Study on the link between gum disease and heart problems using artificial intelligence
| ISRCTN | ISRCTN15877121 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN15877121 |
| Secondary identifying numbers | 1737/CEL, ADMA-PERIO-CVD-2024 |
- Submission date
- 04/09/2025
- Registration date
- 23/09/2025
- Last edited
- 09/09/2025
- Recruitment status
- No longer recruiting
- Overall study status
- Completed
- Condition category
- Other
Plain English summary of protocol
Background and study aims
Many people have both gum disease and heart problems, but we don't fully understand how these conditions affect each other. This study uses artificial intelligence (computer technology) to find patterns that doctors might miss, helping us better understand who is at risk. This research looks at how gum disease and heart disease might work together to affect blood vessel health. We measure a substance in the blood called ADMA, which tells us how healthy blood vessels are. High ADMA levels mean blood vessels aren't working properly, which can lead to heart problems.
Who can participate?
Adults aged 18-75 years, divided into four groups:
1. People with healthy gums and heart
2. People with only gum disease
3. People with only heart disease
4. People with both conditions
What does the study involve?
Participants come for ONE visit (2-3 hours) where we:
1. Check your teeth and gums
2. Take dental X-rays
3. Collect a small blood sample (two teaspoons)
4. Collect saliva
5. Check blood pressure
6. Ask about medical history
This is an observational study - we only measure and analyze, we don't provide any treatment.
What are the possible benefits and risks of participating?
The study will:
1. Help identify people at higher risk for heart problems
2. Show if saliva tests could replace blood tests
3. Create a simple scoring system for dentists to spot at-risk patients
4. Improve understanding of how gum disease affects heart health
The procedures are the same as routine dental and medical check-ups. The only discomforts are minor - like having your gums examined or giving a blood sample. No payment is provided, but you receive a free comprehensive dental examination and information about your cardiovascular risk markers. All information is kept confidential. You get a study number instead of using your name. Only the research team can access your data, which is stored securely following privacy laws. Results will be published in medical journals, but no participant will be identifiable.
Where is the study run from?
IRCCS Istituto Tumori "Giovanni Paolo II" (Italy)
When is the study starting and how long is it expected to run for?
July 2024 to April 2025
Who is funding the study?
IRCCS Istituto Tumori "Giovanni Paolo II" (Italy)
Who is the main contact?
Prof. Francesco Inchingolo, francesco.inchingolo@uniba.it
Contact information
Public, Scientific, Principal investigator
Piazza Giulio Cesare, 11
Bari
70124
Italy
 ORCID ID |
0000-0003-3797-5883 |
|---|---|
| Phone | +39 (0)80 559 1111 |
| francesco.inchingolo@policlinico.bari.it |
Study information
| Study design | Observational cross-sectional case-control study |
|---|---|
| Primary study design | Observational |
| Secondary study design | Cross sectional study |
| Study setting(s) | Dental clinic |
| Study type | Diagnostic |
| Participant information sheet | Not available in web format, please use the contact details to request a participant information sheet |
| Scientific title | Machine learning analysis of asymmetric dimethylarginine (ADMA) levels in patients with periodontitis and cardiovascular disease: a cross-sectional study |
| Study acronym | ADMA-PERIO-CVD |
| Study objectives | 1. To investigate the synergistic effects of periodontitis and cardiovascular disease on serum asymmetric dimethylarginine (ADMA) levels and to develop a machine learning algorithm capable of predicting ADMA concentrations from clinical and radiographic parameters. 2. To quantify the correlation between periodontal inflammatory burden (measured by PISA) and ADMA levels in patients with and without cardiovascular disease. 3. To identify differential ADMA regulatory pathways between cardiovascular and non-cardiovascular cohorts through biomarker pattern analysis. 4. To validate the accuracy of artificial intelligence-based ADMA prediction compared to conventional clinical assessment. 5. To develop a simplified clinical risk score (Periodontal-ADMA Risk Score [PARS]) for identifying patients at risk of elevated ADMA levels. 6. To evaluate the relationship between salivary and serum ADMA levels as a potential non-invasive diagnostic approach. 7. To assess the association between radiographic periodontal parameters (bone loss patterns, furcation involvement) and systemic endothelial dysfunction markers. |
| Ethics approval(s) |
Approved 15/07/2024, Comitato Etico Locale IRCCS Istituto Oncologico "Gabriella Serio" (IRCCS Istituto Tumori "Giovanni Paolo II", Bari, 1001, Italy; +39 (0)80 555 5111; comitatoetico@oncologico.bari.it), ref: 1737/CEL |
| Health condition(s) or problem(s) studied | Periodontitis, cardiovascular disease, endothelial dysfunction, asymmetric dimethylarginine (ADMA) elevation |
| Intervention | This is an observational cross-sectional study with no interventions. Participants undergo a single visit including: 1. Clinical periodontal examination (probing depth, clinical attachment level, bleeding on probing, plaque index) 2. Blood sample collection for ADMA analysis via high-performance liquid chromatography (HPLC) 3. Medical history and cardiovascular assessment Total duration: One visit (approximately 2 hours) No follow-up required (cross-sectional design) |
| Intervention type | Other |
| Primary outcome measure | Serum ADMA levels as a biomarker of endothelial dysfunction measured using high-performance liquid chromatography (HPLC) at baseline (single assessment) |
| Secondary outcome measures | 1. Periodontal parameters (probing depth, clinical attachment level, bleeding on probing, plaque index) measured using UNC-15 probe at baseline (single assessment) 2. Inflammatory markers (hs-CRP, IL-6, TNF-α) measured using ELISA at baseline (single assessment) 3. Machine learning algorithm accuracy for ADMA prediction based on clinical parameters, assessed at study completion |
| Overall study start date | 15/07/2024 |
| Completion date | 30/04/2025 |
Eligibility
| Participant type(s) | Healthy volunteer, Patient |
|---|---|
| Age group | Adult |
| Lower age limit | 18 Years |
| Upper age limit | 75 Years |
| Sex | All |
| Target number of participants | 140 |
| Total final enrolment | 140 |
| Key inclusion criteria | 1. Age 18-75 years 2. Minimum 16 natural teeth present 3. For periodontitis groups: 3.1. ≥40% sites with CAL ≥2mm and PD ≥4mm 3.2. Radiographic evidence of bone loss 3.3. ≥40% sites with bleeding on probing 4. For CVD groups: 4.1. ≥50% stenosis of at least one coronary artery (angiographically verified) 4.2. OR history of documented coronary intervention 5. For healthy controls: 5.1. No systemic disease 5.2. ≤10% sites with bleeding on probing 5.3. No sites with PD ≥4 mm 6. Ability to provide informed consent 7. Willing to complete all study procedures |
| Key exclusion criteria | 1. Antibiotic or anti-inflammatory medication within 3 months prior to enrollment 2. Pregnancy or lactation 3. Uncontrolled diabetes (HbA1c >7.5%) 4. Current smoking >10 cigarettes/day 5. Systemic conditions affecting periodontal health (e.g., immunosuppression) 6. Active cancer treatment 7. Chronic kidney disease (eGFR <30 ml/min/1.73m²) 8. Periodontal treatment within 6 months 9. Unable to provide informed consent 10. Severe cognitive impairment 11. Active substance abuse 12. Participation in other clinical studies within 30 days |
| Date of first enrolment | 01/09/2024 |
| Date of final enrolment | 28/02/2025 |
Locations
Countries of recruitment
- Italy
Study participating centre
Bari
70124
Italy
Sponsor information
Hospital/treatment centre
Viale Orazio Flacco, 65
Bari
70124
Italy
| Phone | +39 (0)80 5593467 |
|---|---|
| comitato.etico@oncologico.bari.it | |
| Website | https://www.oncologico.bari.it |
Funders
Funder type
Hospital/treatment centre
No information available
Results and Publications
| Intention to publish date | 30/11/2025 |
|---|---|
| Individual participant data (IPD) Intention to share | Yes |
| IPD sharing plan summary | Stored in publicly available repository |
| Publication and dissemination plan | Primary Publication Target: Submission to peer-reviewed journal within 6 months of study completion (October 2025) Target journals (in order of preference): Journal of Clinical Periodontology (Impact Factor: 6.7) Journal of Periodontology (Impact Factor: 4.3) Clinical Oral Investigations (Impact Factor: 3.5) Secondary Publications: Machine learning methodology paper for AI/medical informatics journal Clinical validation paper for cardiovascular journal Brief communication on PARS score development Conference Presentations: EuroPerio 2025 (European Federation of Periodontology) International Association for Dental Research (IADR) 2025 European Society of Cardiology Congress 2025 Timeline: Data analysis completion: May 2025 Abstract submissions: June-July 2025 Manuscript preparation: August-September 2025 First submission: October 2025 |
| IPD sharing plan | Will IPD be shared? Yes - De-identified participant data will be made available What data will be shared? De-identified individual participant dataset Data dictionary defining all variables Statistical analysis plan Analytical code (R/Python scripts) When will data become available? 6 months after primary publication (estimated April 2026) For how long? 5 years from publication date With whom will data be shared? Researchers with methodologically sound proposals For meta-analyses and systematic reviews For validation of AI algorithms Upon reasonable request with appropriate ethics approval How to access? Submit proposal to: francesco.inchingolo@uniba.it Proposal must include: Research question Analysis plan Ethics approval (if applicable) Data use agreement must be signed Data provided via secure transfer Repository: Primary: Institutional repository (Policlinico Bari) Secondary: Consider deposit in Zenodo or Figshare for DOI assignment Clinical data: May submit to BioLINCC or similar clinical data repository FAIR Principles Compliance: Findable: DOI assigned, metadata in repositories Accessible: Clear access procedures defined Interoperable: Standard formats (CSV, JSON) Reusable: Clear licensing (CC-BY 4.0 for publications) Restrictions: No attempt to re-identify participants No commercial use without separate agreement Acknowledgment of original study required No data sharing that violates participant consent Additional Dissemination: Study summary for participants (lay language) Press release through institutional communications Social media dissemination (@uniba_it, @policlinicobari) Policy brief if findings have public health implications |
Editorial Notes
04/09/2025: Study's existence confirmed by the Comitato Etico Locale IRCCS Istituto Oncologico "Gabriella Serio".
