Testing
Adjusting the Friedewald LDL formula for Lp(a) does not improve heart disease risk prediction, the Quebec Cardiovascular Study of 2222 men (Atherosclerosis 2002)
Original title: Does correction of the friedewald formula using lipoprotein(a) change our estimation of ischemic heart disease risk? The Quebec Cardiovascular Study
This study tested whether modifying the Friedewald formula for calculating LDL cholesterol to account for Lp(a)-associated cholesterol improves ischemic heart disease (IHD) risk prediction, following 2222 men free of IHD for 5 years, during which 89 had a first IHD event. Both the standard Friedewald formula (relative risk 2.15, 95% CI 1.23-3.75) and the Lp(a)-corrected Dahlen modification (relative risk 2.18, 95% CI 1.25-3.81) showed the highest LDL cholesterol tertile roughly doubled IHD risk compared with the lowest, with no meaningful difference between methods. Lp(a) itself was not an independent predictor of IHD risk. The findings show correcting the Friedewald formula for Lp(a) does not improve IHD risk assessment.
Original abstract
Background: LDL-cholesterol is usually calculated using the Friedewald formula. This calculation method does not take into account the presence of Lp(a), which is associated with LDL-cholesterol. Dahlen has suggested that the Friedewald formula should be modified to account for Lp(a) associated cholesterol. This study was undertaken to determine if correction of the Friedewald formula would result in a better evaluation of ischemic heart disease (IHD) risk.
Methods: 2222 men free from IHD were prospectively followed for 5 years for the appearance of myocardial infarction, coronary insufficiency or coronary death. At the baseline evaluation all had a complete fasting lipid profile which included Lp(a) determinations. LDL-cholesterol levels were calculated from total cholesterol, total triglycerides and HDL-cholesterol using the Friedewald formula and also using the Dahlen modification of the Friedewald formula.
Results: During the follow-up there were 89 first IHD events. Both types of LDL-cholesterol calculations showed that the last tertile of the LDL-cholesterol distribution in comparison to the first tertile, doubles the relative risk (RR: 2.15; 95% confidence limits: 1.23-3.75) using the Friedewald formula (RR: 2.18; 95% confidence limits: 1.25-3.81) using the Dahlen modification. Lp(a) levels were not an independent predictor of IHD risk.
Conclusion: Modification of the Friedewald formula to account for Lp(a) levels does not improve our evaluation of IHD risk.
Summary written by lp-a.org from the published abstract; figures as published. Page updated 18 August 2026. Methods.