Echocardiography in the evaluation of systolic murmurs of unknown cause∗

Article Outline

• Abstract
• Patients and methods
  • Clinical examination
  • Echocardiography
  • Statistical analysis
• Results
  • Diagnostic accuracy of the cardiac examination
  • Discrepancies between clinical and echocardiographic examination
• Discussion
• References
• Copyright

Abstract 

PURPOSE: Systolic murmurs are common, and it is important to know whether physical examination can reliably determine their cause. Therefore, we prospectively assessed the diagnostic accuracy of a cardiac examination in patients without previous echocardiography who were referred for evaluation of a systolic murmur.

SUBJECTS AND METHODS: In 100 consecutive adults (mean [± SD] age of 58 ± 22 years) who were referred for a systolic murmur of unknown cause, the diagnostic accuracy of the cardiac examination by cardiologists (without provision of clinical history, electrocardiogram, or chest radiograph) was compared with the results of echocardiography.

RESULTS: The echocardiographic findings included a normal examination (functional murmur) in 21 patients, aortic stenosis in 29 patients, mitral regurgitation in 30 patients, left or right intraventricular pressure gradient in 11 patients, mitral valve prolapse in 11 patients, ventricular septal defect in 4 patients, hypertrophic obstructive cardiomyopathy in 3 patients, and associated aortic regurgitation in 28 patients. In 28 (35%) of the 79 patients with organic heart disease, more than one abnormality was found; combined aortic and mitral valve disease was the most frequent combination (n = 22). The sensitivity of the cardiac examination was acceptable for detecting ventricular septal defect (100% [4 of 4]), isolated mitral regurgitation (88% [26 of 36]), aortic stenosis (71% [21 of 29]), and a functional murmur (67% [14 of 21]), but not for intraventricular pressure gradients (18% [2 of 11]), aortic regurgitation (21% [6 of 28]), combined aortic and mitral valve disease (55% [6 of 11]), and mitral valve prolapse (55% [12 of 22]). In 6 patients, the degree of aortic stenosis was misjudged on the clinical examination, mainly because of a severely diminished left ventricular ejection fraction. Significant heart disease was missed completely in only 2 patients.

CONCLUSION: In adults with a systolic murmur of unknown cause, a functional murmur can usually be distinguished from an organic murmur. However, the ability of the cardiac examination to assess the exact cause of the murmur is limited, especially if more than one lesion is present. Thus, echocardiography should be performed in patients with systolic murmurs of unknown cause who are suspected of having significant heart disease.

The stethoscope was invented by Laënnec, in 1816, when he had to examine an overweight young woman with a large bosom who was complaining of chest pain (1). The 35-year-old physician was too shy to put his head to the woman’s chest, so he rolled a paper booklet into a long, narrow cylinder, put it onto her chest with his ear at the other end, and heard a loud heart murmur. He called the instrument “stethoscope,” meaning “breast spy.”

Auscultation has been one of the most important means to assess heart disease, but the widespread use of Doppler echocardiography in the last 2 decades has considerably diminished its importance, causing parallel decreases in skill and confidence in its use 2, 3. Auscultation, however, remains the primary diagnostic tool for the interpretation of heart murmurs, and it is essential for making decisions about referring patients for echocardiography 4, 5. The most common abnormal auscultatory finding on cardiac examination is a systolic murmur, which occurs in 80% to 96% of children and 15% to 44% of adults 6, 7, 8, 9, 10, 11, 12. The two main causes of systolic murmurs—functional and organic—have different diagnostic and prognostic implications. The ability of physicians to identify and interpret heart murmurs is therefore essential, but it has not been studied widely. We prospectively assessed the diagnostic accuracy of cardiac palpation and auscultation in 100 patients who had a systolic murmur of unknown cause, using echocardiography as the gold standard.

Back to Article Outline

Patients and methods 

We included 100 consecutive patients who were referred to our echocardiography laboratory because of a systolic murmur of unknown cause and who had not had a prior echocardiographic examination. Their mean (± SD) age was 58 ± 22 years (range 17 to 92 years); 57% were women. Immediately before the echocardiographic examination, the patient was examined by 2 cardiologists who were blinded to the patient’s history, electrocardiogram, and other medical data. The examiners were randomly recruited from a group of 8 cardiologists (4 staff cardiologists, 4 cardiology associates) in our division.

Clinical examination 

The cardiac examination included estimation of jugular venous pressure, assessment of apical impulse and carotid artery upstroke, and auscultation at rest during quiet respiration, with assessment of heart sounds and murmurs and their radiation. Associated findings, such as thrills and systolic clicks, were noted. The Valsalva maneuver was done in every patient; other dynamic maneuvers were added if thought necessary.

The characteristics of the murmurs were classified at the point of maximal intensity. Murmurs were located in the aortic area, pulmonic area, base of the heart, or apex. Loudness was graded from 1 to 6 according to Levine (11); for analysis, murmurs were classified as soft (grades 1 or 2) or loud (grades 3 to 6). Timing and duration of murmurs were classified as early systolic, late systolic, midsystolic, prolonged midsystolic, or holosystolic. Shape was categorized into crescendo-decrescendo or plateau-type. Changes with the Valsalva maneuver and respiration were recorded.

The examiner had to state if the murmur was functional, with normal cardiac anatomy, or organic. If the murmur was thought to be organic, the examiner had to classify the underlying heart disease as significant or insignificant. Significant heart disease was defined as moderate or severe valvular heart disease (as would have been ascertained echocardiographically), congenital shunts, or intraventricular gradients. An “isolated” valvular lesion was defined if there was no clinical evidence of other types of heart disease.

Echocardiography 

All subjects underwent transthoracic two-dimensional and Doppler echocardiography in the left supine position, using a Hewlett Packard 2500 (Andover, Massachusetts) or Vingmed CFM 800 (Horten, Norway) echocardiographic machine. Valvular regurgitation was graded as trivial, mild, moderate, or severe based on a combination of factors, especially the vena contracta for the atrioventricular valves 13, 14 and the ratio of the regurgitant jet height to the outflow tract height for the semilunar valves (15). Aortic stenosis was classified as severe (mean systolic gradient ≥50 mm Hg or aortic valve area ≤0.8 cm²), moderate (mean gradient 30 to 49 mm Hg or valve area 0.8 to 1.0 cm²), mild (mean gradient 10 to 29 mm Hg or valve area 1.1 to 1.9 cm²), or trivial (mean gradient <10 mm Hg, valve area ≥2.0 cm², but with thickening of bicuspid or tricuspid aortic valve) (16).

An intraventricular gradient (left or right ventricle) was defined as a peak systolic gradient ≥10 mm Hg at rest or with Valsalva within the left ventricular outflow tract, or midventricular by continuous-wave Doppler with the typical shape (left convex) and the peak velocity occurring in late systole.

Statistical analysis 

Continuous variables are expressed as mean ± SD. For analysis of the performance of the cardiac examination, only the findings of the staff cardiologists were used. In an additional analysis, however, comparison with the findings of the younger cardiology associates showed no striking differences. Discrete findings on the cardiac examination were compared between patients with and without a specific echocardiographic finding using the chi-square or Fisher’s exact test, as appropriate. Continuous variables were compared using the Mann-Whitney test. Stepwise logistic regression was performed to find the independent predictors of echocardiographic diagnoses. P values of <0.05 (two-sided) were considered significant.

Back to Article Outline

Results 

Among the 100 patients (Table 1), mild to severe aortic stenosis was the most frequent echocardiographic diagnosis (29%), with the mean systolic gradient ranging from 7 to 95 mm Hg. A bicuspid aortic valve was present in 9 patients. Regurgitation through the mitral valve (n = 30), aortic valve (n = 28), and tricuspid valve (n = 28) were frequent; however, regurgitation was mild in most patients. Eleven patients had mitral valve prolapse. A late-peaking intraventricular pressure gradient was noted in 11 patients. A left intraventricular pressure gradient was present in 9 of these patients; of those, 2 had hypertrophic cardiomyopathy (pressure gradients of 7 and 30 mm Hg), 4 had hypertensive heart disease (gradients of 4 to 63 mm Hg), and 3 had aortic valve disease (gradients of 7 to 26 mm Hg). In 2 patients, a gradient was found on the right ventricular side only; this was the only pathology in a 19-year-old man. Another patient had a midventricular dynamic gradient on the right side with mild pulmonary stenosis. The mean shortening fraction in patients with left or right ventricular gradients was significantly greater (43% ± 8%) than in those without a gradient (37% ± 8%, P <0.0001). Six patients had a reduced left ventricular ejection fraction (<50%). In 28 (35%) of the 79 patients with organic heart disease, more than one abnormality was found. The most frequent combination was the association of aortic and mitral valve disease.

Table 1. Echocardiographic Findings in 100 Patients Referred for Evaluation of Systolic Murmurs

Twenty-one patients had a completely normal echocardiographic examination (functional murmur). These patients were much younger than patients with organic heart disease (33 ± 14 years versus 65 ± 18 years, P <0.0001).

Diagnostic accuracy of the cardiac examination 

Generally, the diagnostic accuracy of the clinical examination for any kind of heart disease was good, ranging from 70% to 97%, with the exception of lesions of the aortic or mitral valves in combination with other heart disease (Table 2). The sensitivity to detect significant heart disease was 79%, and a functional murmur could be recognized with a sensitivity of 67%. All 4 patients with a ventricular septal defect were identified with the cardiac examination. The sensitivities to recognize isolated aortic stenosis or mitral regurgitation were good (Table 2), whereas the sensitivity of the cardiac examination to detect mitral valve prolapse was only 55%, albeit with excellent specificity. Only 5 of the 9 patients with a systolic click had mitral valve prolapse; 2 had a bicuspid aortic valve, 1 had an atrial septal aneurysm, and in 1 patient the cause was unknown.

Table 2. Sensitivity, Specificity, and Diagnostic Accuracy of the Cardiac Examination for Various Causes of Systolic Murmurs

Diagnosing combined disease of the aortic and mitral valves was difficult (sensitivity of 55%). Sensitivities were also low for the recognition of intraventricular gradients and aortic regurgitation. The sensitivity to diagnose aortic regurgitation (21%) was significantly less than the sensitivity to diagnose aortic stenosis (71%, P = 0.0002). The diagnosis of significant aortic regurgitation was missed or underestimated in 4 of the 6 patients with that lesion.

In univariate analyses, aortic stenosis was associated with older age, a loud ejection murmur at the aortic area with radiation into the carotid artery, a thrill, and a diminished carotid upstroke (Table 3). A diminished aortic component of the second heart sound was noted in only 5 (33%) of the 15 patients with isolated aortic stenosis. Mitral regurgitation, and combined aortic and mitral valve disease, were especially common in elderly patients.

Table 3. Clinical Findings and Patient Age, by Cause of the Systolic Murmur

A “functional murmur” was never associated with a thrill or diminished carotid upstroke. It was most commonly a midsystolic murmur in a young person with maximal intensity at the base. Significant heart disease was typically associated with a loud murmur that was not midsystolic. The point of maximal intensity and the shape of the murmur were not predictive of significant heart disease. There were no characteristic findings for intraventricular gradients, which had several causes in addition to hypertrophic obstructive cardiomyopathy. With the Valsalva maneuver, a decrease in murmur loudness was heard in 51% of patients, an increase was heard in 9%, and no change was noted in 40%. The Valsalva maneuver was not helpful in discriminating between the different cardiac pathologies; its effects on patients with a late-peaking systolic murmur were not significantly different from those on patients with other murmurs (P = 0.35).

In multivariate analyses, younger age and a midsystolic murmur were associated with functional murmurs. Independent predictors of moderate or severe aortic stenosis were a diminished aortic closure sound (A2), a diminished carotid upstroke, increased age, and especially, a loud murmur at the aortic location that radiated into the carotid artery (Table 4). A plateau-shaped systolic murmur and older age were independent predictors of mitral regurgitation. There were no independent predictors of combined aortic and mitral valve disease other than older age.

Table 4. Predictors of Specific Causes of Systolic Murmurs, as Determined in Multivariate Analysis

Discrepancies between clinical and echocardiographic examination 

Eleven patients had substantial discrepancies between the clinical and echocardiographic examination (Table 5). In 4 patients, moderate or severe aortic stenosis (with an aortic valve area ≤1.0 cm²) was missed. Three of these patients had a reduced left ventricular ejection fraction, and 3 also had mild or moderate mitral regurgitation. Another patient with clinically suspected moderate (or greater) aortic stenosis actually had mild mitral stenosis, moderate mitral regurgitation, and only mild aortic stenosis. Significant aortic stenosis was diagnosed clinically in 1 patient who had hypertrophic obstructive cardiomyopathy and mild mitral regurgitation. Hypertrophic cardiomyopathy was missed in 2 other patients, both of whom had mitral regurgitation. One of these patients had a left ventricular outflow tract gradient of 63 mm Hg at rest that increased to 122 mm Hg with the Valsalva maneuver. Significant tricuspid regurgitation and mitral valve prolapse were suspected in another patient, in whom mitral valve prolapse was confirmed, but the patient had only trivial mitral regurgitation. Severe aortic regurgitation was missed in a 63-year-old woman in whom tricuspid and mitral regurgitation were diagnosed; her heart murmur was atypical for aortic regurgitation. A clinically suspected ventricular septal defect was not confirmed in 1 patient. However, the overall accuracy of the clinical examination for detecting significant cardiac disease was excellent; only 2 patients, who were considered to have trivial aortic stenosis, had significant heart disease.

Table 5. Characteristics of the 11 Patients Who Had Significant Discrepancies between the Clinical and Echocardiographic Examinations

Back to Article Outline

Discussion 

About 1 in 5 patients in our study who were referred for evaluation of a systolic murmur had a normal echocardiographic examination (a functional murmur). Cardiologists are quite proficient at discriminating these murmurs 17, 18, as they were in our study. Skilled examiners can thus decrease the number of unneeded echocardiographic examinations. In children, functional systolic murmurs are especially common, perhaps originating from the right ventricular outflow tract or representing a vibratory systolic murmur within the left ventricle (Still’s murmur) 7, 8, 9. Functional murmurs in adults have been related to aortic flow, increased intraventricular velocities, a venous hum, and vibratory phenomena 11, 19, 20, 21, 22, 23.

Among elderly patients (>60 years old) with an aortic systolic ejection murmur, 18% to 54% have aortic stenosis 24, 25. In our study, the sensitivity of the cardiac examination for detecting aortic stenosis was 71%. The best indicators were older age and a loud, prolonged midsystolic murmur that radiated into the carotid arteries and that was associated with a delayed carotid upstroke. The best discriminators between mild versus moderate or severe aortic stenosis were delayed or diminished carotid upstroke and a decreased or absent aortic closure sound (A2). However, the quality of carotid upstroke is influenced by the anatomy of the neck, the degree of atherosclerosis, and age. In 4 (27%) of our 15 patients with moderate or greater aortic stenosis, aortic stenosis was missed or considerably underestimated by the clinical examination. In 3 of these patients, the murmur was thought to be due to mitral regurgitation. It is well known that the murmur of aortic stenosis may masquerade as mitral regurgitation [Gallavardin’s dissociation 19, 26], perhaps especially in patients with decreased left ventricular ejection fraction 25, 27, 28, 29.

Another difficulty is detecting combined disease of the aortic and mitral valves, for which the sensitivity of the cardiac examination was only 55%. Not only may aortic stenosis mimic mitral regurgitation 19, 30, but also patients with mitral regurgitation can occasionally have murmurs that radiate atypically to the aortic area and base, especially if there is papillary muscle dysfunction or chordal rupture. When the posterior leaflet is involved, the jet is eccentric along the anterior leaflet, shifting the point of maximal intensity of the murmur toward the aorta (25). Generally, if there is a combination of different valvular lesions, as occurred in about one third of our patients, an exact diagnosis is difficult because the turbulent flow has multiple sources.

The clinical examination was excellent for detecting isolated mitral regurgitation (a sensitivity of 88%), but it was somewhat less sensitive for detecting regurgitation in the presence of other heart disease (64%). The probability of mitral regurgitation increases if the murmur is heard in the mitral area or if the murmur is holosystolic or late systolic (25). However, the maximal intensity of the murmur was found in the mitral area (apex) in only 5 (18%) of the 22 patients in our series who had mitral regurgitation and associated heart disease, and that may explain the diminished diagnostic accuracy. Although a holosystolic murmur is typical, mitral regurgitation can also cause a midsystolic murmur, especially in patients with ischemic heart disease (19).

The sensitivity of the cardiac examination for detecting mitral valve prolapse was low (55%), but the specificity was excellent (96%). Only 4 (36%) of the 11 patients had mild (or greater) mitral regurgitation associated with the prolapse. Thus, for diagnosis, the examiner had to rely mainly on the presence of clicks. Among our patients, however, only 5 of 9 patients with a click had mitral valve prolapse.

We detected gradients in the left (or, rarely, right) ventricular outflow tract, or a midventricular gradient, in 11% of our patients. These findings are usually considered to be typical of hypertrophic obstructive cardiomyopathy. As in our patients, however, left ventricular gradients can also be associated with hypertensive heart disease or aortic stenosis. In addition, patients with hypertrophic obstructive cardiomyopathy often have associated mitral regurgitation. Left ventricular gradients may be associated with a variety of cardiac lesions. The Valsalva maneuver was not helpful for diagnosing ventricular gradients or hypertrophic obstructive cardiomyopathy in our patients, despite our experience with its use (31).

Our ability to hear an additional diastolic murmur in the presence of a systolic murmur was poor, with a sensitivity of 21% for aortic regurgitation, and 40% for moderate (or greater) aortic regurgitation. These results are similar to those of a prior study, in which only 38% of physicians recognized murmurs of significant aortic regurgitation, and primary care physicians missed almost half of all diastolic murmurs (32). In the presence of a systolic murmur, aortic regurgitation cannot be excluded by the cardiac examination.

Our study had several limitations. We did not measure aortic peak flow velocities, flow integrals, or evidence of vibratory phenomena, which have been described as common causes of functional murmurs. The cardiac examinations in our study consisted strictly of inspection, palpation, and auscultation. Thus, our study had a somewhat unrealistic setting; clinicians almost always have other information, such as history, vital signs, and the results of electrocardiography, to help with interpreting the cardiac examination. It was our goal, however, to evaluate the usefulness of the cardiac examination.

Another important limitation is that phonocardiograms were not used as a gold standard for assessing murmurs (33). Phonocardiograms, however, are rarely used in practice. In addition, the study sample included only patients who were referred for echocardiographic evaluation of a systolic murmur of unknown cause; thus, echocardiography was thought to be indicated by the referring physician. Accordingly, patients with more typical auscultatory findings due to isolated lesions, or subjects with easily identifiable functional murmurs, were probably underrepresented. However, this underrepresentation probably reduced the diagnostic accuracy that we measured in the current study.

We recommend performing an echocardiographic examination of all patients with systolic murmurs who have cardiac symptoms, such as heart failure, dyspnea, chest pain, or strongly suspected endocarditis. The limitations of the cardiac examination are especially evident in patients with severe aortic stenosis and low cardiac output and in patients with more than one valvular lesion. Although the cardiac examination remains the first-line tool of every clinician, and auscultation is a cost-effective diagnostic skill (34), physicians should be aware of these limitations. Echocardiography should be used as an additional diagnostic tool when there is any uncertainty about the clinical examination.

Back to Article Outline

References 

1. Laennec RTH. Traité de l’Auscultation Médiate. 2nd ed.. Paris: Brosson et Chaude; 1826;
   • View In Article
2. Mangione S, Nieman LZ. Cardiac auscultatory skills of internal medicine and family practice trainees. JAMA. 1997;278:717–722
   • View In Article
   • MEDLINE
3. Tavel ME. Cardiac auscultation. A glorious past—but does it have a future?. Circulation. 1996;93:1250–1253
   • View In Article
   • MEDLINE
4. Pelech AN. The cardiac murmur. When to refer?. Ped Clin North Am. 1998;45:107–122
   • View In Article
5. Etchells E, Bell C, Robb K. Does this patient have an abnormal systolic murmur?. JAMA. 1997;277:564–571
   • View In Article
   • MEDLINE
6. McLaren MJ, Lachman AS, Pocock WA, Barlow JB. Innocent murmurs and third heart sounds in black schoolchildren. Br Heart J. 1980;43:67–73
   • View In Article
   • MEDLINE
   • CrossRef
7. Gardiner HM, Joffe HS. Genesis of Still’s murmurs (a controlled Doppler echocardiographic study). Br Heart J. 1991;66:217–220
   • View In Article
   • MEDLINE
   • CrossRef
8. Mannheimer E. Phonocardiography in children. Adv Pediatr. 1955;7:171
   • View In Article
   • MEDLINE
9. Freeman AR, Levine SA. The clinical significance of the systolic murmur. A study of 1000 consecutive “non-cardiac” cases. Ann Intern Med. 1933;6:1371–1385
   • View In Article
10. Bruns DL, Van der Hauwaert LG. The aortic systolic murmur developing with increasing age. Br Heart J. 1958;20:370–378
    • View In Article
    • MEDLINE
    • CrossRef
11. Levine SA. The systolic murmur. Its clinical significance. JAMA. 1933;101:436–438
    • View In Article
12. Aronow WS, Schwartz KS, Koenigsberg M. Correlation of aortic cuspal and aortic root disease with aortic systolic ejection murmurs and with mitral annular calcium in persons older than 62 years in a long-term health care facility. Am J Cardiol. 1986;58:651–652
    • View In Article
    • MEDLINE
    • CrossRef
13. Mele D, Vandervoort P, Palacios I, et al.  Proximal jet size by Doppler color flow mapping predicts severity of mitral regurgitation. Clinical studies. Circulation. 1995;91:746–754
    • View In Article
    • MEDLINE
14. Baumgartner H, Schima H, Kuhn P. Value and limitations of proximal jet dimensions for the quantitation of valvular regurgitation (an in vitro study using Doppler flow imaging). J Am Soc Echocardiogr. 1991;4:57–66
    • View In Article
    • MEDLINE
15. Dolan MS, Castello R, St.Vrain JA, et al.  Quantitation of aortic regurgitation by Doppler echocardiography (a practical approach). Am Heart J. 1995;129:1014–1020
    • View In Article
    • Abstract
    • Full-Text PDF (5690 KB)
    • CrossRef
16. ACC/AHA guidelines for the management of patients with valvular heart disease. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 1998;32:1486–1588.
    • View In Article
17. Roldan RA, Shively BK, Crawford MH. Value of the cardiovascular physical examination for detecting valvular heart disease in asymptomatic subjects. Am J Cardiol. 1996;77:1327–1331
    • View In Article
    • Abstract
    • Full-Text PDF (568 KB)
    • CrossRef
18. Bloch A, Crittin J, Jaussi A. The diagnosis of functional heart murmurs (the role of echocardiography). Révues Médicale de la Suisse Romande. 1998;118:599–602
    • View In Article
19. Perloff JK, Braunwald E. Physical examination of the heart and circulation. In: Braunwald E, ed. Heart Disease. 4th ed. Philadelphia: W. Saunders; 1996:15–49.
    • View In Article
20. Bujack W, Gioia F, Cayler GG. An innocent thrill (a common finding with an innocent murmur). JAMA. 1976;235:2417
    • View In Article
    • MEDLINE
21. Harpin VA, Ostman-Smith I. The association of left ventricular tissue bands with vibratory functional murmurs. Arch Dis Child. 1987;62:650–651
    • View In Article
22. Spooner PH, Perry P, Brandenburg RO, Pennock GD. Increased intraventricular velocities. An unrecognized cause of systolic murmur in adults. J Am Coll Cardiol. 1998;32:1589–1595
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (164 KB)
    • CrossRef
23. Murgo JP. Systolic ejection murmurs in the era of modern cardiology. What do we really know?. J Am Coll Cardiol. 1998;32:1596–1602
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (220 KB)
    • CrossRef
24. Christen R, Saner H, Romanens M, Pirovino M. Valvular aortic stenosis in elderly patients (clinical and echocardiographic aspects). Schweiz Med Wochenschr. 1994;124:305–312
    • View In Article
    • MEDLINE
25. Aronow WS, Kronzon I. Correlation of prevalence and severity of valvular aortic stenosis determined by continuous-wave Doppler echocardiography with physical signs of aortic stenosis in patients aged 62–100 years with aortic systolic ejection murmurs. Am J Cardiol. 1987;60:399–401
    • View In Article
    • MEDLINE
    • CrossRef
26. Burch GE, Phillips JH. Murmurs of aortic stenosis and mitral insufficiency masquerading as one another. Am Heart J. 1963;66:439–442
    • View In Article
    • MEDLINE
    • CrossRef
27. Forssell G, Janosson R, Orinius E. Identifying severe aortic valvular stenosis by bedside examination. Acta Med Scand. 1985;219:397–400
    • View In Article
28. Rispler S, Rinkevich D, Markiewiscz W, Reisner SA. Missed diagnosis of severe symptomatic aortic stenosis. Am J Cardiol. 1995;76:728–730
    • View In Article
    • Abstract
    • Full-Text PDF (1096 KB)
    • CrossRef
29. Morgan DJR, Hall RJC. Occult aortic stenosis as cause of intractable heart failure. BMJ. 1979;1:784–787
    • View In Article
    • CrossRef
30. Roberts WC, Perloff JK, Costantino T. Severe valvular aortic stenosis in patients over 65 years of age. Am J Cardiol. 1971;27:497–506
    • View In Article
    • Abstract
    • Full-Text PDF (5516 KB)
    • CrossRef
31. Brunner-LaRoca HP, Attenhofer CH, Jenni R. Can an elevation of the left ventricular end-diastolic pressure be predicted by changes of the Doppler-echocardiographic pattern during the Valsalva manoeuvre?. Echocardiography. 1998;15:211–218
    • View In Article
    • CrossRef
32. Paauw DS, Wenrich MD, Curtis JR, et al.  Ability of primary care physicians to recognize physical findings associated with HIV infection. JAMA. 1995;274:1380–1382
    • View In Article
    • MEDLINE
33. Leatham A. Auscultation and phonocardiography (a personal view of the past 40 years). Br Heart J. 1987;57:397–403
    • View In Article
    • MEDLINE
    • CrossRef
34. Shaver JA. Cardiac auscultation (a cost-effective diagnostic skill). Curr Probl Cardiol. 1995;20:447–530
    • View In Article
    • Full-Text PDF (25379 KB)
    • CrossRef

• ∗ Access the “Journal Club” discussion of this paper at http://www. elsevier.com/locate/ajmselect/