Additional heart sounds during early diastole in a patient with hypertrophic cardiomyopathy and atrioventricular block

Abstract

A third heart sound (S3) and a fourth heart sound (S4) are similarly perceived as low-pitched sounds and can be difficult to distinguish in some conditions, such as tachycardia or arrhythmia. We report a case with hypertrophic cardiomyopathy and Wenckebach second-degree atrioventricular block, in which the S4 was initially misdiagnosed as an S3 on auscultation and later confirmed using a phonocardiogram with an apexcardiogram. Interestingly, the amplitude of the S4 dynamically and regularly fluctuated in proportion to the interval between the S4 and the preceding ventricular contraction. These findings were associated with transmitral inflow patterns assessed by Doppler echocardiography, highlighting the importance of not only transmitral A but also E waves for the amplitude of S4 in patients with arrhythmias.

<Learning objective: A fourth heart sound could be difficult to diagnose and dynamically fluctuate in amplitude in patients with arrhythmias, such as Wenckebach second-degree atrioventricular block. These findings are likely to be associated with changes in transmitral inflow patterns assessed by Doppler echocardiography.>

Introduction

A fourth heart sound (S4) is sometimes difficult to differentiate from a third heart sound (S3) in some conditions, such as tachycardia or arrhythmia, because both S3 and S4 are perceived as a low-pitched sound and can be distinguished mainly based on the timing in the cardiac cycle [1]. We report a case with hypertrophic cardiomyopathy and Wenckebach second-degree atrioventricular block, in which the S4 was initially misdiagnosed as an S3 on auscultation and later confirmed using a phonocardiogram with an apexcardiogram.

Case report

A 77-year-old man with nonobstructive hypertrophic cardiomyopathy underwent phonocardiography in Matsushita Memorial Hospital. Wenckebach second-degree atrioventricular block had developed several years earlier, but he had been doing well without any symptoms or cardiovascular events. The patient had dyslipidemia and Dubin–Johnson syndrome. His mother reportedly had left ventricular hypertrophy. Medications included mexiletine hydrochloride, warfarin potassium, diltiazem hydrochloride, rebamipide, and pravastatin sodium.

On examination, the blood pressure was 140/70 mmHg and the pulse was irregular, with a rate of 78 beats per minute. First heart sound (S1) and second heart sound (S2) were normal with no murmurs. An S4 was faintly audible at the apex in the left lateral decubitus position. Interestingly, other additional heart sounds during early diastole were alternately heard best at the apex, but also at the right and left sternal borders. A presumed diagnosis of S3 was made. The remainder of the examination was normal. An electrocardiogram showed normal sinus rhythm with a rate of 75 beats per minute, right axis deviation, ST-T changes, and isolated premature ventricular contractions; progressively increased PR intervals, sometimes followed by a dropped QRS, were diagnostic of Wenckebach second-degree atrioventricular block that was unchanged from previous findings.

A phonocardiogram with microphones placed at the second right sternal border and the apex in the left lateral decubitus position was obtained at a paper speed of 50 mm/s using a commercially available device (MES-1000, Fukuda-Denshi Co., Tokyo, Japan). As shown in Fig. 1, an electrocardiogram in lead II showed regular P and two types of ventricular contraction, which were diagnosed as Wenckebach second-degree atrioventricular block and isolated premature ventricular contractions by long-term monitoring. Low-frequency phonocardiogram showed additional heart sounds before the S1 and 140 ms after the onset of P, findings suggestive of an S4. Of note, additional heart sounds were alternately observed 260 ms after the S2; the loud sounds were low-pitched and described best at the apex.

Fig. 1.

Phonocardiogram. An S4 is observed (arrowheads). Note that other additional sounds are alternately detected (arrows). P indicates a P wave. The second and fourth QRS are premature ventricular contractions.

An apexcardiogram and a phonocardiogram with microphones placed at the second left sternal border and near the apex were subsequently obtained at a paper speed of 50 mm/s in the left lateral decubitus position (Fig. 2). Interestingly, the apexcardiogram showed an association of the loud heart sounds (i.e. presumed S3 on auscultation) with tall A waves, a finding indicating that the additional sounds were not S3 but S4. The magnitude of the S4 was largest after P with a short interval from the preceding ventricular contraction, moderate after P with an intermediate interval from the ventricular contraction, and smallest after P with a longer interval. No significant association was observed between the S4 amplitude and A wave morphology on the apexcardiogram.

Fig. 2.

Apexcardiogram and phonocardiogram. Additional heart sounds after S2 (arrows) coincide with tall A waves (arrowheads), a finding suggestive of an S4. Phonocardiogram is not identical to that in Fig. 1 because the microphone was not placed exactly at the apex; apexcardiogram was recorded at the true apex. P indicates a P wave. The second and fourth QRS are premature ventricular contractions.

We reviewed echocardiographic findings, obtained three months before presentation, showing asymmetric septal hypertrophy (maximum wall thickness of 19 mm) and an apical aneurysm with a left atrial dimension of 42 mm, a left ventricular ejection fraction of 58%, and a left ventricular end-diastolic volume of 138 mm³. Wenckebach second-degree atrioventricular block was also present during echocardiographic examinations, as during recording of the apexcardiogram and the phonocardiogram, except for a lack of premature ventricular contractions. As shown in Fig. 3, pulsed Doppler images through a mitral valve included three types of pattern, probably due to the degree of fusion of transmitral E on A waves: tall inflows after P with a short interval from the preceding QRS, medium-sized inflows with an intermediate interval from QRS, and small inflows after P with a long interval. These echocardiographic findings were consistent with changes in the amplitude of the S4 on the phonocardiogram.

Fig. 3.

Left ventricular inflow. Pulsed Doppler imaging through a mitral valve shows three types of pattern: inflow waves are tallest after P with a short interval from the preceding QRS (arrows), followed by medium-sized waves with an intermediate interval from QRS (arrowheads), and smallest after P with a long interval (asterisks). Note that the medium-sized inflow waves have a small notch on the upward slope (arrowheads), suggestive of the peak of a transmitral E wave. An E wave seems to be completely hidden in the tallest wave (arrows, i.e. fused E and A waves) and to be absent in or around the smallest waves (asterisks, i.e. isolated A waves). P indicates a P wave.

Discussion

The present case had S4 and additional heart sounds during early diastole on auscultation. The differential diagnoses for additional sounds during diastole include split S2, S3, mitral opening sound, and pericardial knock sound. The additional heart sounds in our case were initially misdiagnosed as an S3 because the sounds were low-pitched, lacking in respiratory changes, and best heard at the apex in a situation without clinical signs of mitral valve stenosis or pericardial effusion. It was, however, later confirmed on a phonocardiogram with an apexcardiogram that the presumed S3 was an S4. Of note, the amplitude of the S4 dynamically and regularly fluctuated, in proportion to the interval between the S4 and the preceding ventricular contraction.

S4 seems to be classified into 2 groups, having a right-sided origin or a left-sided origin, according to the interval from the onset of the preceding P (i.e. 90–160 ms for the former and 140–240 ms for the latter) [2], although complete discrimination may be difficult in some cases. The S4 in the present case can originate both in the right-sided heart and in the left-sided heart because the S4 occurred 140 ms after P. It is reasonable that the present case had an S4 with an origin at the left-sided heart because the patient was diagnosed with hypertrophic cardiomyopathy. In general, a left-sided S4 is associated with elevated end-diastolic stiffness in the left ventricle [3], which is commonly observed in patients with hypertrophic cardiomyopathy due to prolonged relaxation and decreased compliance [4]. The presence of S4 in the second right sternal border on a phonocardiogram, however, suggests that the S4 could also originate in the right-sided heart.

An S4 has been thought to be a result of atrial contraction causing vibrations of the ventricular muscle, atrioventricular valve apparatus, and ventricular blood mass 5, 6. Determinants of an S4 include atrial contractility, blood flow to the atrium, ventricular compliance, and patient's position (i.e. left ventricular movements close to the stethoscope) [1]. Given that the apexcardiogram and the phonocardiogram were recorded in the same conditions, the patient's position is unlikely to have been a cause of the dynamic changes in S4 amplitude in the present case.

The amplitude of the S4 increased in proportion to the interval between the S4 and the preceding ventricular contraction; the largest S4 was associated with the tallest transmitral inflow or completely fused E and A waves on echocardiogram whereas the smallest S4 coincided with the smallest transmitral inflow or isolated A waves without E waves. We may safely consider that the fluctuation of the S4 amplitude is provoked by comorbidities including Wenckebach second-degree atrioventricular block as well as premature ventricular contractions because these arrhythmias can affect the combination of E waves and A waves. This speculation is supported by the fact that the medium-sized S4 was related to the medium-sized transmitral inflow due to partially fused E and A waves. Thus, we believe that these findings highlight the importance of not only transmitral A but also E waves for the amplitude of S4 in patients with arrhythmias. It remains unclear, however, why the A waves on the apexcardiogram did not vary in size although the amplitude of the S4 dynamically and regularly fluctuated in our case. Previous studies showed that A wave height on an apexcardiogram was not always a reliable marker of the amplitude of S4 7, 8. Further studies are warranted to clarify the association of S4 amplitude with A wave morphology, for example, the rate of the rise or its first derivative, on an apexcardiogram.

Conflict of interest

None declared.