Approach to diagnosis of asymptomatic left ventricular systolic dysfunction

INTRODUCTION — Numerous cardiac disorders can lead to a decline in left ventricular ejection fraction (LVEF) without heart failure (HF) symptoms, which is referred to as asymptomatic LV systolic dysfunction (ALVSD). The finding of ALVSD requires further evaluation to identify its cause and may require treatment.

The epidemiology, clinical presentation, evaluation, and diagnosis of ALVSD will be reviewed here.

The management of ALVSD is presented separately. (See "Management and prognosis of asymptomatic left ventricular systolic dysfunction".)

DEFINITION — ALVSD is defined as the presence of an LVEF <55 percent in the absence of HF symptoms. ALVSD is synonymous with the American College of Cardiology Foundation/American Heart Association’s Stage B HF (table 1).

The diagnosis of ALVSD is dependent on the LVEF; we do not recommend using other measures of systolic function (eg, qualitative LVEF, global longitudinal strain, fractional shortening) to establish the diagnosis of ALVSD, though these measures can be used to corroborate the finding of a low EF. (See 'Review of echocardiography' below.)

EPIDEMIOLOGY — The prevalence of ALVSD is between 1 to 8 percent but varies with the method used to define LV systolic dysfunction (eg, echocardiography, cardiovascular magnetic resonance [CMR] imaging), the choice of LVEF used to define LV systolic dysfunction, and the population under study, as follows:

●LV systolic dysfunction detected by echocardiography – In studies that used an echocardiographic definition of LV systolic dysfunction, the prevalence of ALVSD is reported as follows:

•In the Strong Heart Study, the prevalence and severity of ALVSD based on LVEF were [1]:

-Any decrease in LVEF (ie, 54 percent or lower) – 5.2 percent

-LVEF 45 to 54 percent – 3.7 percent

-LVEF 35 to 44 percent – 1 percent

-LVEF ≤34 percent – <1 percent

When wall motion abnormalities were used to define LV systolic dysfunction, 5 percent of patients had ALVSD.

•In a study that screened residents of Minnesota for LV systolic dysfunction, the prevalence of ALVSD (LVEF ≤50 percent) in the general population was 4.9 percent (95% CI 4-6 percent) and more severe ALVSD (ie, LVEF ≤40 percent) was present in 1.1 percent (95% CI 0.7-1.7 percent) [2]. Male patients had an increased risk of ALVSD (7.9 versus 4.9 percent in females), as did patients with one or more of the following risk factors: age ≥65 years, history of hypertension, and history of coronary artery disease (CAD; 4.9 versus 10.9 percent among persons with at least one risk factor; odds ratio [OR] 0.42, 95% CI 0.29-0.62).

•In the Echocardiographic Heart of England screening, 1 percent of screened patients who had an LVEF less than 40 percent did not have HF symptoms [3].

•The Cardiovascular Health Study included patients ≥65 years of age and used an LVEF cutoff of ≤55 percent to define LV systolic dysfunction [4]. The prevalence of ALVSD was 7.3 percent.

●LV systolic dysfunction detected by magnetic resonance imaging – In the Framingham Heart Study Offspring Cohort, participants were screened with CMR imaging, and the prevalence of ALVSD was 6.5 percent. This estimate includes patients with global LV systolic dysfunction or a regional wall motion abnormality [5]. Wall motion abnormalities were more likely in patients with male sex (adjusted OR 3.69, 95% CI 2.1-6.6) and hypertension (adjusted OR 2.2, 95% CI 1.2-4.1) [5].

WHEN TO EVALUATE

Patients with examination findings consistent with HF — In asymptomatic patients, the physical examination findings that warrant further assessment of LV function are the same as the findings in symptomatic patients (eg, jugular venous distension, S3 heart sound, or peripheral edema). Additional physical examination findings that require further evaluation are discussed elsewhere. (See "Heart failure: Clinical manifestations and diagnosis in adults", section on 'Physical examination'.)

Patients with cardiac diseases — In patients with known cardiac disease (eg, coronary artery disease, mitral valve disease), the suspicion for LV systolic dysfunction and the need for evaluation is determined by the underlying cardiac disease. Diseases that may require testing for ALVSD include:

●Myocardial infarction. (See "Role of echocardiography in acute myocardial infarction", section on 'Left ventricular systolic function'.)

●Mitral valve regurgitation. (See "Chronic primary mitral regurgitation: General management", section on 'Monitoring'.)

●Aortic valve regurgitation. (See "Clinical manifestations and diagnosis of chronic aortic regurgitation in adults", section on 'Echocardiogram'.)

Additional causes of reduced LVEF can be found elsewhere. (See "Causes of dilated cardiomyopathy".)

Patients with unexplained signs of cardiac injury or disease — Patients without HF symptoms but who have evidence of cardiac injury or disease should undergo an evaluation for their specific abnormality with assessment for LV systolic dysfunction as indicated. Signs of cardiac injury or disease include:

●ECG – Electrocardiogram (ECG) abnormalities, such as Q waves, LV hypertrophy, and left bundle branch block, suggest the presence of LV injury.(See "Left ventricular hypertrophy: Clinical findings and ECG diagnosis", section on 'Clinical findings and evaluation' and "Left bundle branch block", section on 'Evaluation'.)

ECG analysis with artificial intelligence (AI) methods can suggest the presence of LV systolic dysfunction. Similar to other ECG findings, abnormal LV systolic dysfunction identified by AI-based ECG image analysis requires further diagnostic evaluation (eg, echocardiography) to confirm or exclude the presence of LV dysfunction before making the diagnosis of ALVSD and before prescribing treatment. Studies of AI-based ECG analysis include:

•In a cohort of 116,210 patients who had a total of 385,601 12-lead ECGs obtained for various reasons and who had echocardiography within 15 days, analysis by an AI method had a sensitivity of 89 percent and specificity of 77 percent for LVEF <40 percent (area under the receiver operator curve [AUC] 0.91, 95% CI 0.901-0.919) [6]. In external validation cohorts, the diagnostic accuracy of this method was similar to that in the derivation cohort.

In a separate analysis of the same ECGs limited to tracings from lead I, which was performed to simulate tracings produced by wearable devices, a separate AI-based method had 92 percent sensitivity and 68 percent specificity (AUC 0.9, 95% CI 0.89-0.91) for LVEF ≤40 percent [7]. For both studies, it was unclear if the patients were asymptomatic.

•After derivation of a model from 44,959 patients with a 12-lead ECG and echocardiography within 14 days of each other, an AI method had 86 percent sensitivity and 86 percent specificity for LVEF ≤35 percent in a validation cohort of 52,870 similar patients [8]. Using this method, a trial was performed to evaluate the ability of the AI method to detect LVEF ≤50 percent among ECGs obtained during standard clinical practice [9]. The trial included 22,641 patients who did not have a HF diagnosis, and AI-based ECG screening increased the likelihood of detection of LVEF ≤50 percent by a small amount compared with standard practice (1.6 versus 2.1 percent; odds ratio 1.32, 95% CI 1.01-1.61).

●Laboratory abnormalities – Laboratory abnormalities, such as elevated B-type natriuretic peptide (BNP) or N-terminal pro-BNP. (See "Natriuretic peptide measurement in heart failure", section on 'Our approach'.)

●Radiographic abnormalities– Radiographic abnormalities, such as cardiomegaly or pulmonary edema. (See "Heart failure: Clinical manifestations and diagnosis in adults", section on 'Chest radiograph'.)

A comprehensive discussion of the signs of cardiac injury and the appropriate evaluation for these signs is beyond the scope of this topic.

Other patients at risk for reduced ejection fraction — Patients with diseases (eg, human immunodeficiency virus [HIV]) or cardiotoxic exposures (eg, anthracycline-based chemotherapy) associated with LV systolic dysfunction should be tested for LV systolic dysfunction according to the standards for those diseases and exposures. Examples include:

●Patients who received anthracycline-containing chemotherapy. (See "Risk and prevention of anthracycline cardiotoxicity".)

●Patients with HIV. (See "Cardiac and vascular disease in patients with HIV", section on 'Diagnosis and evaluation'.)

WHEN EVALUATION IS UNNECESSARY

Testing in the general population — We do not perform routine screening for ALVSD in unselected adults with either echocardiography or natriuretic peptide testing. Screening for ALVSD has not been shown to be effective in the general population, which is likely a result of the relatively low prevalence of ALVSD, the inability of less expensive tests (ie, serum B-type natriuretic peptide [BNP]) to accurately identify patients with ALVSD, and the unclear benefit of early treatment of ALVSD [10]. While therapy for patients with an LVEF is ≤40 percent is effective, patients with an LVEF in this range are rarely detected by screening. (See 'Epidemiology' above.)

Of the two proposed methods for population screening, BNP screening has been studied more extensively than echocardiography, which is likely due to the prohibitively high cost of echocardiography as a screening test. In general, BNP has a sensitivity <70 percent, which is unacceptably low for population screening. The studies of BNP screening include:

●A prospective cohort study from the Framingham Heart Study included 3177 asymptomatic participants (mean age 58) of whom 5.6 percent had LV systolic dysfunction [11]. Plasma BNP values did not achieve a sensitivity for LV systolic dysfunction greater than 30 percent.

●Suboptimal performance of natriuretic peptide screening was also noted in a community-based study of 2042 randomly selected residents of Olmsted County, Minnesota [12]. This sample of asymptomatic patients was screened with plasma BNP measurements and echocardiography, and 1 percent had an LVEF ≤40 percent. The ability of plasma BNP to detect or exclude the presence of LV systolic dysfunction was low (area under the receiving operating characteristic curve 0.51). The authors concluded that BNP screening would lead to nondiagnostic echocardiography in 10 to 40 percent of the screened population and would miss 10 to 60 percent of patients with ALVSD.

●Similar to BNP, screening with plasma N-terminal pro-BNP (NT-proBNP) to detect individuals with ALVSD has not proven effective [13-16]. Screening with NT-proBNP typically has a negative predictive value ≥99 percent but the positive predictive value (eg, 1 to 26 percent) and sensitivity (33 to 80 percent) are too low for screening in the general population.

Patients with hypertension or a CAD equivalent — In patients who do not have HF symptoms or signs of cardiac disease but who have advanced age, hypertension, diabetes, or peripheral vascular disease, we do not routinely screen for ALVSD. While these patients are at higher risk for LV systolic dysfunction, routine testing for LV systolic dysfunction in this population has not proven effective (see 'Epidemiology' above). The studies that describe the effect of routine testing for ALVSD in high-risk populations are difficult to interpret; these studies often included patients who were symptomatic, known to have abnormal BNP test results, or had known cardiac disease [14,17,18].

APPROACH TO PATIENTS WITH SUSPECTED ALVSD

Reassess for heart failure symptoms — Patients with evidence of LV systolic dysfunction should be reevaluated for exertional and rest symptoms to be certain that symptoms are truly absent; the management of patients with symptomatic HF, especially those with an LVEF <40 percent, is markedly different than the management of asymptomatic patients. (See "Management and prognosis of asymptomatic left ventricular systolic dysfunction" and "Overview of the management of heart failure with reduced ejection fraction in adults".)

In patients where management would change based on the presence or absence of HF symptoms (eg, defibrillator placement) and the subjective history is inconclusive, we recommend a treadmill exercise test or a cardiopulmonary exercise test (CPET) to determine whether a cardiac limitation to exercise is present. (See "Exercise assessment and measurement of exercise capacity in patients with coronary heart disease".)

While patient-reported symptoms are required to assign an appropriate New York Heart Association class and choose therapy, there is evidence to suggest that some patients with apparent ALVSD have decreased exercise capacity. In a single-center study, patients with ALVSD and patients with HF with reduced ejection fraction (HFrEF) underwent CPET testing at the discretion of the referring clinician [19]. Peak oxygen consumption (VO₂) was low in patients with ALVSD and patients with HFrEF (16 versus 14.2 mL/kg/min) as was predicted peak VO₂ (59 versus 52 percent).

Obtain echocardiography — In patients who have signs of LV systolic dysfunction (eg, cardiomegaly, edema), we initially test for LV systolic dysfunction with echocardiography. This preference is largely based on our experience and the ability of echocardiography to diagnose ALVSD and identify causes of ALVSD. We do not consider B-type natriuretic peptide testing to be an alternative approach for ALVSD testing; it does not detect all patients who have LV systolic dysfunction and may be elevated due to causes other than ALVSD. (See 'Definition' above and 'Testing in the general population' above.)

Review of echocardiography — For patients who already have an echocardiogram or other imaging study showing an LVEF <55 percent, the diagnostic quality of the study should be evaluated to confirm that the measurement of low LVEF is accurate. Common scenarios that can lead to a falsely low LVEF in patients with normal LV function include (see "Tests to evaluate left ventricular systolic function", section on 'Limitations and trouble shooting'):

●Poor image quality (eg, lack of endocardial definition, poor acoustic windows due to obesity)

●Qualitative or visual assessment of ejection fraction

If the review of echocardiography suggests the presence of one or more of these factors, options include:

●Repeat echocardiography with use of echocardiographic contrast or formal quantitation of LVEF using the existing echocardiographic images. (See "Tests to evaluate left ventricular systolic function", section on 'Diagnostic performance'.)

●Use of other measures of reduced LV systolic function to corroborate a reduced LVEF (eg, LV dilation, global longitudinal strain measurement) [20,21]. (See "Tissue Doppler echocardiography", section on 'Strain and strain rate imaging'.)

Patients with nondiagnostic echocardiography — If diagnostic uncertainty remains after repeat echocardiography with contrast or after review of the primary images, an alternative imaging study should be obtained. In such patients, we typically obtain a CMR imaging study if the patient does not have contraindications to CMR imaging (eg, pacemaker, contraindications to gadolinium administration), though radionuclide ventriculography, invasive contrast ventriculography, or computed tomographic left ventriculography may also be used. These imaging modalities are discussed elsewhere. (See "Tests to evaluate left ventricular systolic function".)

IDENTIFYING THE CAUSE — The evaluation for the cause of ALVSD is similar to the evaluation for the cause of symptomatic HF with reduced ejection fraction (HFrEF). A full discussion of the approach to determining the etiology of reduced LVEF can be found elsewhere. (See "Determining the etiology and severity of heart failure or cardiomyopathy".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Heart failure in adults".)

SUMMARY AND RECOMMENDATIONS

●Definition – Asymptomatic left ventricular systolic dysfunction ALVSD is defined as the presence of an LV ejection fraction (LVEF) <55 percent in the absence of heart failure (HF) symptoms. ALVSD is synonymous with the American College of Cardiology Foundation/American Heart Association’s Stage B HF (table 1). (See 'Definition' above.)

●Epidemiology – The prevalence of ALVSD is between 1 to 8 percent. (See 'Epidemiology' above.)

●When to evaluate

•Patients with examination findings consistent with HF – In asymptomatic patients, the physical examination findings that warrant further assessment of LV function are the same as the findings in symptomatic patients. (See "Heart failure: Clinical manifestations and diagnosis in adults", section on 'Physical examination'.)

•Patients with cardiac diseases – In patients with known cardiac disease (eg, mitral valve disease), the suspicion for LV systolic dysfunction and the need for evaluation is determined by the underlying cardiac disease. (See 'Patients with cardiac diseases' above.)

•Patients with unexplained signs of cardiac injury – Patients without HF symptoms but who have evidence of cardiac injury (eg, Q waves, LV hypertrophy) or dysfunction (eg, chest radiograph showing cardiomegaly) should undergo an evaluation for their specific abnormality with assessment for LV systolic dysfunction as indicated. (See 'Patients with unexplained signs of cardiac injury or disease' above.)

•Other patients at risk for reduced LVEF – Patients with diseases (eg, HIV) or cardiotoxic exposures (eg, anthracycline-based chemotherapy) associated with LV systolic dysfunction should be appropriately tested for LV systolic dysfunction. (See 'Other patients at risk for reduced ejection fraction' above.)

●When evaluation is unnecessary – We do not perform routine screening for ALVSD in unselected adults, patients with hypertension, or in patients who have a disease considered to be a coronary artery disease equivalent (eg, diabetes, peripheral vascular disease). (See 'When evaluation is unnecessary' above.)

●Approach to patients with suspected ALVSD

•Reassess for HF symptoms – Patients with evidence of reduced LV systolic function should be reevaluated for exertional and rest symptoms to be certain that symptoms are truly absent. (See 'Reassess for heart failure symptoms' above.)

•Obtain echocardiography – We initially test for LV systolic dysfunction with echocardiography. This preference is largely based on our experience and the ability of echocardiography to diagnose cardiac conditions. (See 'Obtain echocardiography' above.)

We do not consider B-type natriuretic peptide (BNP) testing an alternative approach to testing for ALVSD; it does not detect all patients who have LV systolic dysfunction, and causes other than HF may lead to BNP level elevation. (See 'Definition' above and 'Testing in the general population' above.)

•Review of echocardiography – For patients who already have an echocardiogram or other imaging study showing an LVEF <55 percent, the images should be evaluated to confirm that the measurement of low LVEF is accurate.

●Identifying the cause – The evaluation of LV systolic dysfunction in asymptomatic patients is similar to the evaluation of patients with symptomatic HF with reduced ejection fraction (HFrEF). (See "Determining the etiology and severity of heart failure or cardiomyopathy".)