Swarm Learning for decentralized and confidential clinical machine learning.

Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine1,2. Patients with leukaemia can be identified using machine learning on the basis of their blood transcriptomes3. However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation4,5. Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning-a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning. To illustrate the feasibility of using Swarm Learning to develop disease classifiers using distributed data, we chose four use cases of heterogeneous diseases (COVID-19, tuberculosis, leukaemia and lung pathologies). With more than 16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as more than 95,000 chest X-ray images, we show that Swarm Learning classifiers outperform those developed at individual sites. In addition, Swarm Learning completely fulfils local confidentiality regulations by design. We believe that this approach will notably accelerate the introduction of precision medicine.

Inflammatory risk and cardiovascular events in patients without obstructive coronary artery disease: the ORFAN multicentre, longitudinal cohort study.

BACKGROUND: Coronary computed tomography angiography (CCTA) is the first line investigation for chest pain, and it is used to guide revascularisation. However, the widespread adoption of CCTA has revealed a large group of individuals without obstructive coronary artery disease (CAD), with unclear prognosis and management. Measurement of coronary inflammation from CCTA using the perivascular fat attenuation index (FAI) Score could enable cardiovascular risk prediction and guide the management of individuals without obstructive CAD. The Oxford Risk Factors And Non-invasive imaging (ORFAN) study aimed to evaluate the risk profile and event rates among patients undergoing CCTA as part of routine clinical care in the UK National Health Service (NHS); to test the hypothesis that coronary arterial inflammation drives cardiac mortality or major adverse cardiac events (MACE) in patients with or without CAD; and to externally validate the performance of the previously trained artificial intelligence (AI)-Risk prognostic algorithm and the related AI-Risk classification system in a UK population. METHODS: This multicentre, longitudinal cohort study included 40 091 consecutive patients undergoing clinically indicated CCTA in eight UK hospitals, who were followed up for MACE (ie, myocardial infarction, new onset heart failure, or cardiac death) for a median of 2·7 years (IQR 1·4-5·3). The prognostic value of FAI Score in the presence and absence of obstructive CAD was evaluated in 3393 consecutive patients from the two hospitals with the longest follow-up (7·7 years [6·4-9·1]). An AI-enhanced cardiac risk prediction algorithm, which integrates FAI Score, coronary plaque metrics, and clinical risk factors, was then evaluated in this population. FINDINGS: In the 2·7 year median follow-up period, patients without obstructive CAD (32 533 [81·1%] of 40 091) accounted for 2857 (66·3%) of the 4307 total MACE and 1118 (63·7%) of the 1754 total cardiac deaths in the whole of Cohort A. Increased FAI Score in all the three coronary arteries had an additive impact on the risk for cardiac mortality (hazard ratio [HR] 29·8 [95% CI 13·9-63·9], p<0·001) or MACE (12·6 [8·5-18·6], p<0·001) comparing three vessels with an FAI Score in the top versus bottom quartile for each artery. FAI Score in any coronary artery predicted cardiac mortality and MACE independently from cardiovascular risk factors and the presence or extent of CAD. The AI-Risk classification was positively associated with cardiac mortality (6·75 [5·17-8·82], p<0·001, for very high risk vs low or medium risk) and MACE (4·68 [3·93-5·57], p<0·001 for very high risk vs low or medium risk). Finally, the AI-Risk model was well calibrated against true events. INTERPRETATION: The FAI Score captures inflammatory risk beyond the current clinical risk stratification and CCTA interpretation, particularly among patients without obstructive CAD. The AI-Risk integrates this information in a prognostic algorithm, which could be used as an alternative to traditional risk factor-based risk calculators. FUNDING: British Heart Foundation, NHS-AI award, Innovate UK, National Institute for Health and Care Research, and the Oxford Biomedical Research Centre.

Achieving Equity in Hospital Performance Assessments Using Composite Race-Specific Measures of Risk-Standardized Readmission and Mortality Rates for Heart Failure.

BACKGROUND: The contemporary measures of hospital performance for heart failure hospitalization and 30-day risk-standardized readmission rate (RSRR) and risk-standardized mortality rate (RSMR) are estimated using the same risk adjustment model and overall event rate for all patients. Thus, these measures are mainly driven by the care quality and outcomes for the majority racial and ethnic group, and may not adequately represent the hospital performance for patients of Black and other races. METHODS: Fee-for-service Medicare beneficiaries from January 2014 to December 2019 hospitalized with heart failure were identified. Hospital-level 30-day RSRR and RSMR were estimated using the traditional race-agnostic models and the race-specific approach. The composite race-specific performance metric was calculated as the average of the RSRR/RMSR measures derived separately for each race and ethnicity group. Correlation and concordance in hospital performance for all patients and patients of Black and other races were assessed using the composite race-specific and race-agnostic metrics. RESULTS: The study included 1 903 232 patients (75.7% White [n=1 439 958]; 14.5% Black [n=276 684]; and 9.8% other races [n=186 590]) with heart failure from 1860 hospitals. There was a modest correlation between hospital-level 30-day performance metrics for patients of White versus Black race (Pearson correlation coefficient: RSRR=0.42; RSMR=0.26). Compared with the race-agnostic RSRR and RSMR, composite race-specific metrics for all patients demonstrated stronger correlation with RSRR (correlation coefficient: 0.60 versus 0.74) and RSMR (correlation coefficient: 0.44 versus 0.51) for Black patients. Concordance in hospital performance for all patients and patients of Black race was also higher with race-specific (versus race-agnostic) metrics (RSRR=64% versus 53% concordantly high-performing; 61% versus 51% concordantly low-performing). Race-specific RSRR and RSMR metrics (versus race-agnostic) led to reclassification in performance ranking of 35.8% and 39.2% of hospitals, respectively, with better 30-day and 1-year outcomes for patients of all race groups at hospitals reclassified as high-performing. CONCLUSIONS: Among patients hospitalized with heart failure, race-specific 30-day RSMR and RSRR are more equitable in representing hospital performance for patients of Black and other races.

Dose-Blinded Myosin Inhibition in Patients With Obstructive Hypertrophic Cardiomyopathy Referred for Septal Reduction Therapy: Outcomes Through 32 Weeks.

BACKGROUND: Septal reduction therapy (SRT) in patients with intractable symptoms from obstructive hypertrophic cardiomyopathy (oHCM) is associated with variable morbidity and mortality. The VALOR-HCM trial (A Study to Evaluate Mavacamten in Adults with Symptomatic Obstructive Hypertrophic Cardiomyopathy Who Are Eligible for Septal Reduction Therapy) examined the effect of mavacamten on the need for SRT through week 32 in oHCM. METHODS: A double-blind randomized placebo-controlled multicenter trial at 19 US sites included patients with oHCM on maximal tolerated medical therapy referred for SRT with left ventricular outflow tract gradient ≥50 mm Hg at rest or provocation (enrollment, July 2020-October 2021). The group initially randomized to mavacamten continued the drug for 32 weeks, and the placebo group crossed over to dose-blinded mavacamten from week 16 to week 32. Dose titrations were based on investigator-blinded echocardiographic assessment of left ventricular outflow tract gradient and left ventricular ejection fraction. The principal end point was the proportion of patients proceeding with SRT or remaining guideline eligible at 32 weeks in both treatment groups. RESULTS: From the 112 randomized patients with oHCM, 108 (mean age, 60.3 years; 50% men; 94% in New York Heart Association class III/IV) qualified for week 32 evaluation (56 in the original mavacamten group and 52 in the placebo cross-over group). After 32 weeks, 6 of 56 patients (10.7%) in the original mavacamten group and 7 of 52 patients (13.5%) in the placebo cross-over group met SRT guideline criteria or elected to undergo SRT. After 32 weeks, a sustained reduction in resting left ventricular outflow tract gradient (-33.0 mm Hg [95% CI, -41.1 to -24.9]) and Valsalva left ventricular outflow tract gradient (-43.0 mm Hg [95% CI, -52.1 to -33.9]) was observed in the original mavacamten group. A similar reduction in resting (-33.7 mm Hg [95% CI, -42.2 to -25.2]) and Valsalva (-52.9 mm Hg [95% CI, -63.2 to -42.6]) gradients was quantified in the cross-over group after 16 weeks of mavacamten. After 32 weeks, improvement by ≥1 New York Heart Association class was observed in 48 of 53 patients (90.6%) in the original mavacamten group and 35 of 50 patients (70%) after 16 weeks in the cross-over group. CONCLUSIONS: In severely symptomatic patients with oHCM, 32 weeks of mavacamten treatment showed sustained reduction in the proportion proceeding to SRT or remaining guideline eligible, with similar effects observed in patients who crossed over from placebo after 16 weeks. REGISTRATION: URL: https://www. CLINICALTRIALS: gov; Unique identifier: NCT04349072.

Historical Redlining, Socioeconomic Distress, and Risk of Heart Failure Among Medicare Beneficiaries.

BACKGROUND: The association of historical redlining policies, a marker of structural racism, with contemporary heart failure (HF) risk among White and Black individuals is not well established. METHODS: We aimed to evaluate the association of redlining with the risk of HF among White and Black Medicare beneficiaries. Zip code-level redlining was determined by the proportion of historically redlined areas using the Mapping Inequality Project within each zip code. The association between higher zip code redlining proportion (quartile 4 versus quartiles 1-3) and HF risk were assessed separately among White and Black Medicare beneficiaries using generalized linear mixed models adjusted for potential confounders, including measures of the zip code-level Social Deprivation Index. RESULTS: A total of 2 388 955 Medicare beneficiaries (Black n=801 452; White n=1 587 503; mean age, 71 years; men, 44.6%) were included. Among Black beneficiaries, living in zip codes with higher redlining proportion (quartile 4 versus quartiles 1-3) was associated with increased risk of HF after adjusting for age, sex, and comorbidities (risk ratio, 1.08 [95% CI, 1.04-1.12]; P<0.001). This association remained significant after further adjustment for area-level Social Deprivation Index (risk ratio, 1.04 [95% CI, 1.002-1.08]; P=0.04). A significant interaction was observed between redlining proportion and Social Deprivation Index (Pinteraction<0.01) such that higher redlining proportion was significantly associated with HF risk only among socioeconomically distressed regions (above the median Social Deprivation Index). Among White beneficiaries, redlining was associated with a lower risk of HF after adjustment for age, sex, and comorbidities (risk ratio, 0.94 [95% CI, 0.89-0.99]; P=0.02). CONCLUSIONS: Historical redlining is associated with an increased risk of HF among Black patients. Contemporary zip code-level social determinants of health modify the relationship between redlining and HF risk, with the strongest relationship between redlining and HF observed in the most socioeconomically disadvantaged communities.

Community-Level Economic Distress, Race, and Risk of Adverse Outcomes After Heart Failure Hospitalization Among Medicare Beneficiaries.

BACKGROUND: Socioeconomic disadvantage is a strong determinant of adverse outcomes in patients with heart failure. However, the contribution of community-level economic distress to adverse outcomes in heart failure may differ across races and ethnicities. METHODS: Patients of self-reported Black, White, and Hispanic race and ethnicity hospitalized with heart failure between 2014 and 2019 were identified from the Medicare MedPAR Part A 100% Files. We used patient-level residential ZIP code to quantify community-level economic distress on the basis of the Distressed Community Index (quintile 5: economically distressed versus quintiles 1-4: nondistressed). The association of continuous and categorical measures (distressed versus nondistressed) of Distressed Community Index with 30-day, 6-month, and 1-year risk-adjusted mortality, readmission burden, and home time were assessed separately by race and ethnicity groups. RESULTS: The study included 1 611 586 White (13.2% economically distressed), 205 840 Black (50.6% economically distressed), and 89 199 Hispanic (27.3% economically distressed) patients. Among White patients, living in economically distressed (versus nondistressed) communities was significantly associated with a higher risk of adverse outcomes at 30-day and 1-year follow-up. Among Black and Hispanic patients, the risk of adverse outcomes associated with living in distressed versus nondistressed communities was not meaningfully different at 30 days and became more prominent by 1-year follow-up. Similarly, in the restricted cubic spline analysis, a stronger and more graded association was observed between Distressed Community Index score and risk of adverse outcomes in White patients (versus Black and Hispanic patients). Furthermore, the association between community-level economic distress and risk of adverse outcomes for Black patients differed in rural versus urban areas. Living in economically distressed communities was significantly associated with a higher risk of mortality and lower home time at 1-year follow-up in rural areas but not urban areas. CONCLUSIONS: The association between community-level economic distress and risk of adverse outcomes differs across race and ethnic groups, with a stronger association noted in White patients at short- and long-term follow-up. Among Black patients, the association of community-level economic distress with a higher risk of adverse outcomes is less evident in the short term and is more robust and significant in the long-term follow-up and rural areas.

Ninety-Day Risk-Standardized Home Time as a Performance Metric for Cardiac Surgery Hospitals in the United States.

BACKGROUND: Assessing hospital performance for cardiac surgery necessitates consistent and valid care quality metrics. The association of hospital-level risk-standardized home time for cardiac surgeries with other performance metrics such as mortality rate, readmission rate, and annual surgical volume has not been evaluated previously. METHODS: The study included Medicare beneficiaries who underwent isolated or concomitant coronary artery bypass graft, aortic valve, or mitral valve surgery from January 1, 2013, to October 1, 2019. Hospital-level performance metrics of annual surgical volume, 90-day risk-standardized mortality rate, 90-day risk-standardized readmission rate, and 90-day risk-standardized home time were estimated starting from the day of surgery using generalized linear mixed models with a random intercept for the hospital. Correlations between the performance metrics were assessed using the Pearson correlation coefficient. Patient-level clinical outcomes were also compared across hospital quartiles by 90-day risk-standardized home time. Last, the temporal stability of performance metrics for each hospital during the study years was also assessed. RESULTS: Overall, 919 698 patients (age 74.2±5.8 years, 32% women) were included from 1179 hospitals. Median 90-day risk-standardized home time was 71.2 days (25th-75th percentile, 66.5-75.6), 90-day risk-standardized readmission rate was 26.0% (19.5%-35.7%), and 90-day risk-standardized mortality rate was 6.0% (4.0%-8.8%). Across 90-day home time quartiles, a graded decline was observed in the rates of in-hospital, 90-day, and 1-year mortality, and 90-day and 1-year readmission. Ninety-day home time had a significant positive correlation with annual surgical volume (r=0.31; P<0.001) and inverse correlation with 90-day risk-standardized readmission rate (r=-0.40; P <0.001) and 90-day risk-standardized mortality rate (r=-0.60; P <0.001). Use of 90-day home time as a performance metric resulted in a meaningful reclassification in performance ranking of 22.8% hospitals compared with annual surgical volume, 11.6% compared with 90-day risk-standardized mortality rate, and 19.9% compared with 90-day risk-standardized readmission rate. Across the 7 years of the study period, 90-day home time demonstrated the most temporal stability of the hospital performance metrics. CONCLUSIONS: Ninety-day risk-standardized home time is a feasible, comprehensive, patient-centered metric to assess hospital-level performance in cardiac surgery with greater temporal stability than mortality and readmission measures.

Prognostic impact of left ventricular systolic dysfunction in patients with mixed aortic valve disease undergoing aortic valve replacement.

BACKGROUND: The implications of left ventricular remodeling and dysfunction before and after aortic valve replacement (AVR) for mixed aortic valve disease (MAVD) are not well understood. This study aims to evaluate the impact of AVR on left ventricular (LV) systolic function in MAVD, and determine the prognostic value of postoperative LV global longitudinal strain (LV-GLS) and LV ejection fraction (LVEF). METHODS: We retrospectively assessed 489 consecutive patients with MAVD (defined as at least moderate aortic stenosis and at least moderate aortic regurgitation) and baseline LVEF ≥50%, who underwent AVR between February 2003 and August 2018. All patients had baseline echocardiography, whereas 192 patients underwent postoperative echocardiography between 3 and 18 months after AVR. The primary endpoint was all-cause mortality. RESULTS: Mean age was 65 ± 15 years, and 65% were male. AVR in MAVD patients has a neutral effect on LV systolic function quantitated by LVEF and LV-GLS. During a median follow-up period of 5.8 years, 65 patients (34%) of 192 patients with follow-up echocardiography died. The patients with postoperative LVEF ≥50% had better survival than those with postoperative LVEF <50% (P < .001). Furthermore, among patients with postoperative LVEF ≥50%, mortality differed between patients with postoperative LV-GLS worse than -15% and those with postoperative LV-GLS better than -15% (P < .001). CONCLUSIONS: In patients with MAVD who underwent AVR, the mean postoperative LV-GLS and LVEF remain at a similar value to baseline. However, worse postoperative LV-GLS and LVEF were both independently associated with higher mortality in this population.

Physical and physiological effects of dobutamine stress echocardiography in low-gradient aortic stenosis.

Dobutamine stress echocardiography (DSE) is a useful tool for assessing low-gradient significant aortic stenosis (AS) and contractile reserve (CR), but its prognostic utility has become controversial in recent studies. We evaluated the impact of DSE on aortic valve physiological, structural, and left ventricular parameters in low-gradient AS. Consecutive patients undergoing DSE for low-gradient AS evaluation from September 2010 to July 2016 were retrospectively studied, and DSE findings were divided into four groups: with and without severe AS and/or CR. Relationships between left ventricular chamber quantification, CR, aortic valve Doppler during DSE, and calcium score [by computerized tomography (CT)] were analyzed. There were 258 DSE studies performed on 243 patients, mean age 77.6 ± 10.8 yr and 183 (70.1%) were males. With increasing dobutamine dose, apart from systolic blood pressure, left ventricular ejection fraction, flow, cardiac power output, and longitudinal strain magnitude, along with aortic valve area and mean aortic gradient were all significantly increased (P < 0.05). Flow and mean gradient increased in both the presence and absence of CR, whereas stroke volume and aortic valve area increased mainly in those with CR only. The aortic valve area increased in both patients with low and high calcium scores; however, the baseline area was lower in those with a higher calcium score. During DSE, aortic valve area increases with increase in aortic valve gradient. Higher calcium score is associated with lower baseline aortic valve area, but the aortic valve area still increases with dobutamine even in presence of a high calcium score.NEW & NOTEWORTHY We show that even in most severe aortic stenosis, there is some residual valve pliability. This suggests that a complete loss of pliability is not compatible with survival.

Comparison of risk scores for predicting outcomes after isolated tricuspid valve surgery.

BACKGROUND: Risk models play important roles in stratification and decision-making towards cardiac surgery. Isolated tricuspid valve surgery is a high risk but increasingly performed the operation, however, the performance of risk models has not been externally evaluated in these patients. We compared the prognostic utility of contemporary risk scores for isolated tricuspid valve surgery. METHODS: Consecutive patients undergoing isolated tricuspid valve surgery at Cleveland Clinic during 2004-2018 were evaluated in this cohort study. EuroSCORE II, Society of Thoracic Surgeon's tricuspid (STS-TVS) score, and the Model for End-stage Liver Disease (MELD) score were retrospectively calculated, and their performance for predicting operative mortality, postoperative complications, and mortality during follow-up was assessed. RESULTS: Amongst 207 patients studied, the mean age was 54.1 ± 17.9 years, 116 (56.0%) were female, 92 (44.4%) had secondary tricuspid regurgitation, and 151 (72.9%) had a surgical repair. Mean EuroSCORE II, STS-TVS, and MELD scores were 6.3 ± 6.6%, 5.5 ± 6.2%, and 9.8 ± 4.7, respectively. C-statistics (95% confidence intervals) for operative mortality were 0.83 (0.74-0.93) for EuroSCORE II, 0.60 (0.45-0.75) for STS-TVS score, and 0.74 (0.58-0.89) for MELD score, while observed/expected ratios were 0.78 and 0.89 for the first two scores. All three scores were associated with mortality during follow-up and discriminated most postoperative complications. CONCLUSION: EuroSCORE II was superior to STS-tricuspid score for isolated TVS risk assessment. Although surgical risk scores traditionally underestimated operative mortality after isolated tricuspid valve surgery, they did not in our cohort, reflecting the excellent surgical results. The simple MELD score performed similarly to the EuroSCORE II, especially for discriminating morbidities.

Expansion of transcatheter aortic valve replacement in the United States.

BACKGROUND: Patterns of diffusion of TAVR in the United States (U.S.) and its relation to racial disparities in TAVR utilization remain unknown. METHODS: We identified TAVR hospitals in the continental U.S. from 2012-2017 using Medicare database and mapped them to Hospital Referral Regions (HRR). We calculated driving distance from each residential ZIP code to the nearest TAVR hospital and calculated the proportion of the U.S. population, in general and by race, that lived <100 miles driving distance from the nearest TAVR center. Using a discrete time hazard logistic regression model, we examined the association of hospital and HRR variables with the opening of a TAVR program. RESULTS: The number of TAVR hospitals increased from 230 in 2012 to 540 in 2017. The proportion of the U.S. population living <100 miles from nearest TAVR hospital increased from 89.3% in 2012 to 94.5% in 2017. Geographic access improved for all racial and ethnic subgroups: Whites (84.1%-93.6%), Blacks (90.0%- 97.4%), and Hispanics (84.9%-93.7%). Within a HRR, the odds of opening a new TAVR program were higher among teaching hospitals (OR 1.48, 95% CI 1.16-1.88) and hospital bed size (OR 1.44, 95% CI 1.37-1.52). Market-level factors associated with new TAVR programs were proportion of Black (per 1%, OR 0.78, 95% CI 0.69-0.89) and Hispanic (per 1%, OR 0.82, 95% CI 0.75-0.90) residents, the proportion of hospitals within the HRR that already had a TAVR program (per 10%, OR 1.07, 95% CI 1.03-1.11), P <.01 for all. CONCLUSION: The expansion of TAVR programs in the U.S. has been accompanied by an increase in geographic coverage for all racial subgroups. Further study is needed to determine reasons for TAVR underutilization in Blacks and Hispanics.

Study design and rationale of VALOR-HCM: evaluation of mavacamten in adults with symptomatic obstructive hypertrophic cardiomyopathy who are eligible for septal reduction therapy.

BACKGROUND: Hypertrophic cardiomyopathy (HCM) is a primary myocardial disorder which frequently leads to symptoms such as dyspnea and exercise intolerance, often due to severe dynamic left ventricular outflow tract obstruction (LVOTO). Current guideline-recommended pharmacotherapies have variable therapeutic responses to relieve LVOTO. In recent phases 2 and 3, clinical trials for symptomatic obstructive HCM (oHCM), mavacamten, a small molecule inhibitor of ß-cardiac myosin has been shown to improve symptoms, exercise capacity, health status, reduce LVOTO, along with having a beneficial impact on cardiac structure and function. METHODS: VALOR-HCM is designed as a multicenter (approximately 20 centers in United States) phase 3, double-blind, placebo-controlled, randomized study. The study population consists of approximately 100 patients (≥18 years old) with symptomatic oHCM who meet 2011 American College of Cardiology/American Heart Association and/or 2014 European Society of Cardiology HCM-guideline criteria and are eligible and willing to undergo septal reduction therapy (SRT). The study duration will be up to 138 weeks, including an initial 2-week screening period, followed by16 weeks of placebo-controlled treatment, 16 weeks of active blinded treatment, 96 weeks of long-term extension, and an 8-week posttreatment follow-up visit. The primary endpoint will be a composite of the decision to proceed with SRT prior to or at Week 16 or remain guideline eligible for SRT at Week 16. Secondary efficacy endpoints will include change (from baseline to Week 16 in the mavacamten group vs placebo) in postexercise LVOT gradient, New York Heart Association class, Kansas City Cardiomyopathy Questionnaire clinical summary score, NT-proBNP, and cardiac troponin. Exploratory endpoints aim to characterize the effect of mavacamten on multiple aspects of oHCM pathophysiology. CONCLUSIONS: In severely symptomatic drug-refractory oHCM patients meeting guideline criteria of eligibility for SRT, VALOR-HCM will primarily study if a 16-week course of mavacamten reduces or obviates the need for SRT using clinically driven endpoints.

A new machine learning approach for predicting likelihood of recurrence following ablation for atrial fibrillation from CT.

OBJECTIVE: To investigate left atrial shape differences on CT scans of atrial fibrillation (AF) patients with (AF+) versus without (AF-) post-ablation recurrence and whether these shape differences predict AF recurrence. METHODS: This retrospective study included 68 AF patients who had pre-catheter ablation cardiac CT scans with contrast. AF recurrence was defined at 1 year, excluding a 3-month post-ablation blanking period. After creating atlases of atrial models from segmented AF+ and AF- CT images, an atlas-based implicit shape differentiation method was used to identify surface of interest (SOI). After registering the SOI to each patient model, statistics of the deformation on the SOI were used to create shape descriptors. The performance in predicting AF recurrence using shape features at and outside the SOI and eight clinical factors (age, sex, left atrial volume, left ventricular ejection fraction, body mass index, sinus rhythm, and AF type [persistent vs paroxysmal], catheter-ablation type [Cryoablation vs Irrigated RF]) were compared using 100 runs of fivefold cross validation. RESULTS: Differences in atrial shape were found surrounding the pulmonary vein ostia and the base of the left atrial appendage. In the prediction of AF recurrence, the area under the receiver-operating characteristics curve (AUC) was 0.67 for shape features from the SOI, 0.58 for shape features outside the SOI, 0.71 for the clinical parameters, and 0.78 combining shape and clinical features. CONCLUSION: Differences in left atrial shape were identified between AF recurrent and non-recurrent patients using pre-procedure CT scans. New radiomic features corresponding to the differences in shape were found to predict post-ablation AF recurrence.

Diagnostic Utility of CT and MRI for Mycotic Aneurysms: A Meta-Analysis.

OBJECTIVE. The purpose of this meta-analysis was to assess the diagnostic performance of CT and MRI in detecting mycotic aneurysm, an infection of high mortality and morbidity. MATERIALS AND METHODS. The PubMed, Cochrane, and Embase databases were searched from January 1, 1980, through June 30, 2019, for diagnostic studies assessing both sensitivity and specificity of CT or MRI for detecting mycotic aneurysms, and studies were pooled by use of random-effects models and freely available meta-analysis software. RESULTS. Among 1507 articles searched, 15 studies of CT (13 studies) or MRI (five studies) for aortic and cerebral mycotic aneurysms were included. The studies evaluated 622 imaging examinations for 249 mycotic aneurysms. The pooled sensitivities and specificities of CT for all mycotic aneurysms were 0.82 (95% CI, 0.77-0.87) and 0.93 (95% CI, 0.89-0.95) and of MRI were 0.79 (95% CI, 0.61-0.91) and 0.89 (95% CI, 0.81-0.95). CT and MRI had pooled sensitivities and specificities of 0.84 (95% CI, 0.78-0.89) and 0.92 (95% CI, 0.89-0.95) for aortic and 0.71 (95% CI, 0.54-0.85) and 0.90 (95% CI, 0.83-0.95) for cerebral mycotic aneurysms. Heterogeneity and publication bias were observed in some pooled analyses. CONCLUSION. CT and MRI had moderately high sensitivities and specificities for mycotic aneurysms. Study heterogeneity, publication bias, and modest sample size were important limitations, warranting larger and higher-quality studies.

Relationships between mitral annular calcification and cardiovascular events: A meta-analysis.

BACKGROUND: Mitral annular calcification (MAC) is prevalent in the aging population, with recent renewed interest regarding its associations with cardiovascular risk factors, outcomes, and influence on valvular heart disease and interventions. This meta-analysis aimed to report the relationships between MAC and cardiovascular mortality and morbidity events. METHODS: Relevant studies were searched from PubMed, Cochrane, and Embase databases until November 30, 2019. Associations between MAC as a binary variable with death and cardiovascular events were pooled using random-effects models. The main outcomes of interest were all-cause and cardiovascular mortality, myocardial infarction, stroke, heart failure, atrial fibrillation, and procedural outcomes. RESULTS: Among 799 article abstracts and 122 full-text articles screened, 26 (16 prospective and 10 retrospective) studies totaling 35 070 subjects were analyzed. MAC was associated with higher all-cause death, hazard ratio (95% confidence interval) 1.76 (1.43-2.22), and cardiovascular mortality 1.85 (1.45-23.5). It also positively correlated with myocardial infarction 1.48 (1.22-1.79), stroke 1.51 (1.22-2.05), incidental heart failure 1.55 (1.30-1.84), atrial fibrillation 1.75 (1.43-2.15), and their composite, major adverse cardiovascular events (MACE). Finally, conversion to mitral valve replacement at time of cardiac surgery was more in patients with MAC than without MAC, with odds ratio (95% confidence interval) 2.82 (1.28-6.18). CONCLUSION: Mitral annular calcification was overall associated with higher rates of death, and both individual and composite cardiovascular events. The presence of increasingly encountered MAC has significant clinical implications for cardiovascular risk assessment and valvular interventions.

After ISCHEMIA: Is coronary CTA the new gatekeeper?

The ISCHEMIA trial investigated two major principles in the therapy of coronary artery disease (CAD), i.e., symptom relief and improvement of prognosis. Specifically, it was designed to answer the question of whether, after ruling out left main stenosis, a routine interventional strategy in addition to optimal medical therapy can improve clinical outcome. Overall, this hypothesis could not be confirmed. Nevertheless, the trial yields interesting new aspects in the field of cardiac imaging. As a noninvasive diagnostic approach for individuals with suspected coronary artery disease, two different concepts are available: stress testing for ischemia (single-photon emission computed tomography, positron emission tomography, cardiac magnetic resonance imaging, stress echocardiography) and anatomic visualization of coronary artery stenosis by coronary computed tomography (CT) angiography (coronary CTA). While there was no randomized comparison between these two approaches in ISCHEMIA, the good outcome achieved by using coronary CTA as a "gatekeeper" to randomization supports the potential of coronary CTA as a diagnostic tool-both as first- and as second-line-when CAD is suspected. However, the trial also raises new questions in the field of cardiac imaging that need to be addressed in future studies.

Radiation-Associated Valvular Disease.

PURPOSE OF REVIEW: Radiation-associated valvular disease (RAVD) is characterized by late valvular manifestations following radiation exposure to the mediastinum. Review of current guidelines was performed to examine best practices to reduce risk and optimize outcomes in this patient population. RECENT FINDINGS: Early and consistent screening and comprehensive and careful planning are critical in managing RAVD. Due to long latency periods, serial screening and targeted evaluation of risk factors are essential to early detection. Varying and complex presentations of RAVD require an integrated team of experienced specialists equipped with multimodality imaging-based screening protocols to stratify risk, plan intervention, and evaluate treatment response. Patients with valvular manifestations associated with radiation therapy call for an individualized plan of care involving longitudinal multimodality imaging-based screening and experienced decision-making regarding timing and strategy of intervention to improve patient outcomes.

Surgical ablation of atrial fibrillation in hypertrophic obstructive cardiomyopathy: Outcomes of a tailored surgical approach.

OBJECTIVE: To assess outcomes of concomitant ablation for atrial fibrillation (AF) in patients with preoperative AF undergoing septal myectomy for hypertrophic obstructive cardiomyopathy. METHODS: From 2005 to 2016, 67 patients underwent concomitant ablation for AF and septal myectomy and had a follow-up beyond a 3-month blanking period. Ablation strategy (pulmonary vein isolation [PVI], modified Cox-maze III [CM-III], or Cox-maze IV [CM-IV]) was tailored to preoperative AF burden, with high AF burden defined as persistent AF or need for cardioversion. AF recurrence was analyzed as a time-related event and predictors of recurrence identified using a random forest methodology. RESULTS: A total of 38 patients (57%) had low AF burden and 29 (43%) high burden. Patients with low AF burden most frequently underwent PVI (68%). Patients with high AF burden more frequently underwent CM-III (62%) or CM-IV (35%). Besides the preoperative AF burden, baseline characteristics were similar between patients receiving CM-III, CM-IV, and PVI. After surgery, the maximum provoked left ventricular outflow tract (LVOT) gradient decreased from 99 ± 34 to 18 ± 11mm Hg (P < .001). Eight patients (12%) required a permanent pacemaker. Cumulative AF recurrence at 1, 2, and 5 years was 11%, 22%, and 48%, respectively. Age, low preoperative resting LVOT gradient, and large left atrial diameter were predictors of AF recurrence. CONCLUSIONS: Surgical outcomes of concomitant ablation for AF and septal myectomy are good, although recurrence of AF by 5 years is frequent.

A novel machine learning-derived radiotranscriptomic signature of perivascular fat improves cardiac risk prediction using coronary CT angiography.

BACKGROUND: Coronary inflammation induces dynamic changes in the balance between water and lipid content in perivascular adipose tissue (PVAT), as captured by perivascular Fat Attenuation Index (FAI) in standard coronary CT angiography (CCTA). However, inflammation is not the only process involved in atherogenesis and we hypothesized that additional radiomic signatures of adverse fibrotic and microvascular PVAT remodelling, may further improve cardiac risk prediction. METHODS AND RESULTS: We present a new artificial intelligence-powered method to predict cardiac risk by analysing the radiomic profile of coronary PVAT, developed and validated in patient cohorts acquired in three different studies. In Study 1, adipose tissue biopsies were obtained from 167 patients undergoing cardiac surgery, and the expression of genes representing inflammation, fibrosis and vascularity was linked with the radiomic features extracted from tissue CT images. Adipose tissue wavelet-transformed mean attenuation (captured by FAI) was the most sensitive radiomic feature in describing tissue inflammation (TNFA expression), while features of radiomic texture were related to adipose tissue fibrosis (COL1A1 expression) and vascularity (CD31 expression). In Study 2, we analysed 1391 coronary PVAT radiomic features in 101 patients who experienced major adverse cardiac events (MACE) within 5 years of having a CCTA and 101 matched controls, training and validating a machine learning (random forest) algorithm (fat radiomic profile, FRP) to discriminate cases from controls (C-statistic 0.77 [95%CI: 0.62-0.93] in the external validation set). The coronary FRP signature was then tested in 1575 consecutive eligible participants in the SCOT-HEART trial, where it significantly improved MACE prediction beyond traditional risk stratification that included risk factors, coronary calcium score, coronary stenosis, and high-risk plaque features on CCTA (Δ[C-statistic] = 0.126, P < 0.001). In Study 3, FRP was significantly higher in 44 patients presenting with acute myocardial infarction compared with 44 matched controls, but unlike FAI, remained unchanged 6 months after the index event, confirming that FRP detects persistent PVAT changes not captured by FAI. CONCLUSION: The CCTA-based radiomic profiling of coronary artery PVAT detects perivascular structural remodelling associated with coronary artery disease, beyond inflammation. A new artificial intelligence (AI)-powered imaging biomarker (FRP) leads to a striking improvement of cardiac risk prediction over and above the current state-of-the-art.

Non-invasive detection of coronary inflammation using computed tomography and prediction of residual cardiovascular risk (the CRISP CT study): a post-hoc analysis of prospective outcome data.

BACKGROUND: Coronary artery inflammation inhibits adipogenesis in adjacent perivascular fat. A novel imaging biomarker-the perivascular fat attenuation index (FAI)-captures coronary inflammation by mapping spatial changes of perivascular fat attenuation on coronary computed tomography angiography (CTA). However, the ability of the perivascular FAI to predict clinical outcomes is unknown. METHODS: In the Cardiovascular RISk Prediction using Computed Tomography (CRISP-CT) study, we did a post-hoc analysis of outcome data gathered prospectively from two independent cohorts of consecutive patients undergoing coronary CTA in Erlangen, Germany (derivation cohort) and Cleveland, OH, USA (validation cohort). Perivascular fat attenuation mapping was done around the three major coronary arteries-the proximal right coronary artery, the left anterior descending artery, and the left circumflex artery. We assessed the prognostic value of perivascular fat attenuation mapping for all-cause and cardiac mortality in Cox regression models, adjusted for age, sex, cardiovascular risk factors, tube voltage, modified Duke coronary artery disease index, and number of coronary CTA-derived high-risk plaque features. FINDINGS: Between 2005 and 2009, 1872 participants in the derivation cohort underwent coronary CTA (median age 62 years [range 17-89]). Between 2008 and 2016, 2040 patients in the validation cohort had coronary CTA (median age 53 years [range 19-87]). Median follow-up was 72 months (range 51-109) in the derivation cohort and 54 months (range 4-105) in the validation cohort. In both cohorts, high perivascular FAI values around the proximal right coronary artery and left anterior descending artery (but not around the left circumflex artery) were predictive of all-cause and cardiac mortality and correlated strongly with each other. Therefore, the perivascular FAI measured around the right coronary artery was used as a representative biomarker of global coronary inflammation (for prediction of cardiac mortality, hazard ratio [HR] 2·15, 95% CI 1·33-3·48; p=0·0017 in the derivation cohort, and 2·06, 1·50-2·83; p<0·0001 in the validation cohort). The optimum cutoff for the perivascular FAI, above which there is a steep increase in cardiac mortality, was ascertained as -70·1 Hounsfield units (HU) or higher in the derivation cohort (HR 9·04, 95% CI 3·35-24·40; p<0·0001 for cardiac mortality; 2·55, 1·65-3·92; p<0·0001 for all-cause mortality). This cutoff was confirmed in the validation cohort (HR 5·62, 95% CI 2·90-10·88; p<0·0001 for cardiac mortality; 3·69, 2·26-6·02; p<0·0001 for all-cause mortality). Perivascular FAI improved risk discrimination in both cohorts, leading to significant reclassification for all-cause and cardiac mortality. INTERPRETATION: The perivascular FAI enhances cardiac risk prediction and restratification over and above current state-of-the-art assessment in coronary CTA by providing a quantitative measure of coronary inflammation. High perivascular FAI values (cutoff ≥-70·1 HU) are an indicator of increased cardiac mortality and, therefore, could guide early targeted primary prevention and intensive secondary prevention in patients. FUNDING: British Heart Foundation, and the National Institute of Health Research Oxford Biomedical Research Centre.