If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author at: Department of Internal and Cardiovascular Medicine, Tarumizu Chuo Hospital, Tarumizu Municipal Medical Center, 1-140 Kinko-cho, Tarumizu City, Kagoshima 891-2124, Japan. Tel.: +81-994-32-5211; fax: +81-994-32-5722.
Chronic heart failure (HF) has various phenotypes. It is accompanied by repeated hospitalizations over a long period. Therefore, accumulating long-term observational data of patients with various backgrounds is important to establish a prediction technology for the exacerbation of HF. In a patient with chronic right-sided HF caused by cor pulmonale, heart sounds and electrocardiograms were recorded at home or our hospital twice a week for 7 months including the stable (31 days), pre-exacerbation (2 weeks just before the onset of exacerbation), and hospitalization periods and quantified as cardiac acoustic biomarkers (CABs) using AUDICOR technology (Inovise Medical, Inc., Portland, OR, USA). The relationship between the change in CABs and hospitalization events due to HF were investigated. During the pre-exacerbation period just before the onset of exacerbation of HF leading to hospitalization, inaudible changes in the third heart sound (S3) strength that were probably derived from the right heart were observed. Although the values of the fourth heart sound (S4) strength were high during the stable and pre-exacerbation period, values decreased markedly during hospitalization. These findings suggest that CABs including S3 and S4 are useful for the early detection of signs of HF exacerbation.
Learning Objective
In a case of chronic right-sided heart failure, the change in the third heart sound (S3) caused by the right ventricle could be detected using cardiac acoustic biomarkers in exacerbations of heart failure. Even if S3 is inaudible by auscultation, it is possible to observe its changes using quantification technology.
] and a poor prognosis. Monitoring of patients with HF outside the hospital for detecting early signs of exacerbation has been expected to allow the provision of appropriate interventions [
]. However, few reports have investigated the noninvasive and long-term data.
We report a case of chronic right-sided HF caused by cor pulmonale, in which heart sounds and ECGs were recorded and quantified as cardiac acoustic biomarkers (CABs) [
]. In addition, the relationship between the change in the parameters and hospitalization events due to HF was investigated.
Case Report
On June 1, 2020, a 68-year-old man with cor pulmonale due to combined pulmonary fibrosis and emphysema (CPFE) visited our hospital for dyspnea on exertion and was admitted with a diagnosis of acute HF of New York Heart Association functional class IV. After treatment with catecholamine and furosemide, his condition improved, and he was discharged on June 16. Then, he started home oxygen therapy.
Thereafter, the patient recieved oral treatment in our hospital (daily medication: 20-mg tadalafil, 60-mg azosemide, 5-mg enalapril, and 50-mg spironolactone along with inhaled 5-μg tiotropium/olodaterol). From September 2020 to March 2021, home-care nurses visited him twice a week to monitor his status, including body weight, heart sounds, and ECG. This study was approved by the Ethics Committee of Tarumizu Chuo Hospital (No. 19–6) and the Research Ethics Committee of Asahi Kasei Corporation (No. 1908). Written informed consent was obtained from the patient before study participation.
At baseline observation, the patient stood 172 cm tall and weighed 59 kg. He had a pulse rate of 100 beats/min, blood pressure of 100/70 mmHg, and SpO2 of 90 % (O2 at 1 L/min, nasal). Regarding the heart sound, the second heart sound (S2) was enhanced but clear. No extra heart sound or heart murmur was observed. Regarding the breathing sounds, a fine crackle was heard at the bottom of both lungs. Plasma brain natriuretic peptide (BNP) concentration was 248 pg/mL. Results of the ECG, echocardiography, chest X-ray imaging, and chest computed tomography at baseline are shown in Fig. 1. During the observation period, he was hospitalized twice because of HF exacerbation (November 12–27, 2020, and December 24, 2020, to January 12, 2021). He was treated with continuous catecholamine infusion and intravenous furosemide infusion and discharged from hospital at each time.
Fig. 1Initial cardiac findings. (a) Electrocardiogram: sinus rhythm, heart rate 96 bpm, right axis deviation, complete right bundle branch block. (b) Echocardiogram. The right ventricle is markedly enlarged, and the left ventricular short axis image shows flattening of the interventricular septum (yellow arrows). Left ventricular end-diastolic diameter, 39 mm; left ventricular ejection fraction, 63 %; left atrial dimension, 27 mm; E/A ratio of transmitral flow velocity pattern, 0.48; E-wave deceleration time, 254 msec; moderate to severe tricuspid regurgitation; estimated right ventricular systolic pressure, 64 mmHg (assuming right atrial pressure = 5 mmHg); inferior vena cava diameter, 10 mm; good respiratory variation; no obvious pericardial effusion or pleural effusion. (c) Chest X-ray. Diffuse ground glass opacities were observed in both lungs. Cardiothoracic ratio, 57 %. (d) Chest computed tomography. Pulmonary destruction due to severe emphysema with bullae and honeycomb, i.e. lung combined pulmonary fibrosis and emphysema was extended from middle (left) to lower (right) zone.
Synchronized waveforms of heart sound and ECG were recorded using a Holter-type acoustic cardiogram device (AUDICOR® AM-RT, Inovise Medical, Inc., Portland, OR, USA) [
]. The recording was performed for 5 min, during which the patient was kept at rest in a semi-reclining/supine position. Of the CABs, the strengths of the third and fourth heart sounds (S3 and S4) were assessed. These biomarkers are probability scores based on acoustic features reflecting the presence of S3 and S4 for each 10-s section. The sections with a poor-quality signal due to noise were excluded, and the medians of CABs in the remaining sections were used as representative values on measurement days.
The changes in S3 strength, S4 strength, and body weight were determined by comparing the distribution of the parameters and scalogram of the heart sound (time waveform and frequency structure) in the stable, pre-exacerbation, and hospitalization periods. The stable period was the period in which the patient's HF condition was stable and consisted of 15 days before and after the day with minimal BNP levels (31 days in total) before the first hospitalization event. Based on a previous study indicating that pathophysiological congestion begins to develop several weeks before HF hospitalization [
], the pre-exacerbation period was defined as 14 days immediately before hospitalization events.
Plasma BNP concentrations during the stable period, the first event, and the second event were 142 pg/mL, 1071 pg/mL, and 2550 pg/mL, respectively (Fig. 2a ). Although the weight fluctuated during the observation period (Fig. 2b), no significant difference was observed between the stable and pre-exacerbation periods (Fig. 2c). S3 strength increased before hospitalization events because of HF, especially before the second hospitalization (Fig. 2d). As shown in Fig. 2e, compared with the mean (1.691) of the S3 strength during the stable period, the values of the first and second pre-exacerbation periods were significantly higher [2.288 (p = 0.011) and 2.869 (p < 0.001), respectively] according to a two-sided Welch t-test performed using R ver. 4.0.4 (R Foundation, Vienna, Austria). As the weight decreased during hospitalization (Fig. 2b), S3 strength also decreased (Fig. 2d). Scalogram revealed similar changes in the heart sound (Fig. 3). In the pre-exacerbation period of the first event, where the elevation of BNP and S3 strength were relatively low, S3-like sounds were not visible on the scalogram. In contrast, the scalogram indicated the occurrence of a S3-like sound with low frequency and impulsive transition immediately after S2 related to the second event (Fig. 3a). S3 disappeared just before discharge (Fig. 3b).
Fig. 2Plasma B-type natriuretic peptide (BNP) concentration (a), body weight (b, c), S3 strength (d, e), and S4 strength (f, g) in the stable period, pre-exacerbation periods, and hospitalization events 1 and 2. The plots on the left side show the temporal change of each parameter. The blue, pink, and orange colors represent the stable, pre-exacerbation, and hospitalization periods, respectively. The strip plots on the right side display the distribution of the observed values at each home visit, the mean and 95 % confidence interval, and p-values <0.05 compared with stable period by Welch t-test.
Fig. 3Scalograms on the measurement during the stable and pre-exacerbation periods and before the discharge dates. White circles indicate representative examples of appearance and vanishment of S3 (a, b) and S4 (c, d). The arrows on the right side of the panels indicate audible (≥20 Hz) and inaudible frequency regions for human hearing. S1, the first heart sound; S2, the second heart sound.
The values of S4 strength were high (>4) during the stable period, and in particular, they increased to >7 in the latter half of the stable period (Fig. 2f). No clear difference was found between the stable period and each pre-exacerbation period (Fig. 2g). However, values decreased markedly during hospitalization (Fig. 2f). Similarly, scalogram showed that the low frequency and short extra S4 sound observed just before hospitalization (Fig. 3c) decreased or disappeared just before discharge (Fig. 3d).
Discussion
The patient had a good left ventricular systolic and diastolic function (left ventricular ejection fraction: 63 %; E/A ratio: 0.48; and E-wave deceleration time of transmitral flow velocity pattern by echocardiology: 254 msec). Meanwhile, the estimated right ventricular systolic pressure was 64 mmHg in the chronic phase, and echocardiographic images in the left ventricular short-axis view demonstrated flattening of the ventricular septum, clearly indicating pulmonary hypertension. Therefore, we considered that right-sided HF was caused by cor pulmonale due to CPFE.
S3 is useful for diagnosing HF because of its high specificity among physical examinations of HF [
Systematic review and individual patient data meta-analysis of diagnosis of heart failure, with modelling of implications of different diagnostic strategies in primary care.
]. A significant increase in S3 strength was observed in the two pre-exacerbation periods, which suggests that CABs allow for the early detection of HF exacerbation. S3 gallop appears to have been generated from the right ventricle. This suggests the usefulness of noninvasive monitoring of S3 at home not only in cases of HF with reduced ejection fraction but also in right-sided HF cases. Furthermore, in this case, S3 could not be detected by auscultation because it occurred in an inaudible range of <20 Hz (Fig. 3a). The increased S3 strength may have reflected changes in the frequency structure of the heart sound induced by hemodynamic deterioration [
]. Therefore, inaudible S3 could also help in detecting HF that progresses gradually. The usefulness of inaudible S3 in detecting worsening signs of HF has been reported [
], and our study findings are the first to add a case of right-sided HF to the evidence.
The patient presented with a surge in S4 strength in the latter half of the stable period and increased mean of the strength during the stable period. Conversely, S3 strength was comparatively low and stable. Right-sided S4 is a clinical finding that indicates the severity of pulmonary hypertension and is more prevalent than right-sided S3 [
ACCF/AHA 2009 expert consensus document on pulmonary hypertension: a report of the American College of Cardiology Foundation task force on expert consensus documents and the American Heart Association: developed in collaboration with the American College of Chest Physicians, American Thoracic Society inc, and the Pulmonary Hypertension Association.
]. Accordingly, the elevated S4 during the stable period was likely related to worsening pulmonary hypertension. However, the patient maintained elevated S4 during the pre-exacerbation periods, and the values decreased during inpatient care. Therefore, S4 is also clearly associated with the exacerbation of right-sided HF in this case.
Because new sensors were used in the acoustic cardiography device for measuring heart sound and ECG data at each visit, CABs may have fluctuated because of differences in sensor position and fixation. Further, because home visits burden the patient, the frequency of visits was set to twice a week. This frequency resulted in an interval of several days in data recording, making it impossible to capture CAB changes that occurred during the interval and limited the analysis of the relationship with hospitalization events. In the future, to accumulate data of heart sound and ECG from various phenotypes of patients, we would like to apply handheld and wearable devices that patients can independently measure every day.
Our findings that CABs, such as S3 strength, vary with right-sided HF exacerbation and treatment suggest that CABs can monitor patients' condition outside the hospital. Moreover, accumulating data to detect various phenotypes of HF at an early stage is essential. Further studies involving patients with other conditions would also help in making better clinical decisions; for example, the assessment of treatment effect in pulmonary hypertension and prediction of worsening events in HF with preserved ejection fraction.
Funding
This study was sponsored by Asahi Kasei Corporation.
Conflict of interest
NY, MI, NM, and TK are employees of Asahi Kasei Corporation. The other authors declare that there is no conflict of interest.
Acknowledgments
We thank Takahiko Murata, PhD of WysiWyg Co. Ltd. for support in manuscript preparation.
References
Nagamine T.
Gillette B.
Pakhomov A.
Kahoun J.
Mayer H.
Burghaus R.
Lippert J.
Saxena M.
Multiscale classification of heart failure phenotypes by unsupervised clustering of unstructured electronic medical record data.
Systematic review and individual patient data meta-analysis of diagnosis of heart failure, with modelling of implications of different diagnostic strategies in primary care.
ACCF/AHA 2009 expert consensus document on pulmonary hypertension: a report of the American College of Cardiology Foundation task force on expert consensus documents and the American Heart Association: developed in collaboration with the American College of Chest Physicians, American Thoracic Society inc, and the Pulmonary Hypertension Association.
00190-6&id=gr1.jpg){kind=link}
00190-6&id=gr2.jpg){kind=link}
00190-6&id=gr3.jpg){kind=link}