Cardiology Letters 2024, 33(1):13-26

Comparison of three methods for LDL-cholesterol calculation for cardiovascular disease risk categorisation in three distinct patient populations

Rudolf Gaško¹, Roman Alberty², Martin Studenčan^{3, 4}, Dáša Albertyová⁵

¹ Oddelenie klinickej biochémie a hematológie, Psychiatrická nemocnica Michalovce, n.o., Michalovce

² Katedra biológie a ekológie, Fakulta prírodných vied, Univerzita Mateja Bela, Banská Bystrica

³ II. Kardiologická klinika UPJŠ a VÚSCH, a. s., Košice

⁴ Centrum preventívnej a športovej kardiológie, VÚSCH, a. s., Košice

⁵ Centrálny laboratórný komplex - PKB, Fakultná nemocnica s poliklinikou F. D. Roosevelta, Banská Bystrica, Slovenská republika

Background: Limitations of the Friedewald equation for low-density lipoprotein cholesterol (F-LDLC) calculation, which is used by ≈ 33% of laboratories in Slovakia and 66% of laboratories in Europe, led to the Martin-Hopkins (M-LDLC), extended Martin-Hopkins (extM-LDLC) and Sampson/National Institutes of Health (S-LDLC) equations. We studied these newer calculations of LDLC for correlation and discordance for stratification into the 2019 ESC/EAS Guidelines' cardiovascular disease (CVD) risk categories.

Methods: We performed analyses on lipid profiles from 3 populations: 873 healthy children from the cross-sectional Slovak Lipid Community Study (population SLCS), records of a 1949 patients of a hospital biochemistry laboratory from from a specialized cardio centre (population VÚSCH), and records of 9113 patients of a hospital biochemistry laboratory from a large general hospital (population FN FDR). Patients were categorized as concordant if LDL-C was <1.8 mmol/l with each pairwise comparison of equations, and as discordant if LDL-C was <1.8 mmol/l for the index equation and ≥1.8 mmol/l for the comparator. Separate comparisons were made for subpopulations with TAG values ≥1.7 mmol/l.

Results: There was very strong correlation among the 3 calculated LDLC in all 3 populations. The extM-LDLC equation consistently estimated higher LDL-C values than the Friedewald and Sampson equations. Discordance rates for for the F-LDLC vs extM-LDLC for population SLCS were 0.9% and 3.7% for subpopulation with TAG values ≥1.7 mmol/l, respectively. Discordance rates for for the F-LDLC vs extM-LDLC for population VÚSCH were 3.4% and 8.9% for subpopulation with TAG values ≥1.7 mmol/l, respectively. Discordance rates for for the F-LDLC vs extMLDLC for population FN FDR were 4.1% and 8.2% for subpopulation with TAG values ≥1.7 mmol/l, respectively.

Conclusions: Switching from F-LDLC to extM-LDLC or S-LDLC can reclassify up to ≈ 3.4% or 8.9% of patients, respectively, to a higher CVD risk category. The difference between F-LDLC and extM-LDLC or S-LDLC is greater with higher TG, and with lower LDLC. Reliance on the Friedewald equation may result in the underestimation and undertreatment of LDL-C in those at increased risk. We recommend, in line with some national/international guidelines, that clinical laboratories switch to counting and reporting results to extM-LDLC.

Keywords: LDL-cholesterol; Friedewald formula; extended Martin-Hopkins formula; Sampson, NIH formula

Published: January 1, 2024  Show citation