Computational apolipoprotein estimates ‘reliable’ for cardiovascular risk
medwireNews: Computational estimates of apolipoprotein B (apoB) and apolipoprotein A1 (apoA1) can be used as reliable markers for atherosclerosis, report researchers.
Estimated apoB and apoA1 values predicted carotid artery intima-media thickness (IMT) just as well as levels that were directly measured in participants from the Young Finns study, say Olli Raitakari (University of Turku, Finland) and colleagues.
Furthermore, the apoB:apoA1 ratio was the strongest predictor for increased carotid IMT of all lipoprotein variables, including low-density lipoprotein (LDL) cholesterol, regardless of whether the estimated or measured ratio levels were used.
The team's computational approach was based on neural network regression models that estimated apoB and apoA1 values from participants' total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglyceride levels.
"Although the introduction of apoB and apoA1 in clinical practice may improve risk prediction, this is unlikely to happen rapidly as physicians have used traditional lipid variables to diagnose and treat dyslipoproteinemias for decades and because many laboratories would not be able to perform these assays," explain the researchers. "An alternative way to improve the predictive value of traditional lipid variables is to use these variables to estimate the quantity of circulating apoB and apoA1 levels."
As reported in Atherosclerosis, the team's analysis of data available for 2166 individuals, including directly measured apoB and apoA1 levels and carotid IMT data, showed that the values for the estimated apolipoprotein levels were comparable with those that were directly measured, with a mean difference of 0.005 g/L for apoB and of -0.085 g/L for apoA1.
The strongest correlation between the estimated and measured apolipoprotein levels was for the apoB values, at a correlation coefficient of 0.98, followed by apoB/apoA1-ratio (0.97), and then apoA1 (0.95).
Further analysis showed that the highest age- and gender-adjusted correlation with carotid IMT was observed for the apoB:apoA1-ratio, at a correlation coefficient of approximately 0.18 for both estimated and direct measures, followed by that for LDL:HDL cholesterol ratio (0.16), and estimated and measured levels of apoB (both approximately 0.15).
In multivariate analysis, when the directly measured apoB:apoA1 ratio, the LDL:HDL cholesterol ratio, total cholesterol, HDL cholesterol, and triglycerides were included in the model, only the apoB:apoA1 ratio emerged as an independent predictor for carotid IMT. And when the directly measured apoB:apoA1 ratio was replaced by the estimated ratio, the only independent predictors that emerged were estimated apoB:apoA1 ratio and triglycerides.
"Increasing amounts of data indicate that apoB and apoA1 should be measured and introduced into guidelines to complement lipoprotein lipids in risk assessment and drug therapy monitoring," say the researchers.
Although this is unlikely to happen rapidly, the inclusion of apoB and apoA1 estimates into laboratory reports along with lipoprotein lipids would likely hasten this transition, suggests the team.
medwireNews (www.medwire-news.md) is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2012
By Sally Robertson, medwireNews Reporter