OBJECTIVE

To compare the Friedewald and modified Friedewald formulae with direct homogeneous assay for serum lowdensity lipoprotein cholesterol (LDL-C) levels estimation.

STUDY DESIGN

Cross-sectional study.

PLACE AND DURATION OF STUDY

Armed Forces Institute of Pathology, Rawalpindi, from June to December 2011.

METHODOLOGY

Healthy subjects of either gender, from Rawalpindi, aged 18-75 years were included by consecutive sampling. Patients with diabetes mellitus, chronic liver disease, chronic kidney disease, those taking lipid lowering drugs and samples with triglyceride (TG) > 4.52 mmol/l were excluded from the study. Total cholesterol, high-density lipoprotein cholesterol, TG and LDL-C were measured on Hitachi 912 chemistry analyzer (Roche). LDL-C levels were also calculated by Friedewald formula (FF) and Vujovic modified formula (VMF). Paired sample t-test and scatter plots were used for statistical analysis.

RESULTS

Although both calculated methods showed good correlation with direct assay (r > 0.93) in 300 subjects, but the difference was statistically significant. The ffLDL-C were 0.12 ± 31 mmol/l (p < 0.001) lower and vmfLDL-C were 0.11 ± 26 mmol/l (p < 0.001) higher than dLDL-C. The difference was not significant between ffLDL-C and dLDL-C at TG levels < 1.70 mmol/l (p = 0.58) and between vmfLDL-C and dLDL-C at TG levels 2.26 - 4.52 mmol/l (p = 0.38). At all other TG levels, the difference between LDL-C calculated by both formulas and dLDL-C was statistically significant (p < 0.001). As compared to direct assay, 11% and 14% subjects were classified in wrong National Cholesterol Education Programm's cardiac risk categories by FF and VMF respectively.

CONCLUSION

LDL-C should be measured by direct homogeneous assay in routine clinical laboratories, as the calculated methods did not have a uniform performance for LDL-C estimation at different TG levels.