Phono- and Electrocardiogram Assisted Detection of Valvular Disease

NCT03458806 | Completed FDA Device

• 1 other identifier: 17-21881
• Study Type: observational
• Target: 156
• Locations: 1 country
• Sites: 1

Brief Summary

The diagnosis of valvular heart disease (VHD), or its absence, invariably requires cardiac imaging. A familiar and inexpensive tool to assist in the diagnosis or exclusion of significant VHD could both expedite access to life-saving therapies and reduce the need for costly testing. The FDA-approved Eko Duo device consists of a digital stethoscope and a single-lead electrocardiogram (ECG), which wirelessly pairs with the Eko Mobile application to allow for simultaneous recording and visualization of phono- and electrocardiograms. These features uniquely situate this device to accumulate large sets of auscultatory data on patients both with and without VHD. In this study, the investigators seek to develop an automated system to identify VHD by phono- and electrocardiogram. Specifically, the investigators will attempt to develop machine learning algorithms to learn the phonocardiograms of patients with clinically important aortic stenosis (AS) or mitral regurgitation (MR), and then task the algorithms to identify subjects with clinically important VHD, as identified by a gold standard, from naïve phonocardiograms. The investigators anticipate that the study has the potential to revolutionize the diagnosis of VHD by providing a more accurate substitute to traditional auscultation.

Conditions

Aortic Valve Stenosis; Mitral Regurgitation; Heart Murmurs; Valvular Heart Disease

Keywords

Auscultation; Phonocardiogram; Machine Learning; Heart Sounds

Outcome Measures

Primary Outcomes (2)

• Differentiation of clinically significant aortic stenosis from structurally normal hearts

  Identification by the trained machine learning algorithm of clinically important aortic stenosis (defined as moderate-to-severe or greater) from control subjects with structurally normal hearts and no greater than mild valvular heart disease, with comparison to the gold standard echocardiogram interpretation. As our algorithm will provide a continuous "score" to determine the likelihood of disease, the data will primarily come in the form of a receiver operating characteristic curve, for which we will calculate accuracy, specificity, and likelihood ratios at sensitivity cutoffs of 0.9, 0.95, and 0.99.

  Close of study (after final enrollment of the aortic stenosis validation set), within 1 year.

• Differentiation of clinically significant mitral stenosis from structurally normal hearts

  Identification by the trained machine learning algorithm of clinically important mitral regurgitation (defined as moderate-to-severe or greater) from control subjects with structurally normal hearts and no greater than mild valvular heart disease, with comparison to the gold standard echocardiogram interpretation. As our algorithm will provide a continuous "score" to determine the likelihood of disease, the data will primarily come in the form of a receiver operating characteristic curve, for which we will calculate accuracy, specificity, and likelihood ratios at sensitivity cutoffs of 0.9, 0.95, and 0.99..

  Close of study (after final enrollment of the mitral regurgitation validation set), within 1 year.

Secondary Outcomes (2)

• Differentiation of clinically significant aortic stenosis from the absence of clinically significant aortic stenosis

  Close of study (after final enrollment of the aortic stenosis validation set), within 1 year.

• Differentiation of clinically significant mitral regurgitation from the absence of clinically significant mitral regurgitation

  Close of study (after final enrollment of the mitral regurgitation validation set), within 1 year.

Study Arms (4)

Control

Subjects with echocardiographically confirmed valvular disease of less than moderate-to-severe grading with regards to aortic stenosis (AS) and mitral regurgitation (MR). Note that within this cohort will be a sub cohort consisting of subjects with structurally normal hearts, with no greater than mild valvular disease of any valve, no prior valvular intervention, and no evidence of congenital heart disease.

Diagnostic Test: AS Algorithm 2; Diagnostic Test: MR Algorithm 2

AS Case

Subjects with echocardiographically confirmed aortic stenosis (AS) of moderate-to-severe or greater grading.

Diagnostic Test: AS Algorithm 1; Diagnostic Test: AS Algorithm 2

MR Case

Subjects with echocardiographically confirmed mitral regurgitation (MR) of moderate-to-severe or greater grading.

Diagnostic Test: MR Algorithm 1; Diagnostic Test: MR Algorithm 2

Control Subgroup

Subjects with structurally normal hearts, with no greater than mild valvular disease of any valve, no prior valvular intervention, and no evidence of congenital heart disease.

Diagnostic Test: AS Algorithm 1; Diagnostic Test: MR Algorithm 1

Interventions

AS Algorithm 1 DIAGNOSTIC_TEST

Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater aortic stenosis from controls having structurally normal hearts with no greater than mild valvular heart disease at any location.

AS Case; Control Subgroup

AS Algorithm 2 DIAGNOSTIC_TEST

Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater aortic stenosis from controls having any findings other than moderate-to-severe or greater aortic stenosis.

AS Case; Control

MR Algorithm 1 DIAGNOSTIC_TEST

Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater mitral regurgitation from controls having structurally normal hearts with no greater than mild valvular heart disease at any location.

Control Subgroup; MR Case

MR Algorithm 2 DIAGNOSTIC_TEST

Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater mitral regurgitation from controls having any findings other than moderate-to-severe or greater mitral regurgitation.

Control; MR Case

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Adults with either moderate-to-severe to severe AS or moderate-to-severe to severe MR (cases) and adults with structurally normal hearts with minimal VHD (controls). In practice, the accessible population will be adults meeting the entry criteria undergoing clinical echocardiograms at the UCSF echocardiography laboratory amenable to participation.

You may qualify if:

• Able to provide consent
• Undergoing a complete echocardiogram

You may not qualify if:

• Refusal to participate

Sponsors & Collaborators

• University of California, San Francisco: lead
• Eko Devices, Inc.: collaborator

Study Sites (1)

University of California San Francisco

San Francisco, California, 94143, United States

Location

Related Publications (10)

• Jones EC, Devereux RB, Roman MJ, Liu JE, Fishman D, Lee ET, Welty TK, Fabsitz RR, Howard BV. Prevalence and correlates of mitral regurgitation in a population-based sample (the Strong Heart Study). Am J Cardiol. 2001 Feb 1;87(3):298-304. doi: 10.1016/s0002-9149(00)01362-x.

  PMID: 11165964 BACKGROUND

• Eveborn GW, Schirmer H, Heggelund G, Lunde P, Rasmussen K. The evolving epidemiology of valvular aortic stenosis. the Tromso study. Heart. 2013 Mar;99(6):396-400. doi: 10.1136/heartjnl-2012-302265. Epub 2012 Sep 2.

  PMID: 22942293 BACKGROUND

• Faxon DP, Williams DO. Interventional Cardiology: Current Status and Future Directions in Coronary Disease and Valvular Heart Disease. Circulation. 2016 Jun 21;133(25):2697-711. doi: 10.1161/CIRCULATIONAHA.116.023551. No abstract available.

  PMID: 27324364 BACKGROUND

• Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin JP 3rd, Guyton RA, O'Gara PT, Ruiz CE, Skubas NJ, Sorajja P, Sundt TM 3rd, Thomas JD; American College of Cardiology/American Heart Association Task Force on Practice Guidelines. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2014 Jun 10;63(22):e57-185. doi: 10.1016/j.jacc.2014.02.536. Epub 2014 Mar 3. No abstract available.

  PMID: 24603191 BACKGROUND

• Mangione S, Nieman LZ. Cardiac auscultatory skills of internal medicine and family practice trainees. A comparison of diagnostic proficiency. JAMA. 1997 Sep 3;278(9):717-22.

  PMID: 9286830 BACKGROUND

• Mangione S. Cardiac auscultatory skills of physicians-in-training: a comparison of three English-speaking countries. Am J Med. 2001 Feb 15;110(3):210-6. doi: 10.1016/s0002-9343(00)00673-2.

  PMID: 11182108 BACKGROUND

• Zoghbi WA, Adams D, Bonow RO, Enriquez-Sarano M, Foster E, Grayburn PA, Hahn RT, Han Y, Hung J, Lang RM, Little SH, Shah DJ, Shernan S, Thavendiranathan P, Thomas JD, Weissman NJ. Recommendations for Noninvasive Evaluation of Native Valvular Regurgitation: A Report from the American Society of Echocardiography Developed in Collaboration with the Society for Cardiovascular Magnetic Resonance. J Am Soc Echocardiogr. 2017 Apr;30(4):303-371. doi: 10.1016/j.echo.2017.01.007. Epub 2017 Mar 14. No abstract available.

  PMID: 28314623 BACKGROUND

• Baumgartner H, Hung J, Bermejo J, Chambers JB, Edvardsen T, Goldstein S, Lancellotti P, LeFevre M, Miller F Jr, Otto CM. Recommendations on the Echocardiographic Assessment of Aortic Valve Stenosis: A Focused Update from the European Association of Cardiovascular Imaging and the American Society of Echocardiography. J Am Soc Echocardiogr. 2017 Apr;30(4):372-392. doi: 10.1016/j.echo.2017.02.009.

  PMID: 28385280 BACKGROUND

• Pretorius E, Cronje ML, Strydom O. Development of a pediatric cardiac computer aided auscultation decision support system. Annu Int Conf IEEE Eng Med Biol Soc. 2010;2010:6078-82. doi: 10.1109/IEMBS.2010.5627633.

  PMID: 21097128 BACKGROUND

• Chorba JS, Shapiro AM, Le L, Maidens J, Prince J, Pham S, Kanzawa MM, Barbosa DN, Currie C, Brooks C, White BE, Huskin A, Paek J, Geocaris J, Elnathan D, Ronquillo R, Kim R, Alam ZH, Mahadevan VS, Fuller SG, Stalker GW, Bravo SA, Jean D, Lee JJ, Gjergjindreaj M, Mihos CG, Forman ST, Venkatraman S, McCarthy PM, Thomas JD. Deep Learning Algorithm for Automated Cardiac Murmur Detection via a Digital Stethoscope Platform. J Am Heart Assoc. 2021 May 4;10(9):e019905. doi: 10.1161/JAHA.120.019905. Epub 2021 Apr 26.

  PMID: 33899504 RESULT

MeSH Terms

Conditions

Aortic Valve Stenosis; Mitral Valve Insufficiency; Heart Murmurs; Heart Valve Diseases

Condition Hierarchy (Ancestors)

Aortic Valve Disease; Heart Diseases; Cardiovascular Diseases; Ventricular Outflow Obstruction; Signs and Symptoms; Pathological Conditions, Signs and Symptoms

Study Officials

• John Chorba, MD

  University of California, San Francisco

  PRINCIPAL INVESTIGATOR

Study Design

• Study Type: observational
• Observational Model: CASE CONTROL
• Time Perspective: CROSS SECTIONAL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: February 19, 2018
• First Posted: March 8, 2018
• Study Start: February 22, 2018
• Primary Completion: November 11, 2019
• Study Completion: November 11, 2019
• Last Updated: July 2, 2021

Record last verified: 2021-06

Data Sharing

• IPD Sharing: Will share

Locations

US (1)