Radiography, Volatile Organic Compounds
Conditions
Keywords
tuberculosis, vulnerable populations, screening algorithms, cost effectiveness, neural networks, artificial intelligence
Brief summary
Two diagnostic tools for TB screening in high risk groups need additional assessment: the AeoNose™, an 'electronic nose device' for breath sampling, and digital chest X-ray (CXR) with computer aided detection with CAD4TB® software. This study will systematically screen prisoners and its' employees for TB, test the diagnostic performance of AeoNose™ and CAD4TB (both individually and together) as a TB screening tool and and establish Mycobacterium tuberculosis epidemiology in Paraguayan prisons.
Detailed description
The aim of the World Health Organization (WHO) End TB Strategy is to end the global pandemic by 2035. To be able to succeed, better point-of-care (POC) tests are urgently needed to improve screening of high-risk populations. Prisons are recognized worldwide as high risk environments for the concentration, amplification and transmission of TB among prisoners and their communities outside. Paraguayan penal institutions are known to have very high incidence rates of active TB (3000-5000/100.000, according to Paraguayan Ministry of Justice). Two diagnostic tools for TB screening in high risk groups will be evaluated: the AeoNose™, an 'electronic nose device' for breath sampling, and digital chest X-ray (CXR) with computer aided detection with CAD4TB® software. This study will systematically screen prisoners and its' employees for TB, test the diagnostic performance of AeoNose™ and CAD4TB® and assess prison Mycobacterium tuberculosis epidemiology through several objectives: Objective: 1. Assessment of the sensitivity and specificity of the AeoNose™ and its utility for mass TB screening in high incidence settings. 2. Evaluation of the feasibility and utility of CAD4TB® digital CXR as a mass screening tool for TB. 3. Identification of factors that affect diagnostic accuracy of AeoNose™ and CAD4TB®. 4. Assessment of TB epidemiology in Paraguayan prisons and identification of mycobacterial strains. Study design: Diagnostic cohort study. Study population: Detainees (PPL) of Paraguayan penal institutions as well as their employees. Nature and extent of the burden and risks associated with participation, benefit and group relatedness. Individual burden: all participants will perform one visit with a medical doctor for medical history, physical exam, digital X-ray (with CAD4TB®) and AeoNose sampling. Participants will be offered voluntary HIV testing and counseling. In case of presumptive TB (estimated 20-30% of participants) two sputum samples will be taken, either spontaneous or saline-induced. One sample is tested with GeneXpert and the other with liquid mycobacterial culture. Cases of presumptive TB with both negative GeneXpert® and culture results will be followed-up after three and six months for repeat testing.
Interventions
DEVICEelectronic nose device
electronic nose device will be tested as a triage instrument for tuberculosis in a high risk population
DEVICEdig chest Xray with automated readings
software for tuberculosis detection on digital Chest Xray providing a probability score from 0-100
Sponsors
European Union
Instituto de Saude Publica da Universidade do Porto
VHIR
University Medical Center Groningen
Laboratorio central de salud publica, Paraguay
Instituto Nacional de Enfermedades Respiratorias y del Ambiente Juan Max Boetner
National TB program Paraguay
IICS Paraguay
Radboud University Medical Center
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
DIAGNOSTIC
Masking
NONE
Intervention model description
outcomes of individual smell prints and digital ChestXrays will be compared with sputum sampling for GenXpert and Mycobacterial culture
Eligibility
Sex/Gender
ALL
Age
16 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* All persons, either being a prisoner or employee, of the involved penal institutions after providing informed consent, including those with known active TB disease and/or currently on TB treatment
Exclusion criteria
* A potential participant who meets any of the following criteria will be excluded from participation in this study: 1. Unable to exhale breath through the AeoNose™ during five minutes due to respiratory insufficiency 2. Unable to stand in an upright position for the CXR 3. Unable to communicate and comply with the instructions of the study team
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Volatile organic compound signal as measured with the AeoNose™ | Baseline/ change from baseline at 3 and 6 months | The exhaled breath 'smell print' of each participant will be compared over time at baseline, 3 months and 6 months with sputum microbiology results and digital Chest Xray to establish sensitivity and specificity of electronic nose signal to diagnose Tuberculosis |
| Tuberculosis probablity score using Computed Automated detection software CAD4TB® on chest Xrays | Baseline/ change from baseline at 3 and 6 months | Ranges from 0 to 100, where 0 would mean the lowest probability of having Tuberculosis and 100 would be the highest probability of having Tuberculosis. |
| Positive mycobacterial sputum test | Baseline/ change from baseline at 3 and 6 months | positive GenXpert test and/or mycobacterial cultures are used to establish Tuberculosis diagnosis |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Comparison of predictive value of volatile organic compound signal with Tuberculosis Probability Score | Baseline/ change from baseline at 3 and 6 months | comparison of correct positive predictions for active TB, with AeoNose (positive/ negative test result) and Cad4TB software from chest Xrays providing probability scores from 0-100 points |
| added value of the use of AeoNose and Cad4TB as a TB screening algorithm in vulnerable populations | Baseline/ change from baseline at 3 and 6 months | evaluation of the combined use of both AeoNose and Cad4TB probability score to predict active TB |
Countries
Paraguay
Contacts
Primary ContactLuis Gomez Pacielo, MD
Outcome results
None listed