Standard LDL Equations Yield Discordant Results

Ashley Lyles

February 08, 2022

Standard equations to calculate LDL cholesterol show clinically meaningful differences in results, potentially affecting decisions about treatment for patients with atherosclerosis, a retrospective study suggests.

Patients assessed using the Sampson and Friedewald equations were estimated to have lower LDL-C levels than those evaluated with the Martin/Hopkins equation, the researchers found. These inconsistencies were especially pronounced among patients with lower LDL-C levels and triglyceride levels of at least 150 mg/dL.

The study, led by Aparna Sajja, MD, a cardiology fellow, Medstar Georgetown University Hospital-Washington Hospital Center, was published in the February 15 issue of the Journal of the American College of Cardiology.

These trends could result in "undertreatment of those at increased risk of ischemic events," Sajja said. "The Martin/Hopkins equation increases identification of patients who may benefit from more intensive treatment."

Clinicians have generally used the Friedewald equation to estimate LDL-C, noted Sajja. "Reduced accuracy of the Friedewald equation in patients with low LDL-C and high triglyceride levels has prompted alternative equations to assess LDL-C, including the Martin/Hopkins and Sampson equations," she said in an email.

The study authors acknowledged that, given these data, providers "may be tempted to use direct LDL-C measurement instead."

However, they point out that direct homogenous LDL-C assays in the clinical setting are significantly different than preparative ultracentrifugation, and that the direct LDL-C measurement can be costly, time-consuming, and could introduce measurement bias, reduced analytic precision, and more overall errors than the Friedewald equation.

"Because the Martin/Hopkins equation is easy to implement and has no additional cost to patients, we believe it to be the most practical means to estimate LDL-C," they add.

The researchers set out to evaluate differences in LDL-C across the Sampson, Martin/Hopkins, and Friedewald equations, especially among patients with lower LDL-C levels and/or triglyceride levels of at least 150 mg/dL.

They retrospectively analyzed electronic health data on patients with atherosclerosis and triglyceride levels below 400 mg/dL.

The analysis ultimately consisted of 146,106 patients with atherosclerotic cardiovascular disease. Mean age of patients was 68 years, men accounted for 56% of the cohort, and 91% of the study group was White.

Patients were classified as discordant when LDL-C was at least 70 mg/dL for the comparator and below 70 mg/dL for the index equation. Concordance was reported for instances in which the LDL-C was below 70 mg/dL for both of the paired equations.

Discordance rates were 7% for the comparison of Sampson to Martin/Hopkins equations, 15% for the comparison of Friedewald and Martin/Hopkins equations, and 9% for the comparison of Friedewald and Sampson equations.

"Discordance increased at lower LDL cutpoints and in those with elevated TG levels," the authors note.

For patients with triglyceride levels of at least 150 mg/dL, the authors report a difference of 10 mg/dL or more in LDL-C in 23% of patients for the Sampson vs Martin/Hopkins comparisons, 27% of patients for the Friedewald vs Sampson comparisons, and 67% of patients for the Friedewald vs Martin/Hopkins comparisons.

Limitations of the study included a lack of data on the timing of the atherosclerotic cardiovascular disease event, the fact that the study group was predominately White, which limited generalizability of the data, and the fact that patients' fasting status was not available, the research group notes.

Other limitations were that comparisons were not drawn with direct measurement and that LDL-C was only assessed as a single measurement, they add.

"Future studies should validate our findings in other populations with greater racial and ethnic diversity. In addition, there is a need to examine downstream effects on adherence to cholesterol-lowering therapy and changes in treatment following a shift in the equation used to estimated LDL-C," Sajja concluded.

Lack of Consistency

"For the clinician, the decision about how to best use reported LDL-C levels in patients with ASCVD, particularly in those with TG ≥150 mg/dL, is further clouded by the lack of consistency in LDL-C estimation methods used in cardiovascular outcomes trials," write Kevin C. Maki, PhD, from Midwest Biomedical Research, Chicago, and colleagues in an accompanying editorial.

In an email, Maki noted that the Martin/Hopkins equation is best validated for cases in which the LDL-C level is low, particularly when below 100 mg/dL and especially at very low levels, below 50 mg/dL, and the Sampson equation is best validated for scenarios in which the triglyceride level is above 500 mg/dL, especially at levels above 1000 mg/dL.

"Both equations are likely better than Friedewald when the LDL-C is low, the triglyceride level is high, or both conditions are present. Incorporating non-HDL-C into the clinical decision-making process is helpful no matter which method is employed for estimating LDL-C," Maki noted.

Sajja and Maki report no relevant financial relationships.

J Am Coll Cardiol. 2022;79:530-541, 542-544. Abstract, Editorial

Ashley Lyles is an award-winning medical journalist. She is a graduate of New York University's Science, Health, and Environmental Reporting Program. Previously, she studied professional writing at Michigan State University, where she also took premedical classes. Her work has taken her to Honduras, Cambodia, France, and Ghana and has appeared in outlets like The New York Times Daily 360, PBS NewsHour, The Huffington Post, Undark, The Root, Psychology Today, Insider, and Tonic (Health by Vice), among other publications.

For more news, follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.