Possibly more than any other area of cardiology, echocardiography has been a beneficiary of the decreasing cost, miniaturization, and increasing power of computing. The development of handheld ultrasound (HHU) devices is perhaps the prime example, with the most recent generation of equipment miniaturized to the extent that all image processing is performed within the transducer, and the attached tablet PC provides only display, archiving, and power. There are extensive published data on the incremental information provided by HHU, compared with the physical examination. Use of HHU can reduce the number of requests for “rarely appropriate” echocardiograms, many for patients who have had a previous definitive echocardiogram. In this context, the proposal by Dr. Almufleh and colleagues that Fellows should be trained in focused cardiac ultrasound is an understandable response to a truly spectacular technical development.

However, the proposal is based on 2 assumptions: 1) focused ultrasound is inherently different from transthoracic ultrasound training, which our Fellows receive in depth; and 2) the information provided by focused ultrasound is satisfactory for our purposes as cardiologists, as distinct from the other primary users of focused ultrasound in the emergency department or intensive care. Both of these assumptions are questionable.

As mentioned by Dr. Almufleh and colleagues, HHU modalities are limited to 2-dimensional imaging and color. We lack evidence that training in these modalities for focused echocardiograms should be inherently different from a standard echocardiogram. A well-trained echocardiographer should know when these modalities are insufficient for making a definitive diagnosis.

Cardiologists are usually the source of a definitive decision in the patients for whom we are consulted. In many situations, that decision requires a full echocardiogram. In addition to ischemic heart disease—where resting echocardiogram has well-known limitations—most of the general cardiology workload comprises the ongoing epidemics of “geriatric cardiology,” including heart failure (HF) (especially HF with preserved ejection fraction), atrial fibrillation, and aortic stenosis. In HF, an HHU can certainly identify the presence of a dilated ventricle and reduced ejection fraction. However, if these features are lacking, the HF diagnosis remains unresolved, and the diagnosis of HF with preserved ejection fraction will require spectral and tissue Doppler for the assessment of diastolic function, and (in my opinion, although not yet in the guidelines) assessment of global longitudinal strain. In atrial fibrillation, an HHU can provide information about left atrial volume, left ventricular size and systolic function, and a possibly causative valve lesion, but assessment of atrial function is not possible, and further assessment of a suspicious valve will need a full echocardiogram. In suspected aortic stenosis, in the absence of a comprehensively normal aortic valve, functional evaluation is required by Doppler.

The wider adoption of HHU is inhibited not by the lack of special training, but more by the attenuation of imaging skills in nonimaging cardiologists. In this context, Dr. Almufleh and colleagues’ statement for “democratizing ultrasound by providing ubiquitous access devoid of formal studies interpreted by echocardiographers” is truly scary. An HHU performed by someone with insufficient skills is probably worse than no information. Perhaps this attenuation of skills will be less marked as the generation of Fellows-in-training who have had access to HHU mature through their careers. Although this development would be welcome, we have not seen evidence of it yet.

—Tom Marwick

Edited for for clarity