A-215 Calculated LDL-cholesterol: Comparability of the extended Martin-Hopkins,Sampson-NIH, Friedewald and four other equations in South African patients

Abstract

Abstract Background Ultracentrifugation for LDL-C (low-density lipoprotein cholesterol) measurement is the gold standard, but it is unsuitable for routine use. Direct LDL-C assays and predictive equations such as the Friedewald equation are thus used as alternatives. The Friedewald equation is known to be error-prone in hypertriglyceridaemic patients, as well as in LDL-C levels < 1.8 mmol/L (70 mg/dL) which has become increasingly attainable with the use of pro-protein convertase subtilisin/kexin type 9 inhibitors. This study evaluates the comparability of the Friedewald, extended Martin/Hopkins, Sampson/NIH and four other equations to a direct LDL-C assay to establish a suitable alternative to the Friedewald equation. Methods We analysed 44 194 lipid profiles from a mixed South African population. The Friedewald, extended Martin/Hopkins and Sampson/NIH LDL-C and four other predictive equations (Vujovic, Puavilai, Delong, Anandaraja) were compared to the Beckman Coulter direct LDL-C assay. Non-parametric statistics were used to analyse the data. Results The extended Martin/Hopkins equation displayed the best correlation with direct LDL-C, having the most values fall within the desirable bias (59%) and total allowable error (89%) specifications. In patients with TG values between 4.5 mmol (400 mg/dL) and 9.0 mmol/L (800 mg/dL), the extended Martin/Hopkins equation was more likely to overestimate LDL-C levels, showing a positive median bias of 7.3%, while Friedewald (−21.2%) and Sampson/NIH (−9.8%) showed larger negative median biases as compared to the direct LDL-C assay. The direct LDL-C assay classified 13.9% of patients in the low LDL-C (1.0–1.8 mmol/L) (39–70 mg/dL) category, in comparison to the Martin/Hopkins equation (13.4%), the Sampson equation (14.6%) and the Friedewald equation (16.0%). The extended Martin/Hopkins equation performed better than Sampson/NIH equation in the cohort. Conclusion We suggest the extended Martin/Hopkins equation as an alternative to Friedewald’s equation and direct LDL-C assay. Careful consideration is advised as the choice of analyser platform for the lipid profile used in the equations may influence their performance.