A computer based technique that can predict heart disease risk by studying lipoprotein profiles has been developed by German researchers.
The technique devised by Katrin Hubner, at the Charite - Universitatsmedizin Berlin and the Max-Delbruck-Center Berlin and colleagues, made use of clinical data of lipoprotein profiles generated in collaboration with the University Medical Center Freiburg.
Touted as the "container ships" in our blood, the lipoproteins transport lipids (fats) such as cholesterol and triglycerides to various human tissues. Abnormalities in the amount of certain lipoprotein fractions are considered a major risk factor for atherosclerosis and CVD.
Doctors regularly monitor patents' lipoprotein profile, i.e. looking at subfractions of "bad" Low Density Lipoproteins (LDL) and "good" High Density Lipoproteins (HDL), which help in identifying the risk for CVD.
The decrease of LDL cholesterol is a principal target in cardiovascular preventive strategies. However, evidence has suggested that detailed evaluation of the lipoprotein profile needs elaborate and expensive work.
Thus, the researchers thought of designing a mathematical model to provide computer calculations of lipoprotein profiles which take into account the entire "fleet" of lipoproteins in blood plasma by simulating every single lipoprotein ("ship").
This would enable the researchers to study lipoprotein profiles in any desired detail. The model may also be broadly applied to infer relationships between a patient's lipoprotein profile and the underlying biochemical processes.
The researchers verified the calculations by comparing them with clinically measured lipoprotein profiles of healthy subjects and pathological cases of known lipid disorders and showed that more detailed lipoprotein profiles can reveal possibly clinically-relevant abnormalities in the lipid values which would remain undetected by evaluating only LDL and HDL.
Combining this technique with independent information on diet and genetic variations may help in devising patient-oriented diagnosis of molecular causes for observed abnormal lipoprotein profiles.
Details of the study are published in the open-access journal PLoS Computational Biology.