Using Machine Learning (XGBoost) to Predict Outcomes After Infrainguinal Bypass for Peripheral Artery Disease.

OBJECTIVE: To develop machine learning (ML) algorithms that predict outcomes after infrainguinal bypass. BACKGROUND: Infrainguinal bypass for peripheral artery disease carries significant surgical risks; however, outcome prediction tools remain limited. METHODS: The Vascular Quality Initiative database was used to identify patients who underwent infrainguinal bypass for peripheral artery disease between 2003 and 2023. We identified 97 potential predictor variables from the index hospitalization [68 preoperative (demographic/clinical), 13 intraoperative (procedural), and 16 postoperative (in-hospital course/complications)]. The primary outcome was 1-year major adverse limb event (composite of surgical revision, thrombectomy/thrombolysis, or major amputation) or death. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained 6 ML models using preoperative features. The primary model evaluation metric was the area under the receiver operating characteristic curve (AUROC). The top-performing algorithm was further trained using intraoperative and postoperative features. Model robustness was evaluated using calibration plots and Brier scores. RESULTS: Overall, 59,784 patients underwent infrainguinal bypass, and 15,942 (26.7%) developed 1-year major adverse limb event/death. The best preoperative prediction model was XGBoost, achieving an AUROC (95% CI) of 0.94 (0.93-0.95). In comparison, logistic regression had an AUROC (95% CI) of 0.61 (0.59-0.63). Our XGBoost model maintained excellent performance at the intraoperative and postoperative stages, with AUROCs (95% CI's) of 0.94 (0.93-0.95) and 0.96 (0.95-0.97), respectively. Calibration plots showed good agreement between predicted and observed event probabilities with Brier scores of 0.08 (preoperative), 0.07 (intraoperative), and 0.05 (postoperative). CONCLUSIONS: ML models can accurately predict outcomes after infrainguinal bypass, outperforming logistic regression.

GLP-1 receptor agonists and atherosclerosis protection: the vascular endothelium takes center stage.

Atherosclerotic cardiovascular disease is a chronic condition that often copresents with type 2 diabetes and obesity. Glucagon-like peptide-1 receptor agonists (GLP-1RAs) are incretin mimetics endorsed by major professional societies for improving glycemic status and reducing atherosclerotic risk in people living with type 2 diabetes. Although the cardioprotective efficacy of GLP-1RAs and their relationship with traditional risk factors are well established, there is a paucity of publications that have summarized the potentially direct mechanisms through which GLP-1RAs mitigate atherosclerosis. This review aims to narrow this gap by providing comprehensive and in-depth mechanistic insight into the antiatherosclerotic properties of GLP-1RAs demonstrated across large outcome trials. Herein, we describe the landmark cardiovascular outcome trials that triggered widespread excitement around GLP-1RAs as a modern class of cardioprotective agents, followed by a summary of the origins of GLP-1RAs and their mechanisms of action. The effects of GLP-1RAs at each major pathophysiological milestone of atherosclerosis, as observed across clinical trials, animal models, and cell culture studies, are described in detail. Specifically, this review provides recent preclinical and clinical evidence that suggest GLP-1RAs preserve vessel health in part by preventing endothelial dysfunction, achieved primarily through the promotion of angiogenesis and inhibition of oxidative stress. These protective effects are in addition to the broad range of atherosclerotic processes GLP-1RAs target downstream of endothelial dysfunction, which include systemic inflammation, monocyte recruitment, proinflammatory macrophage and foam cell formation, vascular smooth muscle cell proliferation, and plaque development.

Restoration of blood vessel regeneration in the era of combination SGLT2i and GLP-1RA therapy for diabetes and obesity.

Ischaemic cardiovascular diseases, including peripheral and coronary artery disease, myocardial infarction, and stroke, remain major comorbidities for individuals with type 2 diabetes (T2D) and obesity. During cardiometabolic chronic disease (CMCD), hyperglycaemia and excess adiposity elevate oxidative stress and promote endothelial damage, alongside an imbalance in circulating pro-vascular progenitor cells that mediate vascular repair. Individuals with CMCD demonstrate pro-vascular 'regenerative cell exhaustion' (RCE) characterized by excess pro-inflammatory granulocyte precursor mobilization into the circulation, monocyte polarization towards pro-inflammatory vs. anti-inflammatory phenotype, and decreased pro-vascular progenitor cell content, impairing the capacity for vessel repair. Remarkably, targeted treatment with the sodium-glucose cotransporter-2 inhibitor (SGLT2i) empagliflozin in subjects with T2D and coronary artery disease, and gastric bypass surgery in subjects with severe obesity, has been shown to partially reverse these RCE phenotypes. SGLT2is and glucagon-like peptide-1 receptor agonists (GLP-1RAs) have reshaped the management of individuals with T2D and comorbid obesity. In addition to glucose-lowering action, both drug classes have been shown to induce weight loss and reduce mortality and adverse cardiovascular outcomes in landmark clinical trials. Furthermore, both drug families also act to reduce systemic oxidative stress through altered activity of overlapping oxidase and antioxidant pathways, providing a putative mechanism to augment circulating pro-vascular progenitor cell content. As SGLT2i and GLP-1RA combination therapies are emerging as a novel therapeutic opportunity for individuals with poorly controlled hyperglycaemia, potential additive effects in the reduction of oxidative stress may also enhance vascular repair and further reduce the ischaemic cardiovascular comorbidities associated with T2D and obesity.

Machine Learning to Predict Outcomes of Endovascular Intervention for Patients With PAD.

Importance: Endovascular intervention for peripheral artery disease (PAD) carries nonnegligible perioperative risks; however, outcome prediction tools are limited. Objective: To develop machine learning (ML) algorithms that can predict outcomes following endovascular intervention for PAD. Design, Setting, and Participants: This prognostic study included patients who underwent endovascular intervention for PAD between January 1, 2004, and July 5, 2023, with 1 year of follow-up. Data were obtained from the Vascular Quality Initiative (VQI), a multicenter registry containing data from vascular surgeons and interventionalists at more than 1000 academic and community hospitals. From an initial cohort of 262 242 patients, 26 565 were excluded due to treatment for acute limb ischemia (n = 14 642) or aneurysmal disease (n = 3456), unreported symptom status (n = 4401) or procedure type (n = 2319), or concurrent bypass (n = 1747). Data were split into training (70%) and test (30%) sets. Exposures: A total of 112 predictive features (75 preoperative [demographic and clinical], 24 intraoperative [procedural], and 13 postoperative [in-hospital course and complications]) from the index hospitalization were identified. Main Outcomes and Measures: Using 10-fold cross-validation, 6 ML models were trained using preoperative features to predict 1-year major adverse limb event (MALE; composite of thrombectomy or thrombolysis, surgical reintervention, or major amputation) or death. The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). After selecting the best performing algorithm, additional models were built using intraoperative and postoperative data. Results: Overall, 235 677 patients who underwent endovascular intervention for PAD were included (mean [SD] age, 68.4 [11.1] years; 94 979 [40.3%] female) and 71 683 (30.4%) developed 1-year MALE or death. The best preoperative prediction model was extreme gradient boosting (XGBoost), achieving the following performance metrics: AUROC, 0.94 (95% CI, 0.93-0.95); accuracy, 0.86 (95% CI, 0.85-0.87); sensitivity, 0.87; specificity, 0.85; positive predictive value, 0.85; and negative predictive value, 0.87. In comparison, logistic regression had an AUROC of 0.67 (95% CI, 0.65-0.69). The XGBoost model maintained excellent performance at the intraoperative and postoperative stages, with AUROCs of 0.94 (95% CI, 0.93-0.95) and 0.98 (95% CI, 0.97-0.99), respectively. Conclusions and Relevance: In this prognostic study, ML models were developed that accurately predicted outcomes following endovascular intervention for PAD, which performed better than logistic regression. These algorithms have potential for important utility in guiding perioperative risk-mitigation strategies to prevent adverse outcomes following endovascular intervention for PAD.

Predicting inferior vena cava filter complications using machine learning.

OBJECTIVE: Inferior vena cava (IVC) filter placement is associated with important long-term complications. Predictive models for filter-related complications may help guide clinical decision-making but remain limited. We developed machine learning (ML) algorithms that predict 1-year IVC filter complications using preoperative data. METHODS: The Vascular Quality Initiative database was used to identify patients who underwent IVC filter placement between 2013 and 2024. We identified 77 preoperative demographic and clinical features from the index hospitalization when the filter was placed. The primary outcome was 1-year filter-related complications (composite of filter thrombosis, migration, angulation, fracture, and embolization or fragmentation, vein perforation, new caval or iliac vein thrombosis, new pulmonary embolism, access site thrombosis, or failed retrieval). The data were divided into training (70%) and test (30%) sets. Six ML models were trained using preoperative features with 10-fold cross-validation (Extreme Gradient Boosting, random forest, Naïve Bayes classifier, support vector machine, artificial neural network, and logistic regression). The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was assessed using calibration plot and Brier score. Performance was evaluated across subgroups based on age, sex, race, ethnicity, rurality, median Area Deprivation Index, planned duration of filter, landing site of filter, and presence of prior IVC filter placement. RESULTS: Overall, 14,476 patients underwent IVC filter placement and 584 (4.0%) experienced 1-year filter-related complications. Patients with a primary outcome were younger (59.3 ± 16.7 years vs 63.8 ± 16.0 years; P < .001) and more likely to have thrombotic risk factors including thrombophilia, prior venous thromboembolism (VTE), and family history of VTE. The best prediction model was Extreme Gradient Boosting, achieving an AUROC of 0.93 (95% confidence interval, 0.92-0.94). In comparison, logistic regression had an AUROC of 0.63 (95% confidence interval, 0.61-0.65). Calibration plot showed good agreement between predicted/observed event probabilities with a Brier score of 0.07. The top 10 predictors of 1-year filter-related complications were (1) thrombophilia, (2) prior VTE, (3) antiphospholipid antibodies, (4) factor V Leiden mutation, (5) family history of VTE, (6) planned duration of IVC filter (temporary), (7) unable to maintain therapeutic anticoagulation, (8) malignancy, (9) recent or active bleeding, and (10) age. Model performance remained robust across all subgroups. CONCLUSIONS: We developed ML models that can accurately predict 1-year IVC filter complications, performing better than logistic regression. These algorithms have potential to guide patient selection for filter placement, counselling, perioperative management, and follow-up to mitigate filter-related complications and improve outcomes.

Vascular Regenerative Cell Deficiencies in South Asian Adults.

BACKGROUND: South Asian individuals shoulder a disproportionate burden of cardiometabolic diseases. OBJECTIVES: The purpose of this study was to determine if vascular regenerative cell content varies significantly between South Asian and White European people. METHODS: Between January 2022 and January 2023, 60 South Asian and 60 White European adults with either documented cardiovascular disease or established diabetes with ≥1 other cardiovascular risk factor were prospectively enrolled. Vascular regenerative cell content in venous blood was enumerated using a flow cytometry assay that is based on high aldehyde dehydrogenase (ALDHhi) activity and cell surface marker phenotyping. The primary outcome was the difference in frequency of circulating ALDHhi progenitor cells, monocytes, and granulocytes between the 2 groups. RESULTS: Compared with White European participants, those of South Asian ethnicity were younger (69 ± 10 years vs 66 ± 9 years; P < 0.05), had lower weight (88 ± 19 kg vs 75 ± 13 kg; P < 0.001), and exhibited a greater prevalence of type 2 diabetes (62% vs 92%). South Asian individuals had markedly lower circulating frequencies of pro-angiogenic ALDHhiSSClowCD133+ progenitor cells (P < 0.001) and ALDHhiSSCmidCD14+CD163+ monocytes with vessel-reparative capacity (P < 0.001), as well as proportionally more ALDHhi progenitor cells with high reactive oxygen species content (P < 0.05). After correction for sex, age, body mass index, and glycated hemoglobin, South Asian ethnicity was independently associated with lower ALDHhiSSClowCD133+ cell count. CONCLUSIONS: South Asian people with cardiometabolic disease had less vascular regenerative and reparative cells suggesting compromised vessel repair capabilities that may contribute to the excess vascular risk in this population. (The Role of South Asian vs European Origins on Circulating Regenerative Cell Exhaustion [ORIGINS-RCE]; NCT05253521).

Icosapent ethyl modulates circulating vascular regenerative cell content: The IPE-PREVENTION CardioLink-14 trial.

BACKGROUND: REDUCE-IT (Reduction of Cardiovascular Events with Icosapent Ethyl-Intervention Trial) showed that icosapent ethyl (IPE) reduced major adverse cardiovascular events by 25%. Since the underlying mechanisms for these benefits are not fully understood, the IPE-PREVENTION CardioLink-14 trial (ClinicalTrials.gov: NCT04562467) sought to determine if IPE regulates vascular regenerative (VR) cell content in people with mild to moderate hypertriglyceridemia. METHODS: Seventy statin-treated individuals with triglycerides ≥1.50 and <5.6 mmol/L and either atherosclerotic cardiovascular disease or type 2 diabetes with additional cardiovascular risk factors were randomized to IPE (4 g/day) or usual care. VR cells with high aldehyde dehydrogenase activity (ALDHhi) were isolated from blood collected at the baseline and 3-month visits and characterized with lineage-specific cell surface markers. The primary endpoint was the change in frequency of pro-vascular ALDHhiside scatter (SSC)lowCD133+ progenitor cells. Change in frequencies of ALDHhiSSCmid monocyte and ALDHhiSSChi granulocyte precursor subsets, reactive oxygen species production, serum biomarkers, and omega-3 levels were also evaluated. FINDINGS: Baseline characteristics, cardiovascular risk factors, and medications were balanced between the groups. Compared to usual care, IPE increased the mean frequency of ALDHhiSSClowCD133+ cells (-1.00% ± 2.45% vs. +7.79% ± 1.70%; p = 0.02), despite decreasing overall ALDHhiSSClow cell frequency. IPE assignment also reduced oxidative stress in ALDHhiSSClow progenitors and increased ALDHhiSSChi granulocyte precursor cell content. CONCLUSIONS: IPE-PREVENTION CardioLink-14 provides the first translational evidence that IPE can modulate VR cell content and suggests a novel mechanism that may underlie the cardioprotective effects observed with IPE in REDUCE-IT. FUNDING: HLS Therapeutics provided the IPE in kind and had no role in the study design, conduct, analyses, or interpretation.

Bridging the gap between in vitro and in vivo models: a way forward to clinical translation of mitochondrial transplantation in acute disease states.

Mitochondrial transplantation and transfer are being explored as therapeutic options in acute and chronic diseases to restore cellular function in injured tissues. To limit potential immune responses and rejection of donor mitochondria, current clinical applications have focused on delivery of autologous mitochondria. We recently convened a Mitochondrial Transplant Convergent Working Group (CWG), to explore three key issues that limit clinical translation: (1) storage of mitochondria, (2) biomaterials to enhance mitochondrial uptake, and (3) dynamic models to mimic the complex recipient tissue environment. In this review, we present a summary of CWG conclusions related to these three issues and provide an overview of pre-clinical studies aimed at building a more robust toolkit for translational trials.