Testing
Kronenberg group finds Lp(a) is truly stable once technical variability is removed, suggesting reported fluctuations are largely assay noise (Atherosclerosis 2025)
Original title: Repeated measurement of lipoprotein(a): technical versus biological variability
This study tested whether recently reported Lp(a) fluctuations over time reflect true biological variability or technical measurement noise, an important distinction for the recommendation that a single lifetime Lp(a) measurement suffices. Lp(a) was measured in duplicate in 715 elderly people at two blood draws 3.2 years apart, with baseline and follow-up samples run on the same assay plate to minimise technical variability. Paired concentrations correlated strongly (r = 0.98); of 715 individuals grouped by baseline risk category (below 30, 30 to below 50, 50 to below 70, and 70 mg/dL or above), 655 (91.6%) stayed in the same category at follow-up, while 33 (4.6%) moved higher and 27 (3.8%) moved lower. Each 5 mg/dL of measurement error translated to only a 1% miscalculation of lifetime cardiovascular risk. The authors conclude clinically relevant biological Lp(a) change is uncommon in older adults, and that larger fluctuations reported elsewhere likely reflect technical assay variability, arguing for better laboratory quality control over routine repeat testing.
Original abstract
Background And Aims: Repeated measurements of Lp(a) concentrations are not considered relevant since Lp(a) is expected to remain relatively stable during life through the strong genetic determination. This has recently been questioned by reports showing major fluctuations of Lp(a) over time. However, these studies have not distinguished between biological or technically caused variability. Therefore, this study aimed to assess the biological variability of Lp(a) concentrations over time while minimizing technical variability to evaluate the clinical relevance of repeated Lp(a) measurements.
Methods: Lp(a) concentrations were measured in duplicates in 715 elderly people with two blood collections 3.2 years apart and having the sample pairs (baseline and follow-up) on the same assay plate.
Results: The Lp(a) concentrations of sample pairs were strongly correlated (r = 0.98). We grouped individuals according to risk categories from the baseline Lp(a) measurement of <30, 30 to <50, 50 to <70 and ≥ 70 mg/dL and observed that between baseline and follow-up 655 of the 715 individuals (91.6 %) remained in their risk category, while 33 (4.6 %) moved to a higher and 27 (3.8 %) to a lower category. We calculated that each 5 mg/dL incorrectly measured Lp(a) concentration results in a 1 % miscalculation of the lifetime cardiovascular risk. This is relatively small considering that roughly a third of the population is dying from cardiovascular disease.
Conclusions: Clinically relevant changes of Lp(a) concentrations by biological variability do not occur frequently in an elderly population. Major changes of Lp(a) as reported in the literature might be caused by technical rather than biological fluctuations and argues for repeated Lp(a) measurements in specific cases and an improvement of the quality control measures in laboratory practice.
Summary written by lp-a.org from the published abstract; figures as published. Page updated 17 August 2026. Methods.