PURPOSE:To evaluate coronary microvascular function using 13N-ammonia positron emission tomography/ computed tomography in individuals with pathogenic transthyretin (TTR) gene mutations, with and without cardiac involvement. This study is the first to assess coronary flow reserve (CFR) in this population before overt cardiac amyloidosis (CA) is detectable by conventional imaging. METHODS:We evaluated microvascular impairment by measuring CFR in 20 patients with and 20 patients without cardiac involvement due to TTR amyloidosis (ATTR), all presumed to be free from epicardial coronary artery disease and carrying TTR gene mutations. RESULTS:The study revealed a significantly reduced mean global CFR in the cardiac involvement group (1.849 ± 0.379 vs. 2.952 ± 0.7, P < 0.001). Global CFR inversely correlated with age, functional class, troponin, and B-type natriuretic peptide while positively correlating with the 6-minute walk test distance, mean blood pressure, and global longitudinal strain. Receiver operating characteristic curve analysis identified an optimal cutoff value of global CFR < 2.58, yielding a sensitivity of 100% and a specificity of 75% for detecting cardiac involvement. CONCLUSION:In patients with ATTR CA, coronary microvascular dysfunction emerges as a clinically relevant marker of cardiac involvement, even in the absence of structural abnormalities or obstructive coronary disease. CLINICAL SIGNIFICANCE:CFR assessment may aid in diagnostic suspicion, risk stratification, and understanding of angina symptoms in this population.

In transthyretin cardiac amyloidosis (ATTR-CA), static 18F-flutemetamol PET-derived SUVs do not adequately capture intricate tracer kinetics, limiting the accurate quantification of amyloid burden. We developed dynamic parametric PET imaging methods to improve the quantification of myocardial amyloid burden in ATTR-CA. Methods: Twelve treatment-naïve ATTR-CA patients underwent 60-min dynamic cardiac 18F-flutemetamol PET/CT at baseline and after 6 mo of treatment with tafamidis. Image-derived input functions were corrected for blood-to-plasma ratios and metabolites. Myocardium blood volume fraction was estimated using a 1-tissue compartment model (0-10 min), and volume-of-distribution (V T) images were generated using the multilinear analysis 1 method (2-20 min). V T was correlated with echocardiography, 82Rb myocardial blood flow, and biomarkers. Results: The mean myocardial blood volume fraction was 22% ± 6%. A 2-tissue reversible model with metabolite- and plasma-corrected input functions provided optimal kinetic fits. Multilinear analysis 1 using 2- to 20-min PET data produced V T images with the lowest variance. V T decreased significantly after 6 mo of treatment with tafamidis (from 2.11 ± 0.33 to 1.96 ± 0.20, P = 0.046). Conclusion: Dynamic 18F-flutemetamol PET enabled robust quantification of myocardial amyloid burden using metabolite-corrected 2-compartment modeling. V T imaging demonstrated sensitivity to treatment-related changes in ATTR-CA.

BACKGROUND:Aortic enlargement is a powerful predictor of dissection and rupture, yet it is rarely evaluated during routine myocardial perfusion imaging, despite the widespread availability of computed tomography (CT) attenuation correction scans. The aim of this study was to determine whether fully automated, opportunistically derived, artificial intelligence-based aortic measurements from myocardial perfusion imaging CT attenuation correction scans are associated with adverse outcomes in a large multicenter cohort. METHODS:CT attenuation correction scans from patients undergoing positron emission tomography/CT and single-photon emission CT/CT myocardial perfusion imaging across 10 centers were included. A deep learning model automatically segmented the thoracic aorta, and a postprocessing algorithm extracted maximum ascending and descending diameters. The aortic size index was calculated by indexing the diameter to body surface area. RESULTS:A total of 29 339 patients (56% men; median age, 66 years [interquartile range, 58-75 years]) were included. Over a median follow-up of 3.5 years (interquartile range, 1.9-5.0 years), 5083 (17.3%) patients died. Median ascending and descending aortic size index values were 1.8 cm/m2 (interquartile range, 1.6-2.0) and 1.5 cm/m2 (interquartile range, 1.4-1.6), respectively, with an increase with age and higher values in females. Elevated aortic size index thresholds (ascending >2.2 cm/m2; descending >1.6 cm/m2) were significantly associated with increased all-cause mortality (ascending: adjusted hazard ratio, 1.16 [95% CI, 1.07-1.26], P<0.001; descending: adjusted hazard ratio, 1.23 [95% CI, 1.14-1.31]; P<0.001). Notably, the prognostic value of an abnormal aortic size index persisted independent of age, sex, and perfusion abnormalities. CONCLUSIONS:Artificial intelligence can unlock previously unused information within routine myocardial perfusion imaging CT attenuation correction scans by rapidly and automatically quantifying aortic size at scale. Opportunistic aortic measurements derived from CT attenuation correction may serve as an adjunctive risk biomarker and could add prognostic value to standard myocardial perfusion imaging without additional imaging or radiation.

Introduction: Transthoracic echocardiography (TTE) is the standard-of-care imaging modality for management of heart failure (HF). Transthyretin amyloid cardiomyopathy (ATTR-CM) is an important cause of HF that can be difficult to identify and follow longitudinally by TTE, owing to imprecision of measurements and variability in interpretation. Analysis of standard TTE variables using Artificial Intelligence (AI) could improve precision and reproducibility while reducing interpretation time. Aims: To assess performance of a previously validated, fully automated AI algorithm for TTE interpretation (Us2.ai) in patients with heart failure at risk for ATTR-CM. We hypothesized that AI measurements would be strongly correlated with human interpretation and identify differences between participants with and without ATTR-CM. Methods: TTE images from participants in the prospective Screening for Cardiac Amyloidosis with Nuclear Imaging in Minority Populations (SCAN-MP) study (n = 586, 36 ATTR-CM cases) were analyzed independently by a human expert and an AI algorithm (Us2.ai). Intraclass correlation and Bland-Altman agreement between human and AI were assessed for clinically relevant TTE variables including left ventricular ejection fraction (LVEF), global longitudinal strain (GLS), maximal wall thickness (MWT), and mitral E/e’ ratio. Additionally, human and AI parameter distributions were compared between participants with and without ATTR-CM. Results: Comparison of human and AI distributions demonstrated differences in participants with ATTR-CM for most variables, with similar magnitude and directionality (Table 1) . Correlation between AI and human measurements was high for the Doppler measure E/e’ (ICC: 0.89, 95% CI: [0.86-0.90]) and moderate for LVEF [0.72 (0.27-0.86)], MWT [0.68 (0.63-0.73)], and GLS [0.64 (0.54-0.72)]. Similarly, Bland-Altman agreement was highest for E/e’, while LVEF, GLS, and MWT had relatively wider limits of agreement and greater bias. AI underestimated most parameters, including LVEF (bias AI -5.3%), MWT (bias AI -0.15cm), and E/e’ (bias AI -0.49), but overestimated GLS (bias AI +1.1%). Conclusions: Fully automated TTE measurements were correlated with a human reference and reproduced differences between ATTR-CM cases and controls. AI TTE interpretation is a promising tool to democratize echocardiography for clinical care and research, and may be interchangeable with human readers for some measurements relevant to ATTR-CM.

Cardiac imaging is a cornerstone in the initial diagnosis, management, and follow-up of cardiac sarcoidosis. However, ordering thresholds, access, and follow-up imaging vary across the globe. A Delphi study was conducted to define areas of consensus and areas requiring further study in the use of cardiac imaging in suspected or established cardiac sarcoidosis. An international, multidisciplinary panel of experts in cardiac sarcoidosis completed a modified 2-round Delphi study. The study evaluated clinical decision making regarding the use of cardiac imaging, including indication thresholds, interpretation, and interval follow-up imaging. Consensus was defined a priori as ≥70% agreement or disagreement. A total of 89 experts in cardiac sarcoidosis (89 in round 1 and 75 in round 2) participated, representing 61 centers in 13 countries. Consensus was reached on 22 of 46 items (48%) in round 1 and 21 of 29 items (72%) in round 2. There was a low threshold to order advanced cardiac imaging for new rhythm abnormalities or ventricular dysfunction detected on echocardiography in patients with established extracardiac sarcoidosis. 18F-fluorodeoxyglucose (FDG) positron emission tomography was an important co-primary modality with cardiac magnetic resonance (CMR) for initial diagnosis. If CMR was the first test, there was consensus to proceed to FDG-positron emission tomography after any abnormal CMR result or even after normal CMR result in the setting of moderate or high pretest probability for cardiac sarcoidosis. There was consensus that late gadolinium enhancement quantification was important, but there was no consensus on the threshold of risk or on how best to quantify late gadolinium enhancement. Similarly, reduction in FDG uptake was an important factor in guiding treatment response, but there was no consensus on how to best quantify FDG uptake or what constituted an adequate radiographic response. Several consensus areas for cardiac imaging in suspected and established cardiac sarcoidosis were identified. This consensus study identified areas of priority for future prospective, controlled, multicenter research studies.

BACKGROUND:Insufficient health literacy negatively impacts outcomes for heart failure (HF). Older adults with HF face significant barriers, including prevalence of multiple comorbidities, frailty and deficits in physical function, which can impact HF outcomes. Examination of the association between health literacy and physical function remains limited in older adults with HF. We investigated the cross-sectional associations of health literacy and physical function in a cohort of older self-identified Black and Hispanic patients with HF. METHODS AND RESULTS:The Screening for Cardiac Amyloidosis with Nuclear Imaging in Minority Populations (SCAN-MP) study recruited participants (age ≥ 60 years) of self-identified Black race or Hispanic ethnicity in New York City, Boston and New Haven. We measured health literacy by using the Newest Vital Sign, and we measured physical function by using the 6-minute walk duration (6MWD) and the Short Physical Performance Battery test (SPPB). Linear regression models evaluated the association between participants' health-literacy levels and physical function, as defined by continuous measures of 6MWD and SPPB. Only 12.7% of the cohort (n = 433) had adequate health literacy. After adjusting for age, sex, number of comorbidities, and neighborhood social vulnerability, those with adequate (compared to limited) health literacy demonstrated better performance on the 6MWD (β = 37.5 m, 95% CI [1.49, 73.5]; P = 0.04) and SPPB (β = 1.18 (0.41, 1.95); P = 0.002). CONCLUSION:Older Black or Hispanic participants with HF have an extremely high prevalence of limited health literacy, which is associated with poor physical function. Interventions targeting health literacy may represent an avenue to improve HF outcomes for patients with minoritized backgrounds.

We previously demonstrated that a deep learning (DL) model of myocardial perfusion SPECT imaging improved accuracy for detection of obstructive coronary artery disease (CAD). We aimed to improve the clinical translatability of this artificial intelligence (AI) approach using the results to derive enhanced total perfusion deficit (TPD) and 17-segment summed scores. Methods: We used a cohort of patients undergoing myocardial perfusion imaging within 180 d of invasive coronary angiography. Obstructive CAD was defined as any stenosis of at least 70% or at least 50% in the left main coronary artery. We used per-vessel DL predictions to modulate polar map pixel scores. These transformed polar maps were then used to derive TPD-DL and summed stress score-DL. We compared diagnostic performance using area under the receiver operating characteristic curve (AUC). Results: In the 555 patients held out for testing, the median age was 65 y (interquartile range, 57-73 y), and 381 (69%) were male. Obstructive CAD was present in 329 (59%) patients. The prediction performance for obstructive CAD of stress TPD-DL (AUC, 0.837; 95% CI, 0.804-0.870) was higher than AI prediction alone (AUC, 0.795; 95% CI, 0.758-0.831; P = 0.005) and traditional stress TPD (AUC, 0.737; 95% CI, 0.696-0.778; P < 0.001). Summed stress score-DL had the second highest prediction performance (AUC, 0.822; 95% CI, 0.788-0.857) and higher AUC than traditional quantitative summed stress score (AUC, 0.728; 95% CI, 0.686-0.769; P < 0.001). At a threshold of 5%, the sensitivity and specificity of TPD rose from 72% to 79% and from 62% to 70%, respectively. Conclusion: Integrating AI predictions with traditional quantitative approaches leads to a simplified AI approach, presenting clinicians with familiar measures but operating with higher accuracy than traditional quantitative scoring. This approach may facilitate integration of new AI methods into clinical practice.

Myocardial perfusion imaging (MPI) is frequently used to improve cardiac risk prediction in patients undergoing non-cardiac surgery. However, the data supporting this practice is derived from single center studies and predates current therapy for coronary artery disease (CAD). We evaluated the association between pre-operative testing indication and outcomes as well as predictors of death or myocardial infarction (MI) in these patients. We include patients from the international, multicenter REFINE-SPECT registry (13 sites). Based on referral indication, patients were classified as pre-operative testing indications and non-pre-operative testing. We evaluated the associations between pre-operative testing and incidence of death or MI. We then evaluated associations with death or MI in patients referred for pre-operative testing compared to other patients. In total, 32,711 patients were included with pre-operative testing as the indication in 2,173(6.6

Low-dose computed tomography attenuation correction (CTAC) scans are used in hybrid myocardial perfusion imaging (MPI) for attenuation correction and coronary calcium scoring, and contain additional anatomic and pathologic information not utilized in clinical assessment. We seek to uncover the full potential of these scans utilizing a holistic artificial intelligence (AI) approach. A multi-structure model segmented 33 structures and quantified 15 radiomics features in each organ in 10,480 patients from 4 sites. Coronary calcium and epicardial fat measures were obtained from separate AI models. The area under the receiver-operating characteristic curves (AUC) for all-cause mortality prediction of the model utilizing MPI, CT, stress test, and clinical features was 0.80 (95% confidence interval [0.74–0.87]), which was higher than for coronary calcium (0.64 [0.57–0.71]) or perfusion (0.62 [0.55–0.70]), with p < 0.001 for both. A comprehensive multimodality approach can significantly improve mortality prediction compared to MPI information alone in patients undergoing hybrid MPI.

Importance:Transthyretin cardiac amyloidosis (ATTR-CA) is an underdiagnosed but treatable cause of heart failure (HF) in older individuals that occurs in the context of normal wild-type (ATTRwt-CA) or an abnormal inherited (ATTRv-CA) TTR gene variant. While the most common inherited TTR variant, V142I, occurs in 3% to 4% of self-identified Black Americans and is associated with excess morbidity and mortality, the prevalence of ATTR-CA in this at-risk population is unknown. Objective:To define the prevalence of ATTR-CA and proportions attributable to ATTRwt-CA or ATTRv-CA among older Black and Caribbean Hispanic individuals with HF. Design, Setting, and Participants:This prospective, multicenter, cross-sectional study was conducted in several major US cities (Boston, Massachusetts; New York, New York; and New Haven, Connecticut) among individuals who self-identified as Black or Caribbean Hispanic older than 60 years with HF. Participants were enrolled between May 2019 and June 2024, and data analysis was conducted from June 2024 to May 2025. Main Outcomes and Measures:ATTR-CA was determined by radionuclide imaging, with blood testing to exclude light-chain amyloidosis and genotyping to determine TTR gene variant. Echocardiographic, biochemical, physical performance, and quality-of-life data were collected. Results:Among 646 participants, median (IQR) participant age was 73 (66-80) years, 329 (50.6%) were women, 550 (85.1%) identified as Black, and 186 (28.8%) identified as Caribbean Hispanic. Median (IQR) left ventricular wall thickness was 13 (12-14) mm, and median (IQR) left ventricular ejection fraction was 61% (55%-66%). Overall prevalence of ATTR-CA was 6.66% (95% CI, 4.73%-8.58%), of whom 24 (55.8%) had ATTRwt-CA and 19 (44.2%) had ATTRv-CA owing to V142I. Overall prevalence of V142I allele was 5.6%, and of those, 19 (52.8%) had ATTRv-CA. Prevalence of ATTR-CA was 8.15% (95% CI, 5.15%-11.15%) in men and 5.20% (95% CI, 2.79%-7.61%) in women (P = .13). Prevalence of ATTR-CA was 7.82% (95% CI, 5.57%-10.06%) in Black participants and 2.15% (95% CI, 0.07%-4.24%) in Hispanic participants (P = .004). Among Black participants aged 75 years or younger, ATTR-CA was observed in 3.42% of participants (95% CI, 1.43%-5.40%) compared to 14.04% (95% CI, 9.53%-18.54%) of those older than 75 years (P < .001). Among Black male participants older than 75 years, prevalence of ATTR-CA was 17.17% (95% CI, 9.74%-24.60%). Conclusions and Relevance:In this cross-sectional study, ATTR-CA was an important cause of HF in older Black individuals with HF, particularly in men older than 75 years. Approximately half of V142I carriers with HF had ATTR-CA, while 55.8% of all ATTR-CA cases had normal TTR genotype. Trial Registration:ClinicalTrials.gov Identifier: NCT03812172.

Background: Cardio-Kidney-Metabolic (CKM) Syndrome represents a progressive multisystem disorder associated with increasing cardiovascular risk. However, the evolution of coronary microvascular dysfunction across CKM stages and the prognostic value of myocardial flow reserve (MFR) within the CKM construct remain incompletely understood. Hypothesis: Stress myocardial blood flow (MBF) and MFR, as measured by 82 Rb PET/CT, progressively decline with advancing CKM stage. Abnormal MFR remains independently predictive of adverse cardiovascular outcomes. Methods: We retrospectively analyzed 5,749 total patients who underwent rest/stress cardiac 82 Rb PET/CT at our institution between 2016 and 2022. Patients were then defined as having CKM stages 0–4 according to AHA criteria. MFR was calculated as the ratio of stress to rest MBF. Comparisons of stress MBF and MFR across CKM stages were performed using ANOVA. Reduced stress MBF was defined as <1.8 mL/g/min and abnormal MFR as <2.0. Associations between CKM stage, abnormal MFR, and clinical outcomes were evaluated using unadjusted (CKM 0–4) and multivariable-adjusted (CKM 0–4, age, sex, abnormal MFR) Cox regression models. The co-primary endpoints were major adverse cardiovascular events (MACE: death, myocardial infarction, stroke, or revascularization) and heart failure or death (HF/Death). Results: Among 5,624 CKM patients (median age 54 years [IQR 55–73]; 52% women), 1,178 MACE and 1,329 HF/Death events occurred over a median follow-up of 3.9 years (IQR 2.4–5.6). Distribution across CKM stages 0 to 4 was: 47 (0.8%), 191 (3.4%), 849 (15.1%), 1,469 (26.1%), and 3,068 (54.6%), respectively. Rest MBF, stress MBF, and MFR progressively declined with advancing CKM stage (p for trend <0.001), and the prevalence of abnormal MFR increased (p for trend <0.001; Table 1). CKM stage 4 and abnormal MFR were independently associated with higher risk of MACE and HF/Death (Table 2), with event rates rising across the CKM spectrum (Figure 1). Abnormal MFR discriminated high-risk populations within each CKM stage, even in early (0 or 1) stages of CKM (Figure 1). Conclusions: CKM Syndrome is associated with worsening coronary microvascular function as reflected by PET-derived MBF and MFR. Abnormal MFR provides additional prognostic discrimination across all CKM stages and may help identify patients at particularly high cardiovascular risk.

The Society of Nuclear Medicine and Molecular Imaging (SNMMI), founded in 1954 and headquartered in Reston, Virginia, USA, is a nonprofit scientific and professional organization that promotes the science, technology, and practical application of nuclear medicine and molecular imaging. SNMMI strives to be a leader in unifying, advancing, and optimizing molecular imaging with the ultimate goal of improving human health. With 13,000 members worldwide, SNMMI represents nuclear and molecular imaging professionals, all of whom are committed to the advancement of the field. The European Association of Nuclear Medicine (EANM) is a professional, nonprofit medical association that facilitates communication worldwide between individuals pursuing clinical and research excellence in nuclear medicine. The EANM was founded in 1985. The American College of Nuclear Medicine (ACNM) is a professional organization providing education, training, and advocacy for the most sought-after and trusted experts in nuclear medicine who deliver state-of-the-art and innovative care and service to patients and referring physicians. The ACNM's mission is to foster the highest standards in nuclear medicine consultation and service to referring physicians, hospitals, and the public, and to advance the science of nuclear medicine through a program of continuing professional development emphasizing high standards of nuclear medicine practice. The American Society of Nuclear Cardiology (ASNC) is the international home for nuclear cardiology and the largest professional society devoted exclusively to the field. ASNC membership includes more than 5700 nuclear cardiology professionals from around the world. Founded in 1993, ASNC's mission is to improve cardiovascular outcomes through image-guided patient management. ASNC's official publication is the Journal of Nuclear Cardiology. The SNMMI/EANM/ACNM will periodically define new standards/guidelines for nuclear medicine practice to help advance the science of nuclear medicine and to improve the quality of service to patients. Existing standards/guidelines will be reviewed for revision or renewal, as appropriate, on their fifth anniversary or sooner, if indicated. Each standard/guideline, representing a policy statement by the SNMMI/EANM/ACNM, has undergone a thorough consensus process in which it has been subjected to extensive review. The SNMMI, EANM, and ACNM recognize that the safe and effective use of diagnostic nuclear medicine imaging and therapy requires specific training, skills, and techniques, as described in each document. This document was jointly developed with ASNC. ASNC endorses this guideline, and it will be published alongside the Journal of Nuclear Medicine in the Journal of Nuclear Cardiology. The SNMMI, EANM, ASNC, and ACNM have written and approved these standards/guidelines to promote the use of high-quality nuclear medicine procedures. These standards/guidelines are intended to assist practitioners in providing appropriate care for patients. They are not inflexible rules or requirements of practice and are not intended, nor should they be used, to establish a legal standard of care. For these reasons and those set forth below, the SNMMI, EANM, ASNC, andACNM caution against the use of these standards/guidelines in litigation in which the clinical decisions of a practitioner are called into question. The ultimate judgment regarding the propriety of any specific procedure or course of action must be made by medical professionals considering the unique circumstances of each case. Thus, there is no implication that an approach differing from the standards/guidelines, standing alone, is below the standard of care. To the contrary, a conscientious practitioner may responsibly adopt a course of action different from that set forth in the standards/guidelines when, in the reasonable judgment of the practitioner, such a course of action is indicated by the condition of the patient, limitations of available resources, or advances in knowledge or technology subsequent to publication of the standards/guidelines. The practice of medicine involves not only the science but also the art of dealing with the prevention, diagnosis, alleviation, and treatment of disease. The variety and complexity of human conditions make it impossible to always reach the most appropriate diagnosis or to predict with certainty a particular response to treatment. Therefore, it should be recognized that adherence to these standards/guidelines will not ensure an accurate diagnosis or a successful outcome. All that should be expected is that the practitioner will follow a reasonable course of action based on current knowledge, available resources, and the needs of the patient to deliver effective and safe medical care. The sole purpose of these standards/guidelines is to assist practitioners in achieving this objective. Qualitycontrol andimprovement,safety,infectioncontrol, andpatienteducationconcerns See the SNMMI Guideline for General Imaging. Policies and procedures related to quality, patient education, infection control, and safety should be developed and implemented in accordance with good quality control practices. Quality control should also be done regularly to assure consistent, accurate physician interpretation of results. Equipment performance monitoring should be in accordance with a recognized accrediting organization. Documentation/reporting See the SNMMI Guideline for General Imaging for documentation/reporting requirements. Radiationsafety inimaging See the SNMMI Guideline for General Imaging. It is the position of SNMMI that patient exposure to ionizing radiation should be at the minimum level consistent with obtaining a diagnostic examination. Reduction in patient radiation exposure may be accomplished by administering less radiopharmaceutical when the technique or equipment used for imaging can support such an action. Each patient procedure is unique and the methodology to achieve minimum exposure while maintaining diagnostic accuracy needs to be viewed in this light. Radiopharmaceutical activity ranges outlined in this document should be considered as a guide. Dose-reduction techniques should be utilized when appropriate. The same principles should be applied when CT is used in a hybrid imaging procedure. CT acquisition protocols should be optimized to provide the information needed while minimizing patient radiation exposure. Minimizing radiation dose is especially important in children. Thebreastfeedingpatient See the Advisory Committee on the Medical Uses of Isotopes (ACMUI) guidelines.

Background: Observational data have suggested that patients with moderate to severe ischemia benefit from revascularization. However, this was not confirmed in a large, randomized trial. Objectives: Using a contemporary, multicenter registry, the authors evaluated differences in the association between quantitative ischemia, revascularization, and outcomes across important subgroups. Methods: Patients who underwent myocardial perfusion imaging in 12 centers were included in this retrospective analysis. The population was divided into original (2009-2014) and recent (2014-2021) registry sites. Early revascularization was defined as any revascularization within 90 days of myocardial perfusion imaging. A propensity score was developed to adjust for nonrandomization. Propensity score-adjusted survival analyses were used to evaluate the associations between quantitative ischemia, early revascularization, and death or myocardial infarction (MI) to identify at what severity of ischemia the HR for early revascularization crosses 1 (threshold for potential benefit). Results: Overall, 40,449 patients were included with a median follow-up of 3.5 (IQR: 2.4-4.6) years, during which death or MI occurred in 2,797 (6.9%). Early revascularization was associated with reduced death or MI in patients with >9.0% myocardial ischemia (95% upper CI: 11.2%, interaction P < 0.001). The threshold for ischemia, above which patients may benefit from revascularization, was higher in more recent patients (14.0% vs 6.5%), but similar in female (>10.0%) and male patients (>8.6%). Conclusions: Early revascularization was associated with reduced risk in patients with a higher burden of quantitative ischemia in more recent populations. These findings suggest that methods integrating more factors than just ischemia are needed to improve patient selection for revascularization.

BACKGROUND:Automated computation of the HEART score has the potential to facilitate clinical decision support and safety interventions. The goal of this study was to assess the performance of the GPT-4 large language model (LLM) in computation of the HEART score and prediction of 60-day major adverse cardiac events (MACE). METHODS:In this retrospective cohort study from February 2022 to September 2023, patients admitted to a chest pain observation unit were identified. HEART scores were calculated by a physician assistant or nurse practitioner (APP) as part of routine care. Separately, the LLM calculated a HEART score utilizing an iteratively developed prompt from deidentified chart documentation. Any cases of disagreement with the APP score were adjudicated by an emergency physician blinded to clinical outcomes. Agreement on HEART score was assessed, and 60-day MACE was obtained via linkage to an institutional registry. RESULTS:Of the 601 participants, 50 were utilized for prompt development. Among the remaining 551 participants, agreement by Cohen's weighted kappa between the LLM and adjudicators was 0.67 which was similar to the agreement of 0.66 between the APP and adjudicators. The LLM predicted a higher average HEART score (mean 5.06) compared to the adjudicators (mean 4.69) or APP (mean 4.23). No significant difference was seen in diagnostic performance for 60-day MACE by DeLong pairwise comparison (all p > .05). CONCLUSIONS:Automated risk score computation with language models has the potential to power interventions such as clinical decision support but has systematic differences from physician judgment. Prospective investigation is needed.