Cardiac auscultation is the primary tool for detection of valvular heart diseases (VHD). Although prior research has focused on the role of clinical training in auscultation proficiency,1,2 it remains unclear why even experienced practitioners achieve only 40% to 70% sensitivity in detecting VHD compared with echocardiography.3-5
It is possible that in addition to auscultatory training, the quality of heart sounds also plays a role. Clinically undetectable or distant heart sounds have been anecdotally reported to occur in some settings. We sought to quantify the extent to which undetectable heart sounds occur in hospitalized patients who are undergoing echocardiography and evaluate the association between patient factors and missed VHD diagnoses.
In a cross-sectional study approved by the University of Michigan institutional review board, we identified hospitalized patients who were undergoing transthoracic echocardiography within 72 hours of heart sound collection. Participants provided written informed consent. Heart sounds were recorded with an electronic stethoscope by placing the stethoscope at the second right intercostal (aortic) and fifth mid intercostal (mitral) locations for 30 seconds. Recordings were linked to patient data and evaluated using a standardized process by 3 trained annotators (eMethods in the Supplement). Heart sounds were defined as clinically undetectable if they could not be discerned by at least 2 of 3 annotators.
To evaluate patient factors that are associated with undetectable heart sounds, we fit univariable and multivariable logistic regression models with age, sex, body mass index (BMI; calculated as weight in kilograms divided by height in meters squared), congestive heart failure, and chronic obstructive pulmonary disease as covariates. We also evaluated the proportions of undetectable heart sounds in patients with valvular pathologies that were diagnosed as mild or greater with echocardiography.
We enrolled and analyzed auscultation recordings from 200 patients. The median age was 65 years (IQR, 55-73 years), 74 (37%) were women, and 27 (14%) were Black. The median BMI was 29 (IQR, 25-34; missing for 14 patients [7%]). Congestive heart failure was present in 144 (74%), and chronic obstructive pulmonary disease was present in 100 (50%).
Heart sounds were clinically undetectable at the aortic location in 30 (15%) patients and at the mitral location in 130 (65%) patients. Male sex was associated with undetectable aortic sounds (Table 1), and factors associated with undetectable mitral sounds were female sex and higher BMI (Table 2). Heart sounds were undetectable in 0 of 17 patients with aortic stenosis (0/9 moderate or greater), 5 of 29 (17%) with aortic regurgitation (0/5 moderate or greater), 3 of 6 (50%) with mitral stenosis (1/1 moderate or greater), and 41 of 66 (62%) with mitral regurgitation (7/12 moderate or greater).
The annotators had an intraclass correlation of 0.75 (95% CI, 0.71-0.78). In a sensitivity analysis, when we restricted the definition of undetectable heart sounds to those cases in which all annotators agreed, 18 of 200 (9%) aortic and 100 of 200 (50%) mitral sounds remained undetectable.
In this cross-sectional study, the results suggest that clinically undetectable heart sounds may play a role in the low sensitivity of auscultation for detecting VHD among hospitalized patients who are undergoing echocardiography. This problem is particularly evident in mitral valve regurgitation, for which most patients who received a diagnosis based on echocardiography had clinically undetectable mitral sounds. This study is limited by the absence of a criterion standard for undetectable heart sounds, the use of a convenience sample in which rates of undetectable heart sounds may be higher than the general population, and the lack of physiologic maneuvers. Although the data collectors underwent auscultatory training, our findings may not apply to clinicians with extensive auscultation experience. The role of the physical examination in patient care is complex. While it is a cornerstone of the patient-physician relationship,6 this study highlights the need to continually evaluate the diagnostic value and possible limitations of all aspects of clinical practice, including components of the physical examination.
Accepted for Publication: September 20, 2021.
Published Online: November 15, 2021. doi:10.1001/jamainternmed.2021.6594
Corresponding Author: Jonathon McBride, BS, MS, University of Michigan Medical School, 1301 Catherine St, Ann Arbor, MI 48109 (jpmcbrid@med.umich.edu).
Author Contributions: Dr Klapman and Mr McBride served as co–senior authors for this article. Mr McBride had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Jariwala, Singh, Klapman, McBride.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Jariwala, McBride.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Jariwala, Singh, McBride.
Obtained funding: Klapman.
Administrative, technical, or material support: Klapman, McBride.
Supervision: Singh, Klapman, McBride.
Conflict of Interest Disclosures: Dr Singh reported grants from Blue Cross Blue Shield of Michigan and Teva Pharmaceutical Industries Ltd outside the submitted work. No other disclosures were reported.
Funding: This study was funded by a Capstone for Impact grant from the University of Michigan (Dr Klapman).
Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.