An Artificial Intelligence-based Approach in Total Knee Arthroplasty: From Inflammatory Responses to Personalized Medicine

NCT ID: NCT06634654

Last Updated: 2025-02-21

Basic Information

• Recruitment Status: RECRUITING
• Clinical Phase: NA
• Total Enrollment: 197 participants
• Study Classification: INTERVENTIONAL
• Study Start Date: 2024-10-14
• Study Completion Date: 2029-12-31

Brief Summary

Goal: The goal of this interventional study is to understand how multimodal preoperative data can predict outcomes after Total Knee Arthroplasty (TKA) and improve personalized medicine practices.

Participant Population: The study will enroll 197 patients suffering from symptomatic, end-stage knee osteoarthritis, who are above 18 years old and have functionally intact ligaments.

Main Questions:

• Can multimodal preoperative data, genetic predisposition, and psycho-behavioral characteristics predict outcomes after TKA?
• Can AI models effectively use this data to customize prostheses and surgical interventions, and predict patient outcomes? Comparison Group Information (If applicable): Not specified in the provided details.

Participant Tasks:

• Undergo TKA as per the normal clinical routine.
• Participate in pre- and post-surgical follow-ups including:
• Clinical-functional assessments.
• Administration of clinical scores.
• Collection of biological samples.
• Biomechanical analysis using a stereophotogrammetric system.
• Provide data for the comprehensive multimodal indexed database.

Detailed Description

Osteoarthritis is one of the most common causes of knee disorders, leading to pain, reduced mobility, and a decline in quality of life. Total knee arthroplasty (TKA) is one of the most established treatments for end-stage osteoarthritis. Despite advancements in surgical techniques, patient dissatisfaction remains high. After surgery, patients often experience swelling, pain, and difficulty with daily activities. Revision surgery is a major challenge, with aseptic loosening occurring in 15-20% of cases. Given the high disability rates and healthcare costs associated with TKA, optimizing patient care is crucial.

Artificial intelligence (AI) offers the potential to identify new care profiles. For the first time, AI can integrate multimodal datasets. This approach could lead to personalized treatment for knee osteoarthritis patients, in line with precision medicine principles. This study takes a multidisciplinary approach to better understand the causes of failure and dissatisfaction following TKA.

The primary aim of this study is is to create a multimodal database. This database will include structural, genetic, biomechanical, clinical, psychological, biological, stress-related, inflammatory, and demographic data. Using AI, the study aims to build predictive models for post-TKA outcomes. Insights from this research could improve patient management and lead to new therapeutic approaches.

Patients suffering from knee osteoarthritis at Fondazione Policlinico Universitario Campus Bio-Medico will be enrolled in this study if they meet the inclusion/exclusion criteria described above.

There are no risks for the patients recruited in the study. The total duration of the study is 5 years. The enrolment of patients will start on the 01/10/2024 and will last 12 months for each patient.

The Italian Ministry of Health and the Fondazione Policlinico Universitario Campus Bio-Medico supported this study.

The PI and also the main contact of this study is professor Umile Giuseppe Longo.

Conditions

Knee Osteoarthritis

Study Design

• Allocation Method: NA
• Intervention Model: SINGLE_GROUP
• Primary Study Purpose: TREATMENT
• Blinding Strategy: NONE

Study Groups

Patients who undergo total total knee arthroplasty

The study population comprises 197 patients who require Total Knee Arthroplasty (TKA) due to symptomatic, end-stage knee osteoarthritis. Eligible participants are adults over the age of 18 years with functionally intact ligaments. Exclusion criteria include individuals with neurological or other conditions that affect their ability to participate in walking trials, those with inflammatory or infectious arthritis, previous significant knee surgeries such as articular fractures (excluding knee arthroscopy and meniscal surgery), and those with active tumors or who are pregnant. This population selection is aimed at assessing the efficacy of AI-integrated interventions in improving surgical outcomes and postoperative recovery in a homogeneous group affected by severe knee degeneration.

Group Type: EXPERIMENTAL

Total Knee Arthroplasty

Intervention Type: PROCEDURE

Total Knee Arthroplasty is performed using conventional surgical techniques.

Multifaceted diagnostic assessments

Intervention Type: DIAGNOSTIC_TEST

Multifaceted diagnostic assessments involving genetic analysis, biomechanical data collection, radiographic imaging, and psychological evaluations.

Follow-ups

Intervention Type: BEHAVIORAL

Postoperative follow-up includes behavioral interventions, such as lifestyle counseling and rehabilitation programs, tailored based on AI-driven insights into individual patient recovery profiles.

Genetic screening

Intervention Type: GENETIC

Genetic screening and analysis, including whole exome sequencing, are conducted to identify genetic markers that might influence the outcomes of knee arthroplasty. This data is utilized within AI models to predict patient-specific surgical outcomes and recovery processes.

Interventions

Total Knee Arthroplasty

Total Knee Arthroplasty is performed using conventional surgical techniques.

Intervention Type: PROCEDURE

Multifaceted diagnostic assessments

Multifaceted diagnostic assessments involving genetic analysis, biomechanical data collection, radiographic imaging, and psychological evaluations.

Intervention Type: DIAGNOSTIC_TEST

Follow-ups

Postoperative follow-up includes behavioral interventions, such as lifestyle counseling and rehabilitation programs, tailored based on AI-driven insights into individual patient recovery profiles.

Intervention Type: BEHAVIORAL

Genetic screening

Genetic screening and analysis, including whole exome sequencing, are conducted to identify genetic markers that might influence the outcomes of knee arthroplasty. This data is utilized within AI models to predict patient-specific surgical outcomes and recovery processes.

Intervention Type: GENETIC

Eligibility Criteria

Inclusion Criteria

1. Symptomatic, end-stage knee osteoarthritis

2. Ligaments functionally intact

3. Age: older than18 years old

Exclusion Criteria

1. Neurological or other conditions affecting patients ability to join walking trials

2. Inflammatory or infectious arthritis

3. Previous articular fracture or knee surgery (excluding knee arthroscopy and meniscal surgery)

4. Active tumors or pregnancy.

• Minimum Eligible Age: 18 Years
• Eligible Sex: ALL
• Accepts Healthy Volunteers: No

Sponsors

Fondazione Policlinico Universitario Campus Bio-Medico

OTHER

Sponsor Role: lead

Responsible Party

Umile Giuseppe Longo

Professor

Responsibility Role: PRINCIPAL_INVESTIGATOR

Locations

Fondazione Policlinico Universitario Campus Bio-Medico

Rome, Italy, Italy

Site Status: RECRUITING

Countries

Italy

Central Contacts

Umile Giuseppe Longo, MD, MSc, PhD

Role: CONTACT

g.longo@policlinicocampus.it

+39 06225418816

References

Spallone G, Mancini L, Carnevale A, Campi S, Schena E, D'Hooghe P, Hirschmann MT, Papalia R, Longo UG. Joint modeling and marker set selection significantly influence functional biomechanics in end-stage knee osteoarthritis: evidence from the sit-to-stand task. Front Bioeng Biotechnol. 2025 Oct 13;13:1677244. doi: 10.3389/fbioe.2025.1677244. eCollection 2025.

Reference Type: DERIVED

PMID: 41158195 (View on PubMed)

Other Identifiers

PNRR-MCNT2-2023-12378237

Identifier Type: OTHER_GRANT

Identifier Source: secondary_id

179.24 CET2 cbm

Identifier Type: -

Identifier Source: org_study_id