Screening for Left Ventricular Systolic Dysfunction: Is Imaging a Solution?

Transcription

1 JACC: CARDIOVASCULAR IMAGING VOL. 3, NO. 4, BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION ISSN X/10/$36.00 PUBLISHED BY ELSEVIER INC. DOI: /j.jcmg Screening for Left Ventricular Systolic Dysfunction: Is Imaging a Solution? John J., MBBS, PHD Queensland, Australia To address the heart failure burden, our focus needs to shift to disease prevention. Strategies to initially screen for heart failure precursors such as asymptomatic left ventricular systolic dysfunction have been evaluated, including clinical scores, the 12-lead electrocardiogram, and natriuretic peptides. However, their specificity limits their broad application as screening tools in asymptomatic populations. High-quality images are now available from hand-carried cardiac ultrasound devices, at a fraction of the capital cost of standard echocardiography with favorable diagnostic performance, especially when experienced staff perform the imaging. Questions that remain to be addressed include how we should select the target population to screen, who should perform the screening studies, how much training is required, and how often screening studies should be performed. (J Am Coll Cardiol Img 2010;3:421 8) 2010 by the American College of Cardiology Foundation Despite modest reductions in mortality over the last 2 decades, there has been little change in the age-adjusted incidence of heart failure in community-based cohorts; hence, overall and age-adjusted prevalence rates continue to rise (1,2). Current treatment strategies are usually applied at later stages in the disease trajectory, with modest health outcome gains. Our focus needs to shift to an earlier stage, to allow disease prevention and early detection of patients before they develop clinical congestive heart failure. Echocardiographic markers associated with an increased risk of developing heart failure include left ventricular systolic dysfunction (LVSD), left ventricular (LV) diastolic dysfunction, increased LV mass, and increased left atrial size (3 5). This review will focus on asymptomatic LVSD, which holds the most promise as a potentially modifiable echocardiographic risk factor. The Diagnostic Assessment Framework: How Should We Evaluate the Benefit of Diagnostic Tests? An ideal diagnostic test should be safe, accurate, clinically effective, and cost effective. Fineberg et al. (6) initially proposed a hierarchic approach to the evaluation of diagnostic imaging technologies subsequently extended by others, including appraisal of technical efficacy, diagnostic accuracy, impact on diagnostic thinking, therapeutic planning and patient outcomes, and societal considerations (7). Diagnostic accuracy refers to the ability of the test to distinguish the presence or absence of disease. Relevant measures include sensitivity, specificity, and measures related to receiver-operatorcharacteristic curves. To be clinically effective, the information provided by the test should result in a change in both the clinician s diagnostic thinking (by raising or lowering pre-test From the Royal Brisbane and Women s Hospital and University of Queensland, Brisbane, Queensland, Australia. Manuscript received September 17, 2009; revised manuscript received October 30, 2009, accepted November 12, 2009.

2 422 JACC: CARDIOVASCULAR IMAGING, VOL. 3, NO. 4, 2010 ABBREVIATIONS AND ACRONYMS ACE angiotensin-converting enzyme ECG electrocardiogram HCU hand-carried cardiac ultrasound LV left ventricular LVSD left ventricular systolic dysfunction diagnostic probabilities) and subsequent therapeutic planning (by adding, changing, or removing treatments) and henceforth overall improved health outcomes. Ideally the impact of a diagnostic test on clinical outcomes is directly assessed in a randomized controlled trial. This approach is least prone to systematic bias and allows evaluation of both the benefits of identifying subclinical disease and dangers of performing diagnostic imaging whether directly related to the test or through the detection of nontarget findings (8). However, such studies may not be feasible for ethical or logistic reasons, and their selected design may limit our ability to translate any observed treatment efficacy to clinical effectiveness in a broader population. Furthermore, the time required to perform such studies may limit their relevance given the rapid evolution of new technologies (8). Alternatively, if there is evidence from adequately powered randomized controlled trials that treatment of cases leads to better outcomes, one may be able to link this to evidence from diagnostic accuracy studies to infer clinical effectiveness with a few important caveats (Fig. 1) (9). First, the diagnostic test should be applied at a similar stage in the diagnostic pathway and to a similar population as the subjects assessed in the diagnostic accuracy studies. Second, the cases identified by the diagnostic test should represent the same spectrum of disease as those subjects shown to benefit in the treatment efficacy studies. Societal considerations include equity, access, and economic impact. The seemingly exponential rise in health care expenditure on the background of constrained clinical and economic resources has focused our attention on the evaluation of cost effectiveness, using metrics such as cost per case diagnosed and cost per quality-adjusted life year gained. The input variables may be obtained from randomized trials or a decision analytic approach that combines diagnostic accuracy, treatment efficacy, and health resource utilization data from different sources. When it is not economically viable to apply a test to a broad population, we may be able to identify subgroups in which the clinical and cost effectiveness are maximized. Before advocating a screening program, we need to quantify the disease burden, identify whether available treatments lead to better outcomes, and determine whether safe and effective diagnostic testing strategies are available. What Is the Prevalence of Asymptomatic LVSD? Using the heart failure classification proposed by the American College of Cardiology and the American Heart Association, asymptomatic structural heart disease (stage B) was 2.8-fold more common than symptomatic heart failure (stages C and D) in a community-based sample of residents age 45 years (10). This forms the basis for incident heart failure cases in the future. Furthermore, depending on the population studied and the cutoff chosen to define LVSD, approximately 2% to 8% of adults have LVSD, of whom approximately one-half have no prior or current history of heart failure (Table 1). Men are 2 to 3 times more likely to be affected with a strong age predilection. In selected high-risk groups such as patients with vascular disease or diabetes mellitus, up to 20% may have LVSD (11). What Are the Morbidity and Mortality Associated With Asymptomatic LVSD? Longitudinal cohort studies have demonstrated a 3- and 8-fold increased risk of developing heart failure in individuals with mild and moderate to severe asymptomatic LVSD, respectively (12). In the SOLVD (Studies of Left Ventricular Dysfunction)-Prevention study, 15.8% of individuals with asymptomatic LVSD died within 37 months, with approximately 90% of these being cardiovascular deaths (13). In the Framingham cohort, individuals with asymptomatic LVSD had a 60% higher risk of dying (Fig. 2) (12). Most of these deaths were attributed to vascular disease (40% coronary artery disease, 21% other vascular disease), with 43% of the coronary artery disease deaths being sudden. Importantly, 56% of the individuals with asymptomatic LVSD who died did not develop antecedent heart failure, indicating that waiting for someone to develop symptomatic heart failure may be a missed opportunity to intervene and potentially avoid premature death. Can We Improve Outcomes in Individuals With Asymptomatic LVSD? Studies in subjects with LVSD following myocardial infarction with or without heart failure mandate the use of beta-adrenergic receptor blockers and either angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers to retard progressive adverse LV remodeling and improve outcomes (14). There is also randomized controlled trial evidence from the SOLVD-Prevention inves-

3 JACC: CARDIOVASCULAR IMAGING, VOL. 3, NO. 4, Randomized Trial Test Accuracy Study Plus Randomized Trial of Treatment Efficacy Target Population Target Population plus Detected* New Test Old Test New Test Accuracy Old Test Accuracy Treatment No Treatment Detected Detected Detected Detected Patient Outcomes Patient Outcomes Treatment Patient Outcomes Treatment Patient Outcomes * detected by the new and old test may not show similar response to treatment Figure 1. Trial Evidence Versus Linked Evidence of Test Accuracy and Treatment Efficacy Ideally, the impact of a diagnostic test on clinical outcomes is directly assessed in a randomized controlled trial. Alternatively, if there is evidence from randomized controlled trials that treatment of cases leads to better outcomes, it may be appropriate to link this to evidence from diagnostic accuracy studies to infer clinical effectiveness, provided the cases identified by the diagnostic test represent the same spectrum of disease as the subjects in the treatment efficacy studies. Reprinted, with permission, from Lord et al. (9). tigators that ACE inhibitors reduced the combined end point of death or hospitalization for heart failure in patients with asymptomatic LVSD over 3 years of follow-up (13). Despite most subjects being prescribed open-label ACE inhibitors after the study was completed, there was a significant mortality benefit over 11 years in those subjects who were randomized to receive ACE inhibitor therapy in the original study (Fig. 3) (15). This is particularly relevant given that most individuals with LVSD in the community are not taking ACE inhibitors (16). Furthermore, in patients with vascular disease, antiplatelet therapy, statins, and revascularization may reduce the risk of recurrent ischemic events and subsequent adverse LV remodeling (14). Given that LV ejection fraction usually measured with echocardiography is used to determine eligi- Table 1. Prevalence of LVSD in Community Surveys Involving More Than 1,000 Subjects Study (Ref #) Sample Size Age Range (yrs) Mean Age (yrs) LVSD Definition Prevalence (%) Proportion Symptom-Free (%) Gardin et al. (17) 5, Abnormal 3.7 (U.S.) McDonagh et al. (19) 1, EF 30% (Scotland) EF 35% Schunkert et al. (20) 1, EF 48% (Germany) Mosterd et al. (21) 2, FS 25% (the Netherlands) Davies et al. (22) 3, EF 40% (England) EF 50% Devereux et al. (18) 3, EF 40% (U.S.) EF 54% Redfield et al. (23) 2, EF 40% (U.S.) EF 50% Abhayaratna et al. (24) 1, EF 40% (Australia) EF 50% EF ejection fraction; FS fractional shortening; LVSD left ventricular systolic dysfunction.

4 424 JACC: CARDIOVASCULAR IMAGING, VOL. 3, NO. 4, 2010 Survival p< No ALVD Mild ALVD Mod/sev ALVD Systolic CHF which middle-aged men with coronary artery disease and moderate to severe LVSD dominated (13). That being said, it is generally accepted that this indication for ACE inhibitors should be applied broadly, given the consistent demonstration of efficacy in studies that enrolled subjects with LVSD (14) Years Figure 2. Kaplan-Meier Curves for Survival of Participants From the Framingham Study The referent group consisted of subjects with normal left ventricular (LV) systolic function (ejection fraction [EF] 50%) and no history of congestive heart failure (CHF). Reprinted, with permission, from Wang et al. (12). Mild ALVD mild asymptomatic LV systolic dysfunction (EF 40% to 50%); Mod/ Sev ALVD moderate to severe asymptomatic LV systolic dysfunction (EF 40%); Systolic CHF CHF with EF 50%. Survivors (%) bility status in treatment efficacy studies and as the reference standard in diagnostic accuracy studies, it would seem reasonable to infer clinical effectiveness with strategies that improve our ability to detect LVSD. However, most individuals with asymptomatic LVSD in the community would not have been eligible for the SOLVD-Prevention trial, in In-trial period Trial Termination year 77 Enalapril Placebo p=0.001 Number at Risk Enalapril Placebo year Figure 3. Survival Curves for the Enalapril and Placebo Groups in the SOLVD-Prevention Trial Despite most subjects being prescribed open-label angiotensin-converting enzyme inhibitors after the original study was completed at 3 years of follow-up, there was a mortality benefit seen at 11 years in those subjects who were randomized to enalapril in the original study. Numbers beside the curves denote the percentage of survivors at termination, 5 years after randomization, and 12 years after randomization, calculated by the Kaplan- Meier method. Reprinted, with permission, from Jong et al. (15) How Should We Screen for Asymptomatic LVSD? In asymptomatic populations with low disease prevalence, diagnostic screening tests with high specificity and positive predictive value are preferred to allow a rule-in strategy, thereby minimizing false positives and unnecessary downstream testing. The low prevalence of LVSD in the general community (17 24), the inability of a physical examination to reliably detect LVSD (25), and the limited availability of standard echocardiography mandate a selective approach to identify which individuals should undergo echocardiography. Three options that have been widely evaluated are the use of clinical risk factors, the 12-lead electrocardiogram (ECG), and plasma natriuretic peptide levels. The use of clinical characteristics incorporated into a multivariable logistic regression model to detect individuals with LVSD applied to the Framingham cohort had an area under the receiveroperator-characteristic curve of 0.75 in men and 0.72 in women; however, this model did incorporate 12-lead ECG findings and would be difficult to use in clinical practice (26). The 12-lead ECG has been shown to be highly sensitive in identifying LVSD in some but not all studies (27). However, it lacks the specificity and therefore the positive predictive value to be useful as a screening tool, with most studies reporting specificities below 80% (27). In symptomatic patients, recent meta-analyses have demonstrated that plasma natriuretic peptides are better indicators of clinical heart failure than of LVSD (28,29). This reflects their correlation with raised intracardiac filling pressure, which occurs in LVSD and various other forms of heart disease (30). Studies that have focused on the isolated detection of LVSD in asymptomatic individuals have been heterogeneous, with indications of publication bias in at least 1 systematic review (Table 2) (27 29). In part, this may be explained by the different populations evaluated, the various assays and thresholds used to define abnormal levels, and the different cutoffs used to define the reference standard (i.e., any or moderate to severe LVSD). In an economic evaluation based upon the diagnostic performance of plasma natriuretic pep-

5 JACC: CARDIOVASCULAR IMAGING, VOL. 3, NO. 4, Table 2. A Systematic Review of Diagnostic Accuracy Studies of B-Type Natriuretic Peptide According to the Reference Standard Used Outcome n DOR (95% CI) p Value (Homogeneity) AUC (SE) p Value (Publication Bias) Heart failure ( ) (0.03) * 28.9 ( ) (0.02) 0.82 LVSD ( ) (0.03) *Excluding 2 outlier studies. Modified, with permission, from Latour-Perez et al. (29). AUC area under the receiver-operator characteristic curve; CI confidence intervals; DOR diagnostic odds ratio; LVSD left ventricular systolic dysfunction; SE standard error. tides in the Framingham cohort (26) and health outcome data extrapolated from randomized controlled trials (13), it appeared that screening for LVSD may be cost effective, provided the prevalence of LVSD was at least 1% in the target population (Fig. 4) (31). However, by far, the majority of individuals with abnormal natriuretic peptide levels will have normal LV systolic function, which limits the use of natriuretic peptide levels as a screening tool for LVSD in asymptomatic individuals (32). Attempts to improve the specificity by incorporating clinical variables, the 12-lead ECG, measurement of urinary natriuretic peptide levels, and additional inflammatory biomarkers have met with variable success (27,33,34). It has been argued that limiting the use of natriuretic peptides to the detection of LVSD ignores their full diagnostic and prognostic utility. Raised natriuretic peptide levels in individuals with normal LV systolic function (who are labeled as false positives ) may indicate other pathology such as LV diastolic dysfunction, atrial fibrillation, coronary artery disease, valvular heart disease, or cardiomyopathy (30,32). However, it is not clear whether further investigation and treatment of such pathology in the context of screening in an asymptomatic population will lead to better outcomes. Given that the accepted reference standard to identify subjects with LVSD is an imaging method, another approach to screening is to perform a focused and limited imaging study to answer the specific question being asked. Hand-Carried Cardiac Ultrasound (HCU) Devices to Screen for Asymptomatic LVSD Given the limited availability and cost of standard echocardiography, portable HCU devices providing high image quality at substantially reduced capital cost have been developed, with some recent handheld devices weighing 1 kg. Although a number of studies have addressed how HCU may reduce diagnostic error when performed in addition to physical examination (35), there has been growing interest in evaluating its role as a screening tool in community-based populations (36 38). The sensitivity and specificity of HCU to identify LVSD is approximately 90% when performed by experienced operators (36 38). In a consecutive series of 533 subjects, the comparative diagnostic performance 80,000 Incremental C-E ($/QALY Gained) 60,000 40,000 20,000 0 Base case (Women) Base case (Men) 0% 1% 2% 3% 4% 5% Prevalence of LV Systolic Dysfunction (EF<35%) Figure 4. Impact of Prevalence of Low EF on the Cost Effectiveness of Screening for Men and Women Using BNP Followed by Echocardiography in Those With a Positive Test The cost-effectiveness (C-E) ratio drops below $100,000 per quality-adjusted life year (QALY) gained at a prevalence over 0.5% and drops below $50,000 at a prevalence over 1%. For any given prevalence of disease, the C-E ratio is lower for women because the accuracy of B-type natriuretic peptide (BNP) levels is slightly greater for women then men. Open circles men (BNP vs. no screen); closed circles women (BNP vs. no screen). Reprinted, with permission, from Heidenreich et al. (31). EF ejection fraction; LV left ventricular.

6 426 JACC: CARDIOVASCULAR IMAGING, VOL. 3, NO. 4, 2010 favored HCU compared with the 12-lead ECG and plasma natriuretic peptide levels, driven largely by its higher specificity (Table 3) (38). A further small improvement in specificity was achieved by performing HCU only in those individuals who had either an abnormal 12-lead ECG or elevated natriuretic peptide levels, but at the expense of a reduction in sensitivity. The reduction in unnecessary downstream testing driven by the higher positive predictive value of HCU makes this a cost-effective method to determine which patients should have a standard echocardiogram, especially when performed in subjects at the highest risk of developing LVSD, such as those with vascular disease, diabetes mellitus, or previous heavy alcohol exposure (Table 3) (38). Limitations of the Evidence The best method to screen for asymptomatic LVSD is still not clear, with multiple or sequential biomarker strategies and HCU holding the most promise (33,34,36 38). HCU appears to be cost effective; however, a number of questions remain. What is the ideal target population? Screening high-risk populations appears most appropriate (38), but what is the best method to identify these individuals, and should we include age and sex descriptors? Who should perform HCU? The procedure is highly dependent on the operator, and its widespread application may be limited by the finite number of experienced sonographers. Operators with limited echocardiography experience have been trained in standardized programs as short as 3 hours to perform and interpret focused echocardiograms with moderate sensitivity to detect LVSD; however, these studies have largely involved physicians and medical students, who are unlikely to be involved in widespread screening programs (39). Nonetheless, this could allow for opportunistic screening of high-risk patients when they present for clinical evaluation. Furthermore, in a small study of diabetic outpatients, nurses trained to use HCU correctly identified all cases of LVSD, although the false-positive rate was relatively high (40). Whichever method is chosen, appropriate quality control processes need to be established to ensure consistent training, skills maintenance, and acceptable diagnostic accuracy. Should we restrict screening to the detection of asymptomatic moderate to severe LVSD? Mild LVSD and diastolic dysfunction are also associated with increased morbidity and mortality (5,12). However, it is not clear whether treating this would lead to better outcomes in asymptomatic populations. How often should we screen? This will drive cost effectiveness and will be influenced by the incident rate of LVSD in the target population. Finally, screening for asymptomatic LVSD has not been evaluated in a randomized controlled trial; however, a linked evidence approach seems reasonable, given the evidence that we can improve outcomes in subjects with asymptomatic LVSD (13 15). Conclusions The impact of treating symptomatic heart failure on overall disease burden is likely to be limited unless treatment is combined with methods to reduce the frequency of incident heart failure in the community. One method to do this is to screen for recognized precursors of heart failure such as asymptomatic LVSD. A number of strategies have Table 3. The Diagnostic Accuracy and Cost Effectiveness of the 12-Lead ECG, NTproBNP, and HCU,* or Their Combination in Detecting LVSD, With Standard Echocardiography as the Reference Standard Strategy Sensitivity General Population (n 734) High-Risk Group (n 761) Specificity Cost per Case of LVSD Detected ( ) Sensitivity Specificity Cost per Case of LVSD Detected ( ) ECG abnormal , ,006 NTproBNP level abnormal , ,096 ECG or NTproBNP level abnormal , ,485 ECG and NTproBNP level abnormal , ,104 HCU abnormal , HCU abnormal after ECG abnormal HCU abnormal after NTproBNP level abnormal *HCU was performed in 533 consecutive subjects. High-risk refers to individuals with a history of ischemic heart disease, diabetes mellitus, peripheral vascular disease, cerebrovascular disease, or heavy alcohol usage. Modified, with permission, from Galasko et al. (38). ECG electrocardiogram; HCU hand-carried cardiac ultrasound; LVSD left ventricular systolic dysfunction; NTproBNP N-terminal pro B-type natriuretic peptide.

7 JACC: CARDIOVASCULAR IMAGING, VOL. 3, NO. 4, been evaluated, including the use of clinical scores, the 12-lead ECG, natriuretic peptide levels, and more recently, HCU. The HCU holds promise as a screening tool in asymptomatic subjects, with a high specificity if HCU is performed by experienced sonographers. However, the technique is highly operator dependent, which may limit its broader application in resource-constrained environments. That being said, it seems likely that screening for asymptomatic LVSD will have a greater impact on disease burden than our current approach of waiting for individuals to develop clinical heart failure before deciding who and when to treat. Reprint requests and correspondence: Dr. John J., Royal Brisbane and Women s Hospital, Butterfield Street, Herston, Queensland, QLD 4006 Australia. john_atherton@health.qld.gov.au. REFERENCES 1. Roger VL, Weston SA, Redfield MM, et al. Trends in heart failure incidence and survival in a community-based population. JAMA 2004;292: McCullough PA, Philbin EF, Spertus JA, Kaatz S, Sandberg KR, Weaver WD. Confirmation of a heart failure epidemic: findings from the Resource Utilization Among Congestive Heart Failure (REACH) study. J Am Coll Cardiol 2002;39: de Simone G, Gottdiener JS, Chinali M, Maurer MS. Left ventricular mass predicts heart failure not related to previous myocardial infarction: the Cardiovascular Health Study. Eur Heart J 2008;29: Takemoto Y, Barnes ME, Seward JB, et al. Usefulness of left atrial volume in predicting first congestive heart failure in patients or 65 years of age with well-preserved left ventricular systolic function. Am J Cardiol 2005; 96: Aurigemma GP, Gottdiener JS, Shemanski L, Gardin J, Kitzman D. Predictive value of systolic and diastolic function for incident congestive heart failure in the elderly: the Cardiovascular Health Study. J Am Coll Cardiol 2001;37: Fineberg HV, Bauman R, Sosman M. Computerized cranial tomography. Effect on diagnostic and therapeutic plans. JAMA 1977;238: Gazelle GS, McMahon PM, Siebert U, Beinfeld MT. Cost-effectiveness analysis in the assessment of diagnostic imaging technologies. Radiology 2005;235: Douglas PS, Taylor A, Bild D, et al. Outcomes research in cardiovascular imaging: report of a workshop sponsored by the National Heart, Lung, and Blood Institute. J Am Coll Cardiol Img 2009;2: Lord SJ, Irwig L, Simes RJ. When is measuring sensitivity and specificity sufficient to evaluate a diagnostic test, and when do we need randomized trials? Ann Intern Med 2006;144: Ammar KA, Jacobsen SJ, Mahoney DW, et al. Prevalence and prognostic significance of heart failure stages: application of the American College of Cardiology/American Heart Association heart failure staging criteria in the community. Circulation 2007;115: Kelly R, Struthers AD. Screening for left ventricular systolic dysfunction in patients with stroke, transient ischaemic attacks, and peripheral vascular disease. QJM 1999;92: Wang TJ, Evans JC, Benjamin EJ, Levy D, LeRoy EC, Vasan RS. Natural history of asymptomatic left ventricular systolic dysfunction in the community. Circulation 2003; 108: Effect of enalapril on mortality and the development of heart failure in asymptomatic patients with reduced left ventricular ejection fractions. The SOLVD Investigators. N Engl J Med 1992;327: Hunt SA, Abraham WT, Chin MH, et al focused update incorporated into the ACC/AHA 2005 guidelines for the diagnosis and management of heart failure in adults: a report of the American College of Cardiology Foundation/ American Heart Association Task Force on Practice Guidelines, developed in collaboration with the International Society for Heart and Lung Transplantation. J Am Coll Cardiol 2009;53:e Jong P, Yusuf S, Rousseau MF, Ahn SA, Bangdiwala SI. Effect of enalapril on 12-year survival and life expectancy in patients with left ventricular systolic dysfunction: a follow-up study. Lancet 2003;361: Ho SF, O Mahony MS, Steward JA, Burr ML, Buchalter M. Left ventricular systolic dysfunction and atrial fibrillation in older people in the community a need for screening? Age Ageing 2004;33: Gardin JM, Siscovick D, Anton-Culver H, et al. Sex, age, and disease affect echocardiographic left ventricular mass and systolic function in the free-living elderly. The Cardiovascular Health Study. Circulation 1995;91: Devereux RB, Roman MJ, Paranicas M, et al. A population-based assessment of left ventricular systolic dysfunction in middle-aged and older adults: the Strong Heart Study. Am Heart J 2001;141: McDonagh TA, Morrison CE, Lawrence A, et al. Symptomatic and asymptomatic left-ventricular systolic dysfunction in an urban population. Lancet 1997;350: Schunkert H, Broeckel U, Hense HW, Keil U, Riegger GA. Leftventricular dysfunction. Lancet 1998; 351: Mosterd A, Hoes AW, de Bruyne MC, et al. Prevalence of heart failure and left ventricular dysfunction in the general population: the Rotterdam Study. Eur Heart J 1999;20: Davies M, Hobbs F, Davis R, et al. Prevalence of left-ventricular systolic dysfunction and heart failure in the Echocardiographic Heart of England Screening Study: a population based study. Lancet 2001;358: Redfield MM, Jacobsen SJ, Burnett JC Jr., Mahoney DW, Bailey KR, Rodeheffer RJ. Burden of systolic and diastolic ventricular dysfunction in the community: appreciating the scope of the heart failure epidemic. JAMA 2003;289: Abhayaratna WP, Smith WT, Becker NG, Marwick TH, Jeffery IM, McGill DA. Prevalence of heart failure and systolic ventricular dysfunction in older Australians: the Canberra Heart Study. Med J Aust 2006;184: Madhok V, Falk G, Rogers A, Struthers AD, Sullivan FM, Fahey T. The accuracy of symptoms, signs and diagnostic tests in the diagnosis of left ventricular dysfunction in primary care: a diagnostic accuracy systematic review. BMC Fam Pract 2008;9:56.

8 428 JACC: CARDIOVASCULAR IMAGING, VOL. 3, NO. 4, Vasan RS, Benjamin EJ, Larson MG, et al. Plasma natriuretic peptides for community screening for left ventricular hypertrophy and systolic dysfunction: the Framingham Heart Study. JAMA 2002;288: Davenport C, Cheng EY, Kwok YT, et al. Assessing the diagnostic test accuracy of natriuretic peptides and ECG in the diagnosis of left ventricular systolic dysfunction: a systematic review and meta-analysis. Br J Gen Pract 2006;56: Ewald B, Ewald D, Thakkinstian A, Attia J. Meta-analysis of B type natriuretic peptide and N-terminal pro B natriuretic peptide in the diagnosis of clinical heart failure and population screening for left ventricular systolic dysfunction. Intern Med J 2008;38: Latour-Perez J, Coves-Orts FJ, Abad- Terrado C, Abraira V, Zamora J. Accuracy of B-type natriuretic peptide levels in the diagnosis of left ventricular dysfunction and heart failure: a systematic review. Eur J Heart Fail 2006;8: Nakamura M, Endo H, Nasu M, Arakawa N, Segawa T, Hiramori K. Value of plasma B type natriuretic peptide measurement for heart disease screening in a Japanese population. Heart 2002;87: Heidenreich PA, Gubens MA, Fonarow GC, Konstam MA, Stevenson LW, Shekelle PG. Cost-effectiveness of screening with B-type natriuretic peptide to identify patients with reduced left ventricular ejection fraction. J Am Coll Cardiol 2004;43: Redfield MM, Rodeheffer RJ, Jacobsen SJ, Mahoney DW, Bailey KR, Burnett JC Jr. Plasma brain natriuretic peptide to detect preclinical ventricular systolic or diastolic dysfunction: a community-based study. Circulation 2004;109: Ng LL, Pathik B, Loke IW, Squire IB, Davies JE. Myeloperoxidase and C-reactive protein augment the specificity of B-type natriuretic peptide in community screening for systolic heart failure. Am Heart J 2006;152: Ng LL, Loke IW, Davies JE, et al. Community screening for left ventricular systolic dysfunction using plasma and urinary natriuretic peptides. J Am Coll Cardiol 2005;45: Spencer KT, Anderson AS, Bhargava A, et al. Physician-performed pointof-care echocardiography using a laptop platform compared with physical examination in the cardiovascular patient. J Am Coll Cardiol 2001;37: Vourvouri EC, Schinkel AF, Roelandt JR, et al. Screening for left ventricular dysfunction using a handcarried cardiac ultrasound device. Eur J Heart Fail 2003;5: Galasko GI, Lahiri A, Senior R. Portable echocardiography: an innovative tool in screening for cardiac abnormalities in the community. Eur J Echocardiogr 2003;4: Galasko GI, Barnes SC, Collinson P, Lahiri A, Senior R. What is the most cost-effective strategy to screen for left ventricular systolic dysfunction: natriuretic peptides, the electrocardiogram, hand-held echocardiography, traditional echocardiography, or their combination? Eur Heart J 2006;27: Alexander JH, Peterson ED, Chen AY, Harding TM, Adams DB, Kisslo JA Jr. Feasibility of point-of-care echocardiography by internal medicine house staff. Am Heart J 2004; 147: Kirkpatrick JN, Belka V, Furlong K, et al. Effectiveness of echocardiographic imaging by nurses to identify left ventricular systolic dysfunction in high-risk patients. Am J Cardiol 2005;95: Key Words: diagnosis y heart failure y screening y ventricular dysfunction.