A systematic review and utilization study of digital stethoscopes for cardiopulmonary assessments

Literature review

A review of the literature was performed in accordance with a rigorous, systematic process^[17] in February 2021 to identify and assess the methodological quality of published studies that compared the quality of real-time audio and video streaming using digital stethoscopes to conventional stethoscopes used at the bedside. PubMed, Science Direct, CINAHL, TRIP and Open Grey databases, as well as ClincialTrials.gov were searched using broad search terms “digital stethoscope*” to ensure the search was as comprehensive as possible. To improve the relevancy of the articles returned, a more focused approach was followed in Pubmed and Science Direct by adding “cardio*” to the search terms, filtering by article type, and limiting by year (2000-present in Science Direct only). More details about the search are provided in Table 1. Additionally, relevant digital health journals were searched using the words “digital stethoscope,” websites of digital stethoscope manufacturers were searched, and hand searches were conducted.

PICOS (Population, Intervention, Comparator, Outcome, Study Design) criteria were used to guide inclusion/ exclusion criteria [Table 2]. Briefly, only studies involving human adults undergoing cardiopulmonary assessments in trials or meta-analysis were of interest. The justification for excluding pediatric patients is because the systems tested in the utilization study are meant to be used on adult patients with cardiovascular disease. Additionally, only studies that tested the quality of digital stethoscopes in a setting of real-time remote monitoring and transmitting audio and video over a distance were of interest. Quality deteriorates significantly when audio is transmitted over Wi-Fi/Internet, as opposed to audio recordings made at the bedside for review later. Trials were excluded if they compared digital and conventional stethoscopes, but the assessor was in the same location as the patient for both digital and conventional stethoscope assessments. This was a requirement to ensure study conditions of publications resembled the conditions under which the digital stethoscopes were tested in the current study. Additionally, clinical trials without available published full-text articles, articles not in English, and studies without a comparator were also excluded. The outcomes of interest were quality and ease of use of digital stethoscopes for real-time audio and video streaming between different locations in comparison to cardiopulmonary auscultation with conventional stethoscopes and methods.

One reviewer (AL) excluded articles based on title and abstract review, and two reviewers (AL, CK) excluded articles after full-text review. Reviewers were in consensus, and a third adjudicating party was not needed to resolve disagreement. The Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines were followed [Figure 1].^[18] Methodological quality was assessed using the Critical Appraisal Skills Programme (CASP) for Randomized Controlled Trials (RCTs).^[19] Data were extracted by a single reviewer (AL) into a table [Table 3], including information about design, setting, population, intervention, comparator, outcome, findings, and quality.

Close

Figure 1:: PRISMA flow chart.

Export to PPT

CaViAs

Clinical testing of the equipment was carried out to compare the published findings to the clinical test findings. Exhaustive testing of various digital stethoscopes and other required equipment was undertaken to ensure that Fraser Health was adopting the best digital stethoscope set up, and maximizing the audio and video experience for both provider and remote receiver of care. The control comparator was a conventional, non-electronic, non-digital stethoscope: America Diagnostic Corporation (ADC) Adscope 615 Clinical Stethoscope. The primary purpose was to test the equipment’s quality and capability to preserve audio and video integrity when transmitted via real-time streaming. The secondary purpose was to test the ease of utility of the devices and connections from a human factors standpoint, including setting up the equipment, connecting the devices, and initiating the real-time stream. All simulation and testing was conducted with the digital stethoscope directly on the subject’s skin.

The subject and the stethoscope was situated in a different location to where the listener to the heart and lung sounds was situated; sending a transmission to a location remote from where the stethoscope was receiving the sounds. All, but the final test, involved the same operator of the equipment and the same listener of the real-time feed. The equipment operator has 16 years of experience in computer programming and technical leadership, and the listener and evaluator has 18 years of clinical practice in cardiopulmonary assessments. The final test included an independent operator and listener to verify the results obtained in all prior tests. The project was named CaViAs: Cardiac Virtual Health Assessments. CaViAs and included evaluation of quality and utility of three digital stethoscope devices, three digital platforms/applications, three noise-cancelling headsets, and two Internet-enabled devices [Table 4]. Each digital stethoscope was tested with the selection of platforms, headsets, and Internet-enabled devices as outlined in the methods. Each digital stethoscope was optimized to maximum potential in consultation with the device developer or vendor specialist. Scores were assigned from 1 (low) to 10 (high) for quality (audio and video) and ease of use in setting up the equipment and establishing a real-time connection. The equipment choice was based on availability from Canadian vendors.

No ethical clearance was required for the literature review or for the CaViAs project because no patients were involved.

Literature review

The literature search generated 238 articles, of which 139 were duplicates or were excluded based on title. The remaining 99 abstracts were screened and 57 were excluded, leaving 42 articles that underwent full-text review. Of these, 41 were excluded based on PICOS criteria [Figure 1]. Fifteen articles were found through various hand searches, but all were ultimately excluded. Only one study was found that compared the quality of a digital stethoscope (real-time live audio and video streaming over several kilometers) to the quality of using a conventional stethoscope at the bedside.^[20]

The methodological quality (using CASP criteria) of the clinical validation of the stethoscope in this study was relatively poor [Table 3]. The authors rated the acoustic quality of the digital stethoscope as good to very good; intra-observer agreement was mostly good to very good, but poor for some murmurs; there was inter-observer agreement in 8 of 12 patients for respiratory and cardiac sounds; and exam time with the digital stethoscope was 191 seconds compared to 110 seconds with the conventional stethoscope (Classic-II Littman).^[20] Limitations of this study are that it was mainly about the design of the digital stethoscope with only a small section addressing the clinical validation in a group of patients. However, no patient demographic details were provided, the details about the comparative process were limited, and the reported results were not comprehensive.^[20] An additional test was conducted with the digital stethoscope prototype described in this study in Peru over a 180 km distance, and a couple of cardiologists and pulmonologists were satisfied with the digital stethoscope when they tested it in this real-time telemedicine scenario.^[20] Due to the lack of high quality published literature, this systematic literature review could not inform the CaViAs Project, but it highlighted the need for equipment testing in this particular type of setting.

CaViAs testing

Digital stethoscopes that were tested are presented in Table 5. Various combinations of selected equipment were evaluated. Each digital stethoscope device was tested with the Cisco DX80 and Lenovo Laptop, the three headsets, and Zoom. The DX80 has videoconference capability used for video and audio feed. Additional tests of the platforms were limited to device capability and developer recommendations. The Eko ECG App is compatible with the Eko Duo device. The Littman 3200 Bluetooth is the preferred device for the TeleSensi platform.

Table 5:: Digital stethoscopes tested in the CaViAs project as illustrated by the manufacturers.

The CaViAs test results revealed that the best quality of audio, video, and ease of use for all parties was the Eko Duo digital stethoscope in combination with the Eko ECG application, streamed between two Cisco DX80 devices, using the Plantronics Voyager 8200 on the receiver end. This combination scored 9.5 out of 10 for audio quality when compared to the conventional stethoscope. The video quality of the subject and the application video feed was superior to other combinations, and compared to other device setups, ease of use and initiation of the video session was superior to others.^[9-10] The addition of the Eko ECG app provided confirmation of audio signal with visual sound waves (phonocardiography) and a one-lead electrical tracing [Figure 2]. This feature added significant value to the experience that was not matched with the other devices. The Eko Duo used without the Eko ECG application reduced the audio quality to 5 out of 10 due to volume issues, even with all sounds setting optimized. However, not using the Eko ECG app increased the ease of use to 10 out of 10 because using the Eko ECG app adds steps to the user process.

Close

Figure 2:: Cisco DX80 showing visual tracing of the Eko ECG application on a shared screen in the CaViAs project.

Export to PPT

Testing audio quality showed that the Thinklabs One device was comparable to the Eko Duo, however the set up and settings on the device itself were technically challenging. The Littman sound quality, even when paired with the TeleSensi platform that is intended to enhance the experience, was suboptimal and introduced static and suboptimal lung sounds. The Eko Duo was the easiest digital stethoscope to set up and connect to other devices, with no settings on the device itself that would require manipulation, and minimal interference when moving the device over the skin. When comparing the Eko ECG app to TeleSensi and Zoom, the steps to activate the Eko ECG app were significantly less. The Eko ECG app is the only platform that has visual representation of both the sound and electrical cardiac waves. The TeleSensi platform provides audio visualizations of the sound waves (phonocardiography), however, in tests, the audio quality was suboptimal and the steps to connect the digital stethoscope to the platform were cumbersome, necessitating session codes, and on several occasions, connections were not made on the first attempt. TeleSensi does not allow for video streaming of the subject during the session, required a separate connection via Zoom to see the subject under assessment. Zoom used alone lacked the ability to transmit the visual representation of the audio sounds, but provided video streaming of the subject and audio streaming of the stethoscope sound. Zoom setup includes a series of steps including meeting codes and accepting invitations to sessions, which decreased ease of use. The DX80 device was the best option for Internet-based real time streaming of audio and video. The DX80 is a touch screen video conference-enabled device that allows for very easy connecting between DX80 devices. The steps to activate a call are simple and allow very little room for error, and the screen layout is also customizable to the viewer’s preference. Compared to the Lenovo laptop, which needs to be used in conjunction with Zoom for video, the steps with the DX80 were significantly reduced, and user experience and video quality of the DX80 was substantially better. The Plantronics Voyager 8200 headset was the best option for listening remotely to the real-time feed in a different location. This over-ear headset was significantly better at noise cancelling and was more comfortable than the comparators. The addition of Bluetooth connectivity to the Internet-enabled device and the auto on/off feature that is triggered when placing/removing the device from the ear, improved ease of use.

Utilization and adoption of a new practice can be affected by the ease of use. These include factors related to how easy it is to handle the stethoscope, additional settings that need to be manipulated, and ease in connecting the stethoscope to other devices. The ideal set up outlined in the CaViAs project took two steps, compared to the least ideal set up that included more than 10 steps. The best-case set up for each digital stethoscope was utilized to maximize audio and video quality and ease of use. As illustrated in Figure 3, the Eko Duo was ranked the best with an overall score of 18.5 out of 20; the Thinklabs was ranked 15.5 out of 20 and lost points for ease of use; and the Littman was ranked 7 out of 20 and lost points for both audio quality and ease of use. More details can be found in Table 6 outlining the various combinations of devices tested and the strengths and challenges encountered in the CaViAs project test scenarios.

Close

Figure 3:: Best-case scenario of digital stethoscope performance, audio quality, and ease of use.

Export to PPT