• heart disease

The stethoscope is the symbol of a physician. Every patient expects their doctor to listen carefully to their heart and believes that listening with a stethoscope will identify the presence of heart disease. Unfortunately, our patient’s trust in this part of the physical examination is misplaced. There is no doubt that René Laennec’s invention of the stethoscope over 200 years ago transformed medical diagnosis, and this device remains useful for amplifying internal noises from several different organs in health and disease. However, the characteristics of an audible murmur are not reliable for diagnosis, despite the long-standing fascination with cardiac auscultation and valvular heart disease as expressed so eloquently by George Eliot in Middlemarch, ‘He was not much acquainted with valves of any sort, but he knew that valvae were folding-doors, and through this crevice came a sudden light startling him with his first vivid notion of the finely adjusted mechanisms in the human frame’.1

Most heart murmurs are benign, due to normal blood flow patterns or associated with transient increases in cardiac output due to pregnancy, anaemia, fever or other causes. However, some murmurs do indicate serious underlying heart disease, and distinguishing a benign from malignant murmur, or even hearing the murmur at all, is not easy. Diagnosis of heart valve disease is most important when cardiac symptoms are present. Unfortunately, symptoms due to valve disease are non-specific with many patients attributing exertional dyspnoea, dizziness, chest discomfort or fatigue to ageing, deconditioning or comorbid conditions. Furthermore, many patients simply circumvent symptoms by an unconscious gradual adjustment of activity levels in response to a slowly progressive disease process; these patients do not complain of symptoms at all.

Why does it matter? The most common pathological cause of a murmur in adults is calcific aortic valve disease with some degree of valve thickening present in 25% of people over age 65 years. Progression from mild obstruction to severe aortic stenosis (AS) occurs in many patients with an increasing prevalence of severe AS with age. The prevalence of moderate to severe AS is estimated to be only about 0.4% over the entire adult age range but increases to 1.3% in those aged 60–69 years, 3.9% for ages 70–79 years and 9.8% over age 80 years.2 3 Patients with severe symptomatic AS have dismal outcomes with a mortality rate over 60% at 2 years in the absence of intervention, a prognosis that can be reversed by surgical or transcatheter valve replacement.4 Yet, patients may not report symptoms, and healthcare providers may not recognise the murmur.

In their Heart paper, Gardezi and colleagues5 evaluated the accuracy of cardiac auscultation for detection of valve disease by primary care providers, compared with echocardiography, in 251 asymptomatic patients over age 65 years with no prior diagnosis of heart valve disease. Remarkably, significant valve disease was present in 14%, and mild valve disease was present in the majority (68%) of these patients. Auscultation with a stethoscope had a very low sensitivity, identifying only 44% of the patients who had significant valve disease. Furthermore, specificity was suboptimal with a murmur indicating significant valve disease in only 69% of those with this finding. In fact, diagnostic accuracy of the stethoscope was so low that auscultation findings did not significantly increase or decrease the likelihood that the patient had significant valve disease, compared with not listening at all.

The limitation of cardiac auscultation for diagnosis of valve disease is not simply lack of training; numerous studies have documented that more advanced training in cardiology does not improve diagnosis.6 7 The problem is that transmission of heart sounds to the chest wall correlates only modestly with the presence and severity of valve disease. Echocardiography is very sensitive for diagnosis of valvular heart disease, and while many patients have only mild disease that will never need treatment, a substantial number have progressive valve disease that eventually does affect quality of life. In addition, even mild valve thickening is a marker of increased cardiovascular risk, identifying patients who might benefit from more aggressive primary risk factor reduction.8 Rather than continuing to consider cardiac auscultation a key clinical skill with time and effort dedicated to training healthcare providers in the nuances of heart sounds heard with a stethoscope, it is time to turn to more effective technology—ultrasound, not acoustic sound.

One option is a diagnostic echocardiogram in any patient with risk factors or symptoms that might be due to valve disease. A less expensive option might be point-of-care ultrasound (POCUS) by the primary care provider to identify abnormal valve anatomy or motion. Patients with thickened valves or abnormal valve motion would then be referred for a subsequent complete echocardiogram with imaging and Doppler flow studies to determine the cause, severity and haemodynamic consequences of valve disease. In addition to diagnosis of severe valve disease in symptomatic patients who would benefit from valve intervention, this approach also would identify early valve disease, as well as other even more common cardiac conditions such as atrial fibrillation and heart failure with reduced ejection fraction. Patients with mild to moderate valve disease benefit from patient education, healthy lifestyle behaviours, appropriate preventative therapies and prospective monitoring for disease progression and symptoms, allowing intervention before any irreversible cardiac changes. We then might avoid the current situation where many patients present so late in the disease course that intervention is risky and sometimes futile.

Valve disease is as common and as lethal as many cancers for which routine screening is recommended. Why are we not screening for significant valve disease in older adults with something more effective than a stethoscope? Of course, replacing the stethoscope with an ultrasound transducer will only be effective if this technology is used appropriately. Thus, several steps are needed before we consider wide implementation of this approach. We need validated training modules that ensure providers can acquire and interpret images correctly. Clinical practice should include archiving of selected images, continuous quality improvement and adherence to a focused scope of practice. Next, we need population-based studies to ensure that screening adults for valve disease, followed by a complete echocardiogram in those with an abnormal screening test, actually improves diagnosis and clinical outcomes at a reasonable cost. This analysis also will need to consider the types of populations that should be screened in terms of age and other risk factors. Finally, patients who are diagnosed with valve disease need accurate information about their condition, periodic monitoring and medical care by qualified practitioners, and the opportunity to participate meaningfully in decisions about timing and type of intervention, if needed.

Now is the time to start teaching POCUS to healthcare providers. The diagnosis of heart valve disease is missed in too many patients. We need to ‘mind the gap’ by putting our stethoscopes back in our pockets and taking out our ultrasound transducers.