Author + information
- Received December 16, 1997
- Revision received June 24, 1998
- Accepted September 15, 1998
- Published online January 1, 1999.
- Michele Senni, MD∗,
- Richard J Rodeheffer, MD, FACC∗,
- Christophe M Tribouilloy, MD, PhD∗,
- Jonathan M Evans, MD†,
- Steven J Jacobsen, MD, PhD‡,
- Kent R Bailey, PhD‡ and
- Margaret M Redfield, MD, FACC∗,*
- ↵*Address for correspondence: Dr. Margaret M. Redfield, Division of Cardiovascular Diseases and Internal Medicine, Mayo Clinic, 200 First Street Southwest, Rochester, Minnesota 55905
Objectives. We evaluated the use and the impact of echocardiography in patients receiving an initial diagnosis of congestive heart failure in Olmsted County, Minnesota, in 1991.
Background. The American College of Cardiology/American Heart Association clinical practice guidelines recommend echocardiography in all patients with suspected congestive heart failure. No data are available on use and impact of echocardiography in management of congestive heart failure in a community.
Methods. The medical records linkage system of the Rochester Epidemiology Project was used to identify all 216 patients who satisfied the Framingham criteria for congestive heart failure. Of these, 137 (63%) underwent echocardiography within 3 weeks before or after the episode of congestive heart failure (Echo group), and the other 79 patients constitute the No-Echo group.
Results. The No-Echo group patients were older (p = 0.022), were more likely to be female (p = 0.072), had milder symptoms (p = 0.001) and were less often hospitalized at diagnosis (p = 0.001). Fewer patients in the No-Echo group were treated with angiotensin-converting enzyme inhibitors (p = 0.001). Advanced age (≥80 years), lower New York Heart Association functional class, absence of a fourth heart sound on examination, absence of cardiomegaly or signs of congestive heart failure on chest radiography and absence of known valve disease were independently related to the decision not to obtain an echocardiogram. Survival after adjustment for age, functional class and gender was lower in the No-Echo group than the Echo group (risk ratio = 0.607, p = 0.017).
Conclusions. The underuse of echocardiography appears to be associated with poorer survival and underuse of angiotensin-converting enzyme inhibitor therapy.
Congestive heart failure is a syndrome with high morbidity and mortality. It is now recognized that 30% to 50% of patients with the clinical syndrome of congestive heart failure have normal systolic function, implying that abnormal left ventricular diastolic function is the mechanism responsible for producing congestive symptoms in these patients (1). The approach to treatment may differ, depending on whether systolic or diastolic dysfunction is predominant. Because clinical findings fail to distinguish patients with systolic or diastolic dysfunction (2), American College of Cardiology/American Heart Association (ACC/AHA) and Agency for Health Care Policy and Research (AHCPR) guidelines recommend echocardiography, a widely available noninvasive technique, in all patients with suspected congestive heart failure (3–5). Doppler echocardiography is able to assess systolic and diastolic function, valvular heart disease, pulmonary artery pressure and pericardial disease in patients with suspected congestive heart failure. Doppler echocardiography can lead to more accurate diagnosis and more appropriate management in cardiovascular disease (5,6).
However, no data are available on use and outcome of patients studied with echocardiography in the management of congestive heart failure in the community. Furthermore, the effects of management changes on morbidity and mortality as a result of Doppler echocardiography in patients with congestive heart failure are unknown (7). Therefore, using the Rochester Epidemiology Project, we retrospectively evaluated the relationship between echocardiography use and morbidity and mortality in patients receiving an initial diagnosis of congestive heart failure in Olmsted County, Minnesota, in 1991. Specifically, we analyzed the characteristics of patients who did or did not receive an echocardiogram at diagnosis of congestive heart failure and compared morbidity, mortality and therapy in those congestive heart failure patients in whom an echocardiogram was or was not obtained.
In 1990, 106,470 people lived in Olmsted County, Minnesota; 96% were white, 28% were older than age 45 years and 11% were older than age 65 years. The Olmsted County population is unique in that the residents of this county receive their medical care from a limited number of health care providers: Mayo Clinic, Olmsted Medical Center, their affiliated hospitals and a few private practices. Most care is provided through the Mayo Clinic in a division of internists and family practitioners dedicated to the care of patients in Olmsted County. The Rochester Epidemiology Project has developed a diagnostic index that includes all diagnoses for residents of Olmsted County made during outpatient office visits, during clinic consultations, during emergency department visits, in nursing homes, at hospital admissions, at autopsy examination and at death certification. Therefore, the Rochester Epidemiology Project records linkage system captures information on essentially all medical care (diagnosis and procedures) for the local population (8,9).
This study was approved by the Mayo Clinic Institutional Review Board. The clinical records of all residents of Olmsted County who received a diagnosis of congestive heart failure for the first time during 1991 (January 1 to December 31) were reviewed retrospectively. All patients in whom slightly modified Framingham diagnostic criteria (10)for congestive heart failure were met were included in the current study. Physicians’ notes were examined for indication of major and minor criteria for congestive heart failure. The major diagnostic criteria included: paroxysmal nocturnal dyspnea, orthopnea, abnormal jugular venous distension, rales, cardiomegaly, pulmonary edema in the presence of the third heart sound and central venous pressure of more than 16 cm of water. The minor criteria included: edema, night cough, dyspnea on exertion, hepatomegaly, pleural effusion, tachycardia (more than 120 beats per minute) and weight loss of 4.5 kg or more in 5 days. Persons were assigned a diagnosis of congestive heart failure if two major criteria were present or, alternatively, if one major and two minor criteria were present concurrently. In addition to fulfilling diagnostic criteria, persons must have established residence in Olmsted County, Minnesota, at least 1 year before diagnosis for inclusion in the study. This guideline was intended to exclude persons who might have moved into the county for the diagnosis or treatment of their congestive heart failure (11).
The age at diagnosis, gender and specific clinical characteristics were recorded. The presence of underlying cardiovascular disease, such as coronary artery disease (clinically or angiographically defined), systemic hypertension (history of hypertension necessitating antihypertensive therapy, two blood pressure measurements with a systolic pressure of 160 mm Hg or higher or a diastolic pressure of 95 mm Hg or higher at diagnosis) or valvular heart disease (history of clinically significant valve disease in the past or at heart failure episode), was noted. The subsequent medical therapy prescribed after the diagnosis was recorded. Any cardiac surgery within 3 months after the diagnosis of congestive heart failure was noted. All congestive heart failure–related hospitalizations subsequent to the diagnosis were recorded. Retrospective determination of cause of death is unreliable; thus, no effort was made to determine cause of death (12).
Once the date of diagnosis of congestive heart failure was determined, an echocardiogram was considered to reflect the level of systolic function if it was obtained within 3 weeks before or after the diagnosis of congestive heart failure. One patient had assessment of systolic function by radionuclide ventriculography and was excluded from the analysis. No patient had assessment of systolic function by contrast ventriculography alone. Echocardiograms were performed at Mayo Clinic Echocardiographic Laboratory according to the recommendations of the American Society of Echocardiography (13).
The study population was divided into two groups: the No-Echocardiography group included patients who did not undergo echocardiography within 3 weeks of diagnosis (before or after), and the Echocardiography group included patients with echocardiographic assessment within 3 weeks before or after diagnosis of congestive heart failure.
Demographic and clinical characteristics were compared between the Echocardiography and No-Echocardiography groups with the unpaired Student ttest and chi-square tests. The independent association of selected clinical variables with the performance of an echocardiogram was examined by logistic regression analysis. Only variables with p values <0.10 in univariate analyses were included in the final multiple logistic model. Survival was estimated by Kaplan–Meier curves, and differences between curves were assessed by the log-rank test. Expected survival overall or for subgroups was based on age- and gender-matched mortality data for the 1990 Minnesota white population, and comparisons of observed and expected survival were based on the one-sample log-rank test. Proportional hazards models were used to assess predictors of survival. Survival curves were calculated for the Echocardiography and No-Echocardiography groups, adjusted for patient characteristics by the Cox proportional hazards model. From the model estimates of the underlying curve and the Cox regression parameter estimates, the predicted curves for all patients in both groups were averaged, once with the Echocardiography/No-Echocardiography group variable set to Yes and once with this variable set to No (14). The statistical packages S-PLUS and SAS software were used. A p value <0.05 was considered significant.
Two hundred sixteen patients fulfilled study criteria. Of these, 79 patients (37%) did not have an echocardiogram (or other assessment of systolic function) within 3 weeks before or after the diagnosis of congestive heart failure and constitute the No-Echocardiography group. The remaining 137 patients (63%) had an echocardiogram within the specified time and represent the Echocardiography group. The clinical profile of the two groups is summarized in Table 1. The No-Echocardiography group patients were older, were more likely to be female, had milder symptoms and were less likely to be inpatients at diagnosis. The prevalence of coronary artery disease, hypertension and previously known valvular disease was similar in Echocardiography and No-Echocardiography patients.
Objective clinical findings in the two groups are listed in Table 2. On physical examination, third or fourth heart sounds were more common in the Echocardiography group. On chest radiography, interstitial pulmonary edema and bilateral pleural effusions were more common in the Echocardiography group, whereas cardiomegaly was present equally in the two groups. An increased creatinine concentration was found more frequently in the Echocardiography group. Electrocardiographic features were similar in the two groups.
Medical and surgical therapies prescribed in the two groups after the diagnosis are summarized in Table 3. Patients in the No-Echocardiography group were less likely to be treated with angiotensin-converting enzyme inhibitors, digoxin or anticoagulant drugs, but they were as likely to be treated with diuretic agents. Few patients underwent coronary artery bypass graft surgery, but all who did had an echocardiogram. Three patients underwent valve replacement within 3 months after diagnosis of congestive heart failure.
Logistic regression analysis was performed to determine which clinical characteristics influenced the decision to obtain an echocardiogram in patients with a new diagnosis of congestive heart failure. All variables that appeared to have potential discriminatory value were included in the multivariate logistic analyses (Table 4). Advanced age (≥80 years) and lower New York Heart Association (NYHA) class, absence of a fourth heart sound on examination, absence of cardiomegaly or signs of congestive heart failure on chest radiography and absence of known valve disease were independently related to the decision not to obtain an echocardiogram.
To further evaluate the impact of echocardiography on the use of angiotensin-converting enzyme inhibitors in the treatment of patients with congestive heart failure in the community, we compared the use of angiotensin-converting enzyme inhibitors in patients with congestive heart failure and reduced ejection fraction (<40%) and those without an echocardiogram. Seventy-five percent of patients with congestive heart failure in whom an echocardiogram was obtained and an ejection fraction <40% was reported were treated with angiotensin-converting enzyme inhibitors compared with 28% in the No-Echocardiography group.
Both groups showed significantly decreased survival compared with their respective age- and gender-matched expected survival (p = 0.0001) (Fig. 1). Crude survival was not significantly different between the Echocardiography and No-Echocardiography groups. However, by multivariate analysis controlling for age, NYHA functional class and gender, performance of an echocardiogram was associated with a lower risk of death (risk ratio = 0.607, p = 0.017) (Fig. 2). Although the No-Echocardiography group had milder symptoms at diagnosis, 20% were later hospitalized for congestive heart failure. No-Echocardiography group patients had fewer subsequent hospitalizations than Echocardiography group patients (0.3 ± 0.7 compared with 0.7 ± 1.1 hospitalizations/patient, p = 0.0004).
We performed a population-based study to evaluate the use of echocardiography in patients with congestive heart failure and its impact on the outcome. We found that echocardiography was not performed in 37% of patients with clinical congestive heart failure and that those who had an echocardiogram had better survival, after adjusting for their significantly worse symptomatic status at presentation. In addition, we found that patients who did not undergo echocardiography were treated less often with angiotensin-converting enzyme inhibitors, a therapy that has been shown to improve survival in patients with decreased ejection fraction (15,16).
Utility of echocardiography in the management of congestive heart failure
The use of echocardiography is recommended by the ACC/AHA and AHCPR guidelines for the management of patients with suspected congestive heart failure (3,4)and also by the ACC/AHA guidelines for the clinical application of echocardiography (17). The assessment of systolic function provided by echocardiography is critical to the management of patients with heart failure. Angiotensin-converting enzyme inhibitors are indicated in patients with an ejection fraction <40% regardless of their symptoms (18,19). Although the recently published digoxin trial did not demonstrate an effect on survival, digoxin therapy did decrease the subsequent hospitalization for congestive heart failure in patients with abnormal systolic function and symptoms of heart failure (20). Most authorities agree that digoxin probably is not indicated in patients with heart failure and normal systolic function in the presence of normal sinus rhythm.
In patients with heart failure and normal ejection fraction, echocardiography may provide clues to the etiology of their diastolic heart failure by assessing valvular function, left ventricular mass, regional wall motion abnormalities and diastolic function. These clues may guide further workup and treatment. Patients with heart failure also may have valvular dysfunction; echocardiography provides a means to assess primary or secondary valvular lesions (3,17). In rare cases, echocardiography may document more obscure etiologies of heart failure such as hypertrophic cardiomyopathy, pericardial disease or other forms of cardiomyopathy such as amyloid or primary restrictive cardiomyopathy (3). It may guide therapy by identifying patients with advanced diastolic dysfunction who may experience complications if treated aggressively with vasodilator or diuretic agents (21,22).
Data derived from echocardiography also are useful in assessing prognosis in patients with heart failure. Left ventricular size (23), ejection fraction (24)and pulmonary artery systolic pressure, as estimated by the velocity of tricuspid regurgitation (25), and right ventricular function (26)and diastolic function (26), as determined by echocardiography, have been shown to have predictive value in patients with heart failure. Although repeated measurement of systolic function is not recommended in the follow-up of patients with heart failure, there are many instances in which such assessment may be indicated, particularly when there has been a dramatic change in clinical status that may indicate improvement or deterioration in systolic function (27).
Underuse of echocardiography in the community
Despite the recognized utility of echocardiography in the management of patients with heart failure, 37% of patients with an initial diagnosis of heart failure in this community setting were not studied with echocardiography. This frequency is similar to the report of Krumholz et al. (28)that 44% of Medicare patients at risk for left ventricular dysfunction after an acute myocardial infarction did not have any assessment of ejection fraction to determine the suitability for angiotensin-converting enzyme inhibitor therapy. There are many factors that may explain the underuse of echocardiography observed in this study. Patients without echocardiographic examination were older; in fact, with a mean age of 80 years, this population represents the “very old.” These patients are in general understudied in the heart failure literature. A recent editorial (29)underscores the lack of data available to the clinician in managing the very elderly patient with heart failure. These patients are often managed by noncardiologists. The managing physician may think that other concomitant medical conditions are more likely to affect the quality of life and survival in these very elderly patients than their recent diagnosis of heart failure, and therefore a less aggressive approach may be chosen. Elderly patients are less mobile, and in nursing home patients or other institutionalized patients, echocardiography may not be easily available. Cost considerations may come into play. These other factors may influence the management of these very elderly patients; however, age alone should not be an indication to be less aggressive in the management of heart failure (30).
One of the few trials to include elderly patients was the CONSENSUS I Trial, with a mean patient age of 71 years (15). These patients were extremely symptomatic, and this trial of angiotensin-converting enzyme inhibitors was the most positive trial published to date. The recently published ELITE Trial looked specifically at elderly patients (mean age, 74 years) with congestive heart failure and demonstrated a significant impact on survival and symptoms of treatment with angiotensin II blockers (31). Thus, age alone does not preclude an ability to have an impact on survival and symptoms in these patients. Furthermore, congestive heart failure is extremely common in patients age 80 years or older, with a prevalence approaching 10% in the Framingham Study (32)cohort. Congestive heart failure is the most common diagnosis-related group in elderly patients. The majority of the cost related to congestive heart failure is linked to hospitalizations (33), and aggressive treatment to avoid hospitalization may have a positive cost–benefit profile in addition to its important effect on quality of life in the elderly patient.
The patients who did not have an echocardiogram had milder congestive heart failure, as assessed by chest radiography and NYHA functional class. The managing physician may have thought that a less aggressive approach was warranted. However, this assumption would not be valid given the findings of the Studies of Left Ventricular Dysfunction Prevention (18)and Survival and Ventricular Enlargement (19)Trials published in 1992, 1 year after the onset of congestive heart failure in these patients, which indicated that even asymptomatic patients with systolic dysfunction benefit from angiotensin-converting enzyme inhibitor therapy in terms of mortality, hospitalizations and progression to more severe heart failure.
The presence of comorbidities may have affected the physicians’ decision to obtain an echocardiogram, because they may have felt that these comorbidities would have limited potential treatment with angiotensin-converting enzyme inhibitors or other heart failure therapies. Although comorbidity was not extensively or formally assessed in the current study, the fact that fewer patients in the No-Echocardiography group were inpatients and fewer had increased creatinine concentrations indicates that excessive comorbidities may not have been present in these patients.
The managing physician may have thought that these less symptomatic patients were less at risk for systolic dysfunction. However, the clinical characteristics of the No-Echocardiography group do not indicate that they were at low risk for systolic dysfunction. Many of them had abnormal chest radiographs, evidence of previous myocardial infarction or left bundle branch block on their electrocardiogram or coronary artery disease, or were in class III–IV. Furthermore, in 1991 the appreciation that heart failure could be present with normal systolic function was even less widespread than it is currently, and studies have consistently demonstrated that clinical variables are unable to differentiate heart failure due to systolic from that due to diastolic dysfunction (1,2).
Implications of underuse of echocardiography in the community
Previous studies (5,6)reported that echocardiography can provide a new diagnosis in 5% to 12% of patients referred for echocardiographic assessment as a part of the clinical evaluation for different cardiovascular diseases. Furthermore, echocardiography leads to a change in the management in 13% to 37% of patients (5,6). The impact of use of a diagnostic test is often harder to assess than that of a therapeutic agent, procedure or device, especially because the use of the diagnostic test is then linked to therapy. However, whether these changes in the management are beneficial has never been proven. In this uncontrolled study, multiple other factors including referral to a cardiologist could affect outcome. Indeed, the failure to obtain an echocardiogram or to use an angiotensin-converting enzyme inhibitor may be linked to lack of referral to a cardiologist or heart failure specialist. Previous studies (34,35)demonstrated improved outcomes for hospitalized heart failure patients when managed by a cardiologist or heart failure specialist. However, as recently emphasized, the majority of very elderly patients with heart failure are not managed by a cardiologist (29), and if this practice pattern is to continue, appropriate care algorithms including echocardiography or other assessment of ventricular function must be used. The current study indicates that both morbidity and mortality were associated with the lack of the information provided by echocardiography. Survival was worse after controlling for other factors, and 20% of patients were hospitalized subsequently with congestive heart failure, despite the fact that 75% of the patients were in NYHA class II at diagnosis.
Despite the advantages afforded by this community-based study, some limitations must be considered. First, this study includes limitations of retrospective cohort studies and the realities of clinical record keeping. The Framingham criteria are specific but relatively insensitive to define congestive heart failure status, especially early manifestations (36). In some patients, echocardiography could have been planned later and not obtained within three weeks after the diagnosis. Comorbidity, in this very old population, might have affected the mortality of both groups, and detailed assessment of comorbidity was not performed. Retrospective determination of cause of death is difficult, and the degree to which congestive heart failure contributed to death is often impossible to determine. No effort was made to determine cause of mortality, and thus mortality may not truly reflect deaths related to congestive heart failure.
Although an objective evaluation of left ventricular ejection fraction and of other structural and hemodynamic data is indicated in the proper clinical management of patients with congestive heart failure, echocardiography was underused in the patients with new onset of congestive heart failure in the community. The underuse of echocardiography appears to be associated with decreased survival, increased morbidity and underuse of angiotensin-converting enzyme therapy.
☆ This study was funded in part by grants from the Mayo Foundation, the Joseph P. and Jeanne Sullivan Foundation, Chicago, Illinois, the Miami Heart Research Institute, the NHLBI (1R01 HL55502-01A1), the U.S. Public Health Service, National Institutes of Health (AR 30582) and the Federation Française de Cardiologie (Dr. Tribouilloy).
- American College of Cardiology/American Heart Association
- Agency for Health Care Policy and Research
- New York Heart Association
- Received December 16, 1997.
- Revision received June 24, 1998.
- Accepted September 15, 1998.
- American College of Cardiology
- Vasan R.S,
- Benjamin E.J,
- Levy D
- Williams J.F,
- Bristow M.R,
- Fowler M.B,
- et al.
- Konstam M.A,
- Dracup K,
- Bottorff M.B,
- et al.
- Krumholz H.M,
- Douglas P.S,
- Goldman L,
- Waksmonski C
- Douglas P.S
- Schiller N.B,
- Shah P.M,
- Crawford M,
- et al.
- Cheitlin M.D,
- Alpert J.S,
- Armstrong W.F,
- et al.
- Pfeffer M.A,
- Braunwald E,
- Moye L.A,
- et al.
- Stevenson W.G,
- Stevenson L.W,
- Middlekauff H.R,
- Saxon L.A
- Enriquez-Sarano M,
- Rossi A,
- Seward J.B,
- Bailey K.R,
- Tajik A.J
- Pinamonti B,
- Di Lenarda A,
- Sinagra G,
- Camerini F
- Steimle A.E,
- Stevenson L.W,
- Fonarow G.C,
- Hamilton M.A,
- Moriguchi J.D
- Krumholz H.M,
- Radford M.J,
- Vaccarino V,
- Wang Y,
- Hennen J,
- Ellerbeck E
- Parmley W.W
- Fonarow G.C,
- Stevenson L.W,
- Walden J.A,
- et al.
- Reis S.E,
- Holubkov R,
- Edmundowicz D,
- et al.