Bringing doctors together

or

A free, private network for verified physicians

Leading voices

Dr. Spencer Dorn

Functional & Motility Disorders · UNC

View

Dr. Saurabh Sethi

General GI · Oja Health

View

Dr. Sara Horst

IBD · Vanderbilt

View

Dr. Nneka Ufere

Liver Transplant · MGH

View

Dr. Wendi LeBrett

General GI · ModernGut

View

Dr. Treta Purohit

General GI · Oshi Health

View

Dr. Laura Sahyoun

IBD · WCM

View

Dr. Uma Mahadevan

IBD · UCSF

View

Dr. Rushabh Modi

General GI · Kaiser Permanente

View

Dr. Brennan Spiegel

General GI · Cedars-Sinai

View
  • Discover your community

  • Stay current

  • A profile that represents you

  • Find your people

About RoonPrivacyTerms

    Overview

    Association of Lipoprotein(a) With Major Adverse Limb Events and All-Cause Mortality Following Revascularization for Chronic Limb-Threatening Ischemia: A Substudy of the BEST-CLI Trial.

    May 22, 2025DOI: 10.1161/jaha.125.041177Equivalence TrialAHA

    Written by: Alexander E SullivanShi HuangSuman KunduVictoria E ThomasDaniel G ClairAaron W AdayMatthew T MenardAlik FarberKenneth RosenfieldJonathan D NewmanJeffrey S BergerQuinn S WellsMatthew S FreibergMacRae F LintonJoshua A Beckman

    METHODS

    The data that support the findings of this study are available from the corresponding author upon reasonable request.This was a concurrent and embedded substudy (TIDE [The Impact of Diabetes on Revascularization], NCT03085524) within the BEST‐CLI trial aimed at identifying mechanisms of revascularization failure. The BEST‐CLI trial was an investigator‐initiated, international, multispecialty, prospective, randomized, open‐label, pragmatic superiority trial designed to assess the efficacy of a surgery‐ versus endovascular‐first revascularization strategy in patients with CLTI who were candidates for both revascularization strategies.16 Patients were recruited for TIDE at a subset of participating sites and were eligible if they had planned surveillance of the index intervention with duplex ultrasound imaging. All aspects of the substudy protocol, including laboratory and imaging assessment, were synchronous and parallel with the BEST‐CLI trial protocol and collected by the trial team. Blood was collected at baseline and 6‐month follow‐up visits for laboratory assessment. Clinical data and outcomes were collected throughout the study period by the trial team in accordance with the BEST‐CLI protocol. Because all substudy procedures were conducted under the umbrella of the parent study, and subject data were deidentified to the TIDE study team, this substudy was institutional review board exempt. All participants provided written informed consent.Patients underwent fasting venous sampling in accordance with the BEST‐CLI study protocol. Serum was isolated and stored at −80°C. Lp(a) was measured by using the MedTest Dx method, and lipid panels were run using the Alfa Wassermann ACE Axcel.17 LDL‐C was calculated using the Friedewald formula or, in cases where triglyceride levels were >400 mg/dL, direct low‐density lipoprotein was measured using the Alfa Wassermann ACE LDL‐C reagent. Because Lp(a) levels are stable longitudinally, only the first recorded value during the study period was considered.5 Serum creatinine was measured, and estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration formula.18Clinical outcomes paralleled those of the BEST‐CLI trial.16 The primary outcome of interest was a composite of death from any cause or MALE, defined as above the ankle amputation of the index limb or major index‐limb reintervention (new bypass surgery, interposition graft revision, thrombectomy, or thrombolysis). Secondary outcomes included the components of the primary outcome, death from any cause and MALE, as well as the components of MALE, above the ankle amputation of the index limb or major index‐limb reintervention, and major adverse cardiac events. Major adverse cardiac events were defined as the composite of myocardial infarction, stroke, or all‐cause death. Patients were assessed for outcomes through the end of the study period. Major reinterventions, myocardial infarction, and stroke were adjudicated by a multidisciplinary clinical‐events committee.
    Descriptive statistics were reported for demographic and clinical characteristics. Continuous variables were reported as median (interquartile range [IQR]). Between‐Lp(a) group comparisons were conducted using the Mann–Whitney U test. Categorical variables are reported as total number (percent) with comparisons made using Pearson χ2 test or Fisher exact test as appropriate. Lp(a) categorical levels used 50 mg/dL as the upper limit of normal, in accordance with the 2018 American College of Cardiology and American Heart Association Guideline on the Management of Blood Cholesterol.19 Incidence rates of the primary and secondary outcomes were determined per 100‐person years. Kaplan–Meier plots of MALE or death and other secondary outcomes were presented by Lp(a) level categories. Cox proportional hazard regression analyses assessed the association of serum Lp(a) as a continuous variable, with the outcomes of interest adjusted for prespecified covariates. Two models were used to minimize overfitting given the cohort sample size. The first model included age, sex, and race, because all outcomes had sufficient events to avoid overfitting. The second included age, sex, race, eGFR, and statin use. Lp(a) levels rise as eGFR declines, so controlling for renal function and the interaction with Lp(a) was needed to isolate the effects of Lp(a).14 Statin use similarly has been shown to increase Lp(a) levels.7 Sensitivity analyses also included randomization strategy (endovascular versus surgical bypass), which was not included in the prespecified regression model due to concern for overfitting. The relationship between eGFR and Lp(a) was explored using a linear regression model controlling for age, sex, and race. Data were analyzed using R version 4.2.2 (R Foundation for Statistical Computing).

    RESULTS

    A total of 189 patients underwent Lp(a) assessment during the study period and were followed for a median [IQR] of 2.1 [1.2–2.9] years. The median age was 67.3 [61.6–74.1] years and included 52 (27.5%) women (Table 1). The majority of patients self‐identified as White (n=146, 77.2%) or Black (n=27, 14.3%). Among all patients, 124 (65.6%) had diabetes, 157 (83.1%) were either current or former tobacco smokers, and 39 (20.6%) had chronic kidney disease. Initial revascularization strategy was open surgery in 100 (52.9%) patients and endovascular in 84 (44.4%) patients.
    Study Demographics
    BMI indicates body mass index; CABG, coronary artery bypass graft surgery; CKD, chronic kidney disease; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; HDL‐C, high‐density lipoprotein cholesterol; IQR, interquartile range; LDL‐C, low‐density lipoprotein cholesterol; Lp(a), lipoprotein(a); PCI, percutaneous coronary intervention; and SBP, systolic blood pressure.
    Fisher exact test performed.
    Median Lp(a) concentration for the total study population was 27.3 [10.4–65.8] mg/dL, and 127 (67.2%) patients were within the normal range (<50 mg/dL).19 Compared with patients with normal Lp(a) levels, more patients with elevated levels were Black (25.8% versus 8.7%, P=0.019) and had hypertension (95.2% versus 81.1%, P=0.018). Total cholesterol (140 versus 134 mg/dL, P=0.029) and LDL‐C (70.5 versus 58.0 mg/dL, P=0.23) were significantly higher in those with elevated Lp(a) levels compared with those with normal levels. Statin use was similar between the 2 groups, and no patients were treated with PCSK9 (proprotein convertase subtilisin/kexin type 9) inhibition.
    At the 1‐year follow up, the primary outcome of MALE or death had occurred in 47 out of 198 (23.7%) of the study population, yielding a 1‐year event rate of 33.3% (95% CI, 23.7%–42.8%) per 100‐person years (Table 2). At the median follow up of 2.1 years, the primary outcome had occurred in 64 out of 198 (32.3%) of the study population, yielding an event rate of 26.4% (95% CI, 11.9%–32.9%) per 100‐person years. At both time points, this was primarily driven by MALE. Event rates for all secondary end points are shown in Table 2.
    One‐Year and Median Follow‐Up Event Rates of the Primary and Secondary End Points
    Median follow‐up was 2.1 years. Major adverse cardiac events are defined as a composite of myocardial infarction, stroke, or all‐cause death. Major adverse limb events are defined as a composite of major above the ankle amputation or major reintervention.
    Figure 1 shows the Kaplan‐Meier survival curve according to Lp(a) level, demonstrating similar rates of MALE or death among patients with a normal and elevated Lp(a) (P=0.430). In Cox proportional hazards regression models adjusted for age, sex, and race, there was no association between Lp(a) level and the primary outcome of MALE or death (hazard ratio [HR], 0.996 [95% CI, 0.990–1.002]; P=0.186; Table 3). Associations of Lp(a) level with secondary outcomes are shown in Table 3. Increasing Lp(a) was associated with reduced risk for major amputation (HR, 0.982 [95% CI, 0.968–0.997]; P=0.016), but was not associated with any other secondary outcome. Kaplan‐Meier curves for all secondary outcomes are shown in Figures S1 through S5. These results were not modified after controlling for assigned intervention (endovascular versus surgical bypass).
    Lp(a) indicates lipoprotein(a); and MALE, major adverse limb events.
    Association of Lipoprotein(a) With Adverse Events Adjusted for Age, Sex, and Race
    MACE is the composite of myocardial infarction, stroke, and all‐cause death by the end of the study. MALE is the composite of above the ankle amputation and major reintervention. HRs are per milligrams per deciliter change in Lp(a). HR indicates hazard ratio; MACE, major adverse cardiac events; and MALE, major adverse limb events.
    There was no correlation between Lp(a) level and eGFR (R=−0.018, P=0.780, Figure S6). There was no association between Lp(a) and the primary composite outcome of MALE or death (HR, 1.001 [95% CI, 0.986–1.016]; P=0.532, Table 4) after controlling for age, sex, race, renal function measured by eGFR, and statin use. There was a significant association between Lp(a) and all‐cause death (HR, 1.029 [95% CI, 1.007–1.051]; P=0.009), and the relationship between Lp(a) and major amputation no longer reached statistical significance (HR, 0.966 [95% CI, 0.928–1.006]; P=0.098). There was a significant interaction effect between Lp(a) and eGFR in association with all‐cause death (P=0.007 for interaction). When assessing the HRs for death across eGFR, Lp(a) exhibited a protective effect in those with normal renal function, but a significant risk as eGFR approached 30 mL/min per 1.73 m2 (Figure 2). As Lp(a) level increases, the probability of death at 1.5 years increases for patients with reduced eGFR but decreases for patients with normal eGFR (Figure 3). Patients with high Lp(a) (>150 mg/dL) and low eGFR were at the highest risk of death at 1.5 years (Figure 4).
    Association of Lipoprotein(a) With Adverse Events Adjusted for Age, Sex, Race, Glomerular Filtration Rate, and Statin Use
    MACE is the composite of myocardial infarction, stroke, and all‐cause death by the end of the study. MALE is the composite of above the ankle amputation and major reintervention. HRs are per milligrams per deciliter change in Lp(a) and are based on a model with the interaction term for Lp(a) and eGFR. eGFR indicates estimated glomerular filtration rate; HR, hazard ratio; Lp(a), lipoprotein(a); MACE, major adverse cardiac events; and MALE, major adverse limb events.
    eGFR indicates estimated glomerular filtration rate; and Lp(a), lipoprotein(a).
    eGFR indicates estimated glomerular filtration rate; and Lp(a), lipoprotein(a).

    DISCUSSION

    In this post hoc substudy embedded within the BEST‐CLI trial, we found that Lp(a) level was associated with risk for all cause death following revascularization for CLTI. This relationship varied across eGFR, and elevated Lp(a) imposed the highest risk of death for those with eGFR ≤45 mL/min per 1.73 m2. There was no significant association between elevated Lp(a) and MALE in this high‐risk patient population.
    Elevated Lp(a) has been shown to be a significant risk enhancer for cardiovascular events and cardiovascular mortality.20, 21, 22, 23 Similarly, the risk for incident PAD is nearly 3‐fold higher in those with elevated Lp(a) and has been genetically linked to variation in the LPA gene.8, 24 In patients with stable PAD and claudication, Lp(a) has been associated not only with increased risk of major amputation and need for revascularization, but also with revascularization failure and subsequent MALE.9, 10, 11, 13, 25
    Because CLTI is the most severe form of PAD and is associated with the highest rates of amputation, death, and cardiovascular complications, the association between Lp(a) and adverse outcomes has been unexplored in this group of patients.2 Little progress has been made at reducing amputation rates and improving survival despite advances in medical and interventional therapies over the past 20 years, so understanding the association of risk enhancers, such as Lp(a), is a priority. The 2‐year mortality in the first BASIL (Bypass Versus Angioplasty in the Severe Ischaemia of the Leg) trial published in 2005 was ≈25%, and was the same in the BEST‐CLI trial published in 2022.16, 26 We demonstrate that Lp(a) is significantly associated with all‐cause mortality. More than 50% of patients with CLTI will experience mortality within 3 to 4 years of presentation, and so recognition of this association may be increasingly important to direct the use of Lp(a)‐specific therapies.3, 26, 27
    There was no association between Lp(a) and MALE in this cohort. Although stable PAD is primarily an atherosclerotic disease, CLTI is an atherosclerotic, calcific, and thrombotic disease process. Narula et al have demonstrated that 73% of arteries in patients with CLTI will have evidence of thrombi, even in the absence of significant atherosclerosis.28 Large genome‐wide association studies support this observation and have found associations between PAD and variants in genes encoding Lp(a), LDL‐C receptor, and factor V. Given that Lp(a) is composed of both a proatherosclerotic apoB and a prothrombotic apoA component, this molecule appears to be uniquely positioned to be a driver of this disease process. The inability to detect an association between Lp(a) and MALE may be in part due to the relatively low median Lp(a) levels observed in the cohort. Lp(a) was within the normal range specified by the 2018 American Heart Association/American College of Cardiology lipid guidelines in 67.2% of patients, and the median and mean Lp(a) cohort were 27.3 and 44.3 mg/dL, respectively.19 Prior studies that have included patients with CLTI also noted lower Lp(a) levels in this patient population compared with those with claudication or abdominal aortic aneurysm.13 The strongest associations between Lp(a) and outcomes such as myocardial infarction, stroke, incident PAD, and need for revascularization have been seen in large cohort studies comparing those with low Lp(a) levels with those with extreme elevations.11, 20, 23, 24, 29 The detrimental effects of Lp(a) may be more modest and require a larger sample size to detect differences at the event rates observed in the current cohort. Defining these effects specifically in patients with PAD and CLTI is critical, especially as 3 phase III clinical trials (Lp(a) HORIZON [NCT04023552], OCEAN(a) [NCT05581303], ACCLAIM‐Lp(a) [NCT06292013]) for Lp(a)‐targeted therapies are nearing completion.
    Another potential explanation is that the effect of Lp(a) is overshadowed by other stronger atherosclerotic and thrombotic risk factors patients accumulate before developing CLTI. In the present cohort, a high proportion of patients had diabetes (65.6%), current or prior tobacco use (83.1%), and renal dysfunction (19.0%), all of which predict both mortality and MALE in those with CLTI undergoing revascularization.30, 31, 32, 33 These comorbidities contribute to the development of an advanced vascular phenotype perpetuated by a cycle of inflammation, thrombosis, and atherosclerosis, and differentiate CLTI from other atherosclerotic disease processes. Furthermore, the relative importance of Lp(a) at this end stage may be muted compared with earlier in the disease cycle. Moreover, all participants received therapeutic revascularization, which may minimize the impact of Lp(a) in the acute and intermediate term. Although Lp(a) contributes to the initiation of this process, its effects may be modest once the cycle is set in motion and has reached end stage.9, 34, 35
    The relationship between kidney function and Lp(a) levels is complex and not well understood. Several studies suggest the kidney has a role in Lp(a) clearance, especially in those with large apoA isoforms, and have demonstrated an inverse relationship between eGFR and Lp(a) level.14 This is supported by studies showing lower Lp(a) levels in the renal veins than the aorta and lower urinary apoA in patients with chronic kidney disease than matched controls.36, 37 This inverse relationship, however, has not been consistently replicated, and no change in Lp(a) level is seen following kidney donation despite a significant drop in eGFR.38, 39 Others have proposed Lp(a) as an independent driver of the development of chronic kidney disease, particularly in those with diabetes. This relationship has also not been consistently reproduced.40, 41, 42 Similarly, the association between Lp(a) and cardiovascular events among patients with chronic kidney disease has been inconsistent, but those on hemodialysis and with small apoA size are likely to be at highest risk.43, 44, 45, 46 Our data indicate the risk attributable to either extreme Lp(a) or eGFR values may represent shared pathophysiology between the two risk factors.
    This study has several limitations as an exploratory post hoc analysis. Lipid assessment was done on only a small portion of the trial population. The comorbidities and outcomes of our subgroup parallel those of the overall BEST‐CLI trial, suggesting that our sample is representative and without selection bias. To further account for potential confounding from including patients from both cohorts of BEST‐CLI, we adjusted for randomization procedure in our regression model. Patients with CLTI represent a high‐risk cohort, and this was reflected in their LDL‐C levels, which were well controlled regardless of Lp(a) status and unlikely to be confounding the relationship between Lp(a) and our study outcomes. We measured Lp(a) levels, however, in mass concentration (milligrams per deciliter) rather than by particle concentration (nanomoles per liter), which may be less precise given variability in particle size.47 Similarly, our methodology does not allow us to account for genetic Lp(a) regulation and apoA size. Future studies are needed to take this into account. This is one of the smaller cohorts of patients in which the association between Lp(a) and outcomes has been assessed, but is the largest to do so in the CLTI population. Considering our sample size of 189 patients, with 62 of those having an Lp(a) level ≥50 mg/dL, the minimum detectable hazard ratio for the effect of elevated Lp(a) level compared to normal Lp(a) level at 80% power with a 5% type I error rate would be an HR of 2.26. As such, the cohort is not powered to detect small or moderate effect sizes. The protective association between Lp(a) and major amputation in our regression model is counterintuitive to the established body of literature. Our findings may be limited by sample size, other unbalanced risk factors (ie, tobacco use), and should be further explored in larger studies. Although multiple analyses were conducted, all were prespecified and followed the protocol laid out in the BEST‐CLI trial to minimize the risk of finding an association through multiple testing. To ensure our multivariable Cox regression models were not overfit for the population, all models were prespecified in advance without modification after statistical analysis.

    CONCLUSIONS

    In this post hoc subgroup analysis from the BEST‐CLI trial, Lp(a) was associated with an increased risk for all‐cause death following revascularization for CLTI. Larger studies are needed to define the association between Lp(a) and MALE in this high‐risk patient population.

    Sources of Funding

    This work was supported by National Institutes of Health grants T32 GM007569 (A.E.S.), R01 HL165208 (J.D.N.), HL116263 (M.F.L.), and R01 HL131977 (J.A.B.).

    Disclosures

    Dr Aday reports serving on the Clinical Endpoint Committee for Janssen. Dr Farber reports consulting fees from Sanifit, BiogenCell, and Lemaitre and serves on the advisory board for iThera and DialysisX. Dr Linton reports consulting fees from Merck, Amgen, Regenxbio, and Ionis and receives research support from Novartis, Sanofi Aventis, Ionis, Regeneron, Regenxbio, and FH Foundation. Dr Beckman reports consulting fees from JanOne, Merck, and MingSight and serves on the Data and Safety Monitoring Committee for Janssen and Novartis. All other authors have no financial disclosures.

    Supporting information

    Figures S1–S6
    AADr. Ariel Aday

    Dr. Ariel Aday

    Gastroenterology · Functional & Motility Disorders
    The University of Texas Southwestern Medical Center
    arieladay