Validating the NIH LDL-C equation for provincial implementation in Alberta

V Higgins; L Garcia; J L Gifford; N Volodko; D R Beriault; M L Parker; M P Estey; D T Proctor; O Z Ismail

doi:10.1016/j.clinbiochem.2023.110678

Validating the NIH LDL-C equation for provincial implementation in Alberta

Clin Biochem. 2023 Nov:121-122:110678. doi: 10.1016/j.clinbiochem.2023.110678. Epub 2023 Oct 21.

Authors

V Higgins¹, L Garcia², J L Gifford³, N Volodko⁴, D R Beriault⁵, M L Parker⁴, M P Estey⁴, D T Proctor⁴, O Z Ismail⁴

Affiliations

¹ DynaLIFE Medical Labs, Edmonton, AB, Canada; Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, AB, Canada. Electronic address: victoria.higgins@dynalife.ca.
² Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, AB, Canada.
³ DynaLIFE Medical Labs, Calgary, AB, Canada; Department of Pathology and Laboratory Medicine, University of Calgary, Calgary, AB, Canada.
⁴ DynaLIFE Medical Labs, Edmonton, AB, Canada; Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, AB, Canada.
⁵ Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON, Canada; Department of Laboratory Medicine and Pathobiology, St Michael's Hospital, Toronto, ON, Canada.

PMID: 37866698
DOI: 10.1016/j.clinbiochem.2023.110678

Abstract

Background: LDL-C, a cardiovascular disease risk assessment biomarker, is commonly calculated using the Friedewald equation. The NIH equation overcomes several limitations of the Friedewald equation. Consistent with the Canadian Society of Clinical Chemists (CSCC) lipid reporting recommendations, we assessed the NIH LDL-C equation in Alberta prior to its provincial implementation.

Methods: 1-year (01/01/2021-12/31/2021) of lipid results (n = 1,486,584 after data cleaning) were obtained from five analytical instrument groups used across Alberta. Analyses were performed on all data and after separating by age, analytical instrument group, and fasting status. The correlation between Friedewald- and NIH-calculated LDL-C and between Friedewald- and NIH-calculated LDL-C difference and each lipid parameter, was determined. The frequency of unreportable/inaccurate LDL-C results was compared between the two equations. The concordance between the two equations and with non-HDL-C was determined at LDL-C thresholds. Lastly, LDL-C calculated by Friedewald, NIH, and Martin-Hopkins equations was compared to density-gradient ultracentrifugation.

Results: Friedewald- and NIH-calculated LDL-C exhibit the strongest correlation when triglycerides ≤ 4.52 mmol/L. The difference between Friedewald- and NIH-calculated LDL-C increases with decreasing LDL-C concentration. The NIH equation yields fewer inaccurate results (0.35 % vs. 22.0 %). The percent agreement between equations was > 96 % at all LDL-C thresholds, suggesting most patients will not require treatment changes. NIH-calculated LDL-C exhibited better agreement with non-HDL-C when triglycerides ≤ 9.04 mmol/L and better correlated with LDL-C measured by ultracentrifugation (r² = 0.926 vs. 0.775 (Friedewald) and 0.863 (Martin-Hopkins)). Results were consistent across age, analytical instrument group, and fasting status.

Conclusions: Our findings demonstrate the benefits of implementing the NIH equation across Alberta.

Keywords: Cardiovascular disease; LDL-C; Lipid panel; Triglycerides.

MeSH terms

Alberta
Biomarkers
Cholesterol, LDL* / analysis
Humans
Triglycerides
Ultracentrifugation

Substances

Cholesterol, LDL
Triglycerides
Biomarkers