Echocardiographic Examination, Image Transfer, and Interpretation

Echocardiographic examinations were performed using pocket-sized, hand-held cardiac ultrasound units (Vscan and Vivid I or Vivid Q portable cardiac ultrasound systems; GE Medical Systems, Milwaukee, WI). Scans were performed by volunteer sonographers trained to execute a protocol consisting of 11 standard views, including color-flow Doppler images of all valves ( Appendix 2 ). The Vscan is a small, pocket-sized device (135 × 73 × 28 mm), weighs <400 g, and has an 8.9-cm (diagonal) display with a resolution of 240 × 320 pixels. It uses a phased-array transducer (1.7–3.8 MHz) and displays grayscale images with a sector width of 75° and color Doppler images with a fixed sector width of 30°. Current-generation devices do not have the capabilities of spectral Doppler and M-mode imaging. Therefore, patients who needed additional imaging using continuous-wave or pulsed-wave Doppler to arrive at initial diagnoses were further scanned using the Vivid I or Vivid Q system. The Vivid I and Vivid Q are laptop-based, portable systems that allow more comprehensive examinations. All studies were digitally recorded in either mp3 or Digital Imaging and Communications in Medicine format.

On completion of each study, a provisional echocardiographic report was generated by the scanning sonographer and given to the patient for consultation with the on-site physician or cardiologist. Studies from the camp were uploaded to a cloud-based Web server (Studycast; Core Sound Imaging, Inc., Raleigh, NC). Using commercially available software (CoreConnect; Core Sound Imaging, Inc.), the study images were acquired from the modality (GE Vscan devices at the camp) and then transmitted to the image and work flow management component (CoreWeb; Core Sound Imaging, Inc.). The studies were then securely transmitted using a broadband internet connection. CoreConnect ensured the validity of the transmitted data by applying multiple integrity checks during the transmission process. Confidentiality of the transmitted data was ensured using standard Secure Sockets Layer (Transport Layer Security) encryption while the data were in transit between CoreConnect and CoreWeb and between CoreWeb and the user. Once the study images and data were transmitted to CoreWeb, they were available for access (interpretation, report generation, etc.) by any user with valid login credentials. Worldwide interpretations were performed by 75 volunteer physicians with level 2 or 3 or equivalent training who had preregistered with the American Society of Echocardiography ( Supplemental Figure 1 ). The study interpretations were performed using a standardized template that included information about chamber dimensions, valve morphology, color flow, global and regional left ventricular (LV) systolic function, and any apparent congenital cardiac malformations. Any other abnormality, if found, was also recorded. The reports were finalized on the Web-based system, with the goal of accomplishing this within 24 hours of initial scanning. The reports were subsequently downloaded and printed by the local coordinators, who distributed these reports to the patients. The remote readers were blinded to the interpretations made by the on-site readers.

For the purposes of analysis and interpretation, readers were requested to give only visual, qualitative insights (mild, moderate, or severe) on specific pathologic issues: LV dilation, LV wall hypertrophy (concentric or asymmetric), reduction of LV systolic function (visual LV ejection fraction [LVEF]), right ventricular dilation, left atrial dilatation, aortic root dilatation, valve calcification, pericardial effusion, pleural effusion, and dilation with reduced inspiratory reactivity of the inferior vena cava. LVEF was considered low if it was <55% by visual estimation and graded by American Society of Echocardiography–recommended definitions for LV dysfunction as mild (LVEF, 45%–54%), moderate (LVEF, 30%–44%), or severe (LVEF < 30%) LV dysfunction. We also noted segmental wall motion abnormality (yes or no) and the presence of pericardial effusion (clinically significant or not clinically significant). The presence of valvular abnormalities (regurgitant or stenotic) and their grades (mild, moderate, or severe) were also recorded. The severity of regurgitant lesions was based on two-dimensional findings (atrial or ventricular enlargement, hyperdynamic left ventricle) and qualitative color Doppler findings (width of vena contracta and jet area), whereas the severity of stenotic lesions was based on two-dimensional findings of valve opening and leaflet mobility, thickness, and calcification alongside chamber changes (hypertrophy in aortic stenosis, atrial dilatation in mitral stenosis). An abnormality was considered major if any of the following was found: valvular regurgitation of moderate or greater severity, any valvular stenosis, all congenital heart defects (except bicuspid aortic valves in the absence of any other associated significant abnormality), any LV systolic dysfunction or wall motion abnormality, and any other moderate or severe abnormality (e.g., moderate aortic root dilatation, moderate LV hypertrophy). All other echocardiographic abnormalities were deemed to be minor. The quality of echocardiographic images was graded by off-site readers on a scale ranging from 1 to 4 (1 = excellent, 2 = good, 3 = fair, and 4 = poor). In addition, images were labeled as (1) technically challenging and diagnostic or (2) technically challenging and nondiagnostic.

Cardiology Consultations

Patients with abnormal echocardiographic results were examined by the on-site cardiologists, who advised patients of the appropriate treatment on the basis of the clinical findings and the provisional echocardiographic reports. If required, immediate medical attention was facilitated with the help of the local administrative and medical staff members. The initial treatment advice was later modified, if necessary, once the final echocardiographic reports became available.

Follow-Up

Patients were asked to provide their contact phone numbers (if available) at the time of enrollment. Between 6 and 7 months after the initial evaluation, we contacted by telephone the cohort of patients who had registered their phone numbers and were found to have significant cardiac abnormalities during the initial echocardiographic examinations. We inquired about their overall well-being, the response to the treatment advice given, and whether they had sought further medical attention as advised.

Data Analysis and Interpretation

All data were managed and analyzed using a Microsoft Excel 2007 spreadsheet (Microsoft Corporation, Redmond, WA). Continuous data are reported as mean ± SD (or as medians and interquartile ranges if not normally distributed), and categorical data are reported as numbers and percentages. Descriptive analysis was performed to summarize the abnormal echocardiographic findings. The time intervals from scanning to study upload or interpretation were calculated and correlated with the image file size using Spearman’s rank correlation coefficient. The on-site interpretation was compared with the subsequent, formal expert interpretation to determine the diagnostic accuracy of the on-site interpretation. Discordance between on-site and expert readings was recorded when an abnormality was not reported or was overreported or when difference of more than one level of severity existed. Discordance was considered as major when the discrepancy related to a major abnormality (not stated, underrated, or overreported). Kappa coefficients were calculated as the measure of agreement between the two. P values < .05 were considered significant.