Assessing the Accuracy of LDL-Cholesterol Estimation Formulas in the Tunisian Population

International guidelines for the management of coronary artery disease highlight the importance of LDL-C(low density lipoprotein-cholesterol)measurements in assessing cardiovascular risk and monitoring patients. In clinical practice,its direct measurement is complex and expensive. Therefore,calculating LDL-C using several formulas is simpler and less costly.

T o evaluate the accuracy of LDL-C estimation formulas of Friedewald,Sampson and Martin/Hopkins compared to its direct measurement in Tunisian patients.

A cross-sectional study was conducted including patients following in the Endocrinology Department of Farhat Hached University Hospital of Sousse.Clinical data, metabolic parameters,and lipid profiles including Direct LDL-cholesterol measurement were analyzed. LDL-C was estimated using Friedewald,Martin/Hopkins, and Sampson formulas.

A total of 148 patients were included in the study.The male-to-female ratio was 0.51.The mean age of the patients was 58 ± 14.8.The majority of our patients were diabetic (84.5%). Most patients had uncontrolled diabetes (81.8%).Type 2 diabetes was the predominant form (85.6%).The three formulas,Friedewald,Sampson,and Martin/Hopkins provided comparable fasting correlation values with measured LDL-C(r=0.52, 0.55, and 0.47, respectively). One hour after meal intake, LDL-C estimation using the three formulas was similar, showing a strong correlation with measured LDL-C (r > 0.7 for all three formulas). At two hours postprandial, the three formulas showed a moderate correlation with the measured LDL-C value.

In this study,all three formulas Friedewald, Sampson, and Martin/Hopkins demonstrated acceptable accuracy in estimating LDL-C compared with direct measurement in Tunisian patients. Overall, these findings support the usefulness of formula-based LDL-C estimation as a practical and cost-effective alternative in routine clinical practice.