The use of ultrasound as a widely available bedside tool for cardiovascular evaluation has grown in parallel with technical developments that have permitted better image resolution, miniaturization, and automation. In the new generation of handheld devices, most of the image processing occurs in the ultrasound probe, with the image display, storage, and communication provided by a tablet computer or a smart phone.

A literature has developed around the validation of assessment and measurement using handheld devices. There is good evidence that their use is a valuable addition to the physical examination, and a number of validation studies have shown analogous diagnostic performance to a full-size echocardiographic system. Apart from the restrictions imposed by the number of modalities they offer, the main limitations of these devices relate to the humans controlling them and interpreting the findings. The acquisition of diagnostic images by nonexperts is a disruptive challenge, which has been provoked by the technical success of this methodology. Most of the studies of this topic have shown that relatively unsophisticated users, such as medical students and trainee physicians, can be taught to provide interpretations analogous to those obtained by experts. Although encouraging, the performance of young clinicians in the course of training, who are facile with digital devices and able to devote time and effort to learning a new technique, represents perhaps the “best case” scenario for the evaluation of this technology.

In this issue of JASE , Kimura et al. have extended the training discussion to examine retention of ultrasound skills. The movement of echocardiography away from the laboratory to the bedside has spawned a number of basic echocardiography courses, including one organized for all internal medicine residents over a number of years by the authors of this study. Specifically, this 50-hour program provided didactic and bedside instruction on seven core views to identify nine evidence-based signs of disease. A comprehensive scoring system evaluated image acquisition, knowledge of diagnostic criteria, and image interpretation, and an 80% threshold was used to represent competency.

The investigators have leveraged the valuable resource of these personnel and their records to improve the understanding of how the knowledge might attenuate with disuse over time. Successful graduates of the training program, most of whom were unable to continue their echocardiographic practice after graduation, and who were working locally, were approached to undertake follow-up evaluation at an average of nearly 3 years from their last use of ultrasound. In a subgroup of physicians who had no refresher training before reassessment, those physicians who were within 1 year of their last use of ultrasound performed better than those with a longer interval of nonuse. Although the numbers were small, the one third of the group who were provided with the opportunity for refresher training to prepare them for the second evaluation performed better than those without preparation.

These observations are extremely important in the context of attempts to incorporate handheld ultrasound as part of routine practice. It seems clear that, in common with many technical skills in medicine, attenuation occurs with disuse. Had the opportunity for ongoing use of this modality been provided to the physicians, this process of deskilling could have been avoided. Inherent in access to devices is the conflict between echocardiography as a bedside component of the physical examination (which is why we have started referring handheld ultrasound using the acronym PUPED as shorthand for “portable ultrasound as physical examination device”) and sonographer-performed, laboratory-based echocardiography. In the context of an ongoing and seemingly inevitable 20% rate of rarely appropriate testing, perhaps a different model should be constructed whereby the performance of PUPED becomes the gateway to entry into the echocardiography laboratory. Indeed, several studies have started to propose this model. In our experience, this approach satisfies the clinician requesting the test, avoids an inappropriate echocardiographic study, and sometimes identifies an unanticipated problem that justifies full echocardiography. However, for handheld ultrasound to satisfactorily fill this role, there will be a need to provide additional modalities as well as encourage broader use. Perhaps even a greater barrier in this respect is an economic one, related to the lack of reimbursement for an activity that will involve capital cost—perhaps the reason the former trainees mentioned in this report did not have ongoing access to PUPED. The solution to justifying the capital cost will have to be based on the provision of improved outcomes, shorter length of stay, increased referral of the appropriate tests for definitive diagnostic imaging modalities, and avoidance of nonreimbursement for rarely appropriate tests. In health care systems that are increasingly based on capitation, saving unnecessary imaging tests and ensuring appropriate ones will enable substantial cost saving.

The universal application of PUPED as part of a cardiologist’s practice is the final frontier for ultrasonography. For nonimaging cardiology specialists who have been practicing for >5 to 10 years, this is not a trivial barrier. However, perhaps this barrier can be scaled down by improved automation. For example, the report of Kimura et al. relates how grossly malpositioned transducer probe placement was the source of underperformance on follow-up testing. An automated process, based on pattern recognition, could be used to prompt the user that orientation was unacceptable or to lock at an appropriate position. Carotid ultrasound machines have already adapted such processes. Similarly, segmentation of the blood pool could be used to calculate chamber dimensions and volumes, speckle-tracking could be used for quantification of function, and automated Doppler tracing could be used for quantifying valvular disease. Keeping in mind that testing would be used to confirm normality or lack of change in most instances, the level of sophistication of each of the steps must not necessarily need to be high.

The development of PUPED may represent an example of disruptive innovation. Introduction of a relatively simple application at the most unsophisticated end of the market may yet alter the approach to and performance of echocardiography on a wider scale. It is becoming increasingly apparent that traditional approaches to clinical practice, rather than technical challenges, are the main barriers to this progression.