Translating AI for the Clinician.

This Viewpoint explores how artificial intelligence technologies can adopt a clinical practice framework to identify use cases and outline the technology's objectives and potential uses in modern health care.

Real-World Exploration of LDL-Cholesterol Management in Patients with Atherosclerotic Cardiovascular Disease.

BACKGROUND: Although guidelines recommend low-density lipoprotein cholesterol (LDL-C) to be <70mg/dL in patients with atherosclerotic cardiovascular disease (ASCVD), the rate of achieving this goal remains suboptimal. We sought to understand real world contemporary practice patterns of LDL-C management in patients with ASCVD, and whether LDL-C testing influenced management across US health systems. METHODS: A retrospective cohort study utilizing electronic medical record data from five health systems participating in the CardioHealth Alliance was performed on patients with an LDL-C measurement in 2021 and prior ASCVD. Multivariable regression modeling was used to determine the relationship of clinical factors with achievement of guideline directed LDL-C target. Changes in lipid lowering therapy (LLT) after LDL-C testing were also described. RESULTS: Among 216,074 patients with ASCVD, 129,886 (60.1%) had uncontrolled LDL-C (i.e. ≥70 mg/dL). Compared with participants with controlled LDL-C (<70mg/dL), those with uncontrolled LDL-C were more frequently female (50.9% vs 35.1%), or Black (13.7% vs. 10.3%), and less commonly had coronary artery disease as the form of vascular disease (73.0% vs. 83.5% %), heart failure (21.3% vs. 29.1% %), diabetes (34.1% vs. 48.2%), atrial fibrillation (19.3% vs. 26.1%), or chronic kidney disease (25.1% vs. 32.2%). In multivariable analyses, the factors most strongly associated with failure to achieve LDL-C control were female sex (RR 1.13 [95% CI 1.12-1.14] p <0.001) and Black race (1.15 [1.14-1.17] p <0.001). Among the 53,957 (41.5%) of those with uncontrolled LDL-C ≥70 mg/dL not on lipid lowering therapy (LLT) at baseline, only 21% were initiated on any LLT within 6 months of the uncontrolled LDL-C value. CONCLUSIONS: Within five diverse large health systems in the CardioHealth Alliance, more than half of the patients with ASCVD had uncontrolled LDL-C with significant disparities based on sex and race at baseline. The vast majority were not initiated on any lipid lowering therapy within 6 months of an elevated test result indicating persistent gaps in care that will likely worsen health inequities in outcomes. This highlights the urgent need for implementation efforts to improve equitable care.

Endovascular Ablation of the Greater Splanchnic Nerve in Heart Failure With Preserved Ejection Fraction: The REBALANCE-HF Randomized Clinical Trial.

Importance: Greater splanchnic nerve ablation may improve hemodynamics in patients with heart failure and preserved ejection fraction (HFpEF). Objective: To explore the feasibility and safety of endovascular right-sided splanchnic nerve ablation for volume management (SAVM). Design, Setting, and Participants: This was a phase 2, double-blind, 1:1, sham-controlled, multicenter, randomized clinical trial conducted at 14 centers in the US and 1 center in the Republic of Georgia. Patients with HFpEF, left ventricular ejection fraction of 40% or greater, and invasively measured peak exercise pulmonary capillary wedge pressure (PCWP) of 25 mm Hg or greater were included. Study data were analyzed from May 2023 to June 2024. Intervention: SAVM vs sham control procedure. Main Outcomes and Measures: The primary efficacy end point was a reduction in legs-up and exercise PCWP at 1 month. The primary safety end point was serious device- or procedure-related adverse events at 1 month. Secondary efficacy end points included HF hospitalizations, changes in exercise function and health status through 12 months, and baseline to 1-month change in resting, legs-up, and 20-W exercise PCWP. Results: A total of 90 patients (median [range] age, 71 [47-90] years; 58 female [64.4%]) were randomized at 15 centers (44 SAVM vs 46 sham). There were no differences in adverse events between groups. The primary efficacy end point did not differ between SAVM or sham (mean between-group difference in PCWP, -0.03 mm Hg; 95% CI, -2.5 to 2.5 mm Hg; P = .95). There were also no differences in the secondary efficacy end points. There was no difference in the primary safety end point between the treatment (6.8% [3 of 44]) and sham (2.2% [1 of 46]) groups (difference, 4.6%; 95% CI, -6.1% to 15.4%; P = .36). There was no difference in the incidence of orthostatic hypotension between the treatment (11.4% [5 of 44]) and sham (6.5% [3 of 46]) groups (difference, 4.9%; 95% CI, -9.2% to 18.8%; P = .48). Conclusions and Relevance: Results show that SAVM was safe and technically feasible, but it did not reduce exercise PCWP at 1 month or improve clinical outcomes at 12 months in a broad population of patients with HFpEF. Trial Registration: ClinicalTrials.gov Identifier: NCT04592445.

Relationship Between Infarct Artery, Myocardial Injury, and Outcomes After Primary Percutaneous Coronary Intervention in ST-Segment-Elevation Myocardial Infarction.

BACKGROUND: The extent to which infarct artery impacts the extent of myocardial injury and outcomes in patients with ST-segment-elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention is uncertain. METHODS AND RESULTS: We performed a pooled analysis using individual patient data from 7 randomized STEMI trials in which myocardial injury within 30 days after primary percutaneous coronary intervention was assessed in 1774 patients by cardiac magnetic resonance (n=1318) or technetium-99m sestamibi single-photon emission computed tomography (n=456). Clinical follow-up was performed at a median duration of 351 days (interquartile range, 184-368 days). Infarct size and outcomes were assessed in anterior (infarct vessel=left anterior descending) versus nonanterior (non-left anterior descending) STEMI. Median infarct size (percentage left ventricular myocardial mass) was larger in patients with anterior compared with nonanterior STEMI (19.7% [interquartile range, 9.4%-31.7%] versus 12.6% [interquartile range, 5.1%-20.5%]; P<0.001). Patients with anterior compared with nonanterior STEMI were at higher risk for 1-year all-cause mortality (6.2% versus 3.6%; adjusted hazard ratio [HR], 1.66 [95% CI, 1.02-2.69]; P=0.04) and heart failure hospitalization (4.4% versus 2.6%; adjusted HR, 1.96 [95% CI, 1.15-3.36]; P=0.01). Infarct size was a predictor of subsequent all-cause mortality or heart failure hospitalization in anterior STEMI (adjusted HR per 1% increase, 1.05 [95% CI, 1.03-1.07]; P<0.001), but not in nonanterior STEMI (adjusted HR, 1.02 [95% CI, 0.99-1.05]; P=0.19). The P value for this interaction was 0.04. CONCLUSIONS: Anterior STEMI was associated with substantially greater myonecrosis after primary percutaneous coronary intervention compared with nonanterior STEMI, contributing in large part to the worse prognosis in patients with anterior infarction.

Asundexian versus Apixaban in Patients with Atrial Fibrillation.

BACKGROUND: Stroke prevention with direct-acting oral anticoagulant agents in patients with atrial fibrillation confers a risk of bleeding and limits their use. Asundexian, an activated factor XI (XIa) inhibitor, is an oral anticoagulant that may prevent strokes with less bleeding. METHODS: In a phase 3, international, double-blind trial, we randomly assigned high-risk patients with atrial fibrillation in a 1:1 ratio to receive asundexian at a dose of 50 mg once daily or standard-dose apixaban. The primary efficacy objective was to determine whether asundexian is at least noninferior to apixaban for the prevention of stroke or systemic embolism. The primary safety objective was to determine whether asundexian is superior to apixaban with respect to major bleeding events. RESULTS: A total of 14,810 randomly assigned patients were included in the intention-to-treat population. The mean (±SD) age of the patients was 73.9±7.7 years, 35.2% were women, 18.6% had chronic kidney disease, 18.2% had a previous stroke or transient ischemic attack, 16.8% had received oral anticoagulants for no more than 6 weeks, and the mean CHA2DS2-VASc score (range, 0 to 9, with higher scores indicating a greater risk of stroke) was 4.3±1.3. The trial was stopped prematurely at the recommendation of the independent data monitoring committee. Stroke or systemic embolism occurred in 98 patients (1.3%) assigned to receive asundexian and in 26 (0.4%) assigned to receive apixaban (hazard ratio, 3.79; 95% confidence interval [CI], 2.46 to 5.83). Major bleeding occurred in 17 patients (0.2%) who received asundexian and in 53 (0.7%) who received apixaban (hazard ratio, 0.32; 95% CI, 0.18 to 0.55). The incidence of any adverse event appeared to be similar in the two groups. CONCLUSIONS: Among patients with atrial fibrillation at risk for stroke, treatment with asundexian at a dose of 50 mg once daily was associated with a higher incidence of stroke or systemic embolism than treatment with apixaban in the period before the trial was stopped prematurely. There were fewer major bleeding events with asundexian than with apixaban during this time. (Funded by Bayer; OCEANIC-AF ClinicalTrials.gov number, NCT05643573; EudraCT number, 2022-000758-28.).

Heart Failure Risk Assessment Using Biomarkers in Patients With Atrial Fibrillation: Analysis From COMBINE-AF.

BACKGROUND: Heart failure (HF) is common among patients with atrial fibrillation (AF), and accurate risk assessment is clinically important. OBJECTIVES: The goal of this study was to investigate the incremental prognostic performance of N-terminal pro-B-type natriuretic peptide (NT-proBNP), high-sensitivity cardiac troponin T (hs-cTnT), and growth differentiation factor (GDF)-15 for HF risk stratification in patients with AF. METHODS: Individual patient data from 3 large randomized trials comparing direct oral anticoagulants (DOACs) with warfarin (ARISTOTLE [Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation], ENGAGE AF-TIMI 48 [Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation-Thrombolysis In Myocardial Infarction 48], and RE-LY [Randomized Evaluation of Long-Term Anticoagulation Therapy]) from the COMBINE-AF (A Collaboration Between Multiple Institutions to Better Investigate Non-Vitamin K Antagonist Oral Anticoagulant Use in Atrial Fibrillation) cohort were pooled; all patients with available biomarkers at baseline were included. The composite endpoint was hospitalization for HF (HHF) or cardiovascular death (CVD), and secondary endpoints were HHF and HF-related death. Cox regression was used, adjusting for clinical factors, and interbiomarker correlation was addressed using weighted quantile sum regression analysis. RESULTS: In 32,041 patients, higher biomarker values were associated with a graded increase in absolute risk for CVD/HHF, HHF, and HF-related death. Adjusting for clinical variables and all biomarkers, NT-proBNP (HR per 1 SD: 1.68; 95% CI: 1.59-1.77), hs-cTnT (HR: 1.39; 95% CI: 1.33-1.44), and GDF-15 (HR: 1.20; 95% CI: 1.15-1.25) were significantly associated with CVD/HHF. The discrimination of the clinical model improved significantly upon addition of the biomarkers (c-index: 0.70 [95% CI: 0.69-0.71] to 0.77 [95% CI: 0.76-0.78]; likelihood ratio test, P < 0.001). Using weighted quantile sum regression analysis, the contribution to risk assessment was similar for NT-proBNP and hs-cTnT for CVD/HHF (38% and 41%, respectively); GDF-15 provided a statistically significant but lesser contribution to risk assessment. Results were similar for HHF and HF-related death, individually, and across key subgroups of patients based on a history of HF, AF pattern, and reduced or preserved left ventricular ejection fraction. CONCLUSIONS: NT-proBNP, hs-cTnT, and GDF-15 contributed significantly and independently to the risk stratification for HF endpoints in patients with AF, with hs-cTnT being as important as NT-proBNP for HF risk stratification. Our findings support a possible future use of these biomarkers to distinguish patients with AF at low or high risk for HF.

Safety and Feasibility of an Implanted Inferior Vena Cava Sensor for Accurate Volume Assessment: FUTURE-HF2 Trial.

BACKGROUND: A novel implantable sensor has been designed to measure the inferior vena cava (IVC) area accurately so as to allow daily monitoring of the IVC area and collapse to predict congestion in heart failure (HF). METHODS: A prospective, multicenter, single-arm, Early Feasibility Study enrolled 15 patients with HF (irrespective of ejection fraction) and with an HF event in the previous 12 months, an elevated NT-proBNP level, and receiving ≥ 40 mg of furosemide equivalent. Primary endpoints included successful deployment without procedure-related (30 days) or sensor-related complications (3 months) and successful data transmission to a secure database (3 months). Accuracy of sensor-derived IVC area, patient adherence, NYHA classification, and KCCQ were assessed from baseline to 3 months. Patient-specific signal alterations were correlated with clinical presentation to guide interventions. RESULTS: Fifteen patients underwent implantation: 66 ± 12 years; 47% female; 27% with HFpEF, NT-ProBNP levels 2569 (median, IQR: 1674-5187, ng/L; 87% NYHA class III). All patients met the primary safety and effectiveness endpoints. Sensor-derived IVC areas showed excellent agreement with concurrent computed tomography (R2 = 0.99, mean absolute error = 11.15 mm2). Median adherence to daily readings was 98% (IQR: 86%-100%) per patient-month. A significant improvement was seen in NYHA class and a nonsignificant improvement was observed in KCCQ. CONCLUSIONS: Implantation of a novel IVC sensor (FIRE1) was feasible, uncomplicated and safe. Sensor outputs aligned with clinical presentations and improvements in clinical outcomes. Future investigation to establish the IVC sensor remote management of HF is strongly warranted.

Left Ventricular Unloading Using Intra-aortic Entrainment Pumping Before Reperfusion Reduces Post-AMI Infarct Size.

BACKGROUND: Anterior myocardial infarction standard of care prioritizes swift coronary reperfusion. Recent studies show left ventricular (LV) unloading with transvalvular axial flow pumps for 30 minutes before reperfusion (vs immediate reperfusion) decreases 28-day infarct size. Intra-aortic entrainment pumping, using hardware located away from the heart to provide support throughout the cardiac cycle, decreases effective systemic vascular resistance and augments visceral blood flow and pressure, and may reproduce this benefit with a decreased risk. This study characterized the hemodynamic effects of unloading before and during reperfusion using intra-aortic entrainment pumping and investigated whether unloading decreased anterior myocardial infarction scar size. METHODS AND RESULTS: Yorkshire swine were subjected to 90 minutes of left anterior descending artery balloon occlusion and randomly assigned to immediate reperfusion (n = 6) vs 30 minutes unloading before reperfusion followed by 120 minutes of further unloading (n = 7). Unloading was achieved using percutaneous entrainment pumping in the descending aorta. The anterior myocardial infarction model matches that used in recent transvalvular pumping studies. Mortality before randomization was 22%. After randomization, mortality was 36% for immediate reperfusion and 0% for unloading. Unloading showed immediate hemodynamic benefit that increased through reperfusion and continued support, leading to distinct differences in cardiac function between groups after 30 minutes of reperfusion. Unloading increased stroke volume and cardiac efficiency at this timepoint relative to preocclusion baseline and reduced 28-day LV scar size by 37%-45%. CONCLUSIONS: We present the first preclinical data showing extracardiac LV unloading before coronary reperfusion using intra-aortic entrainment pumping decreases 28-day infarct size. Extracardiac unloading to decrease LV scar size may provide an alternative to transvalvular pumping with potential advantages, including reduced risk.

Carbon emissions and air pollution savings among telehealth visits for cardiology appointments.

Background: Climate change has been associated with adverse cardiovascular health, prompting interest in climate mitigation strategies while improving access for cardiovascular patients. We estimated greenhouse gas and air pollution savings from telehealth use in cardiology. Methods: Using cardiology telehealth visits at a large academic medical center from July 2020 to March 2024, carbon dioxide (CO2), nitrogen oxides (NOx), carbon monoxide (CO), and particulate matter (PM2.5) emissions saved were calculated using U.S. Environmental Protection Agency modeling software. Savings were converted into real-world comparators and differences were assessed by cardiology subspecialty and patient insurance status. Results: Over 45 months, 14,828 telehealth visits among 9942 patients resulted in savings of 484,152 kg of CO2, 5225 kg of CO, 243,491 g of NOx, and 9091 g of PM2.5 with the total carbon saved equivalent to planting 9070 tree saplings over ten years. CO2 emissions saved per visit (kg) differed significantly by payor (Self-pay 24.99, Medicare 19.67, Medicaid 19.54, Private 17.85, Other 17.37, p = 0.004) and by subspecialty (Interventional 23.79, General 19.08, Heart Failure 18.86, Electrophysiology 17.81, Adult Congenital 16.59, p < 0.001). Conclusions: Carbon emission and air pollution savings from telehealth in cardiology were substantial, with an estimated 19.06 kg of CO2 saved per visit and total savings over 45 months equivalent to planting over nine thousand trees.

Low-Dose Rivaroxaban Plus Aspirin in Fragile Patients After Lower Extremity Revascularization.

BACKGROUND: Rivaroxaban 2.5 mg plus aspirin reduced limb and cardiovascular events and increased bleeding in patients with symptomatic peripheral artery disease (PAD) after lower extremity revascularization in the VOYAGER PAD (Efficacy and Safety of Rivaroxaban in Reducing the Risk of Major Thrombotic Vascular Events in Subjects With Symptomatic Peripheral Artery Disease Undergoing Peripheral Revascularization Procedures of the Lower Extremities) study. Fragile patients are at heightened risk for ischemic and bleeding events. OBJECTIVES: The purpose of this study was to investigate the safety and efficacy of rivaroxaban 2.5 mg in fragile patients from VOYAGER PAD. METHODS: Patients were categorized as fragile based on prespecified criteria (age >75 years, weight ≤50 kg, or baseline estimated glomerular filtration rate <50 mL/min/1.73 m2). The primary efficacy outcome was the composite of acute limb ischemia, major amputation of a vascular etiology, myocardial infarction, ischemic stroke, or cardiovascular death. The principal safety outcome was TIMI major bleeding. RESULTS: Of 6,564 randomized patients, a total of 1,674 subjects were categorized as fragile at baseline. In the placebo arm, fragile patients were at higher risk of the primary outcome (HR: 1.34; 95% CI: 1.12-1.61) and TIMI major bleeding (HR: 1.57; 95% CI: 0.83-2.96), compared with nonfragile patients. The effect of rivaroxaban on the primary endpoint was not modified by frailty status (fragile HR: 0.93; 95% CI: 0.75-1.15; nonfragile HR: 0.83; 95% CI: 0.72-0.97; P interaction = 0.37). Rivaroxaban increased TIMI major bleeding in fragile (HR: 1.54; 95% CI: 0.82-2.91) and nonfragile patients (HR: 1.37; 95% CI: 0.84-2.23; P interaction = 0.65). CONCLUSIONS: Patients with PAD after lower extremity revascularization meeting fragile criteria are at higher risk of ischemic complications and bleeding. Rivaroxaban reduces ischemic risk and increases bleeding regardless of frailty status. These data may assist in personalization of antithrombotic therapy in fragile population.

Effect of Peripheral Interventions in Patients with Peripheral Artery Disease Receiving Rivaroxaban and Aspirin: Analyses from the XATOA Registry.

OBJECTIVE: To assess the characteristics and clinical outcomes of patients with lower extremity peripheral artery disease (PAD) in XATOA receiving dual pathway inhibition (DPI) with rivaroxaban 2.5 mg twice daily plus aspirin according to lower extremity revascularisation (LER) history. METHODS: XATOA is an international, multicentre, prospective, single arm registry study. This subanalysis investigated patients with lower extremity PAD according to LER history. Patients with coronary artery disease, PAD, or both, receiving DPI were followed for 12 or more months. Baseline characteristics and clinical outcomes were assessed according to LER history. A time dependency analysis assessed outcomes by time between the most recent LER procedure and the start of DPI. A multivariable analysis assessed the influence of patient characteristics on clinical outcomes. RESULTS: In XATOA (n = 5 532), 2 820 (51.0%) patients had lower extremity PAD, of whom 1 736 (61.6%) had prior LER and 1 084 (38.4%) had no prior LER. Baseline characteristics were generally similar between patients with or without prior LER. A higher proportion of patients with prior LER experienced any treatment emergency clinical events compared with those without prior LER (15.0% vs. 9.4%, respectively), with greater differences observed between incidence rates of limb events, including major adverse limb events (9.06 vs. 4.09 events per 100 patient years, respectively). Similar rates of myocardial infarction, stroke, and major bleeding were observed in both subgroups. Clinical event rates were generally higher in patients who had previous LER for six months or less compared with patients who had previous LET for more than six months before starting DPI, regardless of LER type. Multivariable analyses showed that prior LER was predictive of limb events. CONCLUSION: This subanalysis of XATOA found that prior LER was associated with increased rates of limb events, consistent with results of COMPASS and VOYAGER PAD. Rates of bleeding were also low regardless of LER history and consistent with the findings from these trials.

Completeness of revascularization by FFRCT in stable angina: Association to adverse cardiovascular outcomes.

BACKGROUND: The prognostic impact of complete coronary revascularization relative to non-invasive testing methods is unknown. OBJECTIVES: To assess the association between completeness of revascularization defined by CTA-derived fractional flow reserve (FFRCT) and cardiovascular outcomes in patients with stable angina. METHODS: Multicenter 3-year follow-up study of patients with new onset stable angina and ≥ 30% stenosis by CTA. The lesion-specific FFRCT value (two cm-distal-to-stenosis) was registered in all vessels with stenosis and considered abnormal when ≤ 0.80. Patients with FFRCT ≤ 0.80 were categorized as: Completely revascularized (CR-FFRCT), all vessels with FFRCT ≤ 0.80 revascularized; incompletely revascularized (IR-FFRCT), ≥ 1 vessels with FFRCT ≤ 0.80 non-revascularized. Early revascularization (< 90 days from index CTA) categorized vessels as revascularized. The primary endpoint comprised cardiovascular death and non-fatal myocardial infarction; the secondary endpoint vessel-specific late revascularization and non-fatal myocardial infarction. RESULTS: Amongst 900 patients and 1759 vessels, FFRCT was ≤ 0.80 in 377 (42%) patients, 536 (30%) vessels; revascularization was performed in 244 (27%) patients, 340 (19%) vessels. Risk of the primary endpoint was higher for IR-FFRCT (15/210 [7.1%]) compared to CR-FFRCT (4/167 [2.4%]), RR: 2.98; 95% CI: 1.01-8.8, p â€‹= â€‹0.036, and to normal FFRCT (3/523 [0.6%]), RR: 12.45; 95% CI: 3.6-42.6, p â€‹< â€‹0.001. Incidence of the secondary endpoint was higher in non-revascularized vessels with FFRCT ≤ 0.80 (29/250 [12%]) compared to revascularized vessels with FFRCT ≤ 0.80 (5/286 [1.7%]), p â€‹= â€‹0.001, and to vessels with FFRCT > 0.80 (10/1223 [0.8%]), p â€‹< â€‹0.001. CONCLUSION: Incomplete revascularization of patients with lesion-specific FFRCT ≤ 0.80 is associated to unfavorable cardiovascular outcomes compared to those with complete revascularization or FFRCT > 0.80.

Framework of the strengths and challenges of clinically integrated trials: An expert panel report.

The limitations of the explanatory clinical trial framework include the high expense of implementing explanatory trials, restrictive entry criteria for participants, and redundant logistical processes. These limitations can result in slow evidence generation that is not responsive to population health needs, yielding evidence that is not generalizable. Clinically integrated trials, which integrate clinical research into routine care, represent a potential solution to this challenge and an opportunity to support learning health systems. The operational and design features of clinically integrated trials include a focused scope, simplicity in design and requirements, the leveraging of existing data structures, and patient participation in the entire trial process. These features are designed to minimize barriers to participation and trial execution and reduce additional research burdens for participants and clinicians alike. Broad adoption and scalability of clinically integrated trials are dependent, in part, on continuing regulatory, healthcare system, and payer support. This analysis presents a framework of the strengths and challenges of clinically integrated trials and is based on a multidisciplinary expert "Think Tank" panel discussion that included representatives from patient populations, academia, non-profit funding agencies, the U.S. Food and Drug Administration, and industry.

Assessment of Days Alive Out of Hospital as a Possible End Point in Trials of Stroke Prevention for Atrial Fibrillation: A ROCKET AF Analysis.

BACKGROUND: Days alive out of hospital (DAOH) is an objective and patient-centered net benefit end point. There are no assessments of DAOH in clinical trials of interventions for atrial fibrillation (AF), and it is not known whether this end point is of clinical utility in these populations. METHODS AND RESULTS: ROCKET AF (Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared With Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation) was an international double-blind, double-dummy randomized clinical trial that compared rivaroxaban with warfarin in patients with atrial fibrillation at increased risk for stroke. We assessed DAOH using investigator-reported event data for up to 12 months after randomization in ROCKET AF. We assessed DAOH overall, by treatment group, and by subgroup, including age, sex, and comorbidities, using Poisson regression. The mean±SD number of days dead was 7.3±41.2, days hospitalized was 1.2±7.2, and mean DAOH was 350.7±56.2, with notable left skew. Patients with comorbidities had fewer DAOH overall. There were no differences in DAOH by treatment arm, with mean DAOH of 350.6±56.5 for those randomized to rivaroxaban and 350.7±55.8 for those randomized to warfarin (P=0.86). A sensitivity analysis found no difference in DAOH not disabled with rivaroxaban versus warfarin (DAOH not disabled, 349.2±59.5 days and 349.1 days±59.3 days, respectively, P=0.88). CONCLUSIONS: DAOH did not identify a treatment difference between patients randomized to rivaroxaban versus warfarin. This may be driven in part by the low overall event rates in atrial fibrillation anticoagulation trials, which leads to substantial left skew in measures of DAOH.

Frameworks for procurement, integration, monitoring, and evaluation of artificial intelligence tools in clinical settings: A systematic review.

Research on the applications of artificial intelligence (AI) tools in medicine has increased exponentially over the last few years but its implementation in clinical practice has not seen a commensurate increase with a lack of consensus on implementing and maintaining such tools. This systematic review aims to summarize frameworks focusing on procuring, implementing, monitoring, and evaluating AI tools in clinical practice. A comprehensive literature search, following PRSIMA guidelines was performed on MEDLINE, Wiley Cochrane, Scopus, and EBSCO databases, to identify and include articles recommending practices, frameworks or guidelines for AI procurement, integration, monitoring, and evaluation. From the included articles, data regarding study aim, use of a framework, rationale of the framework, details regarding AI implementation involving procurement, integration, monitoring, and evaluation were extracted. The extracted details were then mapped on to the Donabedian Plan, Do, Study, Act cycle domains. The search yielded 17,537 unique articles, out of which 47 were evaluated for inclusion based on their full texts and 25 articles were included in the review. Common themes extracted included transparency, feasibility of operation within existing workflows, integrating into existing workflows, validation of the tool using predefined performance indicators and improving the algorithm and/or adjusting the tool to improve performance. Among the four domains (Plan, Do, Study, Act) the most common domain was Plan (84%, n = 21), followed by Study (60%, n = 15), Do (52%, n = 13), & Act (24%, n = 6). Among 172 authors, only 1 (0.6%) was from a low-income country (LIC) and 2 (1.2%) were from lower-middle-income countries (LMICs). Healthcare professionals cite the implementation of AI tools within clinical settings as challenging owing to low levels of evidence focusing on integration in the Do and Act domains. The current healthcare AI landscape calls for increased data sharing and knowledge translation to facilitate common goals and reap maximum clinical benefit.

Hypertension, microvascular obstruction and infarct size in patients with STEMI undergoing PCI: Pooled analysis from 7 cardiac magnetic resonance imaging studies.

BACKGROUND: Mortality after ST-segment elevation myocardial infarction (STEMI) is increased in patients with hypertension. The mechanisms underlying this association are uncertain. We sought to investigate whether patients with STEMI and prior hypertension have greater microvascular obstruction (MVO) and infarct size (IS) compared with those without hypertension. METHODS: We pooled individual patient data from 7 randomized trials of patients with STEMI undergoing primary percutaneous coronary intervention (PCI) in whom cardiac magnetic resonance imaging was performed within 1 month after reperfusion. The associations between hypertension and MVO, IS, and mortality were assessed in multivariable adjusted models. RESULTS: Among 2174 patients (61.3 ± 12.6 years, 76% male), 1196 (55.0%) had hypertension. Patients with hypertension were older, more frequently diabetic and had more extensive coronary artery disease than those without hypertension. MVO and IS measured as percent LV mass were not significantly different in patients with and without hypertension (adjusted differences 0.1, 95% CI -0.3 to 0.6, P = .61 and -0.2, 95% CI -1.5 to 1.2, P = .80, respectively). Hypertension was associated with a higher unadjusted risk of 1-year death (hazard ratio [HR] 2.28, 95% CI 1.44-3.60, P < .001), but was not independently associated with higher mortality after multivariable adjustment (adjusted HR 1.04, 95% CI 0.60-1.79, P = .90). CONCLUSION: In this large-scale individual patient data pooled analysis, hypertension was not associated with larger IS or MVO after primary PCI for STEMI.