Laboratory Medicine ›› 2026, Vol. 41 ›› Issue (6): 521-529.DOI: 10.3969/j.issn.1673-8640.2026.06.002

Previous Articles     Next Articles

Consistency of the estimation results of LDL-C based on formula method and the direct determination results of LDL-C by homogeneous enzymatic method in patients with ASCVD

WANG Xiaoling1, SHI Qingyu2, ZHOU Jiaye2, WANG Kouqiong2, ZHU Jing2(), SHEN Yifeng2, SHAO Wenqi2, PAN Baishen2, WANG Beili2, GUO Wei2   

  1. 1 Department of Clinical Laboratory,Zhongshan Hospital(Xiamen),Fudan UniversityXiamen 361006,Fujian, China
    2 Department of Clinical LaboratoryZhongshan Hospital,Fudan UniversityShanghai 200032, China
  • Received:2025-12-23 Revised:2026-02-24 Online:2026-06-30 Published:2026-07-01

Abstract:

Objective To evaluate the correlation between the results of low-density lipoprotein cholesterol(LDL-C) calculated using different estimation formulas and the homogeneous enzymatic method(direct method) in patients with atherosclerotic cardiovascular disease(ASCVD). Methods A total of 101 144 newly diagnosed ASCVD patients from Zhongshan Hospital of Fudan University from January 1,2020 to December 31,2022 were enrolled. The total cholesterol(TC),triglyceride(TG),high-density lipoprotein cholesterol(HDL-C) and LDL-C levels were determined using the homogeneous enzymatic method,and non-high-density lipoprotein cholesterol(non-HDL-C) was calculated. LDL-C results were calculated using the Friedewald formula,Martin-Hopkins formula,Martin-Hopkins extended formula and Sampson formula. All the patients were classified based on the TG determination results:≤1.70 mmol·L-1 group,>1.70-≤4.52 mmol·L-1 group,>4.52-≤9.04 mmol·L-1 group and >9.04 mmol·L-1 group. The determination coefficient(r2),root mean square error(RMSE) and mean absolute difference(MAD),as well as the proportion of patients whose classification changed at the LDL-C treatment target values(1.4,1.8 and 2.6 mmol·L-1),were used to evaluate the differences in LDL-C estimation results between the formula method and the direct method. Results Among the 101 144 ASCVD patients,the Sampson formula,Martin-Hopkins formula,Martin-Hopkins extended formula and Friedewald formula were all correlated with the direct method,with r2 values of 0.956 6,0.948 1,0.949 3 and 0.904 5,respectively,and RMSE values of 0.29,0.27,0.27 and 0.44 mmol·L-1,respectively. As the TG level of patients increased,the r2 values of each estimation formula and the direct method decreased,and the RMSE and MAD increased. According to different LDL-C treatment target values,the classification of ASCVD patients based on the direct method LDL-C results was used as the standard,and the proportion of patients whose classification changed was the lowest for the Martin-Hopkins formula and Martin-Hopkins extended formula(23.10% and 23.11%),and the highest for the Friedewald formula(34.27%). The 1.4‰ of patients had a classification change amplitude exceeding 3 LDL-C treatment target values. Conclusions In ASCVD patients,the correlation between LDL-C estimation results of the Sampson formula and LDL-C determination results of the direct method is the highest,followed by the Martin-Hopkins formula and Martin-Hopkins extended formula,and the lowest is the Friedewald formula. As the TG level increases,the correlation between estimation formula and direct method results decreases. Therefore,when making treatment decisions based on LDL-C treatment target values in clinical practice,the differences in LDL-C estimation results among different formulas should be fully considered.

Key words: Low-density lipoprotein cholesterol, Friedewald formula, Martin-Hopkins formula, Martin-Hopkins extended formula, Sampson formula, Atherosclerotic cardiovascular disease

CLC Number: