1. A method to estimate the ratio of concentration of serum triglycerides (TG) and cholesterol esters (CE) using infrared spectrophotometry was described. This method based on the difference in wave numbers of maximal absorption of C-O stretching of ester bonds of TG and CE. Due to smallness of this difference, the mixtures of the two compounds (TG and CE) produced apparently a single absorption band in the 1150-1175 cm^-1, although the change in the shape of this band was clearly seen by varying ratio of TG and CE. In order to express this visual change of the shape of the absorption band quantitatively, the difference of the transmittance between the site of the maximal absorption of this band and that of 1154 cm^-1 was measured and expressed as ΔT_1154-c. The site of 1154 cm^-1 was decided referring one of the absorption bands of polystylene film. A linear relationship was observed between the ratio of TG/CE and logarithm of ΔT_1154-c as follows:
log ΔT_1154-C=1.23-0.81×TG/CE
(when TG/CE 1.0)
2. In order to apply this procedure to serum lipid extracts, possible interference of other lipid conponents than TG and CE must be considered. Free fatty acid and unesterified cholesterol can be neglected in the absence of any absorption bands in 1150-1175 cm^-1 region. Mono- and di-glycerides are also negligible because of their minor quantity in the serum, although they have absorption bands in this region. As phospholipids have absorption bands in this region, they were removed by zeolite adsorption method prior to infrared analysis. A similar correlation between ΔT_1154-c and TG/CE was obtained as follows:
logΔT_1154-C=1.04-0.74×TG/CE
(when TG/CE 1.0)
TG was determined by Van Handel and Zilversmidtmethod and CE was determined by Zak method.
3. Although phospholipide contain ester group, the absorption in 1150-1175 cm^-1 region was weak, compared with P-O-C and P=0 absorptions. Amounts of phospholipids up to twice of CE gave no appreciable changes in ΔT_1154-C of CE and TG mixture. Therefore the TG/CE ratios can be estimated by the analysis of ΔT_1154-C of total lipids without removal of phospholipids, when high accuracy is not needed.
4. Finally ΔT_1154-c of total serum lipids in 149 diabetics were measured and compared to clinical features. It was found that, following diabetic treatment, serum TG decreased more rapidly than CE (P<0.05). No significant difference was detected between the mean values of ΔT_1154-C of those patients with or without vascular lesions.

View full abstract