A Comparison between Eight Formulas for the Estimation of Plasma Low-Density Lipoprotein Cholesterol in the Saudi Arabian Population

Abstract

CONTEXT: Low-density lipoprotein cholesterol (LDL-C) is the classical target in cardiovascular (CV) disease management and is usually estimated by Friedewald's formula. However, this formula may over- or underestimate LDL-C levels. AIMS: Our aim is to compare eight LDL-C-estimating formulas to the direct LDL-C measurement and validate their use in the Saudi population. SETTINGS AND DESIGN: This was a retrospective study. SUBJECTS AND METHODS: A blood sample of fasting 669 Saudi subjects was tested for lipid profiles in King Abdulaziz University Hospital Laboratory, from which directly measured LDL-C was obtained. LDL-C was then estimated from eight different formulas: Friedewald's, Cordova's, Hata's, Puavilai's, Chen's, Ahmadi's, Hattori's, and Vujovic's, which were all compared to direct LDL-C. STATISTICAL ANALYSIS USED: Mean and standard deviation, paired t-test, and Pearson's correlation were used for statistical analysis. RESULTS: The mean differences between the direct LDL-C and Hattori and Chen's LDL-C were 0.03 and 0.08 mmol/L, respectively; P < 0.001, while the mean difference observed with Hata, Friedewald, Puavilai, and Vujovic's formulas were higher in comparison: 0.15, 0.24, 0.29, and 0.33 mmol/L, respectively, P < 0.001. Ahmadi and Cordova's LDL-C were estimated to be 0.60 and 0.64 mmol/L more than direct LDL-C levels, respectively, which showed the highest discordance with direct LDL-C, P < 0.001. At a triglyceride (TG) level of <4.5 mmol/L, Hattori also had the best agreement with direct LDL-C, with a mean difference of 0.04 mmol/L, and with TG >4.5 mmol/L, their mean difference was 0.21 mmol/L. All estimated LDL-C strongly correlated with direct LDL-C, except for Ahmadi's. CONCLUSIONS: Hattori's LDL-C had the best agreement with the direct LDL-C, and across all TG levels. However, we recommend directly measuring LDL-C in those with high CV risk.