Chapter 11 Heart Sounds and Extra Sounds

“… the gallop stroke is diastolic and is due to the beginning of sudden tension in the ventricular wall as a result of blood flowing into the cavity. It is more pronounced if the wall is not distensible and the failure of distensibility may depend on either a sclerotic thickening of the heart wall (hypertrophy) or to a decrease in muscular tonicity … the sound is dull, much more so than the normal sound. It affects the tactile sensation, more perhaps than the auditory sense. If one attempts to hear it with a flexible stethoscope, it lacks only a little, almost always, of disappearing completely.”

–Potain PC: Note sur les dedoublements normaux des bruits du coeur. Bull Mem Soc Med Hop Paris 3:138, 1866.

A. Generalities

Conventional teaching has long recognized auscultation of the heart as the centerpiece of physical diagnosis. Indeed, proper identification of the various findings can still allow the prompt recognition of many important cardiac diseases. This is particularly true in the area of sounds and extra sounds, a field that has fascinated physicians since the introduction of stethoscopy. A plethora of gallops, clicks, snaps, knocks, and plops has since entered our everyday vocabulary. Accordingly, we have granted all but a few of these sounds a “high pass” in our conventional teaching test. The few that failed did so, not because of the paucity of information they deliver, but because of the rarity of the disease processes they represent.

B. Cardiac Auscultation: Some Suggestions

1 Why is cardiac auscultation so difficult?

Cardiac sounds are often at the threshold of audibility. The human ear can only perceive sounds between 20 and 20,000 Hz: it can neither reach beyond (as dolphins and whales do) nor go below it (as elephants often do). Yet, in this range, it has a preferential bandwidth of 1000–5000 Hz, corresponding to that of the human voice. Yet, most cardiac sounds are <500 Hz. In fact, many are so low pitched to be almost inaudible (S₃ or S₄ can be <100 Hz).

Cardiac sounds are crammed in a very little time interval. At a rate of 70/minute, a cycle of 0.8 seconds can easily harbor four to five sounds, many barely detectable.

Hospital rooms are noisy.

Patients’ hair and respiration create misleading artifacts.

The obesity epidemic has given many patients a much fattier chest muffler.

Pathology has shifted from rheumatic to coronary, thus reducing the pool of teaching patients.

Our ever-increasing fascination with the inanimate and the machine (and the sophistication of technology), compounded by:

Medicolegal issues (which have made imaging a self-protecting necessity)

Patients’ demands and expectations

Reduced emphasis on bedside skills during training andreduced availability of teachers

Time constraints (which make resorting to technology an ever-increasing need)

2 How can you make auscultation a little easier?

Take your time.

Be thorough.

Control noise in the room.

Separate systole from diastole (easily done in normal rates by recognizing the acoustic differences of S₁ and S₂ plus the long and short intervals; in faster hearts, it may require simultaneous assessment of the arterial pulse or precordial impulse).

“Inch” (move your stethoscope inch by inch, from auscultatory area to auscultatory area).

Know how to use your tool: (1) bell versus diaphragm; (2) patient’s position (supine, seated, and left lateral decubitus); (3) changes with respiration; and (4) dynamic bedside maneuvers (straight-leg raising, squatting-standing, Valsalva, hand-gripping, exercise).

When challenged by feeble and crammed signals, focus on one sound at a time.

Develop pattern recognition. This means practice, practice, practice…. In fact, you may need to hear an individual acoustic event as many as 500 times before you can master it.

C. Normal Heart Sounds

3 What are the normal heart sounds?

They are the first (S₁) and second (S₂) heart sounds.

4 What are the hemodynamic and acoustic characteristics of the cardiac cycle?

The cardiac cycle starts with contraction of the atria (S₄), which completes the ventricular diastolic filling and results in electrical activation (and contraction) of the ventricles themselves. This, in turn, closes the atrioventricular valves (S₁) and starts the isometric phase of systole. Opening of the semilunar valves (which may cause ejection sounds/clicks) signals the beginning of isotonic contraction, with expulsion of ventricular content into the great vessels. Closure of the semilunar valves (S₂), and subsequent reopening of the atrioventricular valves, restarts diastole, and the cycle begins anew. Note that diastole is always longer than systole, unless the heart rate exceeds 120/minute. Knowledge of the interrelationship between intracardiac pressure and valve motions is crucial for understanding heart sounds and murmurs.

(1) First Heart Sound (S₁)

6 Where is S₁ best heard?

At the apex (for its mitral component) and over the subxiphoid/epigastrium (for the tricuspid).

8 Which characteristics of S₁ are clinically valuable and should therefore be identified?

The most valuable is intensity (and variations thereof). The next most valuable is splitting.

9 How do you tell S₁ from S₂?

The area of greatest intensity is different (apical for S₁ and basilar for S₂), and so is the timing (beginning of the short interval for S₁ versus beginning of the long interval for S₂). Finally, S₁ is lower pitched and longer than S₂, but still high pitched enough to require the diaphragm.

10 What is the significance of S₂ being louder than S₁ at the apex?

It suggests two possible explanations:

S₂ is indeed louder than S₁ (usually as a result of either pulmonary or systemic hypertension).

S₂ is normal, whereas S1 is softer.

11 Which factors are responsible for the loudness of S₁?

In addition to shape and thickness of the chest wall, three major factors play a role:

1. The rate of rise in left ventricular pressure: This is a function of ventricular contractility, with stronger contractions causing a faster rise in left ventricular pressure and thus brisker and more forceful A-V closure. Hence, a loud S₁ is typical of the hyperkinetic heart syndrome, whereas a soft (muffled) S₁ is instead common in congestive heart failure, whose failing ventricles can only generate a slow rise in systolic pressure.

2. The separation between atrioventricular leaflets at the onset of ventricular systole: The closer the leaflets, the softer S₁ is; conversely, the wider apart the leaflets, the louder S₁ is. This mechanism feeds into two other important variables:

The duration of the P-R interval: A short P-R forces the ventricles to contract while the leaflets are still widely separated, so that their closure occurs on a steeper part of the left ventricular pressure curve. This, in turn, means a more forceful and louder closure. Conversely, a long P-R provides enough time for the leaflets to come close to each other, thus softening S₁. A muffled S₁ used to be quite common in rheumatic fever with first- degree A-V block. The progressive P-R lengthening of the Wenckebach phenomenon may also gradually (and increasingly) soften S₁.

The atrioventricular pressure gradient: A large A-V pressure gradient keeps the leaflets widely separated until ventricular pressure rises high enough to shut them close. Since the closure takes place on a steeper part of the left ventricular pressure curve, it will be forceful and loud. Hence, the longer the ventricle has to contract in order to close the A-V valve, the louder S₁ will be. This is quite common in mitral stenosis, where it contributes to the loudness of S₁.

3. The thickness of the atrioventricular leaflet: The thicker the leaflets, the louder S₁ is (banging hardbacks against each other generates more noise than banging paperbacks). Still, a soft S₁ may indicate leaflets that are too rigid. Hence, a thickened and stenotic mitral valve may generate a booming S₁ early on in the disease, but a softer (or absent) S₁ when the leaflets get eventually calcified and fixed.

12 What factors can affect the rate of rise of ventricular pressure?

The most important is contractility. An increase in left ventricular contractility (because of exogenous or endogenous inotropics) will intensify the mitral component of S₁. Conversely, a decrease in contractility (because of congestive heart failure) will soften it.

13 Which diseases present with a variable intensity of S₁?

Heart blocks, such as second degree (i.e., Mobitz I or Wenckebach) and third degree:

In second-degree A-V block, there is progressive softening of S₁, while S₂ remains constant. This is due to the increasing P-R lengthening, until a beat is eventually dropped. It is so typical of Mobitz I that Wenckebach could describe it even before electrocardiogram (ECG) availability.

In third-degree A-V block (typical of Morgagni-Adams-Stokes syndrome), the change in S₁ intensity is instead random and chaotic because the atrium and ventricle march to the beat of a different drummer, with rates that are totally independent—when ventricular contraction catches the A-V valves wide apart, S₁ booms; when it catches them partially closed, S₁ softens. The varying S₁ intensity is so typically random to allow the recognition of complete block just on the basis of auscultation (Table 11-1).

Table 11-1 Intensity of S₁

Loud	Variable	Soft
Short P-R interval (<160 msec)	Atrial fibrillation	Long P-R interval (>200 msec)
Increased contractility (hyperkinetic states)	Atrioventricular block (Wenckebach and third degree)	Decreased contractility (left ventricular dysfunction)
Thickening of mitral (or tricuspid) leaflets	Ventricular tachycardia (due to atrioventricular dissociation)	Left bundle branch block
Increased atrioventricular pressure gradient (stenosis of the A-V valves)	Pulsus alternans	Calcification of A-V valve(s)
		Premature closure of mitral valve (acute aortic regurgitation)
		Mitral (or tricuspid) regurgitation

14 What was the role of Morgagni in describing complete heart block?

He had actually reported it almost 100 years before Adams and Stokes, in a merchant from Padua whom he had evaluated: “When visiting by way of consultation, I found with such a rarity of the pulse that within the 60th part of an hour the pulsations were only 22. And this rareness, which was perpetual, was perceived to be even more considerable, as often as many as two (epileptic) attacks were at hand. So that the physicians were never deceived from the increase of the rareness they foretold a paroxysm to be coming on.”

15 Who was Mobitz?

Woldemar Mobitz was a German cardiologist who during the first half of the 20th century linked his name to various arrhythmias and to the eponymous second-degree A-V block.

16 What is the intensity of S₁ in atrial fibrillation?

Variable. This is due to the irregular ventricular rate, which may catch the A-V valves widely open, partially closed, or in between.

17 How can you separate the variable S₁ of atrial fibrillation from that of complete A-V block?

In atrial fibrillation, the rhythm is irregularly irregular, whereas in third degree, A-V block is a regular bradycardia (due to either nodal or ventricular “escape”).

18 How is S₁ in mitral stenosis (MS)?

Booming (in 90% of the patients). A loud S₁ should always alert the clinician to the possibility of MS and thus prompt a search for its associated diastolic rumble. Conversely, a soft S₁ argues against the presence of uncomplicated MS (i.e., one where the valve is still relatively pliable). The loud S₁ is usually the result of:

Thickening of the mitral leaflets: In the late stages of MS, however, leaflets can become stiff and poorly mobile, which, in turn, softens S₁ and eventually eliminates it.

High atrioventricular pressure gradient: This is produced by the stenotic valve and keeps the A-V leaflets maximally separated at the onset of ventricular contraction.

20 Which conditions can be associated with a soft S₁?

Other than calcific mitral stenosis, a soft S₁ is usually heard in either early closure of the mitral valve (aortic regurgitation) or late closure (prolonged P-R interval). Alternatively, a soft or absent S₁ can result from inadequate left ventricular contraction because of congestive heart failure, myocardial infarction, or left bundle branch block (where the left ventricle not only contracts ineffectively, but also late, with M₁ following T₁; “M” for mitral and “T” for tricuspid).

21 Which atrioventricular valve closes first?

The mitral, followed by the tricuspid (high pressure beds always close earlier). Since mitral closure is much louder than tricuspid, the first component of S₁ is usually referred to as M₁ and predominates in the formation of the sound.

22 Which semilunar valve opens first?

The pulmonic, followed by the aortic (low pressure beds always open earlier). As for the intensity, the aortic ejection sound is usually louder than the pulmonic, but still not enough to become audible in the normal patient.

24 What is the significance of a narrowly split S₁?

It reflects the audible separation of M₁ and T₁, a normal phenomenon that may at times be detected by listening over the lower left sternal border/epigastric area (where the tricuspid component is louder and thus easier to separate from its mitral counterpart).

25 Is the tricuspid component of S₁ (T₁) audible at the apex?

No. It is only audible over the lower left sternal border (LLSB). T₁, however, may become audible at the apex in case of (1) thickening of the tricuspid valve leaflets (i.e., early tricuspid stenosis) or (2) right ventricular pressure overload (such as pulmonary hypertension or atrial septal defect).

26 What is the significance of a split S₁ at the base?

It does not indicate the audible separation of M₁ and T₁, but instead the presence of an early ejection sound. This can be of either pulmonic or aortic origin.

27 What is the significance of a widely split S₁ at the LLSB?

It usually indicates a delayed closure of the tricuspid valve, most commonly because of a right bundle branch block. Note that a bundle branch block is also a cause of split S₂.

28 What is the significance of an apparently split S₁ at the apex?

It may represent a normal S₁ that is either preceded by an S₄ or followed by an early systolic (ejection) sound. This is an important differential diagnosis to keep in mind.

29 How can one separate a truly split S₁ from a “pseudo-split” S₁?

A truly split S₁ is usually heard over the lower left sternal border. Conversely, an S₄ of left atrial origin is only audible at the apex, whereas an early systolic click is usually louder over the base. To separate S₄ from an early systolic click, keep in mind that S₄ is lower pitched, best heard with the bell, softer, located before the true S₁, and only heard at the apex. An early ejection click is instead higher pitched, best heard with the diaphragm, louder, located after the true S₁, and best heard at the base (although it can also radiate down to the apex).

(2) Second Heart Sound (S₂)

30 Where is S₂ best heard?

At the base. More specifically, over the second/third left parasternal interspace for its pulmonic component and over the second or third right parasternal interspace for the aortic one. Because of its medium to high frequency, S₂ requires the diaphragm of the stethoscope.

31 How is S₂ generated?

By sudden deceleration of blood following the closure of aortic (A₂) and pulmonic (P₂) valves.

32 Which of the two semilunar valve closes earlier?

The aortic, due to systemic pressure being normally higher than pulmonic pressure.

33 How clinically useful is S₂?

Very useful. In fact, it has been suggested that careful evaluation of S₂ ranks with electrocardiography and radiology as one of the most valuable routine screening tests for heart disease. (Leatham used to call it “the key to auscultation of the heart”.)

34 Which S₂ characteristics are more valuable clinically?

Sound intensity and sound splitting. Of these, splitting (and variations thereof) is the most informative. This is in contrast to S₁, where intensity (and variations) are the most important.

36 What is the effect of exhalation on semilunar valve closure?

The opposite. It delays A₂ (more venous return to the left side) and anticipates P₂ (less venous return to the right side), so that the interval between the two components becomes too narrow for being appreciated by the human ear.

38 Why does S₂ splitting disappear with aging?

Because of senile emphysema, with greater air muffling of the pulmonic component of S₂.

39 How important is a patient’s position on S₂ splitting?

Very important. A supine position increases venous return, lengthens right ventricular systole, and thus widens the physiologic splitting of S₂. Conversely, a sitting (or standing) position decreases venous return, shortens right ventricular systole, and narrows the physiologic split (Fig. 11-2). This is especially important when analyzing an expiratory splitting of S₂. In a study by Adolph and Fowler, 22/200 (11%) normal subjects had an expiratory split while supine, but only 1/22 maintained it upon sitting or standing. Hence, a true expiratory splitting of S₂ is one that is present both in a recumbent and upright position.

Figure 11-2 Evaluation of audible expiratory splitting of S₂. The presence of expiratory splitting in the supine position is usually abnormal. Sometimes expiratory splitting of S₂ in the supine position disappears when the patient is upright and the S₂ becomes single on expiration. This response is normal. Patients should be examined carefully in the sitting and standing positions whenever S₂ appears to be abnormally split during expiration.

(Adapted from Abrams J: Essentials of Cardiac Physical Diagnosis. Philadelphia, Lea & Febiger, 1987.)

41 What is a wide (physiologic) splitting of S₂? What causes it?

It is a splitting so wide as to present throughout respiration, albeit still more marked in inspiration. It occurs in (1) delayed closure of the pulmonic valve (delayed P₂), (2) premature closure of the aortic valve (premature A₂), or (3) a combination thereof.

42 What are the causes of delayed closure of the pulmonic valve?

The classic one is a complete right bundle branch block, which delays both the depolarization of the right ventricle and the closure of the pulmonic valve, making the physiologic splitting of S₂ audible both in inspiration and expiration. Loss of pulmonary recoil (because of idiopathic dilation) or severe impedance to right ventricular emptying also can delay the pulmonic closure. The latter can occur in (1) pulmonic stenosis (where the interval between A₂ and P₂ correlates with the severity of stenosis), (2) massive pulmonary embolism, (3) cor pulmonale with right ventricular failure, and (4) atrial septal defect. In pulmonary embolism, an audible expiratory splitting of S₂ (with a loud and palpable P₂) has both diagnostic and prognostic significance, reflecting acute cor pulmonale and usually resolving in hours or days.

43 What are the causes of premature closure of the aortic valve?

The most common is a rapid emptying of the left ventricle, as in severe mitral regurgitation or ventricular septal defect. A premature closure of the aortic valve also can occur in severe congestive heart failure, usually because of a reduction in left ventricular stroke volume. Finally, a widely split S₂ also may occur in tamponade, where expansion of the two ventricles is limited and fixed. During inspiration, the right ventricle fills relatively more, pushing the septum leftward and thus further impairing left ventricular filling. This reduces left ventricular stroke volume, anticipates A₂, and makes S₂ widely split. The opposite occurs in exhalation.

44 What is a fixed splitting of S₂? What does it mean?

It is an S₂ that remains audibly split throughout respiration, both in the supine and upright positions, and with a consistent interval between its two components. Although encountered in severe ventricular failure, a fixed splitting of S₂ should suggest a septal defect (most often atrial but occasionally ventricular), especially if associated with pulmonary hypertension. The defect (and its shunt) eliminate the respiratory changes in right and left ventricular stroke volume, thus fixing the S₂ splitting (more rarely, a fixed S₂ split will occur in severe impedance to right ventricular emptying, such as that of pulmonary stenosis, pulmonary hypertension, or massive pulmonary embolism—with or without bundle branch block). These patients cannot cope with the increased venous return of inspiration by increasing right ventricular stroke volume. Hence, they maintain their S₂ widely and persistently split throughout respiration (see Fig. 11-3).

Figure 11-3 The increased inflow into the right atrium on inspiration (vertical solid arrows) causes a decreased flow through the atrial septal defect (ASD) and thus increased flow through the mitral valve.

(Adapted from Constant J: Bedside Cardiology. Boston, Little, Brown, 1976.)

45 What is the differential diagnosis of a fixed splitting of S₂?

A late-systolic click (which precedes S₂) and an early diastolic extra sound (which follows S₂):

The late-systolic click varies with bedside maneuvers and is loudest at the apex (conversely, the split S₂ is unchanged with maneuvers and only heard at the base).

The two most common early diastolic extra sounds are the S₃ and the opening snap (OS) of mitral (or tricuspid) stenosis (for a discussion of how to differentiate an opening snap from a widely split S₂ or an S₃, see questions 103, 104, and 130). OS is primarily apical, whereas the split S₂ is basilar. Still, OS can be loud enough to transmit to the base, thus producing a triple lilt in inspiration (OS + split S₂, with a loud P₂ because of pulmonary hypertension). Note that the interval between S₂ and OS is wider than that between the two components of S₂. Finally, an OS is usually (but not necessarily) associated with a diastolic rumble.