Lipemic Plasma Identified Blood Donors: Triglyceride Variability and Exploratory Machine Learning Analysis

Area of Science:

• Cardiovascular disease research
• Metabolic health monitoring
• Blood donor screening

Background:

• Early detection of cardiometabolic issues is key to preventing cardiovascular disease.
• Habitual blood donors are an underutilized resource for metabolic monitoring.
• Lipemic plasma in donors may indicate underlying metabolic irregularities.

Purpose of the Study:

• To investigate metabolic variability in blood donors with lipemic plasma.
• To assess lipid profiles and cardiovascular risk in this selected subgroup.
• To explore the potential of donor data for metabolic surveillance.

Main Methods:

• Lipid profiling and cardiovascular risk assessment were performed on 160 blood donors with lipemic plasma.
• Multivariable and machine-learning analyses were conducted on 90 donors with complete data.
• Random Forest classification was used to predict elevated triglyceride levels.

Main Results:

• Significant triglyceride variability was observed, with higher and more dispersed values in males.
• Triglycerides correlated with Body Mass Index (BMI) and composite cardiovascular risk metrics.
• Age was the primary determinant of the 10-year cardiovascular risk score.
• A Random Forest model achieved an Area Under the Curve (AUC) of 0.86 for predicting elevated triglycerides (exploratory).

Conclusions:

• Clinically relevant hypertriglyceridemia was common in blood donors with lipemic plasma.
• Routine donor data may offer avenues for targeted metabolic monitoring.
• Findings are specific to this subgroup and cannot be generalized; larger studies are needed.