Publication
Article
Author(s):
The results of a recent retrospective analysis of electronic health record (EHR) data found that only 0.4% of patients with atherosclerotic cardiovascular disease (ASCVD) were tested for lipoprotein(a) (Lp[a]) across 5 large US health systems, with disparities seen in testing rates by race and gender.1 Furthermore, data indicated that Lp(a) testing and scores were associated with differences in lipid-lowering therapy (LLT) use and initiation. The findings were presented as a poster, “Lipoprotein(a) Testing in Patients with Atherosclerotic Cardiovascular Disease in Five Large U.S. Health Systems” during a session at the American Heart Association Scientific Sessions 2023 in Philadelphia, Pennsylvania, which ran from November 11 to 13, 2023.1
Lp(a) is an apolipoprotein B-containing lipoprotein that is a genetic, independent, causal risk factor for ASCVD.2-4 Approximately 20% of the general global population have increased Lp(a), making it the most common genetic dyslipidemia.2,3 Measurement of Lp(a) can inform risk assessment for ASCVD events.3
Study investigators conducted a retrospective analysis of 2019-2021 EHR data from Allina Health, Duke University Medical Center, University of Pittsburgh Medical Center, Vanderbilt University Medical Center, and Ochsner Health System for adults with ASCVD to identify patterns in Lp(a) testing based on patient demographics, as well as whether Lp(a) testing is associated with differences in LLT initiation. Data included in the analysis were from patients aged at least 18 years with established ASCVD during an Lp(a) test or a randomly selected outpatient encounter during the study period (if no Lp[a] test), and baseline characteristics and LLT initiations were compared between the 2 groups. To determine the clinical factors associated with Lp(a) testing, investigators used multivariable regression modeling. The index date was defined as the date of the Lp(a) test or random outpatient encounter in those without a test. LLT initiation was defined as no prescription in the year prior to the index date, but a new prescription in the 6 months following the index date.1
The cohort included 595,684 patients, of whom 2587 (0.4%) underwent Lp(a) testing. In the overall population, 45.0% of patients were female and 84.5% were White. The prevalence of Black race was lower among those who were in the Lp(a) tested cohort, while prevalence of Hispanic ethnicity was similar between the tested and untested cohorts. The average age was 70 years in the untested cohort, and 61 years in the tested cohort. Among those who underwent Lp(a) testing, there was a higher prevalence of coronary artery disease, ischemic stroke or transient ischemic attack, and familial hypercholesterolemia, and a lower prevalence of hypertension and diabetes. The median total cholesterol was 156 mg/dL in the untested cohort and 162 mg/dL in the tested cohort.1
Patient demographics associated with lower likelihood of Lp(a) testing included Black race and older age. Among patients in the tested cohort, 75.5% were already prescribed any LLT (ie, statins, high-intensity statins, a PCSK9 inhibitor (PCSK9i), ezetimibe, bempedoic acid, or some combination) at the time of testing, while only 50.2% of those in the untested cohort were prescribed any LLT. Patients who underwent Lp(a) testing were significantly more likely to initiate any LLT than were those who went untested (P < .001), and this difference was significant for each individual LLT therapy except for bempedoic acid, for which the P value was not calculated. Among patients in the tested cohort, those with Lp(a) levels above 50 mg/dL (n = 821; 33.4%) were significantly more likely to initiate any LLT (P = .033), and specifically a PCSK9i or ezetimibe (both P < .001) than were patients with a tested Lp(a) threshold of 50 mg/dL or lower; however, differences in statin therapy initiation were not significant between groups.1
Limitations of this analysis include the potential lack of generalizability of results because the study data included only 5 health systems. Furthermore, the COVID-19 pandemic may have impacted lab frequencies, and data on medication intolerance and adherence was not available in the electronic health record.1
According to the American Association of Clinical Endocrinologists/American College of Endocrinology Guidelines for Management of Dyslipidemia and Prevention of Cardiovascular Disease and a 2022 statement from the National Lipid Association, Lp(a) testing should be considered in patients with clinical ASCVD or a history of aortic valve stenosis, as well as in those with a family history of premature ASCVD, increased Lp(a), or aortic valve stenosis; Lp(a) measurement should also be considered in patients with South Asian or African ancestry, a 10-year ASCVD risk of at least 10% (including those at very high risk of ASCVD to identify those who are more likely to benefit from PCSK9i therapy), primary severe hypercholesterolemia or suspected familial hypercholesterolemia, or refractory elevations of LDL cholesterol (LDL-C) regardless of aggressive LDL-C–lowering therapy (ie, statin resistance).2,3 An Lp(a) of at least 50 mg/dL is a risk-enhancing factor.3,5 For example, an increased risk of recurrent ASCVD events is seen in patients on statin therapy with Lp(a) levels of 50 mg/dL or greater.2
Results of this study provide real-world patterns of Lp(a) testing and LLT initiation in over half a million adults with ASCVD, indicating differences in testing rates based on patient age and race, as well as notable associations between initiation of LLT and Lp(a) testing and values.1