One measurement of serum total cholesterol is enough to predict future levels in healthy postmenopausal women

Abstract

purpose: The purpose of this study was to investigate how well a single or double measurement of serum total cholesterol represents the spontaneous, future level in a particular person. subjects and methods: The spontaneous fluctuations in serum total cholesterol levels in 169 healthy early postmenopausal women were followed during the course of 12 years. Serum total cholesterol was measured enzymatically. results: The initial measurement and the long-term level of serum total cholesterol were highly related (p <0.0001). The long-term level (calculated for each woman as the area under the curve of serum total cholesterol versus time) was not statistically significantly different from the initial level (mean difference: 0.036 ± 0.046 mmol/L [mean ± SEM], NS). The initial serum total cholesterol level was then used to classify each woman into a high or a low cholesterol group, according to the current recommendations. The predictive value of an initial total cholesterol value in the high level (greater than or equal to 6.2 mmol/L) group was 84%, when compared with the long-term level. The predictive value of an initial total cholesterol level below 6.2 mmol/L was 80%. No improvement in these parameters was found when the average of the initial two (or three, when the difference exceeded 0.9 mmol/L) measurements were used as the baseline value. The fluctuations in serum total cholesterol levels were mainly due to short-term variations. conclusion: For screening purposes, one measurement of serum total cholesterol in a woman gives a good estimate of the long-term level. The current data indicate that repeated measurements of serum total cholesterol do not improve the predictability of future cholesterol levels. The data also suggest that, at least in postmenopausal women with an elevated level of serum total cholesterol, one should proceed immediately to lipoprotein analysis for further risk assessment.