You are asked to see a 64-year-old man who has been admitted to the orthopedic service after a packing crate tipped over on his leg, producing an unstable fracture of his distal tibia and fibula. You see him as he is being prepared for surgery. The patient previously had a normal exercise tolerance and no cardiac symptoms. You conduct a complete cardiac examination, observing a grade 2 systolic murmur, loudest at the lower left sternal border, which does not radiate to the right carotid artery. The S2 has normal intensity, and you do not hear a fourth heart sound (S4). The carotid artery pulsation has a normal rate of increase and normal volume. The orthopedic surgeon is concerned about the murmur because a recent patient had a postoperative myocardial infarction (MI) and was subsequently diagnosed with aortic stenosis. The surgeon wonders whether surgery should be delayed until an echocardiogram is obtained to rule out aortic stenosis.
Your next patient is a 34-year-old woman without cardiovascular symptoms who has normal exercise tolerance. She has a grade 2 systolic murmur that begins late in systole and is loudest at the lower left sternal border. When the patient is examined in a standing position, the murmur increases in intensity, and you detect a loud systolic click just before the onset of the murmur. The rest of the cardiovascular examination result is normal. You suspect mitral valve prolapse (MVP), but you wonder how confident you should feel about the diagnosis.
Why Is the Clinical Examination Important in Evaluating Systolic Murmurs?
Systolic murmurs can be an important clue to a structural cardiac abnormality (Table 33-1). The use of noninvasive cardiac testing, such as echocardiography, has increased dramatically. It is estimated that 2% of the general population undergoes noninvasive cardiac diagnostic evaluation.1 In lieu of performing routine echocardiography on patients with systolic murmurs, a careful clinical examination may eliminate the need for additional tests in selected patients.
Table 33-1Selected Causes of Systolic Murmurs |Favorite Table|Download (.pdf) Table 33-1 Selected Causes of Systolic Murmurs
|Abnormal cardiac structure |
|Aortic stenosis |
|Hypertrophic cardiomyopathy |
|Mitral regurgitation |
|Mitral valve prolapse |
|Ventricular septal defect |
|Pulmonic stenosis |
|Tricuspid regurgitation |
|Atrial septal defect |
|Normal cardiac structure, increased flow |
|Renal failure with volume overload |
The Anatomic and Physiologic Origins of Systolic Murmurs
Heart murmurs are produced when turbulent blood flow causes prolonged auditory vibrations of cardiac structures. The intensity of the murmur depends on many factors, including blood viscosity and blood flow velocity and turbulence. In addition, the distance between the vibrations and the stethoscope, the angle at which the vibrations meet the stethoscope, and the transmission qualities of the tissue between the vibration and the stethoscope affect murmur intensity.2
In this article, we will arbitrarily define an abnormal systolic murmur as one associated with abnormal cardiac structure. We will not consider the diagnosis of systolic murmurs caused by abnormally increased blood flow across normal cardiac structures, such as in anemia or thyrotoxicosis. However, clinicians must consider the diagnosis of abnormally increased blood flow in patients with systolic murmurs.
How to Examine for Systolic Murmurs
Most clinicians agree that a complete clinical history and physical examination, including a detailed cardiac examination, is an essential step in the assessment of systolic murmurs. Clinicians will interpret a systolic murmur in an asymptomatic 24-year-old woman with iron deficiency anemia differently from a systolic murmur in a 76-year-old woman with fever, weight loss, and digital infarctions after recent dental surgery.
Although a complete cardiac examination is important, the reliability and accuracy of many components of the cardiac examination for systolic murmurs have not been adequately evaluated. For example, the only adequately evaluated individual element of the cardiac history related to murmurs is effort syncope, which refers to a transient loss of consciousness during effort or exertion. This article focuses on features of the cardiac physical examination for systolic murmurs that have been adequately evaluated for precision and accuracy. A complete description of the cardiac physical examination of systolic murmurs is beyond the scope of this article but can be found in many textbooks.
The cardiac physical examination includes nonauscultatory and auscultatory components. Adequately evaluated nonauscultatory components include carotid artery palpation, apical-carotid delay, and brachioradial delay. To assess the carotid pulse, the clinician applies both light and firm pressure over the artery and assesses both the rate of increase and the pulse volume. Experts suggest that examiners pay special attention to the peak of pulsation. A normal rate of increase feels like a sharp tap, whereas an abnormal rate of increase feels like a nudge. An abnormal rate of increase can also feel like a weak tap, followed by a nudge or push.3 Surprisingly, no clear guidelines exist for interpreting carotid volume. Suggested methods include palpating the artery with both hands and all fingers, or palpating with the thumb only.4 We can only offer that a normal carotid volume is easily felt with light palpation, whereas a reduced carotid volume is difficult to feel even with firm palpation.
Brachioradial delay and apical-carotid delay may be important findings for detecting aortic stenosis. For brachioradial delay, the examiner palpates simultaneously the right brachial artery of the patient with the right thumb and the right radial artery of the patient with the left index and middle finger. The examiner should use only light pressure on the brachial artery to avoid dampening the pulse waveform. The examiner attempts to detect a delay between the brachial artery and the radial artery pulsations; any palpable delay is considered abnormal.5 For apical-carotid delay, the examiner simultaneously palpates the precordial apex pulsation and the right carotid artery. The examiner attempts to detect a delay between the apical and the carotid artery pulsation; any palpable delay is abnormal.6
In contrast to the cardiac history and nonauscultatory examination, many components of routine cardiac auscultation have been adequately evaluated. During routine auscultation, the examiner attempts to detect a systolic murmur, which can be defined as a systolic noise with a duration longer than a heart sound.7 Examiners describe the grade, radiation (Table 33-2), onset, duration, and timing of peak murmur intensity (Figure 33-1). The Levine grading system8 facilitates description of intensity: a grade 1 murmur is not heard immediately on auscultation but only after the examiner has focused on systole for a few seconds, a grade 2 murmur is heard immediately on auscultation but is not loud, a grade 3 murmur is heard immediately on auscultation and is loud, and a palpable precordial vibration, called a thrill, signifies a grade 4 murmur. Other murmur characteristics, such as pitch and tonal quality, have not been adequately evaluated.
Table 33-2Typical Location of Maximal Intensity and Radiation for Various Types of Abnormal Systolic Murmurs |Favorite Table|Download (.pdf) Table 33-2 Typical Location of Maximal Intensity and Radiation for Various Types of Abnormal Systolic Murmurs
|Location of Maximal Intensity ||Radiation ||Typical for |
|Second right intercostal space ||Right carotid artery ||Aortic stenosis |
|Right clavicle |
|Fifth or sixth left intercostal space mid left thorax ||Left anterior axillary line ||Mitral regurgitation (including mitral regurgitation caused by mitral valve prolapse) |
|Left axilla |
|Lower left sternal border ||Lower right sternal border ||Tricuspid regurgitation |
|Fifth intercostal space, mid left thorax |
|Fifth left intercostal space mid left thorax ||Lower left sternal border ||Hypertrophic cardiomyopathy |
Select Features of Systolic Murmurs
In the holosystolic murmur, the murmur begins just after the first heart sound (S1) and continues throughout the systole. In the late systolic murmur, the murmur begins at the middle of the systole or later and ends at the second heart sound (S2). In an early peaking murmur, peak intensity is before the middle of the systole. In a mid- or late-peaking murmur, peak intensity is at the middle of the systole or later.
Other evaluated relevant features on routine auscultation include the intensity of the S2, the S4, and systolic clicks. The intensity of S2 can be graded as normal, decreased, or absent. A normal S2 should be easily heard in the second right and left intercostal spaces next to the sternum and should be louder than the first heart sound (S1) in these areas.3 Abnormal splitting of the S2 in relation to cardiac murmurs has not been adequately evaluated.
An S4 is a low-pitched sound occurring just before systole, sometimes described as a presystolic sound. The S4 from the left ventricle is best heard with the bell of the stethoscope lightly applied to the patient in the left lateral decubitus position.
Systolic clicks are high-pitched sounds with a duration similar to that of heart sounds. Systolic clicks (previously termed nonejection clicks) are associated with MVP. They generally occur later than 40 to 60 ms after the S1, and patient position greatly affects their timing. When a patient stands, a systolic click moves closer to the S1. Ejection sounds (previously termed ejection clicks) come from aortic or pulmonary valves opening in early systole, approximately 40 to 60 ms after the S1. The S1 and an ejection sound together have roughly the cadence of saying “pa-da” or “pa-ta” quickly.3 Patient position causes no appreciable change in the timing of ejection sounds.
After routine auscultation, the clinician may wish to further assess a systolic murmur using special maneuvers. If the maneuver is intended to increase the intensity of the murmur, then the clinician should listen at the edge of the murmur's radiation, where the murmur is barely audible. This will make it easier to detect an increase in murmur intensity. Similarly, if the maneuver is intended to decrease the intensity of the murmur, then the clinician should listen at the point of maximal intensity.
Maneuvers that primarily increase the venous return include quiet inspiration and sustained abdominal pressure. These maneuvers are intended to increase the intensity of right-sided heart murmurs, such as tricuspid regurgitation (TR) or pulmonic stenosis. For the quiet inspiration maneuver, the examiner determines the effect of quiet inspiration on the intensity of the murmur.9 The examiner should not ask the patient to breathe deeply, because the murmur will be obscured by the breath sounds. For the sustained abdominal pressure maneuver, the examiner exerts firm, sustained pressure inward and cephalad below the right costal margin. The intensity of the murmur is observed during several cardiac cycles.10
Transient arterial occlusion primarily increases systemic arterial resistance. This maneuver increases the intensity of left-sided regurgitant murmurs, such as mitral regurgitation (MR) or ventricular septal defect. The examiner inflates simultaneously 2 sphygmomanometers placed around each of the patient's upper arms to approximately 20 to 40 mm above the previously recorded systolic blood pressure of the patient. Twenty seconds after cuff inflation, any changes in murmur intensity are observed.11
Maneuvers that increase both venous return and systemic arterial resistance include standing to squatting and passive leg elevation. These maneuvers are intended to decrease the intensity of the murmur of hypertrophic cardiomyopathy and MVP. For the standing to squatting maneuver, the clinician sits to the side of the patient and instructs him or her to rapidly squat from the standing position. Changes in murmur intensity are noted immediately after squatting.12 For the passive leg elevation maneuver, an assistant passively elevates both of the patient's legs to approximately 45 degrees while the patient is supine. Changes in murmur intensity are observed 15 to 20 seconds after leg elevation.13
The Valsalva maneuver decreases venous return and increases systemic arterial resistance.3 The Valsalva maneuver decreases the intensity of aortic stenosis murmurs. The patient strains against a closed glottis for 20 seconds, and changes in murmur intensity are observed just before the end of the 20-second period.13 Patients may inadvertently do a Valsalva during other maneuvers, such as sustained abdominal pressure or standing to squatting, so clinicians should ensure that patients breathe normally during these latter maneuvers.
Precision of the Examination Related to Systolic Murmurs
Precision refers to agreement among clinicians regarding a particular clinical finding. The precision of the clinical examination for systolic murmurs has been evaluated in usual clinical circumstances by auscultating patients14, 15, 16, and 17 or in controlled nonclinical circumstances by listening to prerecorded audiotapes.18 Studies using audiotapes will yield higher estimates of precision, as will studies consisting of only normal patients or very abnormal patients. Most of the available precision studies include patients with various causes of abnormal systolic murmurs, although one study included only patients with mild or moderate aortic stenosis.17 The experience of observers likely affects precision; all but one study16 used cardiologists as the examiners.
The only evaluated historical variable for diagnosing murmurs is effort syncope, which had a κ of 1.0 (simple agreement, 100) in one small study (n = 22).17 This study excluded patients with other types of syncope that could be confused with effort syncope, so it was relatively easy for the cardiologists to agree on the presence or absence of effort syncope.
One study found that the agreement between cardiology trainees on the carotid upstroke was poor, but data to calculate simple agreement or κ values were not provided.19 The precision of physical findings is summarized in Table 33-3.
Table 33-3Precision of the Clinical Examination of Systolic Murmurs |Favorite Table|Download (.pdf) Table 33-3 Precision of the Clinical Examination of Systolic Murmurs
|Finding ||Examiner ||No. ||κa ||Simple Agreement, % |
|No murmur vs grades 1-4 ||Cardiologists (tapes)18 ||100 ||0.48 ||70 |
|Cardiologists14 ||100 ||0.30 ||54 |
|Cardiologists15 ||80 ||… ||86 |
|Cardiologists16 ||32 ||… ||97 |
|Noncardiologists16 ||32 ||… ||78 |
|No murmur/grade 1 vs grades 2-4 ||Cardiologists (tapes)18 ||100 ||0.74 ||87 |
|Cardiologists14 ||100 ||0.29 ||76 |
|Acoustic shape (late peaking vs not late peaking) ||Cardiologists17 ||22 ||0.74 ||95 |
|Midsystolic click ||Cardiologists15 ||80 ||… ||85 |
The Bottom Line for Precision
The precision of examining for any systolic murmur is moderate using audiotapes (κ, 0.48) but only fair in the clinical setting (κ, 0.30). The precision of examining for a loud (grade 2 or louder) systolic murmur is good using audiotapes (κ, 0.74) but only fair in the clinical setting (κ, 0.29).
The precision of examining for a late-peaking systolic murmur is excellent (κ, 0.74).
The precision of examining for a systolic click is good (simple agreement, 85%).
Accuracy of the Examination Related to Systolic Murmurs
To develop a structured search strategy, we used pertinent articles already in our files. Our strategy was deliberately broad to minimize the possibility of overlooking important articles. We then searched MEDLINE (English language) from 1966 through January 1996, using our structured search strategy (available on request). We manually reviewed potentially relevant articles that we identified; we also reviewed the reference lists of these articles. We contacted authors of relevant studies for additional information.
Table 1-7 for a summary of Evidence Grades and levels).20 Grade summary of Evidence Grades and Levels).20 Grade A studies involve the independent, blind comparison of sign or symptom with a gold standard of diagnosis among a large number of consecutive patients suspected of having the target condition. Grade B studies involve the independent, blind comparison of sign or symptom with a gold standard of diagnosis among a small number of consecutive patients suspected of having the target condition. Grade C studies involve the independent, blind comparison of sign or symptom with a gold standard of diagnosis among nonconsecutive patients suspected of having the target condition; nonindependent comparison of sign or symptom with a gold standard of diagnosis among a sample of patients who obviously had the target condition plus, perhaps, normal individuals; or nonindependent comparison of a sign or symptom with a standard of uncertain validity.
Many of the studies were conducted in cardiology clinics, so the prevalence of abnormalities in these studies will be higher than in usual practice. For example, a study of patients undergoing cardiac catheterization for suspected aortic stenosis found a prevalence of aortic stenosis of 73%, so a positive clinical examination result virtually ruled in aortic stenosis. In usual practice, the prevalence of aortic stenosis would be much lower, so a positive clinical examination result would not rule in aortic stenosis, but rather indicate the need for further testing with echocardiography.
Is This an Abnormal Murmur?
Clinicians are primarily concerned whether a systolic murmur indicates a cardiac abnormality. In this context, the goal of the clinical examination is not an exact diagnosis, but rather identification of patients needing further testing to confirm or quantify an abnormality.
Several studies evaluated the accuracy of the entire clinical examination, including the medical history, physical examination, electrocardiogram, and chest radiograph; none has evaluated the history and physical examination alone.21, 22, 23, and 24 In each study, cardiologists used the clinical examination to classify a systolic murmur as normal, possibly abnormal, or abnormal. Patients then underwent an echocardiogram or cardiac catheterization as the reference standard test. The most common abnormalities detected were valvular stenosis or regurgitation, atrial or ventricular septal defects, MVP, and cardiac hypertrophy. The study results, which are summarized in Table 33-4, indicate that cardiologists are efficient at identifying abnormal and normal murmurs.
Table 33-4Accuracy of Clinical Examination for Detecting Abnormal Systolic Murmur |Favorite Table|Download (.pdf) Table 33-4 Accuracy of Clinical Examination for Detecting Abnormal Systolic Murmur
|Overall Clinical Assessment ||LR (95% CI) ||Quality Gradea |
|Abnormal Murmur |
|Study 121b ||∞ (14-∞) ||A |
|Study 223c ||∞ (2.8-∞) ||C |
|Study 322d ||3.8 (2.8-5.4) ||C |
|Possibly Abnormal Murmur |
|Study 121b ||2.3 (0.7-5.9) ||A |
|Study 224e ||1.3 (1.2-1.4) ||C |
|Normal Murmur |
|Study 121b ||0 (0-0.4) ||A |
|Study 222c ||0.01 (0-0.02) ||C |
|Study 324e ||0.05 (0.01-0.20) ||C |
|Study 423c ||0.3 (0.1-0.6) ||C |
The Bottom Line for Abnormal Murmur
A clinical assessment of “normal murmur” by a cardiologist significantly reduces the likelihood of a cardiac abnormality.
A clinical assessment of “abnormal murmur” by a cardiologist significantly increases the likelihood of a cardiac abnormality.
Effort syncope is the only adequately studied individual historical variable. Presence of effort syncope in patients with a systolic murmur effectively rules in aortic stenosis (positive likelihood ratio [LR+], ∞; 95% confidence interval [CI], 1.3-∞) but absence of effort syncope is not helpful (negative likelihood ratio [LR–], 0.76; 95% CI, 0.67-0.86) (grade C study).17
Several studies have examined the accuracy of the physical examination for detecting aortic stenosis. In these studies, echocardiography or cardiac catheterization confirmed aortic stenosis. Definitions of aortic stenosis varied, with peak instantaneous gradients ranging from as low as 25 mm Hg to as high as 50 mm Hg or aortic valve areas ranging from as low as 0.7 cm2 to as high as 1.1 cm2.
Many physical findings may increase or decrease the likelihood of aortic stenosis.17, 25, 26 Table 33-5 lists the findings beginning with the highest positive LRs from the largest studies with the best methodologic quality. All of the studies used cardiologist examiners.
Table 33-5Accuracy of the Physical Examination for Detecting Aortic Stenosis |Favorite Table|Download (.pdf) Table 33-5 Accuracy of the Physical Examination for Detecting Aortic Stenosis
|Finding ||Reference Standard (No. of Patients) ||LR+ (95% CI) ||LR– (95% CI) ||Quality Gradea |
|Slow rate of increase of carotid pulse |
|Study 125 ||Cardiac catheterization (781) ||130 (33-560) ||0.62 (0.51-0.75) ||A |
|Study 226 ||Cardiac catheterization (231) ||2.8 (2.1-3.7) ||0.18 (0.11-0.30) ||Cb |
|Study 317 ||Cardiac catheterization (106) ||6.4 (0.8-45) ||0.73 (0.59-0.90) ||C |
|Timing of peak murmur intensity |
|Late peaking25 ||Cardiac catheterization (781) ||101 (25-410) ||0.31 (0.22-0.44) ||A |
|Midpeaking17 ||Cardiac catheterization (106) ||8.0 (2.7-23) ||0.13 (0.07-0.24) ||C |
|Decreased intensity or absent second heart sound |
|Study 125 ||Cardiac catheterization (781) ||50 (24-100) ||0.45 (0.34-0.58) ||A |
|Study 226 ||Cardiac catheterization (231) ||3.1 (2.1-4.3) ||0.36 (0.26-0.49) ||Cb |
|Apical carotid delay6 ||Cardiac catheterization (44) ||∞ (2.4-∞) ||0.05 (0.01-0.31) ||C |
|Brachioradial delay5 ||Echocardiogram (58) ||6.8 (3.2-14) ||0.0 (0.0-0.3) ||C |
|Fourth heart sound25 ||Cardiac catheterization (781) ||2.5 (2.1-3.0) ||0.26 (0.14-0.49) ||A |
|Presence of any murmur25 ||Cardiac catheterization (781) ||2.4 (2.2-2.7) ||0 (0-0.13) ||A |
|Reduced carotid volume |
|Study 126 ||Cardiac catheterization (231) ||2.3 (1.7-3.0) ||0.31 (0.21-0.46) ||C b |
|Study 217 ||Cardiac catheterization (106) ||2.2 (1.2-4.2) ||0.39 (0.22-0.69) ||C |
|Radiation to right carotid |
|Study 125 ||Cardiac catheterization (781) ||1.4 (1.3-1.5) ||0.10 (0.13-0.40) ||A |
|Study 226 ||Cardiac catheterization (231) ||1.5 (1.3-1.7) ||0.05 (0.01-0.20) ||Cb |
|With Valsalva maneuver intensity is decreased27 ||Cardiac catheterization (50) ||1.2 (0.8-1.6) ||0 (0-1.6) ||C |
Two studies are notable for their high methodologic quality and large sample sizes. The first study25 involved 781 consecutive, unreferred elderly patients who were nursing home residents. Each study participant received an examination by a single senior cardiologist, followed by an echocardiogram. Overall, 68 patients (9%) had aortic stenosis defined as a peak instantaneous Doppler gradient of 25 mm Hg or greater. This study provides a reasonable estimate of the accuracy of the clinical examination in an elderly population. Many of the patients had no symptoms and no audible murmur, which may have elevated the estimates of specificity and the positive LRs for some of the findings.
The second study26 evaluated 231 consecutive patients referred for cardiac catheterization for various reasons, including suspected aortic stenosis. Cardiology fellows or cardiologists examined patients before cardiac catheterization. Overall, 113 patients (49%) had aortic stenosis, defined as a valve area of 0.8 cm2 or less or a peak gradient of 50 mm Hg or greater, at cardiac catheterization. This study population was highly selected, so the prevalence of aortic stenosis was much higher than would be expected in usual clinical practice.
The accuracy of special maneuvers was evaluated by 2 trained cardiologists, who examined 50 nonconsecutive participants with a variety of heart diseases, including aortic stenosis, MR, ventricular septal defect, hypertrophic cardiomyopathy, pulmonic stenosis, and TR.27 No maneuver was useful for ruling in aortic stenosis (data not shown), but certain findings from the Valsalva maneuver reduced the likelihood of aortic stenosis (Table 33-5).
A potentially useful multivariate decisional aid for diagnosing aortic stenosis was developed using split-sample validation (Table 33-6).26 The study showed an excellent positive LR for patients with point scores higher than 10. One of the variables in this model was aortic valve calcification on the lateral chest radiograph.
Table 33-6Multivariable Decision Rule for Suspected Aortic Stenosis26 |Favorite Table|Download (.pdf) Table 33-6 Multivariable Decision Rule for Suspected Aortic Stenosis26
|Point Score ||Aortic Stenosisa ||LR (95% CI)b |
|Yes ||No |
|14 ||7 ||0 ||∞ (0.6-∞) |
|10-13 ||22 ||1 ||8.0 (1.6-46) |
|7-9 ||22 ||3 ||2.7 (1.0-8.0) |
|2-6 ||11 ||15 ||0.27 (0.15-0.49) |
|0 ||1 ||4 ||0.10 (0.01-0.58) |
|Total ||63 ||23 ||… |
|Variable ||Point Score |
|Reduced carotid volume ||2 |
|Slow rate of increase of carotid pulse ||3 |
|Murmur loudest at second right intercostal space ||2 |
|Decreased or absent second heart sound ||3 |
|Valve calcification on chest radiograph ||4 |
|Maximum score ||14 |
Although the preceding results are encouraging, 2 small studies had less impressive results. The first study included 75 patients with severe multivalvular disease who were undergoing cardiac catheterization and found that a cardiologist's clinical diagnosis of aortic stenosis was only reasonably accurate (LR+, 3.7; 95% CI, 2.2-7.0; LR–, 0.23; 95% CI, 0.11-0.44).28 Many of these patients had severe multivalvular disease, which may have made an exact diagnosis more difficult. A study on 35 elderly patients with systolic murmurs who were examined by a geriatrician found that a clinical diagnosis of neither “aortic stenosis present” (LR+, 2.4; 95% CI, 0.72-6.9) nor “aortic stenosis absent” (LR, 0.7; 95% CI, 0.30-1.1) was accurate.29 This study suggests that assessments by cardiologists may be better than assessments by noncardiologists.
The Bottom Line for Aortic Stenosis
The presence of any of the following clinical findings significantly increases the likelihood of aortic stenosis: effort syncope, slow rate of increase of the carotid pulse, timing of peak murmur intensity in late or midsystole, decreased intensity or absent S2, apical-carotid delay, or brachioradial delay.
The absence of any of the following clinical findings significantly reduces the likelihood of aortic stenosis: any systolic murmur or murmur radiation to the right carotid artery.
Combinations of the following clinical variables can be useful to rule in or rule out aortic stenosis: decreased carotid volume, delayed carotid upstroke, decreased or absent S2, murmur loudest at second right intercostal space, and valve calcification on chest radiograph.
We report the accuracy of the clinical examination for detecting moderate to severe regurgitation confirmed through echocardiography or cardiac catheterization (Table 33-7). Detection of moderate to severe MR, even in asymptomatic patients, may influence recommendations for echocardiographic monitoring30 or medical treatment.31
Table 33-7Accuracy of the Clinical Examination for Detecting Mitral Regurgitation |Favorite Table|Download (.pdf) Table 33-7 Accuracy of the Clinical Examination for Detecting Mitral Regurgitation
|Finding ||Reference Standard (No. of Patients) ||LR+ (95% CI) ||LR– (95% CI) ||Quality Gradea |
|Murmur in mitral area ||Study 133 ||Echocardiogram: moderate to severe MR (394) ||3.9 (3.0-5.1) ||0.34 (0.23-0.47) ||C |
|Study 232 ||Cardiac catheterization: moderate to severe MR (35) ||3.6 (1.9-7.7) ||0.12 (0.02-0.50) ||C |
|Late or holosystolic murmur15 ||Echocardiogram: moderate to severe MR (80) ||1.8 (1.2-2.5) ||0 (0-0.8) ||C |
|Any murmur during acute MI34 ||Cardiac catheterization: moderate to severe MR (206) ||4.7 (1.3-11) ||0.66 (0.25-1.0) ||C |
|With transient arterial occlusion, murmur increases in intensity27 ||Cardiac catheterization: severity not statedb ||7.5 (2.5-23) ||0.28 (0.13-0.60) ||C |
If a cardiologist hears a murmur in the mitral area (mid left thorax, fifth intercostal space), then the likelihood of MR is increased slightly, but absence of a murmur significantly reduces the likelihood of MR.15, 32, 33 Similarly, a late systolic or holosystolic murmur slightly increases the likelihood of MR, but absence of such a murmur significantly reduces the likelihood of MR. In the setting of acute MI, absence of a murmur is less useful for ruling out acute MR (LR–, 0.66; 95% CI, 0.25-1.0).34 Transient arterial occlusion was accurate for ruling in and ruling out left-sided regurgitant murmurs, such as MR and ventricular septal defect.27
Internal medicine house staff are less accurate than cardiologists for detecting the murmur of MR, with positive LRs ranging from 1.1 (for interns) to 4.6 (for medical students) and negative LRs ranging from 0.7 (for junior residents) to 1.0 (for interns and senior residents)35 (grade A study).
The Bottom Line for Mitral Regurgitation
For cardiologists, absence of a mitral area murmur or a late systolic/holosystolic murmur significantly reduces the likelihood of MR, except in the setting of acute MI.
Cardiologists can accurately distinguish left-sided regurgitant murmurs, such as MR and ventricular septal defect, using transient arterial occlusion.
Noncardiologists’ assessments for MR are considerably less accurate.
Cardiologists are reasonably accurate for diagnosing the murmur of moderately severe to severe TR in patients (n = 21, with TR; n = 295, without TR) referred for echocardiography (LR+, 10.1; 95% CI, 5.8-18; LR–, 0.41; 95% CI, 0.24-0.70) (grade C).33 Special maneuvers may also be helpful for diagnosing TR and other right-sided lesions such as pulmonic stenosis. One study (n = 10, with TR or pulmonic stenosis; n = 40, without TR or pulmonic stenosis) using cardiologist examiners found that an increase in murmur intensity with inspiration significantly increased the likelihood of a right-sided valvular lesion, whereas the absence of increased intensity made these conditions less likely (LR+, 8.0; 95% CI, 3.5-18; LR–, 0.0; 95% CI, 0-0.43) (grade C).27 In another study, patients with severe MR (n = 15) or TR (n = 15) were examined by experienced cardiologists before cardiac catheterization.10 To distinguish TR from MR, increased murmur intensity on inspiration had a positive LR of ∞ (95% CI, 3.1-∞) and a negative LR of 0.20 (95% CI, 0.07-0.45). For the finding of increased murmur intensity with sustained abdominal pressure, the positive LR was ∞ (95% CI, 2.5-∞) and the negative LR was 0.33 (95% CI, 0.15-0.58) (grade C).
The Bottom Line for Tricuspid Regurgitation
There are limited data on the accuracy of clinical examination for hypertrophic cardiomyopathy (also termed idiopathic hypertrophic subaortic stenosis). Many studies evaluate phonocardiography or intracardiac tracings rather than auscultation,36, 37, 38, 39, and 40 whereas others include fewer than 15 patients.41, 42, 43, 44, and 45 One study evaluated carotid sinus pressure, which is not routinely recommended for the clinical examination.46
Special maneuvers may help distinguish the murmur of hypertrophic cardiomyopathy.27 Using cardiologist examiners, if a murmur decreased in intensity with passive leg elevation, then hypertrophic cardiomyopathy was significantly more likely (LR+, 8.0; 95% CI, 3.0-21), whereas if the murmur did not decrease in intensity, the likelihood was significantly reduced (LR–, 0.22; 95% CI, 0.06-0.77). If murmur intensity was decreased or unchanged with standing to squatting, then hypertrophic cardiomyopathy was significantly more likely (LR+, 4.5; 95% CI, 2.3-8.6), whereas if the murmur increased in intensity, the likelihood of hypertrophic cardiomyopathy was significantly reduced (LR–, 0.13; 95% CI, 0.02-0.81) (grade C).
The Bottom Line for Hypertrophic Cardiomyopathy
Cardiologists can rule in or rule out hypertrophic cardiomyopathy by evaluating for decreased murmur intensity with passive leg elevation or increased murmur intensity when the patient goes from a squatting to standing position.
The accuracy of the clinical examination for diagnosing MVP cannot be defined, because clinical findings alone are sufficient for the diagnosis of MVP. A patient with a systolic click and a systolic murmur meets the diagnostic criteria for MVP even if the patient has a normal echocardiogram result.47, 48
However, we can examine the relationship between clinical findings and echocardiographic findings (Table 33-8).49, 50, 51, 52, and 53 With cardiologist examiners, a systolic click accompanied by a systolic murmur helped to rule in echocardiographic MVP. The accuracy of an isolated systolic click is variable, possibly because of unreliability of the clinical examination and differences between studies regarding the definition of echocardiographic MVP. An isolated systolic murmur has little effect on the likelihood of echocardiographic MVP, whereas absence of both a systolic click and a murmur appears to reduce the likelihood of echocardiographic MVP. Noncardiologists are less accurate than cardiologists for all of these findings.
Table 33-8Accuracy of the Clinical Examination for Detecting Echocardiographic Mitral Valve Prolapse |Favorite Table|Download (.pdf) Table 33-8 Accuracy of the Clinical Examination for Detecting Echocardiographic Mitral Valve Prolapse
|Finding ||Clinician (No. of Patients) ||LR (95% CI) ||Quality Gradea |
|Systolic click and murmur |
|Study 149 ||Cardiologists (401) ||19 (4.6-80) ||C |
|Study 250 ||Noncardiologists (104) ||2.4 (1.0-5.7) ||C |
|Systolic click |
|Study 149 ||Cardiologists (401) ||12 (5.4-25) ||C |
|Study 250 ||Noncardiologists (104) ||1.3 (0.7-2.2) ||C |
|Nonejection click, with or without a murmur |
|Study 151 ||Cardiologists (155) ||3.8 (2.3-6.8) ||A |
|Study 252 ||Cardiologists (140) ||1.7 (1.3-2.1) ||C |
|Murmur, with or without a systolic click |
|Study 152 ||Cardiologists (140) ||1.9 (1.3-3.0) ||C |
|Study 253 ||Noncardiologists (259) ||1.2 (0.9-1.5) ||C |
|Murmur only |
|Study 150 ||Cardiologists (401) ||2.4 (1.0-5.7) ||C |
|Study 251 ||Noncardiologists (104) ||0.7 (0.3-1.3) ||C |
|No murmur, no systolic click |
|Study 151 ||Cardiologists (155) ||0.04 (0.02-0.11) ||A |
|Study 252 ||Cardiologists (140) ||0.26 (0.12-0.54) ||C |
|Study 349 ||Cardiologists (401) ||0.21 (0.15-0.29) ||C |
|Study 450 ||Noncardiologists (104) ||0.53 (0.23-1.20) ||C |
Mitral valve leaflet redundancy or thickening is the echocardiographic variable most strongly associated with adverse clinical outcomes.54, 55 In one study, neither a systolic click (LR+, 2.8; 95% CI, 1.8-4.6; LR–, 0.76; 95% CI, 0.69-0.84) nor a systolic murmur (LR+, 1.3; 95% CI, 1.1-1.5; LR–, 0.57; 95% CI, 0.43-0.76) affected the likelihood of echocardiographic mitral valve leaflet thickening or redundancy (grade C study).56
Several longitudinal studies of patients with echocardiographic MVP have related baseline clinical findings to the development of adverse clinical events, including cardiac death, progressive MR requiring surgery, endocarditis, and systemic embolism.57, 58 A holosystolic murmur without a systolic click significantly increased the likelihood of an adverse event, whereas absence of both a systolic click and murmur was associated with no adverse events. Other clinical findings had little effect on the likelihood of adverse events (Table 33-9).
Table 33-9Accuracy of the Clinical Examination for Predicting Adverse Clinical Outcomes Related to Mitral Valve Prolapsea |Favorite Table|Download (.pdf) Table 33-9 Accuracy of the Clinical Examination for Predicting Adverse Clinical Outcomes Related to Mitral Valve Prolapsea
|Finding ||Clinician (No. of Patients) ||LR (95% CI) ||Quality Gradeb |
|Holosystolic murmur |
|Study 156 ||Cardiologists (316) ||18 (6.6-51) ||C |
|Study 257 ||Cardiologists (321) ||5.1 (2.2-9.9) ||C |
|Late systolic murmur or click and murmur |
|Study 158 ||Cardiologists (316) ||1.2 (0.7-1.7) ||C |
|Study 257 ||Cardiologists (321) ||0.8 (0.3-1.5) ||C |
|Click and holosytolic murmur57 ||Cardiologists (321) ||0.8 (0.2-2.4) ||C |
|Any click or isolated click |
|Study 158 ||Cardiologists (316) ||0.4 (0.2-0.8) ||C |
|Study 257 ||Cardiologists (321) ||0.26 (0.05-1.1) ||C |
|No click/no murmur |
|Study 154 ||Cardiologists (237) ||0 (0-4.1) ||C |
|Study 258 ||Cardiologists (316) ||0 (0-1.4) ||C |
The Bottom Line for Mitral Valve Prolapse
A systolic click, with or without systolic murmur, is sufficient for the diagnosis of MVP.
If a cardiologist hears a systolic click, with or without a murmur, then the likelihood of echocardiographic MVP is significantly increased. The absence of both a systolic click and murmur significantly reduces the likelihood of echocardiographic MVP.
In patients with echocardiographic MVP, a holosystolic murmur without a systolic click significantly increases the likelihood of long-term complications, whereas absence of both a systolic click and murmur significantly reduces the likelihood of long-term complications.
When to Examine for Systolic Murmurs
We are unaware of data by which one might give an evidence-based recommendation regarding the examination for systolic murmurs. Auscultation for systolic murmurs should probably be carried out in any patient for whom a complete cardiovascular database is necessary.
Are Systolic Murmurs Ever Normal?
In unreferred young adults, the prevalence of systolic murmurs ranges from 5% to 52%8, 59, 60, 61; echocardiography result is normal in 86% to 100%.62, 63, 64 Echocardiography result is normal in 90% to 94% of pregnant women with systolic murmurs who are referred for testing.21, 24, 65 In elderly medical outpatients or residents of long-term care facilities, the prevalence of systolic murmurs ranges from 29% to 60%66, 67, 68; echocardiography is normal in 44% to 100%.24, 25, 29, 69, 70 This wide range of normal in the elderly can be partially explained by various study definitions of normal echocardiograms. Commonly detected abnormalities in the elderly were left ventricular systolic dysfunction, aortic stenosis, and MR. Other studies include aortic valve sclerosis as an abnormality, although the clinical importance of aortic valve sclerosis is uncertain.
A venous hum71 and a mammary souffle are both normal conditions that present either as systolic murmurs or, more commonly, as continuous murmurs.
How to Improve Skills in Examining This Area
The characteristics of murmurs can be learned using cardiovascular auscultatory tapes or cardiac patient simulators, although the effectiveness of these aids is uncertain.72, 73 Most audiotapes are accompanied by phonocardiographic and expert cardiologist analyses, so these tapes can help clinicians to calibrate their ears to those of experts.
Most commercially available stethoscopes have similar acoustic properties, although some have poor performance at low frequencies.74 Good stethoscope maintenance is essential because dirt or cracked tubing75 will significantly reduce accuracy. Large earpieces are better because small earpieces can be occluded by the sharp bony angle at the external auditory meatus.3
At the bedside, eliminate background noise whenever possible. If background noise is unavoidable, try to repeat your examination in a quieter setting.
Finally, relate your clinical findings to the results of assessments by a colleague, a cardiologist, or an echocardiogram whenever possible. Resolving disagreements between your assessments and those of others is an excellent way of upgrading your clinical skills.
Recommendations for Further Research
Most studies used cardiologists or senior cardiology fellows to conduct the clinical examinations. There are few data on the precision and accuracy of the clinical examination conducted by noncardiologists. Some studies include inappropriately narrow spectrums of patients, such as only patients with moderate and severe aortic stenosis.5, 6, 17 Further studies should focus on a broad spectrum of patients from primary or secondary care settings, particularly patients older than 40 years when the prevalence of abnormal murmurs is significantly increased.
Author Affiliations at the Time of the Original Publication
Division of General Internal Medicine and Clinical Epidemiology, Department of Medicine, University of Toronto, Toronto, Ontario, Canada; and The Toronto Hospital, Toronto, Ontario, Canada.
We thank Wilbert Aronow, MD, for providing additional data and methodologic information about his studies, Eugene Oddone, MD, MHS, for his helpful comments on earlier drafts of this article, and Ms Sharon Smith for her assistance in the preparation of the manuscript and tables.
Your first patient, who is awaiting urgent surgery for an open fracture, had a systolic murmur that did not radiate to the right carotid artery. The likelihood of aortic stenosis is significantly reduced by this finding. In addition, the carotid artery pulsation had normal volume, the S2 intensity was normal, and there was no S4. These findings also help to reduce the likelihood of aortic stenosis. You are confident in your assessment because it was conducted in a quiet room with a comfortable and cooperative patient. You can advise the surgeon that aortic stenosis is unlikely.
Your second patient has a systolic click and a systolic murmur, strongly suggesting MVP. If you are an experienced auscultator, then these findings significantly increase the likelihood that the echocardiogram will show evidence of MVP. However, even if the echocardiogram result is normal, you already have enough evidence to diagnose MVP. You may wish to obtain an echocardiogram at a later date to determine the severity of the valvular abnormality.
VP. Melton LJ. Population-based study of echocardiography: time trends in utilization and diagnostic profile of an evolving technology. Int J Technol Assess Health Care.
et al.. Variability in detection and interpretation of heart
murmurs. Am J Dis Child.
J. Essentials of Bedside Cardiology. 2nd ed. Totowa, NJ: Humana Press; 2003:30, 121–153.
HJ. Bedside Cardiac Diagnosis. Philadelphia, PA: JB Lippincott; 1991:19.
DJ. Brachioradial delay: a new clinical indicator of the severity of aortic stenosis. Lancet.
SA. The clinical significance of the systolic murmur: a study of 1000 consecutive cardiac cases. Ann Intern Med.
JM. Signo para el diagnostico de las insufficiencias tricuspideas. Arch Inst Cardiol Mex.
AL. Hepatojugular reflux: useful in the bedside diagnosis of tricuspid regurgitation. Ann Intern Med.
RA. Diagnosis of left-sided regurgitant murmurs of transient arterial occlusion: a new maneuver using blood pressure cuffs. Ann Intern Med.
EP. Effects of squatting on the normal and failing circulation. BMJ.
D. Auscultation of the Heart.
3rd ed. Chicago, IL: Mosby-Year Book; 1977:2:45–67.
JJ. Auscultation of the heart
by physicians and by computer. In: Data Acquisition and Processing in Biology and Medicine, Vol 3.
New York, NY: Macmillan Publishing Co; 1964:23–52.
and severity of mitral regurgitation in the mitral valve prolapse syndrome
: a Doppler echocardiographic study of 80 patients. J Am Coll Cardiol.
WW. Examination into the heart
: an investigation into variation. Am J Epidemiol.
E. Identifying severe aortic valvular stenosis by bedside examination. Acta Med Scand.
CA. A study of physician variation in heart
sound interpretation. Med Annu Dist Columbia
BG. Rate of rise of the carotid pulse: an investigation of observer error in a common clinical measurement. Am J Cardiol.
G. Murmurs in pregnancy: an audit of echocardiography. BMJ.
PH. The reliability
of the medical history in the identification of patients at risk for infective endocarditis. J Am Dent Assoc.
DJ. Nonspecific systolic murmurs: an audit of the clinical value of echocardiography. N Z Med J.
and severity of valvular aortic stenosis determined by continuous-wave Doppler echocardiography with physical signs of aortic stenosis in patients aged 62 to 100 years with aortic systolic ejection murmurs. Am J Cardiol.
L. Value of noninvasive testing in adults with suspected aortic stenosis. Am J Med.
RA. Bedside diagnosis of systolic murmurs. N Engl J Med.
JA. Greene ER. Clinical evaluation versus Doppler echocardiography in the quantitative assessment of valvular heart
D. Doppler echocardiography in elderly patients with ejection systolic murmurs. Postgrad Med J.
GP. Managing asymptomatic patients with chronic mitral regurgitation. Chest.
GL. Duplex pulsed Doppler echocardiography in mitral regurgitation. J Clin Ultrasound.
of regurgitant murmurs in patients with valvular regurgitation detected by Doppler echocardiography. Ann Intern Med.
Study Group. Mitral regurgitation in early myocardial infarction: incidence
, clinical detection, and prognostic implications. Ann Intern Med.
EL. Causes of false-negative auscultation of regurgitant lesions: a Doppler echocardiographic study of 294 patients. J Gen Intern Med.
D. The Mueller maneuver in muscular subaortic stenosis [abstract]. Circulation. 1977;56(suppl 3):III–138.
V. Effects of prompt squatting on the systolic murmur in idiopathic hypertrophic obstructive cardiomyopathy. BMJ.
III. The circulatory response of patients with idiopathic hypertrophic subaortic stenosis to nitroglycerin and the Valsalva maneuver. Circulation.
JM. The postextrasystolic murmur response to gradient in hypertrophic cardiomyopathy. Ann Intern Med.
ME. Idiopathic hypertrophic subaortic stenosis as observed in a large community hospital. J Am Geriatr Soc.
AJ. Valsalva's maneuver and the systolic murmur of hypertrophic subaortic stenosis. Am J Cardiol.
MJ. Paradoxic response of the murmur of idiopathic hypertrophic subaortic stenosis to the Valsalva maneuver. Am J Cardiol.
A. Normalization of the external carotid pulse tracing of hypertrophic subaortic stenosis during Muller's maneuver. Chest.
R. The effect of isometric exercise on the systolic murmur of patients with idiopathic hypertrophic subaortic stenosis. Chest.
LM. The older patient with idiopathic hypertrophic subaortic stenosis. Geriatrics. 1977;1:61–69.
A. Kaplinsky E. Increased intensity of the murmur of hypertrophic obstructive cardiomyopathy with carotid sinus pressure. Chest.
JE. New guidelines for the clinical diagnosis of mitral valve prolapse. Am J Cardiol.
R. Savage DD. Diagnosis and classification of severity of mitral valve prolapse: methodologic, biologic, and prognostic consideration. Am Heart J.
et al.. Relation between clinical features of the mitral valve prolapse syndrome
and echocardiographically documented mitral valve prolapse. J Am Coll Cardiol.
ED. Mitral valve prolapse: comparison of diagnosis by physical examination and echocardiography. South Med J.
L. Mitral valve prolapse: comparative value of M-mode, two-dimensional and Doppler echocardiography. J Am Coll Cardiol.
PR. Comparison of auscultation with two-dimensional and Doppler echocardiography in patients with suspected mitral valve prolapse. Clin Cardiol.
RH. The impact of echocardiographic results on treatment decisions for patients suspected of mitral valve prolapse. Med Care.
AJ. Echocardiographically documented mitral-valve prolapse: long-term follow-up
of 237 patients. N Engl J Med.
AE. Identification of high-risk and low-risk subgroups of patients with mitral valve prolapse. N Engl J Med.
GH. Use of auscultation to follow patients with mitral systolic clicks and murmurs. Am J Cardiol.
JE. An investigation of the incidence
of cardiac murmurs in young healthy women. Practitioner.
JP. Blood pressure and heart
murmurs in a rural population in the United Republic of Tanzania. Bull World Health Organ.
A. Innocent systolic murmurs in healthy 40-year-old men. J R Coll Physicians Lond.
RL. Mitral valve prolapse in one hundred presumably healthy young females. Circulation.
TM. Innocent systolic murmurs in healthy young males: a study of the characteristics and prevalence
in the SAF pre-enlistee population. Singapore Med J.
D. Systolic murmurs in pregnancy: value of echocardiographic assessment. Clin Cardiol.
MG. The clinical significance of systolic murmurs in the elderly. Age Ageing.
D. Echocardiographic diagnoses in elderly patients with systolic murmurs and cardiac disease. Age Ageing.
JR. Teaching cardiovascular examination skills: results from a randomized controlled trial. Am J Med.
AW. Waugh RA. Test of a cardiology patient simulator with students in fourth year electives. J Med Educ.
DG. Comparison of the acoustic properties of six popular stethoscopes. J Acoust Soc Am.
1992;91(4 pt 1):2224–2228.CrossRef