Lipoprotein(a) testing trends in young ischemic stroke patients from 2015-2024: An analysis of 188,000 individuals

Abstract

Background

Lipoprotein(a) [Lp(a)] is a genetically determined risk factor for myocardial infarction and stroke. Elevated Lp(a) >50 mg/dL (>125 nmol/L) is common and present in about 1 in 5 individuals. Although Lp(a) may be a cause of young ischemic stroke (age ≤60), limited data on national testing trends in this population are available, testing in the general population remains low overall, and different organizations have varying guidelines for testing. By determining the degree to which this population is tested, information on national testing trends of Lp(a) in young ischemic stroke patients may influence future guideline recommendations to increase Lp(a) testing. This study aims to use a large, real-world dataset to assess trends of Lp(a) testing in young ischemic stroke patients in the United States from 2015-2024.

Methods

We performed a retrospective analysis of Lp(a) testing in young ischemic stroke patients across the United States from January 1, 2015 to December 31, 2024 using Epic Cosmos, a nationwide, de-identified electronic health record (EHR) dataset comprising over 300 million patient records from over 1,715 hospitals and 41,000 clinics, including from all 50 states, Washington D.C., Lebanon, and Saudi Arabia. The current count values for patients, hospitals, and clinics are available on the Epic Cosmos website. Although the Epic Cosmos data dictionary includes Lebanon and Saudi Arabia as standardized site locations, no patients from these countries were present in our analytic cohort; thus, all analyses were restricted to individuals within the United States. We evaluated the number of young ischemic stroke patients, defined as age ≤60 with history of an ischemic cerebrovascular accident (CVA), who had ever undergone Lp(a) testing, the testing rate per annual young ischemic stroke patients, geographical variation, and percentages of patients tested stratified by age, sex, ethnicity, race, and diagnosis of coronary artery disease (CAD). Testing rates were calculated as the number of distinct patients tested per year and as the testing rate per annual patient population. For each stratum we calculated the proportion tested with Wilson 95 % confidence intervals and assessed between-group differences using chi square or Fisher exact tests as appropriate. Annual trends in the testing proportion were modeled using a binomial generalized linear model with a logit link, treating the annual number tested as the numerator and the annual young ischemic stroke population as the denominator, and we report the odds ratio per calendar year with robust standard errors. Geographical variation was visualized using a heat map of testing by state. All analyses were descriptive and intended to characterize population-level patterns of ischemic stroke within the Cosmos network rather than infer causal associations. Given the exploratory design, no additional model-based adjustment for confounding was performed. All data are de-identified in compliance with HIPAA standards and governed under Epic’s “Rules of the Road” for institutional data use.

Results

From 2015 to 2024, out of a total of 188,305 distinct young ischemic stroke patients, 9,226 (4.9 %) underwent Lp(a) testing. Additionally, the annual number of tested patients increased significantly from 179 in 2015 to 1,992 in 2024 (p<0.001), and the annual percentage of patients undergoing Lp(a) testing increased from 4.3 % in 2015 to 9.3 % in 2024. The states with the largest number of tested patients were Ohio (10.4 %), Texas (7.4 %), and Pennsylvania (5.5 %). The rates of testing were significantly different between sexes, with a larger percentage of young women with ischemic strokes tested compared to young men. Analyzing patients with reported racial data, patients who identified as Black or African American underwent testing for Lp(a) at the highest rate, compared with patients who identified as Asian, “None of the above”, White, or Other Race. Among patients undergoing testing with reported ethnic identity, a higher percentage of patients who identified as Hispanic or Latino were tested compared to those who identified as non-Hispanic. Stratifying the total tested patients by age, adults between the ages of 50-60 years made up the largest percentage of patients (4,460; 48.3 %); however, the highest rate of testing occurred in patients aged 5-18. In addition, a higher rate of the young ischemic stroke patients who had ever had a diagnosis of CAD underwent testing compared to patients without CAD.

Discussion

Lp(a) testing among young ischemic stroke patients has increased significantly over the past decade, likely reflecting growing clinical recognition of its causal role in atherosclerotic disease. The rise parallels key updates in lipid management and stroke prevention guidelines, including the 2019 European Society of Cardiology and 2024 National Lipid Association recommendations advocating at least once-in-a-lifetime Lp(a) measurement. Increasing assay availability and heightened awareness of the causal relationship of Lp(a) with atherosclerotic disease may also have contributed to the observed upward trend. Despite this, only about one in twenty young ischemic stroke patients had ever been tested, underscoring a substantial implementation gap between evidence and clinical practice.