| ObjectiveThe importance of precisely quantifying low-density lipoprotein cholesterol（LDL-C）has become more pronounced over the years,with the rise of metabolic syndrome in the population and the reduction in LDL-C treatment goals.Clinically,LDL-C testing capabilities are increasing,and the most commonly used assays are the homogeneous direct method（also known as the "direct method"）and the indirect formula method.According to the Clinical Test Center of the National Health Commission of China,the analysis of data related to the inter-room quality evaluation of LDL-C testing in November 2021 showed that a total of 3316 laboratories participated in the inter-room quality evaluation,of which 57 laboratories used the Friedewald formula to detect LDL-C,accounting for 1.7%,and2903 laboratories used the homogeneous direct method to detect LDL-C,accounting for 87.5%.Therefore,it is still necessary to make a comparative study of the two LDL-C detection methods.In this study,the direct method was used to determine low density lipoprotein cholesterol（LDL-C）,and Martin formula,Sampson formula and Friedewald formula were used as experimental methods to calculate LDL-C.The differences between the direct measurement of LDL-C（DLDL）and the calculated values of the three formulas(MLDL,SLDL,and FLDL,respectively.Thus,the compatibility of the three formulas for indirect estimation of LDL-C with the direct method was compared in the population,and the applicability of different low-density lipoprotein cholesterol（LDL-C）calculation formulas in population people was verified for clinical promotion and resource conservation.MethodsFrom October 2020 to November 2021,we collected lipid measurements,including HDL-C（high-density lipoprotein cholesterol）,LDL-C（low-density lipoprotein cholesterol）,TC（total cholesterol）,TG（triglycerides）,from patients admitted to our hospital’s electronic medical record system for cardiovascular diseases,such as acute coronary syndrome,acute cerebrovascular disease,and unstable angina pectoris.（HDL-C）,TC（total cholesterol）,TG（triglycerides）,and basic clinical information including age,gender,previous diabetes,hypertension,coronary artery disease,and whether the patient had received statin lipid-lowering therapy.Finally,the total sample size was 2112 cases,and the LDL-C level（DLDL）was measured by the direct method,which was used as a reference for the evaluation of the LDL-C level calculated by the Sampson formula,Martin formula and Friedewald formula with the direct method using Pearson correlation and linear regression analysis,Blank-Altman scatter plot The overall correlation and consistency of risk stratification of LDL-C levels were evaluated by using paired t-test;the change in the difference between the LDL-C levels calculated by each formula and the LDL-C levels detected by the direct method at different TG levels.ResultsIn this study,A total of 2112 inpatients were collected,including an average age of 64 ±11 years old,1060 male patients,accounting for 50.2%of the total sample,1348 patients with hypertension,accounting for 63.8%of the total sample,671 patients with type 2 diabetes,accounting for 31.8%of the total sample,and 1231 patients with coronary heart disease,accounting for 58.3% of the total sample.There were 1498 patients with cerebrovascular disease,accounting for 70.9% of the total sample,and 1429 patients had received statin lipid-lowering treatment,accounting for 67.7%of the total sample.Comparing the three formulas with the direct method by linear regression analysis,it was found that the LDL-C level calculated by Sampson formula（SLDL）,the LDL-C level calculated by Martin formula（MLDL）and the LDL-C level calculated by Friedewald formula（FLDL）were significantly correlated with the directly detected LDL-C level（DLDL）（r = 0.987,0.985,0.981,all P < 0.001）.Blank-Altman scatter chart analysis shows that the three formulas tend to underestimate the level of LDL-C,but the consistency between MLDL and DLDL was better than that of SLDL and FLDL.The overall risk stratification coincidence rate shows that MLDL is the highest（60.8%）,followed by SLDL（58.7%）,and the lowest is FLDL（51.4%）.With the increase of the level of triglyceride（TG）,the results of paired t-test showed that the difference between SLDL,FLDL and DLDL gradually increased,but the changes of MLDL and DLDL were not obvious.When TG < 150mg/d L,the average differences of SLDL,MLDL,FLDL and DLDL were 13.2±4.5,14.5±5.0,14.8±4.9mg/d L,respectively（P <0.001）.When TG was 150~199mg/d L,the average differences of SLDL,MLDL,FLDL and DLDL were 15.7±5.7,14.2±6.1,19.6±5.8mg/d L,respectively（P<0.001）;when TG was 200~400 mg/d L,the average differences of SLDL,MLDL,FLDL and DLDL were 17.7±8.2,11.4±9.4,24.6±8.2 mg/d L,respectively（P < 0.001）.ConclusionWith comparing the LDL-C estimated values calculated by the above three formulas,namely Martin formula,Sampson formula and Friedewald formula,with the LDL-C values directly detected by our hospital based on the principle of enzyme colorimetry,although the three formulas have a significant correlation with the direct method,the above three formulas all show underestimation of LDL-C level and risk stratification in the population,which would lead to misclassification of high-risk individuals,resulting in inadequate lipid-lowering treatment,so more data are needed to study the clinical use of formulas to estimate LDL-C levels,and their use is not recommended for the time being. |
PDF Full Text Request