The method of choice is based on the analysis of signals in the 1H-NMR spectrum that are related to the lipoproteins. Differences in lipoprotein composition, size and density translate into respective signal line shape differences, which can be used to extract information on lipoprotein main and subclasses (see figure 1).
A regression model had to be developed, using a training data set which consists of:
- Lipoprotein analytes from total plasma and ultra-centrifugation-based main and subclasses
- 1H-NMR spectra of the same sample set
Once the regression model was established, a prediction algorithm calculates the lipoprotein analytes directly from the 1H-NMR spectra of new plasma or serum samples, without further need for ultra-centrifugation.
Using this 1H-NMR approach, information could be extracted about lipoprotein related information on:
- Plasma and serum
- The main VLDL, IDL, LDL and HDL classes
- Six VLDL subclasses VLDL-1 to VLDL-6 (sorted according to increasing density and decreasing size, respectively)
- Six LDL sub-classes LDL-1 to LDL-6
- Four HDL-subclasses HDL-1 to HDL-4
Information consists of concentrations of lipids, i.e.cholesterol, free cholesterol, phospholipids, triglycerides, concentrations of apolipoproteins Apo-A1, ApoA2 and Apo-B and the LDL particle numbers. Table 1 shows all parameterscalculated by the 1H-NMR lipoprotein analysis. Figure 1 shows the front page of the automatic report (B.I.-LISA).