Evaluation of equations for calculating LDL-cholesterol in patients with concomitant hypertriglyceridemia and low LDL-cholesterol: is there a preferred equation? / Hypertriglycéridémie et cholestérol-LDL bas : quelle équation estime mieux le cholestérol-LDL ?

It is critical to reliably estimate Low Density Lipoprotein Cholesterol (LDL-C) in patients with concomitant hypertriglyceridemia and low LDL-C. We retrospectively compared the performances of the Friedewald (LDL-F), Martin-Hopkins (LDL-MH) and Sampson (LDL-SA) equations against a direct homogeneous LDL-C assay (dLDL-C) on observations presenting mild hypertriglyceridemia (triglycerides between 1.69 and 3.9 mmol/L) and low LDL-C (< 2.58 mmol/L). Observations were stratified according to their LDL-C. Agreement of the equations with dLDL-C was assessed using Intraclass Correlation Coefficients (ICC) with an agreement cut-off of 0.9, and analysis of Bland-Altman plots. Independently of the LDL-C stratum evaluated, the three equations failed to meet the 0.9 ICC cut-off, although their agreement with dLDL-C improves as LDL-C increases. Analysis of Bland-Altman plots shows a downwards discordance of LDL-F with dLDL-C, and an upwards discordance of LDL-MH and LDL-SA with direct LDL-C. LDL-MH resulted in the least observations outside the Bland-Altman limits of agreement. While no equation can be deemed satisfactory enough to replace direct assays in patients with low LDL-C and concomitant hypertriglyceridemia, LDL-MH seems to perform better than the other equations in estimating LDL-C in these patients.

Prospective Validation of a Machine Learning Model for Low-Density Lipoprotein Cholesterol Estimation.

OBJECTIVE: We aim to prospectively validate a previously developed machine learning algorithm for low-density lipoprotein cholesterol (LDL-C) estimation. METHODS: We retrospectively and prospectively evaluated a machine learning algorithm based on k-nearest neighbors (KNN) according to age, sex, healthcare setting, and triglyceridemia against a direct LDL-C assay. The agreement of low-density lipoprotein-k-nearest neighbors (LDL-KNN) with the direct measurement was assessed using intraclass correlation coefficient (ICC). RESULTS: The analysis comprised 31,853 retrospective and 6599 prospective observations, with a mean age of 54.2 ±â€…17.2 years. LDL-KNN exhibited an ICC greater than 0.9 independently of age, sex, and disease status. LDL-KNN was in satisfactory agreement with direct LDL-C in observations with normal triglyceridemia and mild hypertriglyceridemia but displayed an ICC slightly below 0.9 in severely hypertriglyceridemic patients and lower in very low LDL-C observations. CONCLUSION: LDL-KNN performs robustly across ages, genders, healthcare settings, and triglyceridemia. Further algorithm development is needed for very low LDL-C observations.

A Tale of Two Approaches.

OBJECTIVES: To summarize and assess the literature on the performances of methods beyond the Friedewald formula (FF) used in routine practice to determine low-density lipoprotein cholesterol (LDL-C). METHODS: A literature review was performed by searching the PubMed database. Many peer-reviewed articles were assessed. RESULTS: The examined methods included direct homogeneous LDL-C assays, the FF, mathematical equations derived from the FF, the Martin-Hopkins equation (MHE), and the Sampson equation. Direct homogeneous assays perform inconsistently across manufacturers and disease status, whereas most FF-derived methods exhibit variable levels of performance across populations. The MHE consistently outperforms the FF but cannot be applied in the setting of severe hypertriglyceridemia. The Sampson equation shows promise against both the FF and MHE, especially in severe hypertriglyceridemia, but data are still limited on its validation in various settings, including disease and therapeutic states. CONCLUSIONS: There is still no consensus on a universal best method to estimate LDL-C in routine practice. Further studies are needed to assess the performance of the Sampson equation.

Is Machine Learning-derived Low-Density Lipoprotein Cholesterol estimation more reliable than standard closed form equations? Insights from a laboratory database by comparison with a direct homogeneous assay.

BACKGROUND: There is no consensus on the best method to estimate Low Density Lipoprotein-Cholesterol (LDL-C) in routine laboratories. METHODS: We conducted a retrospective study to compare the performances of a Machine Learning (ML) algorithm using the K-Nearest Neighbors (LDL-KNN) method with that of the Friedewald formula (LDL-F), the Martin-Hopkins equation (LDL-NF), the de Cordova equation (LDL-CO) and the Sampson equation (LDL-SA) against direct homogeneous LDL-C assay (LDL-D) in patients who presented to the Laboratories of Hôtel Dieu de France university hospital in Beirut, Lebanon, from September 2017 to July 2020. Agreements between methods were analyzed using Intraclass Correlation Coefficients (ICC) and the Bland-Altman method of agreement. RESULTS: 31,922 observations from 19,279 subjects were included, with a mean age of 52 ± 18 years and 10,075 (52.3%) females. All methods except LDL-F and LDL-CO exhibited an overall ICC beyond the 0.9 cut-off. LDL-SA, LDL-NF and LDL-KNN were less susceptible to triglyceridemia than LDL-F and LDL-CO, with LDL-KNN resulting in the lesser fraction of points beyond the Bland-Altman limits of agreement. CONCLUSION: An ML algorithm using LDL-KNN is promising for the estimation of LDL-C as it agrees better with LDL-D than closed form equations, especially in mild and severe hypertriglyceridemia.