 (52.0%) of the patients with late-onset HF had died. There was no evidence of changes in the annual hazard ratio of all-cause mortality among patients with late-onset HF (HR=1.02, 95%CI: 0.99-1.05) (Table 5). In addition, the 5-year cumulative mortality among patients who developed late-onset HF did not change among men and women from 1985-1992 to 1993-2000 (risk ratios: 1.27 (95%CI: 0.83-2.01) and 0.64 (95%CI: 0.24-1.79), respectively) (Table 6). Patients who died were more likely to have had a recurrent admission for ACS after the occurrence of late-onset HF and a history of DM at the time of their first-ever AMI (Table 5).


View this table:
[in this window]
[in a new window]

  		Table 5 Predictors of death after occurrence of late-onset heart failure following first-ever non-fatal definite myocardial infarction, Perth MONICA 1984-1993: results from Cox proportional hazard models

 


View this table:
[in this window]
[in a new window]

  		Table 6 Temporal trends in cumulative mortality over 5 years of follow-up, among patients with late-onset heart failure, Perth MONICA Project

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
Within this population-based study, our results suggest a decline in the hazard of late-onset HF for patients suffering an AMI in 1989-1993 compared with 1984-1988. However, survival after developing HF remained stable.

The limited data available from other studies do not present consistent figures regarding the trends in the incidence of HF after AMI. The Framingham investigators showed that while there was an increase in the early-onset HF, late-onset HF declined from 1950 through to 1989 [8]. A recent report from the Olmsted County, Minnesota, describes a non-significant decline in the five-year incidence of late-onset HF after AMI from 13% to 9% and 10% in the time periods 1979-1984, 1985-1989 and 1990-1994, respectively [9]. In our study, consistent with a decline in severity of AMI and an improvement in treatment of patients with AMI, the age- and sex-adjusted hazard for late development of HF after AMI showed a downward trend. Theoretically an increase in survival of first-ever AMI events together with an improvement in the treatment of patients with AMI could lead to an increase in the number of survivors of more serious cases, who are at higher risk of HF. However, the absolute number of cases of HF, as well as the hazard of development of late-onset HF, did not show an increase over the study period, contradicting the claim [5] that there is an epidemic of HF driven in part by improved survival after AMI.

Among survivors of AMI in Perth, the cumulative incidence of late-onset HF increased at a rate of 0.87% per year of follow-up. This is lower than the annual increase of 1.3% reported from the CARE study [19], a study of late development of HF among patients with non-fatal AMI, and that of 2% in the Framingham study for the period from1950 to 1989 [8]. The data from Framingham include both early-and late-onset HF after AMI.

Early- and late-onset HF after AMI might have different mechanisms and risk factors. While early-onset HF is related to severity of myocardial infarction (defined by number of involved vessels as well as location of infarction) [7], late-onset HF is a more chronic process related to remodelling and recurrent ACS [6]. Other pre-existing risk factors for HF, such as DM, hypertension and smoking, are associated with an increase in risk of both early- and late-onset HF. In our study, late-onset HF was strongly associated with DM, hypertension, current smoking and recurrent ACS. Length of stay as a possible proxy for severity of the index AMI was not an independent risk factor for late-onset HF. Other indicators of severity of infarction, such as development of a Q-wave and peak CPK ratio, also did not differ between patients with and without late-onset HF. In showing a protective effect associated with CARP, our results were consistent with the report from the Olmsted County, Minnesota, which was limited to CARPs performed in the first 24 h of AMI [9].

The present study shows that despite frequent reports of improvements in survival of patients with HF unselected as to aetiology [11,12], the long-term survival after late-onset HF following AMI did not improve in Perth, in either men or in women, over the 1980s and 1990s (Table 5). More generally, trends in the outcome of patients with HF, which is defined as a pathophysiological state with heterogeneous underlying causes, need further investigation. In our study, we confined the analyses to a group of patients with late-onset HF after a first-ever definite AMI. The prognosis of such patients was poor, 48.0% alive after a median follow-up of 3.2 years. The 5-year cumulative mortality figures in this study (51.0% and 32.0% for the period of 1993-2000 among men and women, respectively) were less than the corresponding values reported from Framingham (59% and 45% for the period of 1990-1999 among men and women, respectively) [12], but our figure for men was comparable with that reported from Olmsted County, (50% and 46% for the period of 1996-2000 among men and women, respectively) [11]. In showing no secular trend to improvement in survival of patients with late-onset HF after AMI, our results are also similar to those on survival of patients from Olmsted County, with early- and late-onset HF after AMI [20].

4.1. Strengths and limitations
The exclusion of older people with an index AMI may limit the generalizability of our estimates of the incidence and survival of patients with late-onset HF. However, with a median follow-up of 14.4 years, patients were aged up to 85 years by the end of the study period. The present study may not reflect current clinical practice, but it does examine a major upstream determinant of the burden of HF that we face at present. This study is one of the few population-based investigations of trends in the incidence of HF after AMI that have been conducted outside of the US.

Changes in the rules governing hospital funding and therefore the nature and number of diagnostic codes recorded for each admission are factors that can affect recording of HF in hospitals. To minimize the effect of these factors, we included the ICD-9-CM code for HF (428) only when it appeared in the first or second diagnostic positions, which are expected to be less affected by changes in hospital practice. We may have under-estimated the incidence of late-onset HF because this study excludes those patients with milder left ventricular dysfunction. The present study was only concerned with those patients who were admitted in hospital with a diagnosis of HF and ignored those with milder presentation of HF who did not need admission. However, it is less likely that any under-enumeration has affected the estimated trends. This study does not take into account the extent that changes in ongoing care may have affected the incidence and survival of patients with HF. Despite the shorter time frame in the present study, the series of late-onset HF we documented in Perth is twice the size of that investigated in Minnesota (follow-up of 215 patients) [9] and six times that reported from the Framingham Heart Study (68 patients) [8].

4.2. Conclusion
The results from this population-based analysis provide evidence that the incidence of late-onset HF has not increased over time in Perth. The survival of such patients also did not change over the period of study in the largest city in Western Australia. With no secular increase in numbers of cases of late-onset HF in Perth as well as the observed decline in hazard of development of late-onset HF, at least for men, these findings contradict the notion that improvement in survival of patients with AMI has made a major contribution to the epidemic of HF.


   Acknowledgement
 
We are grateful to the School of Population Health at the University of Western Australia for providing us with a de-identified copy of MONICA data linked to Mortality and Hospital Morbidity data.


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. Tunstall-Pedoe H., Kuulasmaa K., Mahonen M., Tolonen H., Ruokokoski E., Amouyel P. Contribution of trends in survival and coronary-event rates to changes in coronary heart disease mortality: 10-year results from 37 WHO MONICA project populations. Monitoring trends and determinants in cardiovascular disease. Lancet (1999) 353:1547–1557.[CrossRef][Web of Science][Medline]
  2. Hellermann J.P., Reeder G.S., Jacobsen S.J., Weston S.A., Killian J.M., Roger V.L. Longitudinal trends in the severity of acute myocardial infarction: a population study in Olmsted County, Minnesota. Am J Epidemiol (2002) 156:246–253.[Abstract/Free Full Text]
  3. Stewart S., MacIntyre K., MacLeod M.M., Bailey A.E., Capewell S., McMurray J.J. Trends in hospitalization for heart failure in Scotland, 1990-1996. An epidemic that has reached its peak? Eur Heart J (2001) 22:209–217.[Abstract/Free Full Text]
  4. Schaufelberger M., Swedberg K., Koster M., Rosen M., Rosengren A. 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) 25:300–307.[Abstract/Free Full Text]
  5. Braunwald E. Shattuck lecture—cardiovascular medicine at the turn of the millennium: triumphs, concerns, and opportunities. N Engl J Med (1997) 337:1360–1369.[Free Full Text]
  6. Cohn J.N., Ferrari R., Sharpe N. Cardiac remodeling—concepts and clinical implications: a consensus paper from an international forum on cardiac remodeling. J. Am. Coll. Cardiol. (2000) 35:569–582.[Abstract/Free Full Text]
  7. Stone P.H., Raabe D.S., Jaffe A.S., et al. Prognostic significance of location and type of myocardial infarction: independent adverse outcome associated with anterior location. J. Am. Coll. Cardiol. (1988) 11:453–463.[Abstract]
  8. Guidry U.C., Evans J.C., Larson M.G., Wilson P.W., Murabito J.M., Levy D. Temporal trends in event rates after Q-wave myocardial infarction: the Framingham Heart Study. Circulation (1999) 100:2054–2059.[Abstract/Free Full Text]
  9. Hellermann J.P., Goraya T.Y., Jacobsen S.J., et al. Incidence of heart failure after myocardial infarction: is it changing over time? Am. J. Epidemiol. (2003) 157:1101–1107.[Abstract/Free Full Text]
  10. Spencer F.A., Meyer T.E., Goldberg R.J., et al. Twenty year trends (1975-1995) in the incidence, in-hospital and long-term death rates associated with heart failure complicating acute myocardial infarction: a community-wide perspective. J. Am. Coll. Cardiol. (1999) 34:1378–1387.[Abstract/Free Full Text]
  11. Roger V.L., Weston S.A., Redfield M.M., et al. Trends in heart failure incidence and survival in a community-based population. JAMA (2004) 292:344–350.[Abstract/Free Full Text]
  12. Levy D., Kenchaiah S., Larson M.G., Benjamin E.J., Kupka M.J., Ho K.K.L., Murabito J.M., Vasan R.S. Long-term trends in the incidence of and survival with heart failure. N Engl J Med (2002) 347:1397–1402.[Abstract/Free Full Text]
  13. Haddy F.J. Heart failure—incidence and survival. N Engl J Med (2003) 348:660.[Free Full Text]
  14. Tunstall-Pedoe H., Kuulasmaa K., Amouyel P., Arveiler D., Rajakangas A.M., Pajak A. Myocardial infarction and coronary deaths in the World Health Organization MONICA project. Registration procedures, event rates, and case-fatality rates in 38 populations from 21 countries in four continents. Circulation (1994) 90:583–612.[Abstract/Free Full Text]
  15. Holman C.D., Bass A.J., Rouse I.L., Hobbs M.S. Population-based linkage of health records in Western Australia: development of a health services research linked database. Aust N Z J Public Health (1999) 23:453–459.[Web of Science][Medline]
  16. Bradshaw P.J., Jamrozik K., Le M., Gilfillan I., Thompson P.L. Mortality and recurrent cardiac events after coronary artery bypass graft: long term outcomes in a population study. Heart (2002) 88:488–494.[Abstract/Free Full Text]
  17. Stata Corporation. Stata 9.2 statistics/data analysis special edition. In: College Station (2006) 9 ed. Texas: Stata Corporation.
  18. Gooley T.A., Leisenring W., Crowley J., Storer B.E. Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat. Med. (1999) 18:695–706.[CrossRef][Web of Science][Medline]
  19. Lewis E.F., Moye L.A., Rouleau J.L., et al. Predictors of late development of heart failure in stable survivors of myocardial infarction: the CARE study. J. Am. Coll. Cardiol. (2003) 42:1446–1453.[Abstract/Free Full Text]
  20. Hellermann J.P., Jacobsen S.J., Redfield M.M., Reeder G.S., Weston S.A., Roger V.L. Heart failure after myocardial infarction: clinical presentation and survival. Eur J Heart Fail (2005) 7:119–125.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Eur J Heart FailHome page
F. Najafi, K. Jamrozik, and A. J. Dobson
Understanding the 'epidemic of heart failure': a systematic review of trends in determinants of heart failure
Eur J Heart Fail, May 1, 2009; 11(5): 472 - 479.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Najafi, F.
Right arrow Articles by Jamrozik, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Najafi, F.
Right arrow Articles by Jamrozik, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?