Predictors of Long-Term Evolution in Long COVID; 4-Year Follow-Up. (BioICOPER Follow-up Study)

Carotid-femoral pulse wave velocity will be assessed as the gold-standard measure of arterial stiffness, using the SphygmoCor® system. The study will evaluate the longitudinal change in cfPWV between baseline (BioICOPER inclusion) and the 4-year follow-up visit, in order to determine whether long COVID is associated with accelerated vascular aging.

Fatigue assessed using a validated fatigue scale (Fatigue Severity Scale, FSS). The Fatigue Severity Scale (FSS) is a method of evaluating the impact of fatigue on you. The FSS is a short questionnaire that requires the patient to rate their level of fatigue. The FSS questionnaire contains nine statements that rate the severity of their fatigue symptoms. the patient needs to read each statement and circle a number from 1 to 7, based on how accurately it reflects their condition during the previous week and the extent to which they agree or disagree that the statement applies to them.

A low value (e.g., 1) indicates strong disagreement with the statement, whereas a high value (e.g., 7) indicates strong agreement.

The goal of this outcome is to determine how fatigue severity evolves and its association with biomarkers and vascular parameters.

EQ-5D-5L is a health-related quality of life survey that uses five questions, each with five severity levels, to measure health status across the dimensions of mobility, self-care, usual activities, pain/discomfort, and anxiety/depression.

Health state profile: A person's answers create a five-number profile that describes their current health state.

Scoring: This profile can be used to calculate a single index score that represents health-related quality of life. This score ranges from 0 (death) to 1 (full health).

It assesses eight health concepts: physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotional, and mental health.

The SF-36 survey grades health by scoring responses on eight different health domains, which are then converted to a 0-100 scale where higher scores indicate better health. A score of 100 represents the best possible health in that category, while 0 represents the worst.

Evaluate cognitive impairment associated with long COVID using the Montreal Cognitive Assessment, which is a 10-minute screening test used to detect mild cognitive impairment in adults. It is a 30-point test that evaluates several cognitive domains, including memory, attention, language, visuospatial skills, executive functions, and orientation. A score of 26 or above is considered a normal result, while lower scores may indicate some level of cognitive dysfunction. The goal of this assesment is to identify determinants of neurocognitive persistence or recovery.

Objectively and subjectively measure physical activity levels using both WHO Physical Activity Questionnaire and 7-day pedometer readings. The WHO Physical Activity Questionnaire is a 16 question instrument that assesses physical activity participation in three domains (activity at work, travel to and from places, and recreational activities) as well as sedentary behavior. Physical activity may include planned activity such as walking, running, basketball, or other sports. Physical activity may also include other daily activities such as household chores, yard work, walking the dog, etc. The use of the pedometer only completes the questionnaire, hence both variables need to be meassured simultaneously and interpret them together in order to determine changes in physical activity.

Evaluate diet quality and its relation to metabolic and vascular outcomes. The PREDIMED questionnaire is a screening tool used in the PREDIMED study to assess adherence to the Mediterranean diet. The most common version is a 14-point questionnaire that asks about frequency of consumption of certain foods like olive oil, nuts, fruits, vegetables, and fish, and the limited intake of red meat, butter, and sugary drinks. Participants receive a score of 0 or 1 for each question, and the total score is used to classify their diet as low, moderate, or high in Mediterranean adherence. The goal of this outome is to assess dietary influence on systemic inflammation, endothelial function, and metabolic health.

The InBody 230 Bioimpedance Analysis is a non-invasive test that measures body composition by sending a mild electrical current through the body. It uses a technique called segmental multifrequency bioelectrical impedance to provide detailed information on body fat, skeletal muscle mass, and body water in the torso, arms, and legs. This analysis offers a more comprehensive picture of your health than a standard scale, providing data like Basal Metabolic Rate (BMR) and Body Mass Index (BMI). The goal od this outcome is to explore links between body composition and cardiovascular risk in long COVID.

This Outcome Measure reflects a single assessment in which multiple plasma biomarkers (e.g., ICAM-1, VCAM-1, Endothelin-1, TNF-α, IL-10, and additional cytokines) are quantified within the same measurement framework. All biomarkers are measured using ELISA and proteomic techniques and reported in consistent concentration units (e.g., pg/mL or ng/mL) as appropriate for each assay. Although multiple biomarkers are included, they represent components of one integrated biomarker panel evaluating endothelial and inflammatory activity. No aggregation of different units is performed; instead, each biomarker's concentration is reported separately under this single Outcome Measure to characterize molecular pathways involved in long-term vascular and inflammatory changes in long COVID.

Analysis of proteomic profiling of serum/plasma using advanced mass spectrometry to identify proteins related to fatigue, inflammation, and vascular aging to discover predictive biomarkers for persistent symptoms and disease progression.

In order to explore persistence of viral antigens and relationship with ongoing symptoms and vascular damage, plasma will be qualified using ELISA.

To develop and validate AI-based tools for risk stratification and personalized management using supervised learning (Random Forest, XGBoost, Neural Networks) trained on 70% of the cohort, validated on 30%. The machine learning models will combine clinical, vascular, and proteomic variables to predict disease trajectory.