Clinicians rely heavily on low-density lipoprotein cholesterol (LDL-C) – the main target for lipid-lowering therapies and preventing cardiovascular disease – and in randomised control trials and meta-analyses, statin and non-statin therapy for this has proved to reduce the relative risk of atherosclerotic cardiovascular disease by 20% to 25%.
Writing in the SA Medical Journal, T Pillay, HM Rossouw, N Steyn and A Carelse say accurate LDL-C to guide therapy is crucial, and for this reason, the National Health Laboratory Service (NHLS) chemical expert committee has introduced newer, more accurate equations to replace the older, less reliable Friedewald method, generally used to calculate LDL-C.
If LDL-C is a primary target in dyslipidaemia, they point out, accurate, precise and affordable assessment is vital.
ApoB100 is a test that is more specific and precise, however, only private sector laboratories offer the apoB100 assay.
LDL-C is generally calculated (c-LDL-C) using the Friedewald equation, while some labs routinely measure LDL-C (d-LDL-C) with a direct assay.
Calculating LDL-C is a simple cost-free way to determine routine LDL-C by subtracting high-density lipoprotein (HDL-C) and triglycerides (TG) from total cholesterol when TG is <4.5 moll/L.
Yet the Friedewald equation is inaccurate when TG is >4.5 moll/L or with new lipid-lowering therapies that lower the LDL-C levels to <1.8 mmol/L, and performed poorly in large varied South African (cohorts, including children and patients with diabetes, particularly in samples with hypertriglyceridaemia (TG >4.5 mmol/L.
Thus far, this has been the only equation used in SA laboratories for routine calculation of LDL-C with TG <4.5 mmol/L. Hypertriglyceridaemic samples are referred for direct assay, and this increases turnaround time and has logistical obstacles.
In SA, using big data analysis, differences in instrument performance are also of concern.
The Friedewald equation rated badly compared with the direct LDL-C assay in an outpatient cohort, misclassifying 12% of all patients across different LDL-C cut-offs.
Dissatisfaction with the performance of the equation has led to the development of newer equations, two of which have been found to be robust and accurate: the Martin-Hopkins equation and the Sampson-NIH equation samples.
These both performed better than the Friedewald equation in recent SA cohorts, including diabetics and children.
The newer equation published by Sampson, et al improves accuracy with low LDL-C and hypertriglyceridaemia samples, as these are problematic with the Friedewald equation.
The NHLS committee has implemented the Sampson-NIH2 equation to modernise practice and report more accurate patient results because the superiority of the newer equations is undeniable. They suggest private sector labs and clinicians adopt one of the newer options.
The Martin-Hopkins equation shows excellent comparability with a direct assay, as demonstrated extensively internationally and locally.
The Sampson-NIH and extended Martin-Hopkins demonstrate favourable comparability with direct LDL-C with triglyceride levels up to 9 mmol/L, but this still warrants further investigation in SA cohorts.
Reducing the need for direct LDL-C assays will reduce the overall cost of a lipid profile and lower lab expenditure.
The Friedewald equation is no longer fit for purpose, and laboratories should switch to one of the newer equations.
Pillay is with the Department of Chemical Pathology, Faculty of Medicine, University of Pretoria; H M Rossouw is with Ampath Laboratories, Pretoria; N Steyn and A Carelse are with the Department of Chemical Pathology, Faculty of Medicine, University of Pretoria and National Health Laboratory Service, Steve Biko Academic Hospital, Pretoria.
See more from MedicalBrief archives:
More focus on lipid profiles urged for KZN’s many diabetics – SA study
LDL cholesterol levels must be lowered in high CVD risk patients
More aggressive statin therapy sometimes needed to fight ‘bad’ cholesterol