Abstract
Despite significant advances in its treatment, acute myocardial infarction (AMI) remains an important cause of heart failure (HF). Contemporary data remain lacking, however, describing long-term trends in the incidence rates, demographic and clinical profile, and outcomes of patients who develop HF as a complication of AMI. Our study sample consisted of 11,061 residents of the Worcester (MA) metropolitan area hospitalized with AMI at all greater Worcester hospitals in 15 annual study periods between 1975 and 2005. Overall, 32.4% (n=3,582) of patients with AMI developed new onset HF during their acute hospitalization. Patients who developed HF were generally older, more likely to have pre-existing cardiovascular disease, and were less likely to receive cardiac medications or undergo revascularization procedures during their hospitalization than patients who did not develop HF (p < 0.001). Incidence rates of HF remained relatively stable between 1975 and 1991 at 26%, but declined thereafter. Declines were also noted in the hospital and 30-day death rates among patients with acute HF (p < 0.001). However, patients who developed new-onset HF remained at significantly higher risk for dying during their hospitalization (21.6%) than patients who did not develop this complication (8.3%) (p <0.001). Our large community-based study of patients hospitalized with AMI demonstrates that the incidence rates of, and mortality attributable to, HF have declined over the last 3 decades. In conclusion, HF remains a common and frequently fatal complication of AMI to which increased surveillance and treatment efforts should be directed.
Keywords: epidemiology, heart failure, myocardial infarction
Introduction
In this investigation, we provide a 30-year (1975–2005) perspective into changing trends in the incidence rates of heart failure (HF), hospital treatment practices, and short-term death rates associated with HF complicating acute myocardial infarction (AMI).
Methods
The Worcester (MA) Heart Attack Study is a population-based investigation examining long-term trends in the incidence and death rates of greater Worcester (MA) (2000 census = 478,000) residents hospitalized with AMI at all metropolitan Worcester medical centers. The methods used in this study have been previously described.1–3 Patients with possible AMI were identified through the review of computerized hospital databases for patients with discharge diagnoses consistent with possible AMI or coronary heart disease-related rubrics. Trained study physicians and nurses individually reviewed the medical records of all geographically eligible patients in a standardized manner.1 The diagnosis of AMI was confirmed according to pre-established criteria and a patient was included in the study if they met at least 2 of 3 pre-defined criteria. Autopsy-proven cases of AMI were included in the sample, irrespective of the other criteria. Patients with peri-operative or trauma-related AMI were not included.
Our study population consisted of 11,061 patients without a history of HF who developed a validated AMI at all hospitals in the Worcester metropolitan area during the 15 study years of 1975 (n= 680), 1978 (n=722), 1981 (n=830), 1984 (n=609), 1986 (n=642), 1988 (n=589), 1990 (n=641), 1991 (n=702), 1993 (n=772), 1995 (n=786), 1997 (n=849), 1999 (n=766), 2001 (n=947), 2003 (n=874), and 2005 (n=652).
Study physicians and nurses abstracted clinical and demographic data from the medical records of patients with confirmed AMI. Abstracted data included patient’s age, sex, race, medical history, AMI order (initial vs. prior), type of AMI (Q wave vs. non-Q wave), physiologic factors (e.g., blood pressure) laboratory measures (e.g., serum total cholesterol, glucose), as well as length of hospital stay, presenting symptoms, and discharge status. Use of cardiac medications, thrombolysis, cardiac catheterization, intra-aortic balloon counterpulsation, percutaneous coronary intervention (PCI), and coronary artery bypass surgery, as well as development of important clinical complications, was determined by trained reviewers using pre-established criteria.3 Patients with a medical history of HF were excluded. New-onset HF was considered to be present if a patient was described in his/her medical record as having clinical or radiographic evidence of pulmonary edema or bilateral basilar rales with an S3 gallop on admission or at any time during his/her hospitalization.4 Survival status after hospital discharge was determined through a review of hospital medical records and a search of death certificates and social security files for residents of the Worcester metropolitan area. Follow-up after discharge was obtained for more than 99% of discharged patients.
Differences in the demographic and clinical characteristics as well as in the receipt of various treatment practices among AMI patients who did and did not develop new onset HF were examined using chi-square tests for discrete variables and Student’s t test for continuous variables. Similar methods were used to compare differences between those with HF who survived, as compared to those who did not survive, to hospital discharge. Short-term prognosis was examined in each study year and overall by calculating in-hospital and 30-day case fatality rates (CFRs) separately for patients who did, as well as for those who did not, develop HF.
Multivariate logistic regression modeling was used to evaluate the influence of potential confounding and/or mediating factors on the odds of developing HF. We examined the relation of incident HF to the following factors: age, sex, AMI type, AMI order, body mass index, history of stroke, hypertension, diabetes and angina, hospital development of complete heart block, atrial fibrillation, stroke, and cardiogenic shock, and hospital survival status. Although body mass index and estimated glomerular filtration (eGFR) rates differed between our respective comparison groups, these variables were not included in our multivariable adjusted models because information about these factors was missing in a large percentage of hospitalized patients. Multivariable logistic regression analyses were also used to assess the overall effect of HF on in-hospital and 30-day mortality, while controlling for similar potentially confounding prognostic factors.
Changes over time in the incidence rates of HF complicating AMI, as well as in-hospital and 30-day post-admission CFRs, were examined using the Mantel-Haenszel chi square test for trends. Logistic regression models, controlling for previously described covariates, were utilized to examine differences in the incidence rates of HF and short-term CFRs in those with HF during the 30-year period under study.
Results
A total of 11,061 greater Worcester residents without previously diagnosed HF were hospitalized with confirmed AMI between 1975 and 2005. The sample was elderly and predominantly Caucasian, with a high prevalence of cardiovascular risk factors (Table 1). Thirty-two percent of patients (n=3,582) experienced a first episode of HF during their hospitalization for AMI.
Table 1.
Characteristics of Patients With AMI According to the Presence of New-Onset Heart Failure
| Characteristic | Total Sample | Patients Hospitalized in 2001, 2003, and 2005 | p Value | |||
|---|---|---|---|---|---|---|
| Heart Failure Present (+)(n= 3,582) | Heart Failure Absent (−) (n=7,479) | p Value | Heart Failure Present (+)(n= 702) | Heart Failure Absent (−) (n= 1770) | ||
| Age (mean, yrs) | 71.7 (12.4) | 64.9 (14.0) | <0.001 | 75.5 (12.1) | 66.9 (14.5) | <0.001 |
| Age (years) | ||||||
| <55 | 9.6% | 23.4% | -- | 6.6% | 23.0% | -- |
| 55–64 | 17.0% | 24.1% | -- | 12.5% | 20.8% | -- |
| 65–74 | 28.3% | 25.6% | -- | 22.1% | 21.5% | -- |
| 75–84 | 30.7% | 19.5% | -- | 34.1% | 23.0% | -- |
| ≥85 | 14.5% | 7.5% | <0.001 | 24.8% | 11.7% | <0.001 |
| Male | 54.4% | 65.1% | <0.001 | 47.9% | 62.4% | <0.001 |
| Race | ||||||
| White | 96.7% | 94.6% | <0.001 | 91.3% | 90.5% | 0.53 |
| Other | 3.4% | 5.4% | <0.001 | 8.7% | 9.5% | 0.46 |
| Body mass index (Kg/m2) | ||||||
| <25 | 37.6% | 33.1% | -- | 34.1% | 30.7% | -- |
| 25–29.9 | 35.5% | 38.6% | -- | 35.3% | 39.3% | -- |
| ≥30 | 27.1% | 28.3% | 0.002 | 30.5% | 30.0% | 0.46 |
| Pre-Existing Conditions | ||||||
| Angina pectoris | 25.1% | 21.2% | <0.001 | 17.7% | 17.8% | NS |
| Diabetes mellitus | 30.3% | 21.4% | <0.001 | 35.9% | 25.2% | <0.001 |
| Hypertension | 59.3% | 52.7% | <0.001 | 74.5% | 66.2% | <0.001 |
| Stroke | 11.9% | 6.3% | <0.001 | 14.7% | 7.9% | <0.001 |
| Acute Presenting Symptoms | ||||||
| Chest pain* | 64.9% | 81.3% | <0.001 | 64.0% | 81.8% | <0.001 |
| Diaphoresis* | 35.5% | 42.9% | <0.001 | 32.1% | 42.1% | <0.001 |
| Dyspnea* | 67.9% | 47.9% | <0.001 | 68.0% | 51.0% | <0.001 |
| AMI type | ||||||
| Initial | 65.8% | 74.4% | <0.001 | 67.7% | 74.5% | <0.001 |
| Q wave | 48.7% | 43.9% | <0.001 | 24.2% | 24.0% | <0.001 |
| Clinical Complications | ||||||
| Third degree heart block | 6.1% | 3.3% | <0.001 | 2.8% | 2.6% | 0.78 |
| Atrial fibrillation | 24.8% | 10.9% | <0.001 | 29.3% | 15.6% | <0.001 |
| Cardiogenic shock | 12.5% | 3.3% | <0.001 | 12.5% | 2.0% | <0.001 |
| Stroke† | 1.3% | 1.1% | 0.32 | 2.3% | 1.7% | 0.38 |
| Physiologic findings at the time of hospital admission (mean) | ||||||
| Systolic blood pressure (mmHg) | 142.9 (35.2) | 143.4 (32.6) | 0.66 | 142.0 (35.9) | 143.0 (33.0) | 0.48 |
| Diastolic blood pressure (mmHg) | 77.5 (21.9) | 79.6 (19.7) | 0.005 | 75.9 (21.7) | 79.0 (19.9) | <0.001 |
| Heart rate (bpm)† | 91.3 (25.7) | 81.5 (21.8) | <0.001 | 92.8 (26.1) | 82.9 (21.7) | <0.001 |
| Laboratory findings at the time of hospital admission (mean, mg/dl) | ||||||
| Serum glucose§ | 201.0 (124.8) | 168.8 (133.6) | <0.001 | 199.6 (137.8) | 168.2 (161.5) | <0.001 |
| Estimated GFR findings (%)§ | 56.1 (23.3) | 69.0 (26.3) | <0.001 | 53.3 (22.0) | 67.0 (25.6) | <0.001 |
| Serum cholesterol | 209.0 (61.7) | 210.1 (56.0) | 0.050 | 176.7 (60.2) | 181.5 (49.3) | 0.12 |
| Serum hemoglobin | 13.3 (4.2) | 14.0 (3.3) | <0.001 | 13.2 (5.5) | 14.0 (4.2) | <0.001 |
Data available from
1997–2005 only;
1986–2005 only;
1991–2005 only;
1995–2005 only;
AMI = acute myocardial infarction; BPM = beats per minute; GFR = glomerular filtration rate
Patients who developed new-onset HF were on average older, more frequently female, and were more likely to have a history of prior MI, diabetes, angina, hypertension, and stroke than patients who did not develop this complication (Table 1). Patients with HF complicating AMI were also more likely to have developed atrial fibrillation, complete heart block, and cardiogenic shock during their hospitalization. Patients with HF had lower diastolic blood pressure, lower total serum cholesterol levels, lower hemoglobin levels, and a lower eGFR rate, but higher heart rates and blood glucose levels on hospital admission, than patients who did not develop HF. In contrast to patients admitted during earlier study years, patients with HF admitted in 2001, 2003, and 2005 did not differ from those who did not develop HF with respect to a history of angina, total serum cholesterol levels, body mass index, or likelihood of developing complete heart block during their index hospitalization.
Patients who developed new-onset HF were less likely to be treated with aspirin, beta-blockers, thrombolytics, and lipid-lowering agents during their hospitalization for AMI than patients who did not develop HF (p for all < 0.001; Table 2). As compared to those who did not develop HF, patients with HF complicating AMI were more likely to be prescribed calcium channel blockers and either an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker during their hospitalization (p for both < 0.001). Patients with incident HF were less likely to have undergone cardiac catheterization or PCI, but were more likely to have undergone placement of an intra-aortic balloon pump (p < 0.001).
Table 2.
Treatment Practices in Patients Hospitalized With Acute Myocardial Infarction According to the Presence of New-Onset Heart Failure
| Total Sample | Patients Hospitalized in 2001, 2003, and 2005 | |||||
|---|---|---|---|---|---|---|
| Heart Failure Present (+) (n=3582) | Heart Failure Absent (−) (n=7479) | p Value | Heart Failure Present (+) (n=702) | Heart Failure Absent (−) (n=1770) | p Value | |
| Medications | ||||||
| ACE-I or ARB | 56.2% | 38.2% | <0.001 | 68.4% | 62.6% | 0.007 |
| Aspirin | 58.8% | 68.3% | <0.001 | 92.0% | 93.3% | 0.24 |
| Beta blockers | 53.2% | 69.2% | <0.001 | 88.3% | 92.0% | 0.004 |
| Calcium channel blockers† | 39.3% | 34.9% | <0.001 | 27.8% | 21.3% | <0.001 |
| Lipid lowering Agents‡ | 24.1% | 33.3% | <0.001 | 65.4% | 71.9% | 0.002 |
| Thrombolytics‡ | 16.2% | 20.6% | <0.001 | 5.4% | 7.3% | 0.097 |
| Procedures | ||||||
| Cardiac catheterization | 27.7% | 37.2% | <0.001 | 53.1% | 69.4% | <0.001 |
| Coronary artery bypass graft surgery* | 4.2% | 4.5% | 0.58 | 7.0% | 7.3% | 0.79 |
| Percutaneous coronary intervention‡ | 16.3% | 24.9% | <0.001 | 33.3% | 50.7% | <0.001 |
| Intra-aortic balloon counterpulsation | 6.8% | 2.9% | <0.001 | 12.8% | 4.4% | <0.001 |
Data available from
1981–2005 only;
1984–2005 only;
1986–2005 only;
ACEI = Angiotensin converting enzyme inhibitor; ARB = angiotensin receptor blocker
Irrespective of HF status, patients hospitalized with AMI during the 3 most recent study years were more likely to have received effective cardiac medications and were more likely to have undergone cardiac catheterization, PCI, and intra-aortic balloon placement than patients hospitalized with AMI during early study years (Table 2). The prescribing of thrombolytic medications declined during recent years in both study groups. Differences in the use of various treatment practices between patients who did and did not develop HF during recent study years were similar to those observed over the 30-year study period, with the exception of aspirin prescription which did not differ between our 2 primary comparison groups during the most recent study years.
The incidence rates of HF complicating AMI declined significantly over the 30-year study period (p < 0.01). Between 1975 and 1990, the proportion of patients who developed HF remained relatively stable at approximately 35% (Table 3, Figure 1). Beginning in 1991, the incidence rates of acute HF began to decrease, reaching lows of 23.8% in 1997 and 25.8% in 2005. Compared to patients hospitalized in 1975, patients admitted between 1991 and 2005 had a lower unadjusted odds of developing HF (Table 3). Results were qualitatively similar for models adjusting for important demographic and clinical factors (Table 3).
Table 3.
Odds of Developing New-Onset HF in Patients Hospitalized With Acute Myocardial Infarction Over Time
| Study Year | n | Developing HF | Age and Sex Adjusted Odds of Developing HF | Multivariable Adjusted Odds of Developing HF* |
|---|---|---|---|---|
| 1975 | 680 | 35.4% | 1.0 | 1.0 |
| 1978 | 722 | 35.3% | 0.95 (0.76, 1.20) | 0.91 (0.72,1.15) |
| 1981 | 830 | 34.5% | 0.84 (0.68, 1.05) | 0.83 (0.66,1.04) |
| 1984 | 609 | 48.3% | 1.66 (1.31, 2.09) | 1.59 (1.25,2.03) |
| 1986 | 642 | 36.5% | 0.92 (0.73, 1.16) | 0.89 (0.70,1.13) |
| 1988 | 589 | 33.3% | 0.77 (0.61, 0.99) | 0.73 (0.57,0.94) |
| 1990 | 641 | 36.0% | 0.85 (0.67, 1.07) | 0.85 (0.67,1.08) |
| 1991 | 702 | 31.6% | 0.71 (0.56, 0.89) | 0.69 (0.54,0.88) |
| 1993 | 772 | 31.6% | 0.69 (0.55, 0.87) | 0.68 (0.54,0.86) |
| 1995 | 786 | 32.9% | 0.74 (0.59, 0.93) | 0.73 (0.58,0.92) |
| 1997 | 849 | 23.8% | 0.44 (0.35, 0.56) | 0.44 (0.35,0.56) |
| 1999 | 766 | 28.2% | 0.57 (0.46, 0.72) | 0.57 (0.45,0.73) |
| 2001 | 947 | 29.3% | 0.56 (0.45, 0.69) | 0.55 (0.44,0.70) |
| 2003 | 874 | 29.4% | 0.57 (0.46, 0.71) | 0.57 (0.45,0.72) |
| 2005 | 652 | 25.8% | 0.47 (0.37, 0.60) | 0.46 (0.35,0.60) |
Adjusted for patient’s age, sex, and history of angina, hypertension, diabetes, or stroke, acute myocardial infarction order, acute myocardial infarction type, and development of complete heart block, atrial fibrillation, and cardiogenic shock during hospitalization.
Figure 1.
Percentages of patients with heart failure from 1975 to 2005.
Since declining rates of new-onset HF complicating AMI were noted contemporaneously with changes in the proportion of patients developing a Q wave or an initial AMI over time, and because these factors are known to relate to extent of myocardial injury, we examined changes over the period under study in the incidence rates of HF separately for patients with Q wave versus non-Q wave AMI and for patients with an initial versus prior AMI. The results of these analyses revealed similar trends to those noted in the overall study sample. For example, between 1975 and 1990, the proportion of patients with a Q wave MI who developed HF was 38.1% compared to 35.1% in patients with a non-Q wave AMI. Beginning in 1991, the incidence rates of HF declined in both groups reaching rates of 27.9% and 25.2%, respectively, in 2005.
Overall, 21.6% of patients with AMI who developed HF died during their acute hospitalization compared to 8.3% of those who did not develop this complication (p < 0.001). In multivariate models adjusting for previously described demographic, medical history, and clinical factors known to be associated with increased mortality from AMI (Figure 2), development of HF was associated with a two-fold higher odds (OR = 1.99, 95% CI 1.73, 2.28) of dying during hospitalization for AMI. Results remained similar and statistically significant in multivariable-adjusted analyses restricted to patients hospitalized at all metropolitan Worcester medical centers during our 3 most recent study years (OR = 1.84, 95% CI 1.32, 2.56).
Figure 2.
Case fatality rates from 1975 to 2005.
Hospital CFRs in patients with HF declined over the 30-year study period (Table 4, Figure 2). After controlling for previously described demographic and clinical factors (Table 2), the adjusted odds of dying in patients with HF declined for each study year (with the exception of 1978) in relation to 1975 (Table 4). Similar to declining hospital CFR’s, the proportion of patients with HF who died within 30-days of hospitalization declined significantly over time, from 29.1% in 1975 to 16.0% in 2003 (p < 0.001).
Table 4.
Odds of Dying During Hospitalization in Patients with New-Onset Heart Failure Complicating Acute Myocardial Infarction Over Time
| Study Year | n | Hospital Case- Fatality Rates | Age and Sex Adjusted Odds of Dying | Multivariable Adjusted Odds of Dying* |
|---|---|---|---|---|
| 1975 | 241 | 31.1% | 1.0 | 1.0 |
| 1978 | 255 | 32.6% | 1.01 (0.69, 1.49) | 0.89 (0.58,1.36) |
| 1981 | 286 | 28.7% | 0.78 (0.53, 1.14) | 0.70 (0.46,1.06) |
| 1984 | 294 | 21.8% | 0.58 (0.39, 0.86) | 0.54 (0.35,0.83) |
| 1986 | 234 | 26.1% | 0.66 (0.44, 0.99) | 0.59 (0.37,0.93) |
| 1988 | 196 | 28.6% | 0.74 (0.49, 1.13) | 0.65 (0.41,1.05) |
| 1990 | 231 | 19.9% | 0.45 (0.29, 0.69) | 0.46 (0.29,0.74) |
| 1991 | 222 | 17.1% | 0.39 (0.25, 0.61) | 0.38 (0.23,0.62) |
| 1993 | 244 | 19.3% | 0.43 (0.28, 0.66) | 0.37 (0.23,0.60) |
| 1995 | 259 | 14.7% | 0.31 (0.20, 0.48) | 0.26 (0.16,0.42) |
| 1997 | 202 | 17.8% | 0.38 (0.24, 0.60) | 0.32 (0.19,0.53) |
| 1999 | 216 | 16.7% | 0.34 (0.21, 0.53) | 0.36 (0.22,0.60) |
| 2001 | 277 | 18.4% | 0.37 (0.24, 0.56) | 0.35 (0.22,0.56) |
| 2003 | 257 | 11.7% | 0.20 (0.13, 0.33) | 0.23 (0.13,0.38) |
| 2005 | 168 | 17.3% | 0.31 (0.19, 0.51) | 0.28 (0.16,0.49) |
Adjusted for patient’s age, sex, and history of angina, hypertension, diabetes, or stroke, acute myocardial infarction order, acute myocardial infarction type, and development of complete heart block, atrial fibrillation and cardiogenic shock during hospitalization.
Patients who survived an episode of new-onset HF were on average younger, more frequently male, more likely to have a history of hypertension, less likely to have a history of stroke, and less likely to have developed atrial fibrillation, Q waves, or complete heart block during their hospital stay than patients with HF who died (Table 5; p for all ≤ 0.05). Hospital survivors with incident HF had higher admission blood pressure values, higher eGFR rates, and higher admission heart rate values than patients with HF who died. Patients with AMI who survived an episode of HF were also more likely to have received effective cardiac medications and interventional procedures than patients with HF who died.
Table 5.
Characteristics of Patients with Heart Failure Complicating AMI According to Hospital Survival Status
| Characteristic | Survived (n=2810) | Died (n=772) | P Value |
|---|---|---|---|
| Age (mean) (yrs) | 70.7 (12.6) | 75.2 (10.9) | <0.001 |
| Male | 55.3% | 51.3% | 0.05 |
| Body mass index (Kg/m2) | |||
| <25 | 36.3% | 45.3% | |
| 25–29.9 | 36.2% | 30.0% | |
| ≥30 | 27.6% | 24.7% | 0.08 |
| Duration of pre-hospital delay (mean hrs)† | 3.0 (21.5) | 4.1 (7.9) | 0.14 |
| Angina pectoris | 25.0% | 25.8% | 0.63 |
| Diabetes mellitus | 30.8% | 28.6% | 0.24 |
| Hypertension | 60.4% | 55.7% | 0.02 |
| Stroke | 10.9% | 15.5% | <0.001 |
| Acute Presenting Symptoms | |||
| Chest pain* | 22.3% | 13.1% | <0.001 |
| Diaphoresis* | 12.2% | 7.0% | <0.001 |
| Dyspnea* | 22.9% | 15.2% | <0.001 |
| AMI characteristics | |||
| Initial | 66.4% | 63.7% | 0.16 |
| Q wave | 45.9% | 58.9% | <0.001 |
| Clinical Complications | |||
| Atrial fibrillation | 22.7% | 32.5% | <0.001 |
| Third degree heart block | 4.1% | 13.6% | <0.001 |
| Stroke† | 1.1% | 1.9% | 0.066 |
| Physiologic findings at the time of hospital admission (mean) | |||
| Systolic blood pressure (mmHg) | 144.6 (34.8) | 133.8 (35.9) | <0.001 |
| Diastolic blood pressure (mmHg) | 78.8 (21.7) | 70.5 (21.9) | <0.001 |
| Heart rate (bpm)† | 91.4 (25.1) | 90.7 (28.0) | <0.001 |
| Laboratory findings at the time of hospital admission (mean, mg/dl) | |||
| eGFR (%)§ | 57.6 (23.4) | 47.5 (21.0) | <0.001 |
| Serum glucose§ | 201.2 (129.4) | 200.2 (96.8) | 0.88 |
| Serum cholesterol | 208.8 (61.2) | 209.7 (63.4) | 0.79 |
| Hemoglobin | 13.4 (4.5) | 13.1 (2.0) | 0.31 |
| Medications | |||
| Angiotensin converting enzyme inhibitors* | 36.5% | 18.3% | <0.001 |
| Aspirin | 63.9% | 40.3% | <0.001 |
| Beta blockers | 59.9% | 28.8% | <0.001 |
| Calcium channel blockers | 40.6% | 33.6% | 0.003 |
| Lipid lowering agents† | 27.1% | 10.9% | <0.001 |
| Thrombolytics | 17.3% | 11.6% | 0.002 |
| Procedures | |||
| Cardiac catheterization | 31.2% | 15.2% | <0.001 |
| Coronary artery bypass graft | 4.5% | 2.9% | 0.085 |
| Percutaneous coronary intervention | 17.7% | 10.5% | <0.001 |
| Intra-aortic balloon counterpulsation | 5.9% | 10.1% | <0.001 |
Data available from
1990–2005 only;
1986–2005 only;
1991–2005 only;
1995–2005 only;
AMI = acute myocardial infarction; BPM = beats per minute; GFR= glomerular filtration rate
Discussion
The results of this study in more than 11,000 residents of a large, central New England metropolitan area hospitalized with AMI at all area medical centers demonstrate that the incidence rates of, and short-term death rates from, HF declined appreciably between 1975 and 2005. Despite these encouraging trends, short-term death rates remained considerably higher in patients with HF complicating AMI in comparison to those who did not develop this clinical syndrome.
Estimates of the contemporary incidence rates of HF complicating AMI vary widely, ranging from 6% to upwards of 45% in the published literature.5,6 The incidence rates of HF observed in our study were higher than those reported in recent randomized trials.7 Use of more restrictive inclusion criteria in the context of randomized controlled trials may partially account for differences in these findings, a hypothesis supported by the fact that our results are more similar to those reported in other observational studies.6,8 Although a greater proportion of older patients were hospitalized for AMI during our most recent study years, we noted a decline in the frequency of acute HF during the period under study, with the sharpest decline occurring over the most recent decade. These patterns were observed in patients with an initial or prior AMI, as well as in those with a Q wave or non-Q wave AMI.
Although relatively little data exist on contemporary trends in the magnitude of HF complicating AMI, investigators from the multinational GRACE study observed a similar trend toward lower rates of HF complicating AMI over a 10-year period beginning in 1995.4,6,8,9 Our investigation confirms the findings of that large registry study, extends them to a large, community-based population, and places them in the context of a 30-year study period. On the other hand, the incidence rates of HF in the 30 days after an initial AMI in the Framingham Heart Study increased in the 1990’s (1990–1999) as compared to the 1970’s (1970–1979).10 Reasons for the discrepant findings between these 2 community-based studies may be attributed to differences in the clinical and demographic characteristics of the study samples, inclusion of all patients with AMI as opposed to only first events, varying duration of follow-up (hospital versus 30-day follow-up), and time periods under study.
From 1991 on, the odds of developing HF decreased relative to the baseline year of 1975, even after adjustment for important AMI and patient-related characteristics. The lower odds of HF observed in our study’s recent years may, therefore, have resulted from treatment-related factors.11 The observational nature of the present investigation does not allow us to establish causality, but declines in the frequency of HF occurred contemporaneously with increased use of effective cardiac therapies. Since early revascularization has been shown to reduce infarct size, improve cardiac function, and decrease the attack rates of cardiogenic shock, it is possible that fewer AMI patients developed HF as a result of a larger proportion undergoing PCI.12,13 Greater use of effective cardiac medications in hospitalized persons with AMI may also have contributed to the lower incidence rates and odds of acute HF observed.11
Caution in interpreting our findings is warranted, however, since the average length of hospital stay declined over the 30-year period under study and we did not examine HF developing after hospital discharge. Moreover, the majority (83%) of cases of acute HF developed within the first 24 hours of hospitalization and, in some cases, preceded administration of AMI therapies. Further studies are needed to better delineate the relative contributions of patient, disease, and treatment-related factors to declining rates of HF complicating AMI.
Prior studies have reported in-hospital death rates ranging from 20–40% in patients with new-onset HF complicating AMI.5,14 Notably, death rates among patients with HF complicating AMI admitted during our study’s most recent years were lower than those reported in prior investigations. Death rates in patients with incident HF complicating AMI declined by 14% between 1975 and 2005, with the most impressive decreases occurring during the late 1990’s. Declining 30-day death rates after hospital admission for AMI paralleled improved in-hospital survival. As such, improved short-term survival was not attributable to shorter lengths of stay or increased post-discharge death rates. When viewed in the context of recent registry and community-based data showing a decline in age and sex-adjusted hospital or short-term CFRs among patients with AMI, our results suggest that survival gains among patients with AMI may be accounted for by improvements in the prognosis of patients with HF.15,16
Although death rates declined for all AMI patients, the greatest improvement in hospital prognosis occurred in patients with incident HF (Figure 2). This finding is of significance since treatment disparities, including lower rates of PCI, beta-blocker, and lipid-lowering agent use persisted over the period under study between those who developed HF relative to those who did not develop this complication. Although it is possible that the natural history of HF complicating AMI has changed during the years under study, with fewer patients developing “higher risk” HF and/or a greater number of cases identified earlier in the course of their disease, we hypothesize that reductions in death rates among patients with HF are related to improved monitoring and treatment of all patients hospitalized with AMI.17,18 Reasons for under-use of evidence-based cardiac procedures among patients with AMI and new-onset HF are unclear, but ours and other data suggest that this may relate to the age and disease burden of these patients.19 Since AMI patients developing HF were generally older and affected by a greater number of comorbid illnesses, providers may have been less willing to recommend cardiac medications or procedures despite a favorable risk-to-benefit ratio due to concerns regarding adverse effects.19,20
Despite survival gains in patients with HF complicating AMI, new-onset HF remains one of the most powerful prognostic markers of increased mortality in patients hospitalized with AMI. The odds of dying in patients with incident HF was 2-fold higher than in patients not developing this complication. These findings are consistent with those of the NRMI-2 investigators and, when viewed in light of treatment disparities between those with and without HF complicating AMI, suggest that more concerted efforts should be undertaken to enhance the monitoring and treatment of patients with HF as well as those at high-risk for developing HF.21 Particular attention should be devoted to older patients, women, and those with an unfavorable hemodynamic or laboratory profile, because these patients were at the highest risk of dying from HF.
The strengths of the present study include its large sample size, its population-based design, and its multi-decade long perspective. Our study has several limitations. Due to our methods of data abstraction, we could not determine the timing of HF-onset relative to the use of medications or procedures, or other AMI-related complications. Although the use of various treatment approaches might have affected both the odds of developing or dying from HF, we did not control for the hospital use of medications or procedures in our regression models due to concerns about potential confounding by treatment indication and timing. In addition, since data on eGFR was lacking from patients with AMI hospitalized before 1995, we did not adjust for eGFR in our multivariable models. We were unable to differentiate between HF attributed to systolic versus diastolic function due to the infrequent use of echocardiography in our study population. We were also unable to systematically categorize the sample according to the presence of ST-segment elevation, as this data was unavailable prior to 1997. Information on troponins was incorporated into our diagnostic AMI criteria in 2003, perhaps increasing AMI detection and influencing the natural history of AMI, including the odds of developing or dying from HF. Lastly, our study population consisted predominantly of older, Caucasian individuals with a high prevalence of cardiovascular risk factors, and our findings may not be generalizable to other patient groups hospitalized with AMI.
Acknowledgments
This research was made possible by the cooperation of participating hospitals in the Worcester metropolitan area and through funding support provided by the National Institutes of Health (RO1 HL35434).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Floyd KC, Yarzebski J, Spencer FA, Lessard D, Dalen JE, Alpert JS, Gore JM, Goldberg RJ. A 30-year perspective (1975–2005) into the changing landscape of patients hospitalized with initial acute myocardial infarction: Worcester Heart Attack Study. Circ Cardiovasc Qual Outcomes. 2009;2:88–95. doi: 10.1161/CIRCOUTCOMES.108.811828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Goldberg RJ, Spencer FA, Yarzebski J, Lessard D, Gore JM, Alpert JS, Dalen JE. A 25-year perspective into the changing landscape of patients hospitalized with acute myocardial infarction (the Worcester Heart Attack Study) Am J Cardiol. 2004;94:1373–1378. doi: 10.1016/j.amjcard.2004.07.142. [DOI] [PubMed] [Google Scholar]
- 3.Goldberg RJ, Yarzebski J, Lessard D, Gore JM. A two-decades (1975 to 1995) long experience in the incidence, in-hospital and long-term case-fatality rates of acute myocardial infarction: a community-wide perspective. J Am Coll Cardiol. 1999;33:1533–1539. doi: 10.1016/s0735-1097(99)00040-6. [DOI] [PubMed] [Google Scholar]
- 4.Spencer FA, Meyer TE, Goldberg RJ, Yarzebski J, Hatton M, Lessard D, Gore JM. 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. doi: 10.1016/s0735-1097(99)00390-3. [DOI] [PubMed] [Google Scholar]
- 5.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;15:761–768. doi: 10.1093/oxfordjournals.eurheartj.a060583. [DOI] [PubMed] [Google Scholar]
- 6.Fox KA, Steg PG, Eagle KA, Goodman SG, Anderson FA, Jr, Granger CB, Flather MD, Budaj A, Quill A, Gore JM. Decline in rates of death and heart failure in acute coronary syndromes, 19992006. JAMA. 2007;297:1892–1900. doi: 10.1001/jama.297.17.1892. [DOI] [PubMed] [Google Scholar]
- 7.Efficacy and safety of tenecteplase in combination with enoxaparin, abciximab, or unfractionated heparin: the ASSENT-3 randomised trial in acute myocardial infarction. Lancet. 2001;358:605–613. doi: 10.1016/S0140-6736(01)05775-0. [DOI] [PubMed] [Google Scholar]
- 8.Wu AH, Parsons L, Every NR, Bates ER. Hospital outcomes in patients presenting with congestive heart failure complicating acute myocardial infarction: a report from the Second National Registry of Myocardial Infarction (NRMI-2) J Am Coll Cardiol. 2002;40:1389–1394. doi: 10.1016/s0735-1097(02)02173-3. [DOI] [PubMed] [Google Scholar]
- 9.Hellermann JP, Goraya TY, Jacobsen SJ, Weston SA, Reeder GS, Gersh BJ, Redfield MM, Rodeheffer RJ, Yawn BP, Roger VL. Incidence of heart failure after myocardial infarction: is it changing over time? Am J Epidemiol. 2003;157:1101–1107. doi: 10.1093/aje/kwg078. [DOI] [PubMed] [Google Scholar]
- 10.Lloyd-Jones DM, Larson MG, Leip EP, Beiser A, D’Agostino RB, Kannel WB, Murabito JM, Vasan RS, Benjamin EJ, Levy D. Lifetime risk for developing congestive heart failure: the Framingham Heart Study. Circulation. 2002;106:3068–3072. doi: 10.1161/01.cir.0000039105.49749.6f. [DOI] [PubMed] [Google Scholar]
- 11.Spencer F, Scleparis G, Goldberg RJ, Yarzebski J, Lessard D, Gore JM. Decade-long trends (1986 to 1997) in the medical treatment of patients with acute myocardial infarction: A community-wide perspective. Am Heart J. 2001;142:594–603. doi: 10.1067/mhj.2001.117776. [DOI] [PubMed] [Google Scholar]
- 12.Hochman JS, Sleeper LA, Webb JG, Sanborn TA, White HD, Talley JD, Buller CE, Jacobs AK, Slater JN, Col J, McKinlay SM, LeJemtel TH. Early revascularization in acute myocardial infarction complicated by cardiogenic shock. SHOCK Investigators. Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock. N Engl J Med. 1999;341:625–634. doi: 10.1056/NEJM199908263410901. [DOI] [PubMed] [Google Scholar]
- 13.Holmes DR., Jr Cardiogenic shock: a lethal complication of acute myocardial infarction. Rev Cardiovasc Med. 2003;4:131–135. [PubMed] [Google Scholar]
- 14.Torabi A, Cleland JG, Khan NK, Loh PH, Clark AL, Alamgir F, Caplin JL, Rigby AS, Goode K. The timing of development and subsequent clinical course of heart failure after a myocardial infarction. Eur Heart J. 2008;29:859–870. doi: 10.1093/eurheartj/ehn096. [DOI] [PubMed] [Google Scholar]
- 15.Tu JV, Nardi L, Fang J, Liu J, Khalid L, Johansen H. National trends in rates of death and hospital admissions related to acute myocardial infarction, heart failure and stroke, 1994–2004. CMAJ. 2009;180:E118–125. doi: 10.1503/cmaj.081197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Yeh RW, Sidney S, Chandra M, Sorel M, Selby JV, Go AS. Population trends in the incidence and outcomes of acute myocardial infarction. N Engl J Med. 362:2155–2165. doi: 10.1056/NEJMoa0908610. [DOI] [PubMed] [Google Scholar]
- 17.Setoguchi S, Glynn RJ, Avorn J, Mittleman MA, Levin R, Winkelmayer WC. Improvements in long-term mortality after myocardial infarction and increased use of cardiovascular drugs after discharge: a 10-year trend analysis. J Am Coll Cardiol. 2008;51:1247–1254. doi: 10.1016/j.jacc.2007.10.063. [DOI] [PubMed] [Google Scholar]
- 18.Rott D, Behar S, Leor J, Hod H, Boyko V, Mandelzweig L, Gottlieb S. Effect on survival of acute myocardial infarction in Killip classes II or III patients undergoing invasive coronary procedures. Am J Cardiol. 2001;88:618–623. doi: 10.1016/s0002-9149(01)01802-1. [DOI] [PubMed] [Google Scholar]
- 19.Dziewierz A, Siudak Z, Rakowski T, Zdzienicka J, Dykla D, Mielecki W, Dubiel JS, Dudek D. In-hospital management and mortality in elderly patients with non-ST-segment elevation acute coronary syndromes treated in centers without on-site invasive facilities. Cardiol J. 2008;15:451–457. [PubMed] [Google Scholar]
- 20.Dziewierz A, Siudak Z, Rakowski T, Mielecki W, Giszterowicz D, Dubiel JS, Dudek D. More aggressive pharmacological treatment may improve clinical outcome in patients with non-ST-elevation acute coronary syndromes treated conservatively. Coron Artery Dis. 2007;18:299–303. doi: 10.1097/MCA.0b013e32812cb91c. [DOI] [PubMed] [Google Scholar]
- 21.Rogers WJ, Frederick PD, Stoehr E, Canto JG, Ornato JP, Gibson CM, Pollack CV, Jr, Gore JM, Chandra-Strobos N, Peterson ED, French WJ. Trends in presenting characteristics and hospital mortality among patients with ST elevation and non-ST elevation myocardial infarction in the National Registry of Myocardial Infarction from 1990 to 2006. Am Heart J. 2008;156:1026–1134. doi: 10.1016/j.ahj.2008.07.030. [DOI] [PubMed] [Google Scholar]