Accuracy of cardiac auscultation in the era of Doppler-echocardiography: A comparison between cardiologists and internists

doi:10.1016/j.ijcard.2008.06.066

International Journal of Cardiology

Volume 138, Issue 3, 4 February 2010, Pages 308-310

https://doi.org/10.1016/j.ijcard.2008.06.066 Get rights and content

Abstract

To investigate the present accuracy of cardiac auscultation, we asked a group of senior cardiologists and internists to auscultate respectively 72 and 70 selected patients and to give a diagnosis of the type of lesions heard and their degree of severity, using transthoracic Doppler-echocardiography as the standard reference. The percentage of correctly identified auscultations by cardiologists and by internists, particularly for common valvular lesions, such as aortic stenosis and mitral regurgitation, was respectively 76.1 vs 64.9% (P = 0.0787) for all types of lesions taken together, 57.1 vs 48%.0 (P = 0.5057) for mild, 82.4 vs 76.0% (P = 0.3335) for moderate–severe and 81.8 vs 27.3% (P = 0.0300) for lesions without degree of severity, which included cases of atrial septal defect (ASD) and of hypertrophic cardiomyopathiy (HCM). Our findings show that in the Doppler-echocardiographic era overall cardiac auscultatory proficiency for common valvular lesions is similar in cardiologists and internists. Cardiologists perform better than internists only when auscultating more rare cadiac lesions, such as cases of ASD or HCM.

Introduction

Although cardiac auscultation still remains one of the primary clinical tests performed by physicians, its importance has considerably diminished during the last two decades due to the widespread use of Doppler-echocardiography, causing a significant decrease in structured pre- and post-graduate teaching programs of cardiac auscultation [1], [2], [3]. The purpose of our study was to assess the cardiac auscultatory proficiency of senior cardiologists (cardiologists) and of senior internists (internists) by using transthoracic Doppler-echocardiography as the standard reference.

Section snippets

Study population

The study patients were selected from the teaching sessions on cardiac auscultation given by cardiologists and organised for medical students at our institution.

Study design

We asked 10 cardiologists participating to our teaching sessions and 9 internists to perform a careful cardiac auscultation conducted in a quiet room and to provide after 10 min of examination an auscultatory diagnosis of the selected patients, including their corresponding mild, moderate–severe or without degree of severity. The

Statistical analysis

Statistical comparisons between continuous observations were made with unpaired Student t test. Proportions of patients in each physician group were compared with the chi-square or the Fisher's exact probability test. A regression analysis using a robust method according to Huber and White predicted the mean proportion of correct auscultations given by each physician and the number of lesions detected by Doppler-echocardiography. Data are reported as number of the proportion of correct

Patient population and echocardiographic findings

Characteristics of both study patients' groups were similar regarding demographic, clinical and echocardiographic data (Table 1). Echocardiographic diagnoses included aortic stenosis (AS) (32.8 vs 32.4%), mitral regurgitation (MR) (23.0 vs 23.4%), aortic regurgitation (AR) (19.5 vs 19.8%), tricuspid regurgitation (TR) (8.9 vs 8.1%), mitral stenosis (MS) and pulmonary stenosis (PS) (6.2% vs 6–3%). Lesions without degree of severity included cases of atrial septal defect (ASD) and hypertrophic

Discussion

The data of our study showed that overall accuracy of cardiac auscultation was similar for both physician groups (P = 0.0662), the proportion of all correct diagnoses being 0.69 (95%CI, 0.30 to 1.15) for cardiologists and 0.55 (95%CI, 0.25 to 1.10) for internists (P = 0.1689). Accuracy of cardiac auscultation in the two physician groups was highest for AS and MR, lowest for TR and intermediate for AR. Both had a similar, but rather limited performance for identifying lesions of mild degree of

Study limitations

Because of technical reasons a direct comparison of the study patients between the two physician groups could not be performed. Thus, although the patients auscultated by both physician groups were similar, they were not the same.

Conclusions

The data of our study show that in the Doppler-echocardiographic era overall cardiac auscultatory proficiency, particularly for frequent cardiac lesions, such as AS and MR, is similar in cardiologists and internists. Both identify with a rather limited performance valvular lesions of mild degree and with a higher accuracy valvular lesions of moderate–severe degree of severity. Cardiologists perform better than internists only when auscultating more rare cadiac lesions, such as cases of ASD or

Acknowledgement

The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology [9].

References (9)

S. Mangione
Cardiac auscultatory skills of physicians-in-training: a comparison of three English-speaking countries
Am J Med
(2001)
C.A. Roldan et al.
Value of the cardiovascular physical examination for detecting valvular heart disease in asymptomatic subjects
Am J Cardiol
(1996)
C.H. Attenhofer Jost et al.
Echocardiography in the evaluation of systolic murmurs of unknown cause
Am J Med
(2000)
S. Reichlin et al.
Initial clinical evaluation of cardiac systolic murmurs in the ED by noncardiologists
Am J Emerg Med
(2004)

There are more references available in the full text version of this article.

Cited by (26)

A call to consider an aortic stenosis screening program
2023, Trends in Cardiovascular Medicine
Aortic stenosis (AS) is the most common age-related valvular condition with a prevalence of 13.1% in patients older than 75 years of age. Based on the severity of AS and symptoms, current guidelines recommend interval monitoring with transthoracic echocardiogram (TTE). However, no guidelines exist regarding screening asymptomatic persons for AS. Prevalence of AS is comparable to conditions such as colorectal cancer, lung cancer, breast cancer, and abdominal aortic aneurysm where dedicated screening programs are offered resulting in reduction of overall morbidity and mortality. We review recent advancements in treatment options, and we propose an AS screening program for high-risk individuals without known history of AS including all persons over age 75 and persons aged 70 years and older with dialysis dependent end-stage renal disease (ESRD).
Artificial Intelligence-Based Stethoscope for the Diagnosis of Aortic Stenosis
2022, American Journal of Medicine
Citation Excerpt :
In the current era of echocardiography and sophisticated imaging, when physicians are required to spend more time in front of computers and have less time with patients, physical examination skills and, in particular, cardiac auscultation skills, deteriorate.21 Sztajzel et al22 reported the accuracy of cardiac auscultation by senior cardiologists and internists in the echocardiography era. Cardiologists and internists correctly identified 76% and 65% of all murmurs, respectively.
The diagnostic accuracy of the stethoscope is limited and highly dependent on clinical expertise. Our purpose was to develop an electronic stethoscope, based on artificial intelligence (AI) and infrasound, for the diagnosis of aortic stenosis (AS).
We used an electronic stethoscope (VoqX; Sanolla, Nesher, Israel) with subsonic capabilities and acoustic range of 3-2000 Hz. The study had 2 stages. In the first stage, using the VoqX, we recorded heart sounds from 100 patients referred for echocardiography (derivation group), 50 with moderate or severe AS and 50 without valvular disease. An AI-based supervised learning model was applied to the auscultation data from the first 100 patients used for training, to construct a diagnostic algorithm that was then tested on a validation group (50 other patients, 25 with AS and 25 without AS). In the second stage, conducted at a different medical center, we tested the device on 106 additional patients referred for echocardiography, which included patients with other valvular diseases.
Using data collected at the aortic and pulmonic auscultation points from the derivation group, the AI-based algorithm identified moderate or severe AS with 86% sensitivity and 100% specificity. When applied to the validation group, the sensitivity was 84% and specificity 92%; and in the additional testing group, 90% and 84%, respectively. The sensitivity was 55% for mild, 76% for moderate, and 93% for severe AS.
Our initial findings show that an AI-based stethoscope with infrasound capabilities can accurately diagnose AS. AI-based electronic auscultation is a promising new tool for automatic screening and diagnosis of valvular heart disease.
Accuracy and Efficacy of Hand-Held Echocardiography in Diagnosing Valve Disease: A Systematic Review
2018, American Journal of Medicine
Citation Excerpt :
Accordingly, studies have been conducted comparing the diagnostic capabilities of auscultation and HHE. As evidenced in Table1, even experienced cardiologists demonstrated major limitations in accurately detecting various valvular abnormalities by auscultation.1,2,4,5,9 In a study by Patel et al,1 significant variability was observed between auscultation by experienced clinicians and standard echocardiographic findings in detecting valvular lesions.
In recent years, advances in technology have enabled hand-held echocardiography (HHE) to generate high-quality 2-dimensional and color Doppler images. As these devices become smaller, simpler, and more affordable, the question of whether HHE can augment or replace auscultation as the primary mode of cardiovascular diagnosis has become increasingly more relevant. If widely implemented, HHE has the potential for significant cost savings and better resource utilization. This review examines studies comparing the sensitivities of auscultation, HHE, and standard echocardiography in detecting various valvular lesions and discusses why current evidence supports the use of HHE to augment the physical examination, which can lead to more reliable and rapid bedside diagnoses, triage, and appropriate treatment of structural cardiac abnormalities.
Variation in effectiveness of a cardiac auscultation training class with a cardiology patient simulator among heart sounds and murmurs
2017, Journal of Cardiology
Citation Excerpt :
Cardiac auscultation is one of the most important physical examination skills, providing useful information to diagnose patients with a variety of heart diseases [1]. Not only does the identification of a typical heart murmur strongly suggest the existence of a certain valvular heart disease or congenital heart disease [2], but the presence of heart murmurs has also been shown to predict new-onset heart failure in primary care outpatients [3]. An abnormality in the second heart sound is also a helpful clue in the diagnostic process of patients with overloaded right ventricle or electrical conduction disturbances that affect right ventricular performance [1].
Effectiveness of simulation-based education in cardiac auscultation training is controversial, and may vary among a variety of heart sounds and murmurs. We investigated whether a single auscultation training class using a cardiology patient simulator for medical students provides competence required for clinical clerkship, and whether students’ proficiency after the training differs among heart sounds and murmurs.
A total of 324 fourth-year medical students (93–117/year for 3 years) were divided into groups of 6–8 students; each group participated in a three-hour training session using a cardiology patient simulator. After a mini-lecture and facilitated training, each student took two different tests. In the first test, they tried to identify three sounds of Category A (non-split, respiratory split, and abnormally wide split S2s) in random order, after being informed that they were from Category A. They then did the same with sounds of Category B (S3, S4, and S3 + S4) and Category C (four heart murmurs). In the second test, they tried to identify only one from each of the three categories in random order without any category information.
The overall accuracy rate declined from 80.4% in the first test to 62.0% in the second test (p < 0.0001). The accuracy rate of all the heart murmurs was similar in the first (81.3%) and second tests (77.5%). That of all the heart sounds (S2/S3/S4) decreased from 79.9% to 54.3% in the second test (p < 0.0001). The individual accuracy rate decreased in the second test as compared with the first test in all three S2s, S3, and S3 + S4 (p < 0.0001).
Medical students may be less likely to correctly identify S2/S3/S4 as compared with heart murmurs in a situation close to clinical setting even immediately after training. We may have to consider such a characteristic of students when we provide them with cardiac auscultation training.
Misconceptions and Facts About Aortic Stenosis
2017, American Journal of Medicine
Citation Excerpt :
Additionally, widely available technology has taken a toll on skills of physical examination among many physicians, even cardiologists.3,4 A recent study demonstrated similar overall cardiac auscultatory proficiency for common valvular lesions among cardiologists and internists.5 However, there are still reasons to believe that proper physical examination has a role in identifying patients with aortic stenosis and in grading the severity of aortic stenosis.
Aortic stenosis is the most common valvular heart disease leading to intervention, and it is typically a disease of the elderly. Recent clinical advances have expanded the role of transcatheter aortic valve intervention in patients with severe aortic stenosis, making aortic valve intervention feasible and effective even in patients at intermediate, high, and prohibitive surgical risk. With the rapid advances in treatment, proper diagnosis becomes crucial for a wide range of patients with aortic stenosis: from “concordant” high-gradient aortic stenosis to “discordant” low-gradient aortic stenosis. The latter group commonly presents a clinical challenge requiring thoughtful and comprehensive evaluation to determine eligibility for aortic valve intervention. Providers at all levels should be familiar with basic diagnostic caveats and misconceptions when evaluating patients with possible aortic stenosis.
The need for annual echocardiography to detect cabergoline-associated valvulopathy in patients with prolactinoma: A systematic review and additional clinical data
2015, The Lancet Diabetes and Endocrinology
Citation Excerpt :
In clinical practice, these measurements are not markers of echocardiograms that would change medical management. Cardiovascular examination is a sufficiently accurate method to assess clinically significant valvular heart disease when done either by cardiologist or non-cardiologists.9,10,37,38 Of 143 asymptomatic participants (normal volunteers and patients with connective tissue disorders), a cardiovascular examination had positive and negative predictive values of 92%, and specificity of 98% in the diagnosis of valvular heart disease of mild severity as confirmed by echocardiogram.8
Present recommendations by the US Food and Drug Administration advise that patients with prolactinoma treated with cabergoline should have an annual echocardiogram to screen for valvular heart disease. Here, we present new clinical data and a systematic review of the scientific literature showing that the prevalence of cabergoline-associated valvulopathy is very low. We prospectively assessed 40 patients with prolactinoma taking cabergoline. Cardiovascular examination before echocardiography detected an audible systolic murmur in 10% of cases (all were functional murmurs), and no clinically significant valvular lesion was shown on echocardiogram in the 90% of patients without a murmur. Our systematic review identified 21 studies that assessed the presence of valvular abnormalities in patients with prolactinoma treated with cabergoline. Including our new clinical data, only two (0·11%) of 1811 patients were confirmed to have cabergoline-associated valvulopathy (three [0·17%] if possible cases were included). The probability of clinically significant valvular heart disease is low in the absence of a murmur. On the basis of these findings, we challenge the present recommendations to do routine echocardiography in all patients taking cabergoline for prolactinoma every 12 months. We propose that such patients should be screened by a clinical cardiovascular examination and that echocardiogram should be reserved for those patients with an audible murmur, those treated for more than 5 years at a dose of more than 3 mg per week, or those who maintain cabergoline treatment after the age of 50 years.

View all citing articles on Scopus

View full text

Letter to the EditorAccuracy of cardiac auscultation in the era of Doppler-echocardiography: A comparison between cardiologists and internists