Cardiac Echo: There Is an App for That
By Todd Neale, Senior Staff Writer, September 23, 2011
Action Points
Explain that a field study of portable echocardiography in Honduras found that remote expert interpretation could be performed at both a workstation and smartphone with more than 90% agreement.
Note that diagnoses made by a cardiology fellow in the field performing the echocardiograms were overridden by the experts in more than one-third of cases.
Review
Remote interpretation of echocardiograms on a smartphone may help expand the use of such tests to previously inaccessible settings, researchers found.
There was relatively little loss of accuracy when electrocardiographic data transmitted from a remote Honduran village by a nonexpert was interpreted in the U.S. using a specially equipped smartphone instead of a full workstation, according to Brian Choi, MD, MBA, of George Washington University in Washington, D.C., and colleagues.
Among two expert echocardiographers, the intertechnology agreement between interpretations done on the smartphone and the workstation was 90% (P<0.01), the researchers reported online in the Journal of the American Society of Echocardiography.
"Our study does show that although pocket-size cardiac ultrasound may be inferior to complete echocardiographic evaluation, given the concordance of interpretation by expert readers using different methods to review the transmitted images, pocket-size cardiac ultrasound as an adjunct to traditional physical examination and with remote expert overread may provide additional diagnostic capability in remote or hostile environments," they concluded.
To test the accuracy of using a smartphone running medical imaging software compliant with the Health Insurance Portability and Accountability Act to interpret echocardiographic data from a remote location, the researchers conducted a study using 89 patients in a remote Honduran village and two expert echocardiographers in the U.S.
All of the Honduran patients (mean age 54; 58% female) underwent echocardiography by a nonexpert cardiology fellow who was on a humanitarian mission using a pocket-size ultrasound device.
The most common indications for echocardiography were arrhythmia (33%), cardiomyopathy (28%), and syncope or presyncope (15%).
The fellow, who had completed a three-month rotation in an academic echocardiography lab, provided a point-of-care diagnosis before transmitting the images to the U.S. by dial-up modem. The electronic files were sent by broadband from an urban center when field connectivity was unavailable.
The two expert echocardiographers in the U.S. first interpreted the findings on a workstation. After a minimum of four weeks, the studies were randomized and stripped of identifying information and sent to the readers via their smartphones, which had pinch zoom capabilities for both still images and cine loops.
Using the workstation, point-of-care diagnoses from the cardiology fellow were changed in 38% of patients; 41% of the changes were left ventricular function correction and 38% were valvulopathy correction. Another 18% of the changes involved poor image quality that prevented confirmation of the diagnosis.
The two U.S. experts agreed on the interpretation of the point-of-care diagnosis in 82% of cases (P<0.0001), which was excellent, according to the researchers.
"Smartphone interpretations by expert echocardiographers agreed with the point-of-care diagnoses in 61% of cases," they said. Further analysis comparing interpretations on workstations and the smartphone found excellent agreement of 90%.
Choi and colleagues noted that, because of the ubiquity of Internet access in the healthcare environment, the costs of starting a system involving remote interpretation of echocardiographic data on smartphones are limited to the point-of-care device, the smartphone, and the special software.
"Furthermore, as we find ourselves domestically to be in an era of more aggressive cost containment and care shifting more to mid-level practitioners, mobile consultation could provide improved access to specialists with advanced training, which may help limit cost growth without compromise of care," they wrote.
The researchers acknowledged some limitations of the study, including the fact that the indications for echocardiography were heavily weighted toward three conditions; the possibility that the point-of-care diagnostic accuracy would have been improved by having an expert echocardiographer there; significant transmission delays from the field; and the fact that the device used did not have spectral Doppler, electrocardiographic, or M-mode capabilities.
"Although larger, more robust echocardiographic equipment may have been more capable of acquiring superior data, such machines may not be practical in inaccessible locations, whether it be an austerely equipped humanitarian mission environment or in a crowded room at the bedside of a critically ill patient," the authors noted.