Endovascular therapy versus bypass for chronic limb-threatening ischemia in a real-world practice.

OBJECTIVE: The recent Best Endovascular vs Best Surgical Therapy in Patients with Critical Limb Ischemia (BEST-CLI) study showed that bypass was superior to endovascular therapy (ET) in patients with chronic limb-threatening ischemia (CLTI) deemed suitable for either approach who had an available single-segment great saphenous vein (GSV). However, the superiority of bypass among those lacking GSV was not established. We aimed to examine comparative treatment outcomes from a real-world CLTI population using the Vascular Quality Initiative-Medicare-linked database. METHODS: We queried the Vascular Quality Initiative-Medicare-linked database for patients with CLTI who underwent first-time lower extremity revascularization (2010-2019). We performed two one-to-one propensity score matchings (PSMs): ET vs bypass with GSV (BWGSV) and ET vs bypass with a prosthetic graft (BWPG). The primary outcome was amputation-free survival. Secondary outcomes were freedom from amputation and overall survival (OS). RESULTS: Three cohorts were queried: BWGSV (N = 5279, 14.7%), BWPG (N = 2778, 7.7%), and ET (N = 27,977, 77.6%). PSM produced two sets of well-matched cohorts: 4705 pairs of ET vs BWGSV and 2583 pairs of ET vs BWPG. In the matched cohorts of ET vs BWGSV, ET was associated with greater hazards of death (hazard ratio [HR] = 1.34, 95% confidence interval [CI], 1.25-1.43; P < .001), amputation (HR = 1.30, 95% CI, 1.17-1.44; P < .001), and amputation/death (HR = 1.32, 95% CI, 1.24-1.40; P < .001) up to 4 years. In the matched cohorts of ET vs BWPG, ET was associated with greater hazards of death up to 2 years (HR = 1.11, 95% CI, 1.00-1.22; P = .042) but not amputation or amputation/death. CONCLUSIONS: In this real-world multi-institutional Medicare-linked PSM analysis, we found that BWGSV is superior to ET in terms of OS, freedom from amputation, and amputation-free survival up to 4 years. Moreover, BWPG was superior to ET in terms of OS up to 2 years. Our study confirms the superiority of BWGSV to ET as observed in the BEST-CLI trial.

Clinical use of polygenic risk scores for detection of peripheral artery disease and cardiovascular events.

We have previously shown that polygenic risk scores (PRS) can improve risk stratification of peripheral artery disease (PAD) in a large, retrospective cohort. Here, we evaluate the potential of PRS in improving the detection of PAD and prediction of major adverse cardiovascular and cerebrovascular events (MACCE) and adverse events (AE) in an institutional patient cohort. We created a cohort of 278 patients (52 cases and 226 controls) and fit a PAD-specific PRS based on the weighted sum of risk alleles. We built traditional clinical risk models and machine learning (ML) models using clinical and genetic variables to detect PAD, MACCE, and AE. The models' performances were measured using the area under the curve (AUC), net reclassification index (NRI), integrated discrimination improvement (IDI), and Brier score. We also evaluated the clinical utility of our PAD model using decision curve analysis (DCA). We found a modest, but not statistically significant improvement in the PAD detection model's performance with the inclusion of PRS from 0.902 (95% CI: 0.846-0.957) (clinical variables only) to 0.909 (95% CI: 0.856-0.961) (clinical variables with PRS). The PRS inclusion significantly improved risk re-classification of PAD with an NRI of 0.07 (95% CI: 0.002-0.137), p = 0.04. For our ML model predicting MACCE, the addition of PRS did not significantly improve the AUC, however, NRI analysis demonstrated significant improvement in risk re-classification (p = 2e-05). Decision curve analysis showed higher net benefit of our combined PRS-clinical model across all thresholds of PAD detection. Including PRS to a clinical PAD-risk model was associated with improvement in risk stratification and clinical utility, although we did not see a significant change in AUC. This result underscores the potential clinical utility of incorporating PRS data into clinical risk models for prevalent PAD and the need for use of evaluation metrics that can discern the clinical impact of using new biomarkers in smaller populations.

Association between Statin use and Incident Peripheral Artery Disease According to Race, Age, and Presence of Depression in the Multi-Ethnic Study of Atherosclerosis.

BACKGROUND: Peripheral artery disease (PAD) is associated with high morbidity and mortality and has been commonly described as a coronary heart disease equivalent. Statin medications are recommended for primary prevention of atherosclerotic cardiovascular disease (CVD) among other indications. Therefore, understanding the longitudinal relationship of incident PAD is necessary to inform future research on how to prevent the disease. Depression complicates CVD patients' ability to properly adhere to their medications, yet the effect of depression on the relationship between statin use and incident PAD is understudied. People with PAD have a higher incidence of depressive symptoms than people without PAD. Black American and Hispanic populations are disproportionately affected by both PAD and depression yet research on the modifying effect of either race or depression on the relationship between statin use and onset of PAD is minimal. While statin utilization is highest for ages 75-84 years, there is minimal evidence of favorable risk-benefit balance. Consequently, in this project, we examined the relationship between statin use and incident PAD and whether this relationship is modified by race/ethnicity, depressive symptoms, or age. METHODS: We used data on participants from the Multi-Ethnic Study of Atherosclerosis from visit 1 (2000) through study visit 6 (2020) who had three separate measurements of the ankle-brachial index (ABI) taken at visit 1, visit 3, and visit 5. Incident PAD was defined as 1) incident lower extremity amputation or revascularization or 2) ABI less than 0.90 coupled with ABI decrease greater than 0.15 over the follow-up period. Statin use was noted on the study visit prior to incident PAD diagnosis while depressive symptoms were measured at exam 1, visit 3, and visit 5. Propensity score matching was implemented to create balance between the participants in the two treatment groups, that is, statin-treated and statin-untreated groups, to reduce the problem of confounding by indication. Propensity scores were calculated using multivariate logistic regression model to estimate the probability of receiving statin treatment. We used Cox proportional hazards regression to investigate the relationship between time-dependent statin use as well as other risk factors with incident PAD, overall and stratified by 1) race, 2) depression status, and 3) age. RESULTS: A total of 4,210 participants were included in the final matched analytic cohort. There were 810 incident cases (19.3%) of PAD that occurred over an average (mean) of 11.3 years (SD = 5.7) of follow-up time. In the statin-treated group, and with an average follow-up time of 12.5 years (SD = 5.6), there were 281 cases (13.4%) of incident PAD with the average follow-up time of 10.1 years (SD = 5.5), whereas in the statin-untreated group, there were 531 cases (25.2%) (P < 0.001). Results demonstrate a lower risk of PAD event in the statin-treated group compared to the untreated group (hazard ratio [HR] = 0.45, 95% confidence interval [CI]: 0.33-0.62) over the span of 18.5 years. The interactions between 1) depression and 2) race with statin use for incident PAD were not significant. However, other risk factors which were significant included Black American race that had approximately 30% lower hazard of PAD compared to non-Hispanic White (HR = 0.70, 95% CI: 0.58-0.84); age-stratified models were also fitted, and stain use was still a significant treatment factor for ages 45-54 (HR = 0.45, 95% CI: 0.33-0.63), 55-64 (HR = 0.61, 95% CI: 0.46-0.79), and 65-74 years (HR = 0.61, 95% CI: 0.48-0.78) but not for ages 75-84 years. CONCLUSIONS: Statin use was associated with a decreased risk of incident PAD for those under the age of 75 years. Neither race nor depression significantly modified the relationship between statin use and incident PAD; however, the risk of incident PAD was lower among Black Americans. These findings highlight that the benefit of statin may wane for those over the age of 75 years. Findings also suggest that statin use may not be compromised in those living with depression.

Endovascular versus Surgical Lower Extremity Revascularization among Patients with Chronic Kidney Disease.

Introduction: Patients with chronic kidney disease (CKD) have a high prevalence of peripheral artery disease. How best to manage lower extremity peripheral artery disease remains unclear in this patient population. We therefore sought to compare the outcomes after endovascular versus surgical lower extremity revascularization among patients with CKD. Methods: We used data from Optum's de-identifed Clinformatics® Data Mart Database, a nationwide database of commercially insured persons in the United States to study patients with CKD who underwent lower extremity endovascular or surgical revascularization. We used inverse probability of treatment weighting to balance covariates. We employed proportional hazard regression to study the primary outcome of major adverse limb events (MALE), defined as a repeat revascularization or amputation. We also studied each of these events separately and death from any cause. Results: In our cohort, 60,057 patients underwent endovascular revascularization and 9,338 patients underwent surgical revascularization. Endovascular revascularization compared with surgical revascularization was associated with a higher adjusted hazard of MALE (hazard ratio (HR) 1.52; 95% confidence interval (CI) 1.46-1.59). Endovascular revascularization was also associated with a higher adjusted hazard of repeat revascularization (HR 1.65; 95% CI 1.57-1.72) but a lower adjusted risk of amputation (HR 0.71; CI 0.73-0.89). Patients undergoing endovascular revascularization also had a lower adjusted hazard for death from any cause (0.85; CI 0.82-0.88). Conclusions: In this analysis of patients with CKD undergoing lower extremity revascularization, an endovascular approach was associated with a higher rate of repeated revascularization but a lower risk of subsequent amputation and death compared with surgical revascularization. Multiple factors must be considered when counseling patients with CKD, who have a high burden of comorbid conditions. Clinical trials should include more patients with kidney disease, who are often otherwise excluded from participation, to better understand the most effective treatment strategies for this vulnerable patient population.

Physician- and Patient-Elicited Barriers and Facilitators to Implementation of a Machine Learning-Based Screening Tool for Peripheral Arterial Disease: Preimplementation Study With Physician and Patient Stakeholders.

BACKGROUND: Peripheral arterial disease (PAD) is underdiagnosed, partially due to a high prevalence of atypical symptoms and a lack of physician and patient awareness. Implementing clinical decision support tools powered by machine learning algorithms may help physicians identify high-risk patients for diagnostic workup. OBJECTIVE: This study aims to evaluate barriers and facilitators to the implementation of a novel machine learning-based screening tool for PAD among physician and patient stakeholders using the Consolidated Framework for Implementation Research (CFIR). METHODS: We performed semistructured interviews with physicians and patients from the Stanford University Department of Primary Care and Population Health, Division of Cardiology, and Division of Vascular Medicine. Participants answered questions regarding their perceptions toward machine learning and clinical decision support for PAD detection. Rapid thematic analysis was performed using templates incorporating codes from CFIR constructs. RESULTS: A total of 12 physicians (6 primary care physicians and 6 cardiovascular specialists) and 14 patients were interviewed. Barriers to implementation arose from 6 CFIR constructs: complexity, evidence strength and quality, relative priority, external policies and incentives, knowledge and beliefs about intervention, and individual identification with the organization. Facilitators arose from 5 CFIR constructs: intervention source, relative advantage, learning climate, patient needs and resources, and knowledge and beliefs about intervention. Physicians felt that a machine learning-powered diagnostic tool for PAD would improve patient care but cited limited time and authority in asking patients to undergo additional screening procedures. Patients were interested in having their physicians use this tool but raised concerns about such technologies replacing human decision-making. CONCLUSIONS: Patient- and physician-reported barriers toward the implementation of a machine learning-powered PAD diagnostic tool followed four interdependent themes: (1) low familiarity or urgency in detecting PAD; (2) concerns regarding the reliability of machine learning; (3) differential perceptions of responsibility for PAD care among primary care versus specialty physicians; and (4) patient preference for physicians to remain primary interpreters of health care data. Facilitators followed two interdependent themes: (1) enthusiasm for clinical use of the predictive model and (2) willingness to incorporate machine learning into clinical care. Implementation of machine learning-powered diagnostic tools for PAD should leverage provider support while simultaneously educating stakeholders on the importance of early PAD diagnosis. High predictive validity is necessary for machine learning models but not sufficient for implementation.

Artificial intelligence in clinical workflow processes in vascular surgery and beyond.

In the past decade, artificial intelligence (AI)-based applications have exploded in health care. In cardiovascular disease, and vascular surgery specifically, AI tools such as machine learning, natural language processing, and deep neural networks have been applied to automatically detect underdiagnosed diseases, such as peripheral artery disease, abdominal aortic aneurysms, and atherosclerotic cardiovascular disease. In addition to disease detection and risk stratification, AI has been used to identify guideline-concordant statin therapy use and reasons for nonuse, which has important implications for population-based cardiovascular disease health. Although many studies highlight the potential applications of AI, few address true clinical workflow implementation of available AI-based tools. Specific examples, such as determination of optimal statin treatment based on individual patient risk factors and enhancement of intraoperative fluoroscopy and ultrasound imaging, demonstrate the potential promise of AI integration into clinical workflow. Many challenges to AI implementation in health care remain, including data interoperability, model bias and generalizability, prospective evaluation, privacy and security, and regulation. Multidisciplinary and multi-institutional collaboration, as well as adopting a framework for integration, will be critical for the successful implementation of AI tools into clinical practice.

The potential of chatbots in chronic venous disease patient management.

Objective: Health care providers and recipients have been using artificial intelligence and its subfields, such as natural language processing and machine learning technologies, in the form of search engines to obtain medical information for some time now. Although a search engine returns a ranked list of webpages in response to a query and allows the user to obtain information from those links directly, ChatGPT has elevated the interface between humans with artificial intelligence by attempting to provide relevant information in a human-like textual conversation. This technology is being adopted rapidly and has enormous potential to impact various aspects of health care, including patient education, research, scientific writing, pre-visit/post-visit queries, documentation assistance, and more. The objective of this study is to assess whether chatbots could assist with answering patient questions and electronic health record inbox management. Methods: We devised two questionnaires: (1) administrative and non-complex medical questions (based on actual inbox questions); and (2) complex medical questions on the topic of chronic venous disease. We graded the performance of publicly available chatbots regarding their potential to assist with electronic health record inbox management. The study was graded by an internist and a vascular medicine specialist independently. Results: On administrative and non-complex medical questions, ChatGPT 4.0 performed better than ChatGPT 3.5. ChatGPT 4.0 received a grade of 1 on all the questions: 20 of 20 (100%). ChatGPT 3.5 received a grade of 1 on 14 of 20 questions (70%), grade 2 on 4 of 16 questions (20%), grade 3 on 0 questions (0%), and grade 4 on 2/20 questions (10%). On complex medical questions, ChatGPT 4.0 performed the best. ChatGPT 4.0 received a grade of 1 on 15 of 20 questions (75%), grade 2 on 2 of 20 questions (10%), grade 3 on 2 of 20 questions (10%), and grade 4 on 1 of 20 questions (5%). ChatGPT 3.5 received a grade of 1 on 9 of 20 questions (45%), grade 2 on 4 of 20 questions (20%), grade 3 on 4 of 20 questions (20%), and grade 4 on 3 of 20 questions (15%). Clinical Camel received a grade of 1 on 0 of 20 questions (0%), grade 2 on 5 of 20 questions (25%), grade 3 on 5 of 20 questions (25%), and grade 4 on 10 of 20 questions (50%). Conclusions: Based on our interactions with ChatGPT regarding the topic of chronic venous disease, it is plausible that in the future, this technology may be used to assist with electronic health record inbox management and offload medical staff. However, for this technology to receive regulatory approval to be used for that purpose, it will require extensive supervised training by subject experts, have guardrails to prevent "hallucinations" and maintain confidentiality, and prove that it can perform at a level comparable to (if not better than) humans. (JVS-Vascular Insights 2023;1:100019.).

Evaluating prediction model performance.

This article highlights important performance metrics to consider when evaluating models developed for supervised classification or regression tasks using clinical data. When evaluating model performance, we detail the basics of confusion matrices, receiver operating characteristic curves, F1 scores, precision-recall curves, mean squared error, and other considerations. In this era, defined by the rapid proliferation of advanced prediction models, familiarity with various performance metrics beyond the area under the receiver operating characteristic curves and the nuances of evaluating model value upon implementation is essential to ensure effective resource allocation and optimal patient care delivery.

2022 ACC/AHA guideline for the diagnosis and management of aortic disease: A report of the American Heart Association/American College of Cardiology Joint Committee on Clinical Practice Guidelines.

AIM: The "2022 ACC/AHA Guideline for the Diagnosis and Management of Aortic Disease" provides recommendations to guide clinicians in the diagnosis, genetic evaluation and family screening, medical therapy, endovascular and surgical treatment, and long-term surveillance of patients with aortic disease across its multiple clinical presentation subsets (ie, asymptomatic, stable symptomatic, and acute aortic syndromes). METHODS: A comprehensive literature search was conducted from January 2021 to April 2021, encompassing studies, reviews, and other evidence conducted on human subjects that were published in English from PubMed, EMBASE, the Cochrane Library, CINHL Complete, and other selected databases relevant to this guideline. Additional relevant studies, published through June 2022 during the guideline writing process, were also considered by the writing committee, where appropriate. STRUCTURE: Recommendations from previously published AHA/ACC guidelines on thoracic aortic disease, peripheral artery disease, and bicuspid aortic valve disease have been updated with new evidence to guide clinicians. In addition, new recommendations addressing comprehensive care for patients with aortic disease have been developed. There is added emphasis on the role of shared decision making, especially in the management of patients with aortic disease both before and during pregnancy. The is also an increased emphasis on the importance of institutional interventional volume and multidisciplinary aortic team expertise in the care of patients with aortic disease.

Validity of the Global Vascular Guidelines in Predicting Outcomes Based on First-Time Revascularization Strategy.

BACKGROUND: The Global Vascular Guidelines (GVG) recommend selecting an endovascular versus open-surgical approach to revascularization for chronic limb-threatening ischemia (CLTI), based on the Global Limb Anatomic Staging System (GLASS) and wound, ischemia, and foot infection (WIfI) classification systems. We assessed the utility of GVG-recommended strategies in predicting clinical outcomes. METHODS: We conducted a single-center, retrospective review of first-time lower-extremity revascularizations within a comprehensive limb-preservation program from 2010 to 2018. Procedures were stratified by (1) treatment concordance with GVG-recommended strategy (concordant versus nonconcordant groups), (2) GLASS stages I-III, and (3) endovascular versus open strategies. The primary outcome was 5-year freedom from major adverse limb events (FF-MALE), defined as freedom from reintervention or major amputation, and secondary outcomes included 5-year overall survival, freedom from major amputation, freedom from reintervention, and immediate technical failure (ITF) during initial revascularization. Kaplan-Meier (KM) survival analysis and multivariate analysis with Cox proportional hazard models were performed on the primary and secondary outcomes. RESULTS: Of 281 first-time revascularizations for CLTI, 251 (89.3%) were endovascular and 186 (66.2%) were in the concordant group, with a mean clinical follow-up of 3.02 ± 2.40 years. Within the concordant group alone, 167 (89.8%) of revascularizations were endovascular. The concordant group had a higher rate of chronic kidney disease (60.8% vs. 45.3%, P = 0.02), WIfI foot infection grade (0.81 ± 1.1 vs. 0.56 ± 0.80, P = 0.03), and WIfI stage (3.1 ± 0.79 vs. 2.8 ± 1.2, P < 0.01) compared to the non-concordant group. After both KM and multivariate analyses, there were no significant differences in 5-year FF-MALE or overall survival between concordant and non-concordant groups. There was higher freedom from major amputation in the non-concordant group on KM analysis (83.9% vs. 74.2%, P = 0.025), though this difference was non-significant on multivariate analysis (hazard ratio [HR]: 0.49, 95% confidence interval [CI]: 0.21-1.15, P = 0.10). The open group had lower MALE compared to the endovascular group (HR: 0.39, 95% CI: 0.17-0.91, P = 0.029) attributed to a lower reintervention rate in the open group (HR: 0.31, 95% CI: 0.11-0.87, P = 0.026). GLASS stage was not associated with significant differences in outcomes, but the severity of GLASS stage was associated with ITF (2.1% in stage 1, 6.4% in stage 2, and 11.7% in stage 3, P = 0.01). CONCLUSIONS: In this study, CLTI treatment outcomes did not differ significantly based on whether treatment was received in concordance with GVG-recommended strategy. There was no difference in overall survival between the endovascular and open groups, though there was a higher reintervention rate in the endovascular group. The GVG guidelines are an important resource to help guide the management of CLTI patients. However, in this study, both concordance with GVG guidelines and GLASS staging were found to be indeterminate in differentiating outcomes between complex CLTI patients treated primarily with an endovascular-first approach. The revascularization approach for a CLTI patient is a nuanced decision that must take into account patient anatomy and clinical status, as well as physician skill and experience and institutional resources.

APLUS: A Python library for usefulness simulations of machine learning models in healthcare.

Despite the creation of thousands of machine learning (ML) models, the promise of improving patient care with ML remains largely unrealized. Adoption into clinical practice is lagging, in large part due to disconnects between how ML practitioners evaluate models and what is required for their successful integration into care delivery. Models are just one component of care delivery workflows whose constraints determine clinicians' abilities to act on models' outputs. However, methods to evaluate the usefulness of models in the context of their corresponding workflows are currently limited. To bridge this gap we developed APLUS, a reusable framework for quantitatively assessing via simulation the utility gained from integrating a model into a clinical workflow. We describe the APLUS simulation engine and workflow specification language, and apply it to evaluate a novel ML-based screening pathway for detecting peripheral artery disease at Stanford Health Care.

2022 ACC/AHA Guideline for the Diagnosis and Management of Aortic Disease: A Report of the American Heart Association/American College of Cardiology Joint Committee on Clinical Practice Guidelines.

AIM: The "2022 ACC/AHA Guideline for the Diagnosis and Management of Aortic Disease" provides recommendations to guide clinicians in the diagnosis, genetic evaluation and family screening, medical therapy, endovascular and surgical treatment, and long-term surveillance of patients with aortic disease across its multiple clinical presentation subsets (ie, asymptomatic, stable symptomatic, and acute aortic syndromes). METHODS: A comprehensive literature search was conducted from January 2021 to April 2021, encompassing studies, reviews, and other evidence conducted on human subjects that were published in English from PubMed, EMBASE, the Cochrane Library, CINHL Complete, and other selected databases relevant to this guideline. Additional relevant studies, published through June 2022 during the guideline writing process, were also considered by the writing committee, where appropriate. STRUCTURE: Recommendations from previously published AHA/ACC guidelines on thoracic aortic disease, peripheral artery disease, and bicuspid aortic valve disease have been updated with new evidence to guide clinicians. In addition, new recommendations addressing comprehensive care for patients with aortic disease have been developed. There is added emphasis on the role of shared decision making, especially in the management of patients with aortic disease both before and during pregnancy. The is also an increased emphasis on the importance of institutional interventional volume and multidisciplinary aortic team expertise in the care of patients with aortic disease.

2022 ACC/AHA Guideline for the Diagnosis and Management of Aortic Disease: A Report of the American Heart Association/American College of Cardiology Joint Committee on Clinical Practice Guidelines.

AIM: The "2022 ACC/AHA Guideline for the Diagnosis and Management of Aortic Disease" provides recommendations to guide clinicians in the diagnosis, genetic evaluation and family screening, medical therapy, endovascular and surgical treatment, and long-term surveillance of patients with aortic disease across its multiple clinical presentation subsets (ie, asymptomatic, stable symptomatic, and acute aortic syndromes). METHODS: A comprehensive literature search was conducted from January 2021 to April 2021, encompassing studies, reviews, and other evidence conducted on human subjects that were published in English from PubMed, EMBASE, the Cochrane Library, CINHL Complete, and other selected databases relevant to this guideline. Additional relevant studies, published through June 2022 during the guideline writing process, were also considered by the writing committee, where appropriate. Structure: Recommendations from previously published AHA/ACC guidelines on thoracic aortic disease, peripheral artery disease, and bicuspid aortic valve disease have been updated with new evidence to guide clinicians. In addition, new recommendations addressing comprehensive care for patients with aortic disease have been developed. There is added emphasis on the role of shared decision making, especially in the management of patients with aortic disease both before and during pregnancy. The is also an increased emphasis on the importance of institutional interventional volume and multidisciplinary aortic team expertise in the care of patients with aortic disease.

Disparities in peripheral artery disease care: A review and call for action.

Peripheral artery disease (PAD), the pathophysiologic narrowing of arterial blood vessels of the lower leg due to atherosclerosis, is a highly prevalent disease that affects more than 6 million individuals 40 years and older in the United States, with sharp increases in prevalence with age. Morbidity and mortality rates in patients with PAD range from 30% to 70% during the 5- to 15-year period after diagnosis and PAD is associated with poor health outcomes and reduced functionality and quality of life. Despite advances in medical, endovascular, and open surgical techniques, there is striking variation in care among population subgroups defined by sex, race and ethnicity, and socioeconomic status, with concomitant differences in preoperative medication optimization, amputation risk, and overall health outcomes. We reviewed studies from 1995 to 2021 to provide a comprehensive analysis of the current impact of disparities on the treatment and management of PAD and offer action items that require strategic partnership with primary care providers, researchers, patients, and their communities. With new technologies and collaborative approaches, optimal management across all population subgroups is possible.

Use of Multi-Modal Data and Machine Learning to Improve Cardiovascular Disease Care.

Today's digital health revolution aims to improve the efficiency of healthcare delivery and make care more personalized and timely. Sources of data for digital health tools include multiple modalities such as electronic medical records (EMR), radiology images, and genetic repositories, to name a few. While historically, these data were utilized in silos, new machine learning (ML) and deep learning (DL) technologies enable the integration of these data sources to produce multi-modal insights. Data fusion, which integrates data from multiple modalities using ML and DL techniques, has been of growing interest in its application to medicine. In this paper, we review the state-of-the-art research that focuses on how the latest techniques in data fusion are providing scientific and clinical insights specific to the field of cardiovascular medicine. With these new data fusion capabilities, clinicians and researchers alike will advance the diagnosis and treatment of cardiovascular diseases (CVD) to deliver more timely, accurate, and precise patient care.

Development of a polygenic risk score to improve detection of peripheral artery disease.

INTRODUCTION: Peripheral artery disease (PAD) is a major cause of cardiovascular morbidity and mortality, yet timely diagnosis is elusive. Larger genome-wide association studies (GWAS) have now provided the ability to evaluate whether genetic data, in the form of genome-wide polygenic risk scores (PRS), can help improve our ability to identify patients at high risk of having PAD. METHODS: Using summary statistic data from the largest PAD GWAS from the Million Veteran Program, we developed PRSs with genome data from UK Biobank. We then evaluated the clinical utility of adding the best-performing PRS to a PAD clinical risk score. RESULTS: A total of 487,320 participants (5759 PAD cases) were included in our final genetic analysis. Compared to participants in the lowest 10% of PRS, those in the highest decile had 3.1 higher odds of having PAD (95% CI, 3.06-3.21). Additionally, a PAD PRS was associated with increased risk of having coronary artery disease, congestive heart failure, and cerebrovascular disease. The PRS significantly improved a clinical risk model (Net Reclassification Index = 0.07, p < 0.001), with most of the performance seen in downgrading risk of controls. Combining clinical and genetic data to detect risk of PAD resulted in a model with an area under the curve of 0.76 (95% CI, 0.75-0.77). CONCLUSION: We demonstrate that a genome-wide PRS can discriminate risk of PAD and other cardiovascular diseases. Adding a PAD PRS to clinical risk models may help improve detection of prevalent, but undiagnosed disease.

Dynamic changes in chromatin accessibility are associated with the atherogenic transitioning of vascular smooth muscle cells.

AIMS: De-differentiation and activation of pro-inflammatory pathways are key transitions vascular smooth muscle cells (SMCs) make during atherogenesis. Here, we explored the upstream regulators of this 'atherogenic transition'. METHODS AND RESULTS: Genome-wide sequencing studies, including Assay for Transposase-Accessible Chromatin using sequencing and RNA-seq, were performed on cells isolated from both murine SMC-lineage-tracing models of atherosclerosis and human atherosclerotic lesions. At the bulk level, alterations in chromatin accessibility were associated with the atherogenic transitioning of lesional SMCs, especially in relation to genes that govern differentiation status and complement-dependent inflammation. Using computational biology, we observed that a transcription factor previously related to coronary artery disease, Activating transcription factor 3 (ATF3), was predicted to be an upstream regulator of genes altered during the transition. At the single-cell level, our results indicated that ATF3 is a key repressor of SMC transitioning towards the subset of cells that promote vascular inflammation by activating the complement cascade. The expression of ATF3 and complement component C3 was negatively correlated in SMCs from human atherosclerotic lesions, suggesting translational relevance. Phenome-wide association studies indicated that genetic variation that results in reduced expression of ATF3 is correlated with an increased risk for atherosclerosis, and the expression of ATF3 was significantly down-regulated in humans with advanced vascular disease. CONCLUSION: Our study indicates that the plasticity of atherosclerotic SMCs may in part be explained by dynamic changes in their chromatin architecture, which in turn may contribute to their maladaptive response to inflammation-induced stress.

Collagen Fibril Orientation in Tissue Specimens From Atherosclerotic Plaque Explored Using Small Angle X-Ray Scattering.

Atherosclerotic plaques can gradually develop in certain arteries. Disruption of fibrous tissue in plaques can result in plaque rupture and thromboembolism, leading to heart attacks and strokes. Collagen fibrils are important tissue building blocks and tissue strength depends on how fibrils are oriented. Fibril orientation in plaque tissue may potentially influence vulnerability to disruption. While X-ray scattering has previously been used to characterize fibril orientations in soft tissues and bones, it has never been used for characterization of human atherosclerotic plaque tissue. This study served to explore fibril orientation in specimens from human plaques using small angle X-ray scattering (SAXS). Plaque tissue was extracted from human femoral and carotid arteries, and each tissue specimen contained a region of calcified material. Three-dimensional (3D) collagen fibril orientation was determined along scan lines that started away from and then extended toward a given calcification. Fibrils were found to be oriented mainly in the circumferential direction of the plaque tissue at the majority of locations away from calcifications. However, in a number of cases, the dominant fibril direction differed near a calcification, changing from circumferential to longitudinal or thickness (radial) directions. Further study is needed to elucidate how these fibril orientations may influence plaque tissue stress-strain behavior and vulnerability to rupture.

Unsupervised Learning for Automated Detection of Coronary Artery Disease Subgroups.

Background The promise of precision population health includes the ability to use robust patient data to tailor prevention and care to specific groups. Advanced analytics may allow for automated detection of clinically informative subgroups that account for clinical, genetic, and environmental variability. This study sought to evaluate whether unsupervised machine learning approaches could interpret heterogeneous and missing clinical data to discover clinically important coronary artery disease subgroups. Methods and Results The Genetic Determinants of Peripheral Arterial Disease study is a prospective cohort that includes individuals with newly diagnosed and/or symptomatic coronary artery disease. We applied generalized low rank modeling and K-means cluster analysis using 155 phenotypic and genetic variables from 1329 participants. Cox proportional hazard models were used to examine associations between clusters and major adverse cardiovascular and cerebrovascular events and all-cause mortality. We then compared performance of risk stratification based on clusters and the American College of Cardiology/American Heart Association pooled cohort equations. Unsupervised analysis identified 4 phenotypically and prognostically distinct clusters. All-cause mortality was highest in cluster 1 (oldest/most comorbid; 26%), whereas major adverse cardiovascular and cerebrovascular event rates were highest in cluster 2 (youngest/multiethnic; 41%). Cluster 4 (middle-aged/healthiest behaviors) experienced more incident major adverse cardiovascular and cerebrovascular events (30%) than cluster 3 (middle-aged/lowest medication adherence; 23%), despite apparently similar risk factor and lifestyle profiles. In comparison with the pooled cohort equations, cluster membership was more informative for risk assessment of myocardial infarction, stroke, and mortality. Conclusions Unsupervised clustering identified 4 unique coronary artery disease subgroups with distinct clinical trajectories. Flexible unsupervised machine learning algorithms offer the ability to meaningfully process heterogeneous patient data and provide sharper insights into disease characterization and risk assessment. Registration URL: https://www.clinicaltrials.gov; Unique identifier: NCT00380185.