Accuracy of New Estimation Method for Low-Density Lipoprotein Cholesterol Levels

A new method for low-density lipoprotein cholesterol (LDL-C) estimation consisting of an adjustable triglyceride (TG)/very low-density lipoprotein cholesterol (VLDL-C) ratio based on TG and non-high-density lipoprotein cholesterol (HDL-C) levels provides greater accuracy than the fixed Friedewald estimation in nonfasting patients with low LDL-C and high triglycerides, according to a cross-sectional analysis published in Circulation.

The Very Large Database of Lipids (ClinicalTrials.gov identifier: NCT01698489) provided a total of 1,545,634 patients for this analysis. Investigators analyzed patients who had fasted ≥10 to 12 hours (n=959,153) and compared them with nonfasting patients (n=586,481). To measure LDL-C content (LDL-C_D) in this patient population, rapid ultracentrifugation was used. For the new estimation and the fixed Friedewald estimation, accuracy was defined as the LDL-C_D percentage falling within an estimated category of LDL-C (LDL-C_N or LDL-C_F) by specified clinical cut points.

Accuracy was significantly higher for the new estimation method vs the Friedewald estimation (range, 87%-94% vs 71%-93%, respectively; P ≤.001) in fasting and nonfasting samples. The accuracy of nonfasting LDL-C_N was also significantly greater in accuracy than that of LDL-C_F in samples of LDL-C <70 mg/dL (92% vs 71%, respectively; P <.001). In addition, a greater proportion of fasting (19%) and nonfasting (30%) patients undergoing the fixed estimation demonstrated differences of ≥10 mg/dL between LDL-C_F and LDL-C_D vs only 2% and 3% of fasting and nonfasting patients undergoing the novel estimation strategy.

As triglycerides increased, there were reductions in LDL-C <70 mg/dL accuracy, especially as it related to the Friedewald method (range, 37%-96%) compared with the newer estimation (range, 82%-94%). The accuracy of LDL-C_N <70 mg/dL demonstrated significant superiority over LDL-C_F in nonfasting patients with TGs of 200 to 399 mg/dL (82% vs 37%, respectively; P <.001).

The investigators did not collect data on patients’ exact fasting durations, representing a potential study limitation. Other data, including race, obesity levels, and insulin resistance, were also not collected and could have affected TG:VLDL-C ratios.

“[C]linicians and patients can place greater confidence in LDL-C results from non-fasting samples that are calculated using the novel method of LDL-C estimation,” the investigators noted, rather than the Friedewald equation, when making evidence-based decisions regarding lipid-lowering treatment.

Reference

Sathiyakumar V, Park J, Golozar A, et al. Fasting vs non-fasting and low-density lipoprotein- cholesterol accuracy. Circulation. 2018;137(1):10-19.