Deep learning analysis of epicardial adipose tissue to predict cardiovascular risk in heavy smokers.

BACKGROUND: Heavy smokers are at increased risk for cardiovascular disease and may benefit from individualized risk quantification using routine lung cancer screening chest computed tomography. We investigated the prognostic value of deep learning-based automated epicardial adipose tissue quantification and compared it to established cardiovascular risk factors and coronary artery calcium. METHODS: We investigated the prognostic value of automated epicardial adipose tissue quantification in heavy smokers enrolled in the National Lung Screening Trial and followed for 12.3 (11.9-12.8) years. The epicardial adipose tissue was segmented and quantified on non-ECG-synchronized, non-contrast low-dose chest computed tomography scans using a validated deep-learning algorithm. Multivariable survival regression analyses were then utilized to determine the associations of epicardial adipose tissue volume and density with all-cause and cardiovascular mortality (myocardial infarction and stroke). RESULTS: Here we show in 24,090 adult heavy smokers (59% men; 61 ± 5 years) that epicardial adipose tissue volume and density are independently associated with all-cause (adjusted hazard ratios: 1.10 and 1.38; P < 0.001) and cardiovascular mortality (adjusted hazard ratios: 1.14 and 1.78; P < 0.001) beyond demographics, clinical risk factors, body habitus, level of education, and coronary artery calcium score. CONCLUSIONS: Our findings suggest that automated assessment of epicardial adipose tissue from low-dose lung cancer screening images offers prognostic value in heavy smokers, with potential implications for cardiovascular risk stratification in this high-risk population.

Heavy smokers are at increased risk of poor health outcomes, particularly outcomes related to cardiovascular disease. We explore how fat surrounding the heart, known as epicardial adipose tissue, may be an indicator of the health of heavy smokers. We use an artificial intelligence system to measure the heart fat on chest scans of heavy smokers taken during a lung cancer screening trial and following their health for 12 years. We find that higher amounts and denser epicardial adipose tissue are linked to an increased risk of death from any cause, specifically from heart-related issues, even when considering other health factors. This suggests that measuring epicardial adipose tissue during lung cancer screenings could be a valuable tool for identifying heavy smokers at greater risk of heart problems and death, possibly helping to guide their medical management and improve their cardiovascular health.

Deep Learning to Estimate Cardiovascular Risk From Chest Radiographs : A Risk Prediction Study.

BACKGROUND: Guidelines for primary prevention of atherosclerotic cardiovascular disease (ASCVD) recommend a risk calculator (ASCVD risk score) to estimate 10-year risk for major adverse cardiovascular events (MACE). Because the necessary inputs are often missing, complementary approaches for opportunistic risk assessment are desirable. OBJECTIVE: To develop and test a deep-learning model (CXR CVD-Risk) that estimates 10-year risk for MACE from a routine chest radiograph (CXR) and compare its performance with that of the traditional ASCVD risk score for implications for statin eligibility. DESIGN: Risk prediction study. SETTING: Outpatients potentially eligible for primary cardiovascular prevention. PARTICIPANTS: The CXR CVD-Risk model was developed using data from a cancer screening trial. It was externally validated in 8869 outpatients with unknown ASCVD risk because of missing inputs to calculate the ASCVD risk score and in 2132 outpatients with known risk whose ASCVD risk score could be calculated. MEASUREMENTS: 10-year MACE predicted by CXR CVD-Risk versus the ASCVD risk score. RESULTS: Among 8869 outpatients with unknown ASCVD risk, those with a risk of 7.5% or higher as predicted by CXR CVD-Risk had higher 10-year risk for MACE after adjustment for risk factors (adjusted hazard ratio [HR], 1.73 [95% CI, 1.47 to 2.03]). In the additional 2132 outpatients with known ASCVD risk, CXR CVD-Risk predicted MACE beyond the traditional ASCVD risk score (adjusted HR, 1.88 [CI, 1.24 to 2.85]). LIMITATION: Retrospective study design using electronic medical records. CONCLUSION: On the basis of a single CXR, CXR CVD-Risk predicts 10-year MACE beyond the clinical standard and may help identify individuals at high risk whose ASCVD risk score cannot be calculated because of missing data. PRIMARY FUNDING SOURCE: None.

Risk factors for clonal hematopoiesis of indeterminate potential in people with HIV: a report from the REPRIEVE trial.

ABSTRACT: Clonal hematopoiesis of indeterminate potential (CHIP), the clonal expansion of myeloid cells with leukemogenic mutations, results in increased coronary artery disease (CAD) risk. CHIP is more prevalent among people with HIV (PWH), but the risk factors are unknown. CHIP was identified among PWH in REPRIEVE (Randomized Trial to Prevent Vascular Events in HIV) using whole-exome sequencing. Logistic regression was used to associate sociodemographic factors and HIV-specific factors with CHIP adjusting for age, sex, and smoking status. In the studied global cohort of 4486 PWH, mean age was 49.9 (standard deviation [SD], 6.4) years; 1650 (36.8%) were female; and 3418 (76.2%) were non-White. CHIP was identified in 223 of 4486 (4.97%) and in 38 of 373 (10.2%) among those aged ≥60 years. Age (odds ratio [OR], 1.07; 95% confidence interval [CI], 1.05-1.09; P < .0001) and smoking (OR, 1.37; 95% CI, 1.14-1.66; P < .001) associated with increased odds of CHIP. Globally, participants outside of North America had lower odds of CHIP including sub-Saharan Africa (OR, 0.57; 95% CI, 0.4-0.81; P = .0019), South Asia (OR, 0.45; 95% CI, 0.23-0.80; P = .01), and Latin America/Caribbean (OR, 0.56; 95% CI, 0.34-0.87; P = .014). Hispanic/Latino ethnicity (OR, 0.38; 95% CI, 0.23-0.54; P = .002) associated with significantly lower odds of CHIP. Among HIV-specific factors, CD4 nadir <50 cells/mm3 associated with a 1.9-fold (95%CI, 1.21-3.05; P = .006) increased odds of CHIP, with the effect being significantly stronger among individuals with short duration of antiretroviral therapy (ART; OR, 4.15; 95% CI, 1.51-11.1; P = .005) (Pinteraction= .0492). Among PWH at low-to-moderate CAD risk on stable ART, smoking, CD4 nadir, North American origin, and non-Hispanic ethnicity associated with increased odds of CHIP. This trial was registered at www.ClinicalTrials.gov as NCT02344290.

Interference Screw versus Cement Fixation in Anterior Cruciate Ligament Soft Tissue Grafts: A Biomechanical Study.

Shortcomings of fixation have been reported as a source of graft failure in anterior cruciate ligament (ACL) reconstruction. While interference screws have long been used as fixation devices for ACL reconstruction, they are not without complications. Previous studies have highlighted the use of bone void filler as a fixation method; however, no biomechanical comparisons using soft tissue grafts with interference screws exist to our knowledge. The purpose of this study is to evaluate the fixation strength of a calcium phosphate cement bone void filler compared with screw fixation in an ACL reconstruction bone replica model with human soft tissue grafts. In total, 10 ACL grafts were constructed using semitendinosus and gracilis tendons harvested from 10 donors. Grafts were affixed with either an 8-10 mm × 23 mm polyether ether ketone interference screw (n = 5) or with approximately 8 mL of calcium phosphate cement (n = 5) into open cell polyurethane blocks. Graft constructs were tested to failure in cyclic loading under displacement control at a rate of 1 mm per second. When compared with screw construct, the cement construct showed a 978% higher load at yield, 228% higher load at failure, 181% higher displacement at yield, 233% higher work at failure, and a 545% higher stiffness. Normalized data for the screw constructs relative to the cement constructs from the same donor showed 14 ± 11% load at yield, 54 ± 38% load at failure, and 172 ± 14% graft elongation. The results of this study indicate that cement fixation of ACL grafts may result in a stronger construct compared with the current standard of fixation with interference screws. This method could potentially reduce the incidence of complications associated with interface screw placement such as bone tunnel widening, screw migration, and screw breakage.

Biological and Clinical Implications of the Vascular Endothelial Growth Factor Coreceptor Neuropilin-1 in Human Immunodeficiency Virus.

Plasma vascular endothelial growth factor (VEGF) coreceptor neuropilin-1 (NRP-1) had the largest association with coronary plaque in the Randomized Trial to Prevent Vascular Events in HIV (REPRIEVE) proteomics analysis. With little known about NRP-1 in people with human immunodeficiency virus (PWH), we explored its relation to other proteins in REPRIEVE and validated our findings through a Centers for AIDS Research Network of Integrated Clinical Systems (CNICS) case-cohort study by assessing its relation to host factors and incident cardiovascular disease and cancer. Within REPRIEVE, NRP-1 was associated with proteins involved in angiogenesis, signal transduction, immunoregulation, and cell migration/adhesion. Within CNICS, NRP-1 was associated with key host factors, including older age and male sex. NRP-1 was associated with an increased hazard of multiple cancers but a decreased prostate cancer risk. Finally, NRP-1 was most strongly associated with mortality and type 2 myocardial infarction. These data suggest that NRP-1 is part of a clinically relevant immunoregulatory pathway related to multiple comorbidities in PWH. Clinical Trials Registration. NCT02344290.

Pericoronary Adipose Tissue Density, Inflammation, and Subclinical Coronary Artery Disease Among People With HIV in the REPRIEVE Cohort.

BACKGROUND: Pericoronary adipose tissue (PCAT) may influence plaque development through inflammatory mechanisms. We assessed PCAT density, as a measure of pericoronary inflammation, in relationship to coronary plaque among people with human immunodeficiency virus (HIV [PWH]) and to a matched control population. METHODS: In this baseline analysis of 727 participants of the Randomized Trial to Prevent Vascular Events in HIV (REPRIEVE) Mechanistic Substudy, we related computed tomography-derived PCAT density to presence and extent (Leaman score) of coronary artery disease (CAD), noncalcified plaque, coronary artery calcium (CAC), and vulnerable plaque features using multivariable logistic regression analyses. We further compared the PCAT density between PWH and age, sex, body mass index, CAC score, and statin use-matched controls from the community-based Framingham Heart Study (N = 464), adjusting for relevant clinical covariates. RESULTS: Among 727 REPRIEVE participants (age 50.8 ± 5.8 years; 83.6% [608/727] male), PCAT density was higher in those with (vs without) coronary plaque, noncalcified plaque, CAC >0, vulnerable plaque, and high CAD burden (Leaman score >5) (P < .001 for each comparison). PCAT density related to prevalent coronary plaque (adjusted odds ratio [per 10 HU]: 1.44; 95% confidence interval, 1.22-1.70; P < .001), adjusted for clinical cardiovascular risk factors, body mass index, and systemic immune/inflammatory biomarkers. Similarly, PCAT density related to CAC >0, noncalcified plaque, vulnerable plaque, and Leaman score >5 (all P ≤ .002). PCAT density was greater among REPRIEVE participants versus Framingham Heart Study (-88.2 ± 0.5 HU versus -90.6 ± 0.4 HU; P < .001). CONCLUSIONS: Among PWH in REPRIEVE, a large primary cardiovascular disease prevention cohort, increased PCAT density independently associated with prevalence and severity of coronary plaque, linking increased coronary inflammation to CAD in PWH.

Randomized Placebo-Controlled Trial to Evaluate Effects of Eplerenone on Myocardial Perfusion and Function Among Persons With Human Immunodeficiency Virus (HIV)-Results From the MIRACLE HIV Study.

BACKGROUND: Increased renin angiotensin aldosterone system (RAAS) activity may contribute to excess cardiovascular disease in people with HIV (PWH). We investigated how RAAS blockade may improve myocardial perfusion, injury, and function among well-treated PWH. METHODS: Forty PWH, on stable ART, without known heart disease were randomized to eplerenone 50 mg PO BID (n = 20) or identical placebo (n = 20) for 12 months. The primary endpoints were (1) myocardial perfusion assessed by coronary flow reserve (CFR) on cardiac PET or stress myocardial blood flow (sMBF) on cardiac MRI or (2) myocardial inflammation by extracellular mass index (ECMi) on cardiac MRI. RESULTS: Beneficial effects on myocardial perfusion were seen for sMBF by cardiac MRI (mean [SD]: 0.09 [0.56] vs -0.53 [0.68] mL/min/g; P = .03) but not CFR by cardiac PET (0.01 [0.64] vs -0.07 [0.48]; P = .72, eplerenone vs placebo). Eplerenone improved parameters of myocardial function on cardiac MRI including left ventricular end diastolic volume (-13 [28] vs 10 [26] mL; P = .03) and global circumferential strain (GCS; median [interquartile range 25th-75th]: -1.3% [-2.9%-1.0%] vs 2.3% [-0.4%-4.1%]; P = .03), eplerenone versus placebo respectively. On cardiac MRI, improvement in sMBF related to improvement in global circumferential strain (ρ = -0.65, P = .057) among those treated with eplerenone. Selecting for those with impaired myocardial perfusion (CFR <2.5 and/or sMBF <1.8), there was a treatment effect of eplerenone versus placebo to improve CFR (0.28 [0.27] vs -0.05 [0.36]; P = .04). Eplerenone prevented a small increase in troponin (0.00 [-0.13-0.00] vs 0.00 [0.00-0.74] ng/L; P = .03) without effects on ECMi (0.9 [-2.3-4.3] vs -0.7 [-2.2--0.1] g/m2; P = .38). CD4+ T-cell count (127 [-38-286] vs -6 [-168-53] cells/µL; P = .02) increased in the eplerenone- versus placebo-treated groups. CONCLUSIONS: RAAS blockade with eplerenone benefitted key indices and prevented worsening of myocardial perfusion, injury, and function among PWH with subclinical cardiac disease when compared with placebo. CLINICAL TRIALS REGISTRATION: NCT02740179 (https://clinicaltrials.gov/ct2/show/NCT02740179?term=NCT02740179&draw=2&rank=1).

Deep learning to estimate lung disease mortality from chest radiographs.

Prevention and management of chronic lung diseases (asthma, lung cancer, etc.) are of great importance. While tests are available for reliable diagnosis, accurate identification of those who will develop severe morbidity/mortality is currently limited. Here, we developed a deep learning model, CXR Lung-Risk, to predict the risk of lung disease mortality from a chest x-ray. The model was trained using 147,497 x-ray images of 40,643 individuals and tested in three independent cohorts comprising 15,976 individuals. We found that CXR Lung-Risk showed a graded association with lung disease mortality after adjustment for risk factors, including age, smoking, and radiologic findings (Hazard ratios up to 11.86 [8.64-16.27]; p < 0.001). Adding CXR Lung-Risk to a multivariable model improved estimates of lung disease mortality in all cohorts. Our results demonstrate that deep learning can identify individuals at risk of lung disease mortality on easily obtainable x-rays, which may improve personalized prevention and treatment strategies.

Deep Learning to Predict Mortality After Cardiothoracic Surgery Using Preoperative Chest Radiographs.

BACKGROUND: The Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM) estimates mortality risk only for certain common procedures (eg, coronary artery bypass or valve surgery) and is cumbersome, requiring greater than 60 inputs. We hypothesized that deep learning can estimate postoperative mortality risk based on a preoperative chest radiograph for cardiac surgeries in which STS-PROM scores were available (STS index procedures) or unavailable (non-STS index procedures). METHODS: We developed a deep learning model (CXR-CTSurgery) to predict postoperative mortality based on preoperative chest radiographs in 9283 patients at Massachusetts General Hospital (MGH) having cardiac surgery before April 8, 2014. CXR-CTSurgery was tested on 3615 different MGH patients and externally tested on 2840 patients from Brigham and Women's Hospital (BWH) having surgery after April 8, 2014. Discrimination for mortality was compared with the STS-PROM using the C-statistic. Calibration was assessed using the observed-to-expected ratio (O/E ratio). RESULTS: For STS index procedures, CXR-CTSurgery had a C-statistic similar to STS-PROM at MGH (CXR-CTSurgery: 0.83 vs STS-PROM: 0.88; P = .20) and BWH (0.74 vs 0.80; P = .14) testing cohorts. The CXR-CTSurgery C-statistic for non-STS index procedures was similar to STS index procedures in the MGH (0.87 vs 0.83) and BWH (0.73 vs 0.74) testing cohorts. For STS index procedures, CXR-CTSurgery had better calibration than the STS-PROM in the MGH (O/E ratio: 0.74 vs 0.52) and BWH (O/E ratio: 0.91 vs 0.73) testing cohorts. CONCLUSIONS: CXR-CTSurgery predicts postoperative mortality based on a preoperative CXR with similar discrimination and better calibration than the STS-PROM. This may be useful when the STS-PROM cannot be calculated or for non-STS index procedures.

Sex Differences in Subclinical Atherosclerosis and Systemic Immune Activation/Inflammation Among People With Human Immunodeficiency Virus in the United States.

BACKGROUND: Among people with HIV (PWH), sex differences in presentations of atherosclerotic cardiovascular disease (ASCVD) may be influenced by differences in coronary plaque parameters, immune/inflammatory biomarkers, or relationships therein. METHODS: REPRIEVE, a primary ASCVD prevention trial, enrolled antiretroviral therapy (ART)-treated PWH. At entry, a subset of US participants underwent coronary computed tomography angiography (CTA) and immune phenotyping (n = 755 CTA; n = 725 CTA + immune). We characterized sex differences in coronary plaque and immune/inflammatory biomarkers and compared immune-plaque relationships by sex. Unless noted otherwise, analyses adjust for ASCVD risk score. RESULTS: The primary analysis cohort included 631 males and 124 females. ASCVD risk was higher among males (median: 4.9% vs 2.1%), while obesity rates were higher among females (48% vs 21%). Prevalence of any plaque and of plaque with either ≥1 visible noncalcified portion or vulnerable features (NC/V-P) was lower among females overall and controlling for relevant risk factors (RR [95% CI] for any plaque: .67 [.50, .92]; RR for NC/V-P: .71 [.51, 1.00] [adjusted for ASCVD risk score and body mass index]). Females showed higher levels of IL-6, hs-CRP, and D-dimer and lower levels of Lp-PLA2 (P < .001 for all). Higher levels of Lp-PLA2, MCP-1, and oxLDL were associated with higher plaque (P < .02) and NC/V-P prevalence, with no differences by sex. Among females but not males, D-dimer was associated with higher prevalence of NC/V-P (interaction P = .055). CONCLUSIONS: Among US PWH, females had a lower prevalence of plaque and NC/V-P, as well as differences in key immune/inflammatory biomarkers. Immune-plaque relationships differed by sex for D-dimer but not other tested parameters. Clinical Trial Registration. ClinicalTrials.gov; identifier: NCT0234429 (date of initial registration: 22 January 2015).

Cytomegalovirus Immunoglobulin G (IgG) Titer and Coronary Artery Disease in People With Human Immunodeficiency Virus (HIV).

BACKGROUND: Cytomegalovirus (CMV) infection is thought to result in increased immune activation in people with human immunodeficiency virus (HIV, PWH). Although some data have linked asymptomatic CMV infection to cardiovascular disease among PWH, it remains unknown whether CMV is associated with increased or high-risk coronary plaque. METHODS: The Randomized Trial to Prevent Vascular Events in HIV (REPRIEVE) enrolled PWH aged 40-75 years on stable antiretroviral therapy (ART) with low-to-moderate atherosclerotic cardiovascular disease (ASCVD) risk. Among a subset of US REPRIEVE participants, coronary plaque was assessed by coronary computed tomography angiography. Here, we assessed the relationship between CMV immunoglobulin G (IgG) titer and (1) levels of immune activation, (2) inflammatory biomarkers, and (3) coronary plaque phenotypes at study entry. RESULTS: Of 672 participants, mean age was 51 years, 83% were men, median ASCVD risk score was 4.5%, and 66% had current CD4+ T-cell count ≥500 cells/mm3. Higher CMV IgG quartile group was associated with older age and lower current and nadir CD4+ T-cell counts. CMV IgG titer was associated with specific inflammatory biomarkers (sCD163, MCP-1, interleukin [IL]-6, hsCRP) in univariate analysis, but not after controlling for HIV-specific factors. In contrast, CMV IgG titer was not associated with coronary artery disease indexes, including presence of plaque, coronary artery calcium (CAC) score >0, vulnerable plaque presence, or Leaman score >5. CONCLUSIONS: No meaningful association was seen between CMV IgG titer and coronary artery disease indexes among ART-treated PWH at study enrollment. Longitudinal assessments in REPRIEVE will determine the relationship of CMV IgG titer to plaque progression and cardiovascular events. CLINICAL TRIALS REGISTRATION: NCT02344290.

Validation of a Deep Learning-Based Model to Predict Lung Cancer Risk Using Chest Radiographs and Electronic Medical Record Data.

Importance: Lung cancer screening with chest computed tomography (CT) prevents lung cancer death; however, fewer than 5% of eligible Americans are screened. CXR-LC, an open-source deep learning tool that estimates lung cancer risk from existing chest radiograph images and commonly available electronic medical record (EMR) data, may enable automated identification of high-risk patients as a step toward improving lung cancer screening participation. Objective: To validate CXR-LC using EMR data to identify individuals at high-risk for lung cancer to complement 2022 US Centers for Medicare & Medicaid Services (CMS) lung cancer screening eligibility guidelines. Design, Setting, and Participants: This prognostic study compared CXR-LC estimates with CMS screening guidelines using patient data from a large US hospital system. Included participants were persons who currently or formerly smoked cigarettes with an outpatient posterior-anterior chest radiograph between January 1, 2013, and December 31, 2014, with no history of lung cancer or screening CT. Data analysis was performed between May 2021 and June 2022. Exposures: CXR-LC lung cancer screening eligibility (previously defined as having a 3.297% or greater 12-year risk) based on inputs (chest radiograph image, age, sex, and whether currently smoking) extracted from the EMR. Main Outcomes and Measures: 6-year incident lung cancer. Results: A total of 14 737 persons were included in the study population (mean [SD] age, 62.6 [6.8] years; 7154 [48.5%] male; 204 [1.4%] Asian, 1051 [7.3%] Black, 432 [2.9%] Hispanic, 12 330 [85.2%] White) with a 2.4% rate of incident lung cancer over 6 years (361 patients with cancer). CMS eligibility could be determined in 6277 patients (42.6%) using smoking pack-year and quit-date from the EMR. Patients eligible by both CXR-LC and 2022 CMS criteria had a high rate of lung cancer (83 of 974 patients [8.5%]), higher than those eligible by 2022 CMS criteria alone (5 of 177 patients [2.8%]; P < .001). Patients eligible by CXR-LC but not 2022 CMS criteria also had a high 6-year incidence of lung cancer (121 of 3703 [3.3%]). In the 8460 cases (57.4%) where CMS eligibility was unknown, CXR-LC eligible patients had a 5-fold higher rate of lung cancer than ineligible (127 of 5177 [2.5%] vs 18 of 2283 [0.5%]; P < .001). Similar results were found in subgroups, including female patients and Black persons. Conclusions and Relevance: Using routine chest radiographs and other data automatically extracted from the EMR, CXR-LC identified high-risk individuals who may benefit from lung cancer screening CT.

Coronary Atherosclerosis, Cardiac Troponin, and Interleukin-6 in Patients With Chest Pain: The PROMISE Trial Results.

BACKGROUND: Increased inflammation and myocardial injury can be observed in the absence of myocardial infarction or obstructive coronary artery disease (CAD). OBJECTIVES: The authors determined whether biomarkers of inflammation and myocardial injury-interleukin (IL)-6 and high-sensitivity cardiac troponin (hs-cTn)-were associated with the presence and extent of CAD and were independent predictors of major adverse cardiovascular events (MACEs) in stable chest pain. METHODS: Using participants from the PROMISE trial, the authors measured hs-cTn I and IL-6 concentrations and analyzed computed tomography angiography (CTA) images in the core laboratory for CAD characteristics: significant stenosis (≥70%), high-risk plaque (HRP), Coronary Artery Disease Reporting and Data System (CAD-RADS) categories, segment involvement score (SIS), and coronary artery calcium (CAC) score. The primary endpoint was a composite MACE (death, myocardial infarction, or unstable angina). RESULTS: The authors included 1,796 participants (age 60.2 ± 8.0 years; 47.5% men, median follow-up 25 months). In multivariable linear regression adjusted for atherosclerotic cardiovascular disease (ASCVD) risk, hs-cTn was associated with HRP, stenosis, CAD-RADS, and SIS. IL-6 was only associated with stenosis and CAD-RADS. hs-cTn above median (1.5 ng/L) was associated with MACEs in univariable analysis (HR: 2.1 [95% CI: 1.3-3.6]; P = 0.006), but not in multivariable analysis adjusted for ASCVD and CAD. IL-6 above median (1.8 ng/L) was associated with MACEs in multivariable analysis adjusted for ASCVD and HRP (HR: 1.9 [95% CI: 1.1-3.3]; P = 0.03), CAC (HR: 1.9 [95% CI: 1.0-3.4]; P = 0.04), and SIS (HR: 1.8 [95% CI: 1.0-3.2]; P = 0.04), but not for stenosis or CAD-RADS. In participants with nonobstructive CAD (stenosis 1%-69%), the presence of both hs-cTn and IL-6 above median was strongly associated with MACEs (HR: 2.5-2.7 after adjustment for CAD characteristics). CONCLUSIONS: Concentrations of hs-cTn and IL-6 were associated with CAD characteristics and MACEs, indicating that myocardial injury and inflammation may each contribute to pathways in CAD pathophysiology. This association was most pronounced among participants with nonobstructive CAD representing an opportunity to tailor treatment in this at-risk group. (PROspective Multicenter Imaging Study for Evaluation of Chest Pain [PROMISE]; NCT01174550).

Increased Macrophage-Specific Arterial Infiltration Relates to Noncalcified Plaque and Systemic Immune Activation in People With Human Immunodeficiency Virus.

BACKGROUND: Persistent immune activation is thought to contribute to heightened atherosclerotic cardiovascular disease (ASCVD) risk among people with human immunodeficiency virus (PWH). METHODS: Participants (≥18 years) with or without human immunodeficiency virus (HIV) and without history of clinical ASCVD were enrolled. We hypothesized that increased macrophage-specific arterial infiltration would relate to plaque composition and systemic immune activation among PWH. We applied a novel targeted molecular imaging approach (technetium-99m [99mTc]-tilmanocept single photon emission computed tomography [SPECT]/CT) and comprehensive immune phenotyping. RESULTS: Aortic 99mTc-tilmanocept uptake was significantly higher among PWH (n = 20) than participants without HIV (n = 10) with similar 10-year ASCVD risk (P = .02). Among PWH, but not among participants without HIV, noncalcified aortic plaque volume related directly to aortic 99mTc-tilmanocept uptake at different uptake thresholds. An interaction (P = .001) was seen between HIV status and noncalcified plaque volume, but not calcified plaque (P = .83). Systemic levels of caspase-1 (P = .004), CD14-CD16+ (nonclassical/patrolling/homing) monocytes (P = .0004) and CD8+ T cells (P = .005) related positively and CD4+/CD8+ T-cell ratio (P = .02) inversely to aortic 99mTc-tilmanocept uptake volume. CONCLUSIONS: Macrophage-specific arterial infiltration was higher among PWH and related to noncalcified aortic plaque volume only among PWH. Key systemic markers of immune activation relating to macrophage-specific arterial infiltration may contribute to heightened ASCVD risk among PWH. CLINICAL TRIALS REGISTRATION: NCT02542371.

Deep Learning to Optimize Candidate Selection for Lung Cancer CT Screening: Advancing the 2021 USPSTF Recommendations.

Background A deep learning (DL) model to identify lung cancer screening candidates based on their chest radiographs requires external validation with a recent real-world non-U.S. sample. Purpose To validate the DL model and identify added benefits to the 2021 U.S. Preventive Services Task Force (USPSTF) recommendations in a health check-up sample. Materials and Methods This single-center retrospective study included consecutive current and former smokers aged 50-80 years who underwent chest radiography during a health check-up between January 2004 and June 2018. Discrimination performance, including receiver operating characteristic curve analysis and area under the receiver operating characteristic curve (AUC) calculation, of the model for incident lung cancers was evaluated. The added value of the model to the 2021 USPSTF recommendations was investigated for lung cancer inclusion rate, proportion of selected CT screening candidates, and positive predictive value (PPV). Results For model validation, a total of 19 488 individuals (mean age, 58 years ± 6 [SD]; 18 467 [95%] men) and the subset of USPSTF-eligible individuals (n = 7835; mean age, 57 years ± 6; 7699 [98%] men) were assessed, and the AUCs for incident lung cancers were 0.68 (95% CI: 0.62, 0.73) and 0.75 (95% CI: 0.68, 0.81), respectively. In individuals with pack-year information (n = 17 390), when excluding low- and indeterminate-risk categories from the USPSTF-eligible sample, the proportion of selected CT screening candidates was reduced to 35.8% (6233 of 17 390) from 45.1% (7835 of 17 390, P < .001), with three missed lung cancers (0.2%). The cancer inclusion rate (0.3% [53 of 17 390] vs 0.3% [56 of 17 390], P = .85) and PPV (0.9% [53 of 6233] vs 0.7% [56 of 7835], P = .42) remained unaffected. Conclusion An externally validated deep learning model showed the added value to the 2021 U.S. Preventive Services Task Force recommendations for low-dose CT lung cancer screening in reducing the number of screening candidates while maintaining the inclusion rate and positive predictive value for incident lung cancer. © RSNA, 2022 Online supplemental material is available for this article.

Factors Associated With Systemic Immune Activation Indices in a Global Primary Cardiovascular Disease Prevention Cohort of People With Human Immunodeficiency Virus on Antiretroviral Therapy.

BACKGROUND: Among antiretroviral therapy (ART)-treated people with human immunodeficiency virus (PWH), persistent systemic immune activation contributes to atherogenesis atherosclerotic, cardiovascular disease (CVD) events, and mortality. Factors associated with key immune activation indices have not previously been characterized among a global primary CVD prevention cohort of PWH. METHODS: Leveraging baseline Randomized Trial to Prevent Vascular Events in HIV (REPRIEVE) data, we evaluated factors associated with soluble CD14 (sCD14) and oxidized low-density lipoprotein (oxLDL). RESULTS: The primary analysis cohort included 4907 participants from 5 global-burden-of-disease regions (38% female, 48% Black, median age 50 years). In fully adjusted models for sCD14, female sex and White race (among those in high-income regions) were associated with higher sCD14 levels, while higher body mass index (BMI) and current use of nucleoside reverse transcriptase inhibitor + integrase strand transfer inhibitor ART were associated with lower sCD14 levels. In fully adjusted models for oxLDL, male sex, residence in high-income regions, White race (among those in high-income regions), and higher BMI were associated with higher oxLDL levels. In a subanalysis cohort of 1396 women with HIV, increased reproductive age was associated with higher sCD14 levels but not with higher oxLDL levels. CONCLUSIONS: Factors associated with sCD14 and oxLDL, 2 key indices of immune-mediated CVD risk, differ. Future studies will elucidate ways in which medications (eg, statins) and behavioral modifications influence sCD14 and oxLDL and the extent to which dampening of these markers mediates CVD-protective effects. CLINICAL TRIALS REGISTRATION: NCT0234429.

Radiologists can visually predict mortality risk based on the gestalt of chest radiographs comparable to a deep learning network.

Deep learning convolutional neural network (CNN) can predict mortality from chest radiographs, yet, it is unknown whether radiologists can perform the same task. Here, we investigate whether radiologists can visually assess image gestalt (defined as deviation from an unremarkable chest radiograph associated with the likelihood of 6-year mortality) of a chest radiograph to predict 6-year mortality. The assessment was validated in an independent testing dataset and compared to the performance of a CNN developed for mortality prediction. Results are reported for the testing dataset only (n = 100; age 62.5 ± 5.2; male 55%, event rate 50%). The probability of 6-year mortality based on image gestalt had high accuracy (AUC: 0.68 (95% CI 0.58-0.78), similar to that of the CNN (AUC: 0.67 (95% CI 0.57-0.77); p = 0.90). Patients with high/very high image gestalt ratings were significantly more likely to die when compared to those rated as very low (p ≤ 0.04). Assignment to risk categories was not explained by patient characteristics or traditional risk factors and imaging findings (p ≥ 0.2). In conclusion, assessing image gestalt on chest radiographs by radiologists renders high prognostic accuracy for the probability of mortality, similar to that of a specifically trained CNN. Further studies are warranted to confirm this concept and to determine potential clinical benefits.

Deep Learning to Estimate Biological Age From Chest Radiographs.

OBJECTIVES: The goal of this study was to assess whether a deep learning estimate of age from a chest radiograph image (CXR-Age) can predict longevity beyond chronological age. BACKGROUND: Chronological age is an imperfect measure of longevity. Biological age, a measure of overall health, may improve personalized care. This paper proposes a new way to estimate biological age using a convolutional neural network that takes as input a CXR image and outputs a chest x-ray age (in years) as a measure of long-term mortality risk. METHODS: CXR-Age was developed using CXR from 116,035 individuals and validated in 2 held-out testing sets: 1) 75% of the CXR arm of PLCO (Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial) (N = 40,967); and 2) the CXR arm of NLST (National Lung Screening Trial) (N = 5,414). CXR-Age was compared to chronological age and a multivariable regression model of chronological age, risk factors, and radiograph findings to predict all-cause and cardiovascular mortality with a maximum 23 years and 13 years of follow-up, respectively. The primary outcome was observed mortality; results are provided for the testing datasets only. RESULTS: In the PLCO testing dataset, a 5-year increase in CXR-Age carried a higher risk of all-cause mortality than a 5-year increase in chronological age (CXR-Age hazard ratio [HR]: 2.26 [95% confidence interval (CI): 2.24 to 2.29] vs. chronological age HR: 1.77 [95% CI: 1.75 to 1.78]; p < 0.001). A similar pattern was found for cardiovascular mortality (CXR-Age cause-specific HR: 2.45 per 5 years [95% CI: 2.34 to 2.56] vs. chronological age HR: 1.82 per 5 years [95% CI: 1.74 to 1.90]). Similar results were seen for both outcomes in the NLST external testing dataset. Adding CXR-Age to the multivariable model resulted in significant improvements for predicting both outcomes in both testing datasets (p < 0.001 for all comparisons). CONCLUSIONS: Based on a CXR image, CXR-Age predicted long-term all-cause and cardiovascular mortality.

Deep-learning system to improve the quality and efficiency of volumetric heart segmentation for breast cancer.

Although artificial intelligence algorithms are often developed and applied for narrow tasks, their implementation in other medical settings could help to improve patient care. Here we assess whether a deep-learning system for volumetric heart segmentation on computed tomography (CT) scans developed in cardiovascular radiology can optimize treatment planning in radiation oncology. The system was trained using multi-center data (n = 858) with manual heart segmentations provided by cardiovascular radiologists. Validation of the system was performed in an independent real-world dataset of 5677 breast cancer patients treated with radiation therapy at the Dana-Farber/Brigham and Women's Cancer Center between 2008-2018. In a subset of 20 patients, the performance of the system was compared to eight radiation oncology experts by assessing segmentation time, agreement between experts, and accuracy with and without deep-learning assistance. To compare the performance to segmentations used in the clinic, concordance and failures (defined as Dice < 0.85) of the system were evaluated in the entire dataset. The system was successfully applied without retraining. With deep-learning assistance, segmentation time significantly decreased (4.0 min [IQR 3.1-5.0] vs. 2.0 min [IQR 1.3-3.5]; p < 0.001), and agreement increased (Dice 0.95 [IQR = 0.02]; vs. 0.97 [IQR = 0.02], p < 0.001). Expert accuracy was similar with and without deep-learning assistance (Dice 0.92 [IQR = 0.02] vs. 0.92 [IQR = 0.02]; p = 0.48), and not significantly different from deep-learning-only segmentations (Dice 0.92 [IQR = 0.02]; p ≥ 0.1). In comparison to real-world data, the system showed high concordance (Dice 0.89 [IQR = 0.06]) across 5677 patients and a significantly lower failure rate (p < 0.001). These results suggest that deep-learning algorithms can successfully be applied across medical specialties and improve clinical care beyond the original field of interest.

Venous thrombosis, thromboembolism, biomarkers of inflammation, and coagulation in coronavirus disease 2019.

OBJECTIVE: Coronavirus disease 2019 (COVID-19) is associated with abnormal inflammatory and coagulation markers, potentially mediating thrombotic events. Our objective was to investigate the incidence, time course, laboratory features, and in-hospital outcomes of COVID-19 patients with suspected venous thromboembolism (VTE). METHODS: A retrospective observational cohort study was conducted of patients hospitalized with COVID-19 who had undergone ultrasound imaging for suspected VTE from March 13 to May 18, 2020. The medical records of the included patients were reviewed for D-dimer, fibrinogen, prothrombin time, partial thromboplastin time, platelet count, C-reactive protein (CRP), and high-sensitivity troponin T at admission and at up to seven time points before and after ultrasound examination. The clinical outcomes included superficial venous thrombosis, deep vein thrombosis, pulmonary embolism, intubation, and death. Mixed effects logistic, linear, and Cox proportional hazards methods were used to evaluate the relationships between the laboratory markers and VTE and other in-hospital outcomes. RESULTS: Of 138 patients who had undergone imaging studies, 44 (31.9%) had evidence of VTE. On univariable analysis, an elevated admission CRP (odds ratio [OR], 1.05; 95% confidence interval [CI], 1.01-1.09; P = .02; per 10-U increase in CRP), platelet count (OR, 1.48; 95% CI, 1.04-2.12; P = .03; per 1000-U increase in platelet count), and male sex (OR, 2.64; 95% CI, 1.19-5.84; P = .02), were associated with VTE. However only male sex remained significant on multivariable analysis (OR, 2.37; 95% CI, 1.01-5.56; P = .048). The independent predictors of death included older age (hazard ratio [HR], 1.04; 95% CI, 1.00-1.07; P = .04), active malignancy (HR, 4.39; 95% CI, 1.39-13.91; P = .01), elevated admission D-dimer (HR, 1.016; 95% CI, 1.003-1.029; P = .02), and evidence of disseminated intravascular coagulation (HR, 4.81; 95% CI, 1.76-13.10; P = .002). CONCLUSIONS: Male sex, elevated CRP, and elevated platelet count at admission were associated with VTE on univariable analysis. However, only male sex remained significant on multivariable analysis. Older age, active malignancy, disseminated intravascular coagulation, and elevated D-dimer at admission were independently associated with death for patients hospitalized with COVID-19.