Deep learning features encode interpretable morphologies within histological images.

Convolutional neural networks (CNNs) are revolutionizing digital pathology by enabling machine learning-based classification of a variety of phenotypes from hematoxylin and eosin (H&E) whole slide images (WSIs), but the interpretation of CNNs remains difficult. Most studies have considered interpretability in a post hoc fashion, e.g. by presenting example regions with strongly predicted class labels. However, such an approach does not explain the biological features that contribute to correct predictions. To address this problem, here we investigate the interpretability of H&E-derived CNN features (the feature weights in the final layer of a transfer-learning-based architecture). While many studies have incorporated CNN features into predictive models, there has been little empirical study of their properties. We show such features can be construed as abstract morphological genes ("mones") with strong independent associations to biological phenotypes. Many mones are specific to individual cancer types, while others are found in multiple cancers especially from related tissue types. We also observe that mone-mone correlations are strong and robustly preserved across related cancers. Importantly, linear mone-based classifiers can very accurately separate 38 distinct classes (19 tumor types and their adjacent normals, AUC = [Formula: see text] for each class prediction), and linear classifiers are also highly effective for universal tumor detection (AUC = [Formula: see text]). This linearity provides evidence that individual mones or correlated mone clusters may be associated with interpretable histopathological features or other patient characteristics. In particular, the statistical similarity of mones to gene expression values allows integrative mone analysis via expression-based bioinformatics approaches. We observe strong correlations between individual mones and individual gene expression values, notably mones associated with collagen gene expression in ovarian cancer. Mone-expression comparisons also indicate that immunoglobulin expression can be identified using mones in colon adenocarcinoma and that immune activity can be identified across multiple cancer types, and we verify these findings by expert histopathological review. Our work demonstrates that mones provide a morphological H&E decomposition that can be effectively associated with diverse phenotypes, analogous to the interpretability of transcription via gene expression values. Our work also demonstrates mones can be interpreted without using a classifier as a proxy.

Artificial Intelligence-Powered Hematoxylin and Eosin Analyzer Reveals Distinct Immunologic and Mutational Profiles among Immune Phenotypes in Non-Small-Cell Lung Cancer.

The tumor microenvironment can be classified into three immune phenotypes: inflamed, immune excluded, and immune-desert. Immunotherapy efficacy has been shown to vary by phenotype; yet, the mechanisms are poorly understood and demand further investigation. This study unveils the mechanisms using an artificial intelligence-powered software called Lunit SCOPE. Artificial intelligence was used to classify 965 samples of non-small-cell lung carcinoma from The Cancer Genome Atlas into the three immune phenotypes. The immune and mutational profiles that shape each phenotype using xCell, gene set enrichment analysis with RNA-sequencing data, and cBioportal were described. In the inflamed subtype, which showed higher cytolytic score, the enriched pathways were generally associated with immune response and immune-related cell types were highly expressed. In the immune excluded subtype, enriched glycolysis, fatty acid, and cholesterol metabolism pathways were observed. The KRAS mutation, BRAF mutation, and MET splicing variant were mostly observed in the inflamed subtype. The two prominent mutations found in the immune excluded subtype were EGFR and PIK3CA mutations. This study is the first to report the distinct immunologic and mutational landscapes of immune phenotypes, and demonstrates the biological relevance of the classification. In light of these findings, the study offers insights into potential treatment options tailored to each immune phenotype.

Artificial Intelligence-based Radiomics in the Era of Immuno-oncology.

The recent, rapid advances in immuno-oncology have revolutionized cancer treatment and spurred further research into tumor biology. Yet, cancer patients respond variably to immunotherapy despite mounting evidence to support its efficacy. Current methods for predicting immunotherapy response are unreliable, as these tests cannot fully account for tumor heterogeneity and microenvironment. An improved method for predicting response to immunotherapy is needed. Recent studies have proposed radiomics-the process of converting medical images into quantitative data (features) that can be processed using machine learning algorithms to identify complex patterns and trends-for predicting response to immunotherapy. Because patients undergo numerous imaging procedures throughout the course of the disease, there exists a wealth of radiological imaging data available for training radiomics models. And because radiomic features reflect cancer biology, such as tumor heterogeneity and microenvironment, these models have enormous potential to predict immunotherapy response more accurately than current methods. Models trained on preexisting biomarkers and/or clinical outcomes have demonstrated potential to improve patient stratification and treatment outcomes. In this review, we discuss current applications of radiomics in oncology, followed by a discussion on recent studies that use radiomics to predict immunotherapy response and toxicity.

Association of HLA class I homozygosity with unfavorable clinical outcomes in patients with non-small cell lung cancer treated with chemo-immunotherapy or immunotherapy as first-line therapy.

BACKGROUND: Homozygosity at HLA-I locus has been reported to be an unfavorable predictive biomarker of second-line or beyond immunotherapy in patients with different types of cancer. The linkage between HLA-I zygosity and survival in NSCLC patients treated with first-line immunotherapy with or without chemotherapy has not been reported. METHODS: Next generation sequencing with HLA genotyping was performed on patients with advanced NSCLC treated with immune checkpoint inhibitors with or without chemotherapy as first-line (N = 29). Progression free survival was compared between HLA-I homozygous (defined as homozygosity in at least one locus A, B, or C) and heterozygous patients. Kaplan-Meier curves were built, and log-rank test was used. RESULTS: Among 29 enrollees, 25 patients (86.2%) were HLA-I heterozygous and four patients (13.8%) were HLA-I homozygous. Treatment response was not available in five patients with HLA-I heterozygosity. Among 20 patients with HLA-I heterozygosity, five patients (20.0%) had partial response, 10 patients (50.0%) had stable disease, two patients (8.0%) had non-complete response/non-progressive disease, and three patients (12.0%) had progressive disease. Among four patients with HLA-I heterozygosity, one patient (25.0%) had partial response, one patient (25.0%) had stable disease, and two patients (50.0%) had progressive disease. The median progression free survival was not reached in heterozygous group and was 2.97 months in homozygous group (Log-rank p = 0.68). CONCLUSIONS: We observed a trend toward an inverse association between HLA-I homozygosity and survival outcomes in patients with NSCLC treated with first-line therapy in conjunction with immunotherapy. Further prospective studies to validate aforementioned relationship are warranted.

Blood transfusions may adversely affect survival outcomes of patients with lung cancer: a systematic review and meta-analysis.

BACKGROUND: Despite common use in clinical practice, the impact of blood transfusions on prognosis among patients with lung cancer remains unclear. The purpose of the current study is to perform an updated systematic review and meta-analysis to evaluate the influence of blood transfusions on survival outcomes of lung cancer patients. METHODS: We searched PubMed, Embase, Cochrane Library, and Ovid MEDLINE for publications illustrating the association between blood transfusions and prognosis among people with lung cancer from inception to November 2019. Overall survival (OS) and disease-free survival (DFS) were the outcomes of interest. Pooled hazard ratios (HRs) with 95% confidence intervals (CIs) were computed using the random-effects model. Study heterogeneity was evaluated with the I2 test. Publication bias was explored via funnel plot and trim-and-fill analyses. RESULTS: We included 23 cohort studies with 12,175 patients (3,027 cases and 9,148 controls) for meta-analysis. Among these records, 22 studies investigated the effect of perioperative transfusions, while one examined that of transfusions during chemotherapy. Two studies suggested the possible dose-dependent effect in accordance with the number of transfused units. In pooled analyses, blood transfusions deleteriously influenced both OS (HR=1.35, 95% CI: 1.14-1.61, P<0.001, I2=0%) and DFS (HR=1.46, 95% CI: 1.15-1.86, P=0.001, I2=0%) of people with lung cancer. No evidence of significant publication bias was detected in funnel plot and trim-and-fill analyses (OS: HR=1.26, 95% CI: 1.07-1.49, P=0.006; DFS: HR=1.35, 95% CI: 1.08-1.69, P=0.008). CONCLUSIONS: Blood transfusions were associated with decreased survival of patients with lung cancer.

Prognostic Implications of Resistive Reserve Ratio in Patients With Coronary Artery Disease.

Background Resistive reserve ratio is a thermodilution-based index which integrates both coronary flow and pressure. Resistive reserve ratio represents the vasodilatory capacity of interrogated vessels including both epicardial coronary artery and microvascular circulation. We evaluated the prognostic potential of resistive reserve ratio compared with pressure-derived index (fractional flow reserve [FFR]) or flow-derived index (coronary flow reserve [CFR]). Methods and Results A total of 1245 patients underwent coronary pressure and flow measurement using pressure-temperature wire. Resistive reserve ratio was calculated by CFR adjusted using the ratio between resting and hyperemic distal coronary pressure ([resting mean transit time/hyperemic mean transit time]×[resting distal coronary pressure/hyperemic distal coronary pressure]). Clinical outcome was assessed by patient-oriented composite outcome (POCO), a composite of any death, myocardial infarction, and revascularization at 5 years. At 5 years, the cumulative incidence of POCO was significantly different according to quartiles of resistive reserve ratio (9.9%, 11.3%, 17.2%, and 22.7% in quartiles 1 to 4, respectively, log rank P<0.001). Among patients with deferred revascularization, those with depressed resistive reserve ratio (<3.5) showed a significantly higher risk of POCO than those with preserved resistive reserve ratio (≥3.5) in patients with FFR>0.80 or patients with CFR>2.0. (FFR>0.80 group: 14.8% versus 6.0%; log rank P=0.001; CFR>2.0 group: 13.5% versus 7.1%; log rank P=0.045). Adding resistive reserve ratio into the model for 5-year POCO showed significantly higher global Chi square value than FFR or CFR (P<0.001, respectively, for FFR and CFR). Resistive reserve ratio <3.5 was significantly associated with the risk of POCO at 5 years in multivariable model (adjusted hazard ratio 1.597, 95% CI, 1.098-2.271, P=0.014). Conclusions Resistive reserve ratio, which integrated both coronary flow and pressure, showed incremental prognostic implications in patients with coronary artery disease undergoing elective percutaneous coronary intervention guided by invasive physiologic evaluation. Registration URL: https://www.clini​caltr​ials.gov; Unique identifier: NCT03690713.

Sex Differences in Long-Term Outcomes in Patients With Deferred Revascularization Following Fractional Flow Reserve Assessment: International Collaboration Registry of Comprehensive Physiologic Evaluation.

Background Sex-specific differences may influence prognosis after deferred revascularization following fractional flow reserve (FFR) measurement. This study sought to investigate the sex differences in long-term prognosis of patients with deferred revascularization following FFR assessment. Methods and Results A total of 879 patients (879 vessels) with deferred revascularization with FFR >0.75 who underwent FFR and coronary flow reserve measurements were enrolled from 3 countries (Korea, Japan, and Spain). Long-term outcomes were assessed in 649 men and 230 women by the patient-oriented composite outcome (POCO, a composite of any death, any myocardial infarction, and any revascularization). We applied inverse-probability weighting based on propensity scores to account for differences at baseline between women and men (age, hyperlipidemia, diabetes mellitus, diameter stenosis, lesion length, multivessel disease, FFR, coronary flow reserve. The median follow-up duration was 1855 days (745-1855 days). Median FFR values were 0.88 (0.83-0.93) in men and 0.89 (0.85-0.94) in women, respectively. The occurrences of POCO were significantly high in men compared with that in women (10.5% versus 4.2%, P=0.007). Kaplan-Meier analysis revealed that women had a significantly lower risk of POCO (χ2=7.2, P=0.007). Multivariate COX proportional hazards regression analysis revealed that age, male, diabetes mellitus, diameter stenosis, lesion length, and coronary flow reserve were independent predictors of POCO. After applying IPW, the hazard ratio of males for POCO was 2.07 (95% CI, 1.07-4.04, P=0.032). Conclusions This large multinational study reveals that long-term outcome differs between women and men in favor of women after FFR-guided revascularization deferral. Clinical Trial Registration URL: http://www.ClinicalTrials.gov. Unique identifier: NCT02186093.

Prognostic impact of diabetes mellitus and index of microcirculatory resistance in patients undergoing fractional flow reserve-guided revascularization.

BACKGROUND: The prognostic impact of diabetes mellitus (DM) with or without coronary microvascular dysfunction (CMD) in patients undergoing fractional flow reserve (FFR)-guided revascularization has not been clarified. We sought to investigate the clinical outcomes of patients undergoing FFR-guided revascularization according to the existence of DM and CMD. METHODS: A total of 283 patients with available FFR data as well as index of microcirculatory resistance (IMR) were selected from the 3 V FFR-FRIENDS study. CMD was defined as an IMR ≥25U. Patients were grouped according to the presence of DM and CMD into group A (DM-, CMD-), group B (DM-, CMD+), group C (DM+, CMD-), and group D (DM+, CMD+). The primary outcome was a major adverse cardiac event (MACE, a composite of myocardial infarction, ischemia-driven revascularization, and cardiac death) at 2 years. RESULTS: DM patients displayed a notably higher risk of MACEs in comparison with non-DM patients (HR 4.88, 95% CI 1.54-15.48, p = 0.003). MACEs at 2 years among the four groups were 2.2%, 2.0%, 7.0%, and 18.5%, respectively. Group D exhibited a significantly higher risk of MACEs as compared to group A (HR 8.98, 95% CI 2.15-37.41, p = 0.003). Multivariable regression analysis showed that the presence of DM and CMD was an independent predictor of a 2-year MACE (HR 11.24, 95% CI 2.53-49.88, p = 0.002), and integrating CMD into a model with DM increased discriminant ability (C-index 0.683 vs. 0.710, p = 0.010, integrated discrimination improvement 0.015, p = 0.040). CONCLUSION: Among the patients undergoing FFR-guided revascularization, those with DM and CMD were correlated with an augmented risk of MACEs. Integration of CMD improved risk stratification in predicting the occurrence of a MACE.

Anatomical attributes of clinically relevant diagonal branches in patients with left anterior descending coronary artery bifurcation lesions.

AIMS: This study aimed to investigate the anatomical attributes determining myocardial territory of diagonal branches and to develop prediction models for clinically relevant branches using myocardial perfusion imaging (MPI) and coronary CT angiography (CCTA). METHODS AND RESULTS: The amount of ischaemia and subtended myocardial mass of diagonal branches was quantified using MPI by percent ischaemic myocardium (%ischaemia) and CCTA by percent fractional myocardial mass (%FMM), respectively. In 49 patients with isolated diagonal branch disease, the mean %ischaemia by MPI was 6.8±4.0%, whereas in patients with total occlusion or severe disease of all diagonal branches it was 8.4±3.3%. %ischaemia was different according to the presence of non-diseased diagonal branches and dominant left circumflex artery (LCx). In the CCTA cohort (306 patients, 564 diagonal branches), mean %FMM was 5.9±4.4% and 86 branches (15.2%) had %FMM ≥10%. %FMM was different according to LCx dominance, number of branches, vessel size, and relative dominance between two diagonal branches. The diagnostic accuracy of prediction models for %FMM ≥10% based on logistic regression and decision tree was 0.92 (95% CI: 0.85-0.96) and 0.91 (95% CI: 0.84-0.96), respectively. There was no difference in the diagnostic performance of models with and without size criterion. CONCLUSIONS: LCx dominance, number of branches, vessel size, and dominance among diagonal branches determined the myocardial territory of diagonal branches. Clinical application of prediction models based on these anatomical attributes can help to determine the clinically relevant diagonal branches in the cardiac catheterisation laboratory. CLINICAL TRIAL REGISTRATION: NCT03935542

Clinical Outcome of Lesions With Discordant Results Among Different Invasive Physiologic Indices　- Resting Distal Coronary to Aortic Pressure Ratio, Resting Full-Cycle Ratio, Diastolic Pressure Ratio, Instantaneous Wave-Free Ratio, and Fractional Flow Reserve.

BACKGROUND: We evaluated the 2-year clinical outcomes of deferred lesions with discordant results between resting and hyperemic pressure-derived physiologic indices, including resting distal to aortic coronary pressure (resting Pd/Pa), instantaneous wave-free ratio (iFR), resting full-cycle ratio (RFR), diastolic pressure ratio (dPR), and fractional flow reserve (FFR).Methods and Results:The 2-year clinical outcomes of 1,024 vessels (435 patients) with available resting Pd/Pa, iFR, RFR, dPR, and FFR data were analyzed according to a 4-group classification using known cutoff values (resting Pd/Pa ≤0.92, iFR/RFR/dPR ≤0.89, and FFR ≤0.80): Group 1 (concordant normal), Group 2 (high resting index and low FFR), Group 3 (low resting index and high FFR), and Group 4 (concordance abnormal). The primary outcome was vessel-oriented composite outcomes (VOCO) in deferred vessels at 2 years. In the comparison of VOCO risk among 4 groups classified according to FFR and 4 resting physiologic indices, Group 4 consistently showed a significantly higher risk of VOCO than Group 1. Comparison of VOCO risk among 4 groups classified according to iFR and other resting physiologic indices also showed the same results. The presence of discordance, either between hyperemic and resting indices or among resting indices, was not an independent predictor for VOCO. CONCLUSIONS: Discordant results between resting physiologic indices and FFR and among the resting indices were not associated with increased risk of VOCO in deferred lesions.

Physiologic Characteristics and Clinical Outcomes of Patients With Discordance Between FFR and iFR.

OBJECTIVES: This study evaluated the physiologic characteristics of discordant lesions between instantaneous wave-free ratio (iFR) and fractional flow reserve (FFR) and the prognosis at 5 years. BACKGROUND: FFR or iFR have been standard methods for assessing the functional significance of coronary artery stenosis. However, limited data exist about the physiologic characteristics of discordant lesions and the prognostic implications resulting from these lesions. METHODS: A total of 840 vessels from 596 patients were classified according to iFR and FFR; high iFR-high FFR (n = 580), low iFR-high FFR (n = 40), high iFR-low FFR (n = 69), and low iFR-low FFR (n = 128) groups, which were compared with a control group (n = 23). The differences in coronary circulatory indices including the coronary flow reserve (CFR), index of microcirculatory resistance (IMR), and resistance reserve ratio (RRR) (resting distal arterial pressure × mean transit time / hyperemic distal arterial pressure × hyperemic mean transit time), which reflect the vasodilatory capacity of coronary microcirculation, were compared. Patient-oriented composite outcomes (POCO) at 5 years including all-cause death, any myocardial infarction, and any revascularization were compared among patients with deferred lesions. RESULTS: In the low iFR-high FFR group, CFR, RRR, and IMR measurements were similar to the low iFR-low FFR group: CFR 2.71 versus 2.43 (p = 0.144), RRR 3.36 versus 3.68 (p = 0.241), and IMR 18.51 versus 17.38 (p = 0.476). In the high iFR-low FFR group, the CFR, RRR, and IMR measurements were similar to the control group: CFR 2.95 versus 3.29 (p = 0.160), RRR 4.28 versus 4.00 (p = 0.414), and IMR 17.44 versus 17.06 (p = 0.818). Among the 4 groups, classified by iFR and FFR, CFR and RRR were all significantly different, except for IMR. However, there were no significant differences in the rates of POCO, regardless of discordance between the iFR and FFR. Only the low iFR-low FFR group had a higher POCO rate compared with the high iFR-high FFR group (adjusted hazard ratio: 2.46; 95% confidence interval: 1.17 to 5.16; p = 0.018). CONCLUSIONS: Differences in coronary circulatory function were found, especially in the vasodilatory capacity between the low iFR-high FFR and high iFR-low FFR groups. FFR-iFR discordance was not related to an increased risk of POCO among patients with deferred lesions at 5 years. (Clinical, Physiological and Prognostic Implication of Microvascular Status;NCT02186093; Physiologic Assessment of Microvascular Function in Heart Transplant Patients; NCT02798731).

Influence of Local Myocardial Infarction on Endothelial Function, Neointimal Progression, and Inflammation in Target and Non-Target Vascular Territories in a Porcine Model of Acute Myocardial Infarction.

BACKGROUND: Patients with acute myocardial infarction (AMI) have worse clinical outcomes than those with stable coronary artery disease despite revascularization. Non-culprit lesions of AMI also involve more adverse cardiovascular events. This study aimed to investigate the influence of AMI on endothelial function, neointimal progression, and inflammation in target and non-target vessels. METHODS: In castrated male pigs, AMI was induced by balloon occlusion and reperfusion into the left anterior descending artery (LAD). Everolimus-eluting stents (EES) were implanted in the LAD and left circumflex (LCX) artery 2 days after AMI induction. In the control group, EES were implanted in the LAD and LCX in a similar fashion without AMI induction. Endothelial function was assessed using acetylcholine infusion before enrollment, after the AMI or sham operation, and at 1 month follow-up. A histological examination was conducted 1 month after stenting. RESULTS: A total of 10 pigs implanted with 20 EES in the LAD and LCX were included. Significant paradoxical vasoconstriction was assessed after acetylcholine challenge in the AMI group compared with the control group. In the histologic analysis, the AMI group showed a larger neointimal area and larger area of stenosis than the control group after EES implantation. Peri-strut inflammation and fibrin formation were significant in the AMI group without differences in injury score. The non-target vessel of the AMI also showed similar findings to the target vessel compared with the control group. CONCLUSION: In the pig model, AMI events induced endothelial dysfunction, inflammation, and neointimal progression in the target and non-target vessels.

Prognostic Implications of Plaque Characteristics and Stenosis Severity in Patients With Coronary Artery Disease.

BACKGROUND: Although the presence of ischemia is a key prognostic factor in patients with coronary artery disease, the presence of high-risk plaque characteristics (HRPC) is also associated with increased risk of cardiovascular events. Limited data exist regarding the prognostic implications of combined information on physiological stenosis severity assessed by fractional flow reserve (FFR) and plaque vulnerability by coronary computed tomography angiography (CTA)-defined HRPC. OBJECTIVES: The current study aimed to evaluate the: 1) association between physiological stenosis severity and coronary CTA-defined HRPC; and 2) prognostic implications of coronary CTA-defined HRPC according to physiological stenosis severity in patients with coronary artery disease. METHODS: A total of 772 vessels (299 patients) evaluated by both coronary CTA and FFR were analyzed. The presence and number of HRPC (minimum lumen area <4 mm2, plaque burden ≥70%, low attenuating plaque, positive remodeling, napkin-ring sign, or spotty calcification) were assessed using coronary CTA images. The risk of vessel-oriented composite outcome (VOCO) (a composite of vessel-related ischemia-driven revascularization, vessel-related myocardial infarction, or cardiac death) at 5 years was compared according to the number of HRPC and FFR categories. RESULTS: The proportion of lesions with ≥3 HRPC was significantly decreased according to the increase in FFR values (58.6%, 46.5%, 36.8%, 15.7%, and 3.5% for FFR ≤0.60, 0.61 to ≤0.70, 0.71 to ≤0.80, 0.81 to ≤0.90, and >0.90, respectively; overall p value <0.001). Both FFR and number of HRPC showed significant association with the estimated risk of VOCO (p = 0.008 and p = 0.023, respectively). In the FFR >0.80 group, lesions with ≥3 HRPC showed significantly higher risk of VOCO than those with <3 HRPC (15.0% vs. 4.3%; hazard ratio: 3.964; 95% confidence interval: 1.451 to 10.828; p = 0.007). However, there was no significant difference in the risk of VOCO according to HRPC in the FFR ≤0.80 group. By multivariable analysis, the presence of ≥3 HRPC was independently associated with the risk of VOCO in the FFR >0.80 group. CONCLUSIONS: Physiological stenosis severity and the number of HRPC were closely related, and both components had significant association with the risk of clinical events. However, the prognostic implication of HRPC was different according to FFR. Integration of both physiological stenosis severity and plaque vulnerability would provide better prognostic stratification of patients than either individual component alone, especially in patients with FFR >0.80. (Clinical Implication of 3-vessel Fractional Flow Reserve [3V FFR-FRIENDS study]; NCT01621438).

Prognostic Implications of Door-to-Balloon Time and Onset-to-Door Time on Mortality in Patients With ST -Segment-Elevation Myocardial Infarction Treated With Primary Percutaneous Coronary Intervention.

Background In patients with ST-segment-elevation myocardial infarction, timely reperfusion therapy with door-to-balloon (D2B) time <90 minutes is recommended by the current guidelines. However, whether further shortening of symptom onset-to-door (O2D) time or D2B time would enhance survival of patients with ST-segment-elevation myocardial infarction remains unclear. Therefore, the current study aimed to evaluate the prognostic impact of O2D or D2B time in patients with ST-segment-elevation myocardial infarction who underwent primary percutaneous coronary intervention. Methods and Results We analyzed 5243 patients with ST-segment-elevation myocardial infarction were treated at 20 tertiary hospitals capable of primary percutaneous coronary intervention in Korea. The association between O2D or D2B time with all-cause mortality at 1 year was evaluated. The median O2D time was 2.0 hours, and the median D2B time was 59 minutes. A total of 92.2% of the total population showed D2B time ≤90 minutes. In univariable analysis, 1-hour delay of D2B time was associated with a 55% increased 1-year mortality, whereas 1-hour delay of O2D time was associated with a 4% increased 1-year mortality. In multivariable analysis, D2B time showed an independent association with mortality (adjusted hazard ratio, 1.90; 95% CI , 1.51-2.39; P<0.001). Reducing D2B time within 45 minutes showed further decreased risk of mortality compared with D2B time >90 minutes (adjusted hazard ratio, 0.30; 95% CI , 0.19-0.42; P<0.001). Every reduction of D2B time by 30 minutes showed continuous reduction of 1-year mortality (90 to 60 minutes: absolute risk reduction, 2.4%; number needed to treat, 41.9; 60 to 30 minutes: absolute risk reduction, 2.0%; number needed to treat, 49.2). Conclusions Shortening D2B time was significantly associated with survival benefit, and the survival benefit of shortening D2B time was consistently observed, even <60 to 90 minutes.

Diagnostic Agreement of Quantitative Flow Ratio With Fractional Flow Reserve and Instantaneous Wave-Free Ratio.

Background Quantitative flow ratio ( QFR ) has a high diagnostic accuracy in assessing functional stenoses relevance, as judged by fractional flow reserve ( FFR ). However, its diagnostic performance has not been thoroughly evaluated using instantaneous wave-free ratio ( iFR ) or coronary flow reserve as the reference standard. This study sought to evaluate the diagnostic performance of QFR using other reference standards beyond FFR . Methods and Results We analyzed 182 patients (253 vessels) with stable ischemic heart disease and 82 patients (105 nonculprit vessels) with acute myocardial infarction in whom coronary stenoses were assessed with FFR , iFR, and coronary flow reserve. Contrast QFR analysis of interrogated vessels was performed in blinded fashion by a core laboratory, and its diagnostic performance was evaluated with respect to the other invasive physiological indices. Mean percentage diameter stenosis, FFR , iFR , coronary flow reserve, and QFR were 53.1±19.0%, 0.80±0.13, 0.88±0.12, 3.14±1.30, and 0.81±0.14, respectively. QFR showed higher correlation ( r=0.863 with FFR versus 0.740 with iFR , P<0.001), diagnostic accuracy (90.8% versus 81.3%, P<0.001), and discriminant function (area under the curve=0.953 versus 0.880, P<0.001) when FFR was used as a reference standard than when iFR was used as the reference standard. However, when coronary flow reserve was used as an independent reference standard, FFR , iFR , and QFR showed modest discriminant function (area under the curve=0.682, 0.765, and 0.677, respectively) and there were no significant differences in diagnostic accuracy among FFR , iFR , and QFR (65.4%, 70.6%, and 64.9%; all P values in pairwise comparisons >0.05, overall comparison P=0.061). Conclusions QFR has a high correlation and agreement with respect to both FFR and iFR , although it is better when FFR is used as the comparator. As a pressure-derived index not depending on wire or adenosine, QFR might be a promising tool for improving the adoption rate of physiology-based revascularization in clinical practice.

Relevance of anatomical, plaque, and hemodynamic characteristics of non-obstructive coronary lesions in the prediction of risk for acute coronary syndrome.

OBJECTIVES: We explored the anatomical, plaque, and hemodynamic characteristics of high-risk non-obstructive coronary lesions that caused acute coronary syndrome (ACS). METHODS: From the EMERALD study which included ACS patients with available coronary CT angiography (CCTA) before the ACS, non-obstructive lesions (percent diameter stenosis < 50%) were selected. CCTA images were analyzed for lesion characteristics by independent CCTA and computational fluid dynamics core laboratories. The relative importance of each characteristic was assessed by information gain. RESULTS: Of the 132 lesions, 24 were the culprit for ACS. The culprit lesions showed a larger change in FFRCT across the lesion (ΔFFRCT) than non-culprit lesions (0.08 ± 0.07 vs 0.05 ± 0.05, p = 0.012). ΔFFRCT showed the highest information gain (0.051, 95% confidence interval [CI] 0.050-0.052), followed by low-attenuation plaque (0.028, 95% CI 0.027-0.029) and plaque volume (0.023, 95% CI 0.022-0.024). Lesions with higher ΔFFRCT or low-attenuation plaque showed an increased risk of ACS (hazard ratio [HR] 3.25, 95% CI 1.31-8.04, p = 0.010 for ΔFFRCT; HR 2.60, 95% CI 1.36-4.95, p = 0.004 for low-attenuation plaque). The prediction model including ΔFFRCT, low-attenuation plaque and plaque volume showed the highest ability in ACS prediction (AUC 0.725, 95% CI 0.724-0.727). CONCLUSION: Non-obstructive lesions with higher ΔFFRCT or low-attenuation plaque showed a higher risk of ACS. The integration of anatomical, plaque, and hemodynamic characteristics can improve the noninvasive prediction of ACS risk in non-obstructive lesions. KEY POINTS: • Change in FFR CT across the lesion (ΔFFR CT ) was the most important predictor of ACS risk in non-obstructive lesions. • Non-obstructive lesions with higher ΔFFR CT or low-attenuation plaque were associated with a higher risk of ACS. • The integration of anatomical, plaque, and hemodynamic characteristics can improve the noninvasive prediction of ACS risk.

Identification of invasive and radionuclide imaging markers of coronary plaque vulnerability using radiomic analysis of coronary computed tomography angiography.

AIMS: Identification of invasive and radionuclide imaging markers of coronary plaque vulnerability by a single, widely available non-invasive technique may provide the opportunity to identify vulnerable plaques and vulnerable patients in broad populations. Our aim was to assess whether radiomic analysis outperforms conventional assessment of coronary computed tomography angiography (CTA) images to identify invasive and radionuclide imaging markers of plaque vulnerability. METHODS AND RESULTS: We prospectively included patients who underwent coronary CTA, sodium-fluoride positron emission tomography (NaF18-PET), intravascular ultrasound (IVUS), and optical coherence tomography (OCT). We assessed seven conventional plaque features and calculated 935 radiomic parameters from CTA images. In total, 44 plaques of 25 patients were analysed. The best radiomic parameters significantly outperformed the best conventional CT parameters to identify attenuated plaque by IVUS [fractal box counting dimension of high attenuation voxels vs. non-calcified plaque volume, area under the curve (AUC): 0.72, confidence interval (CI): 0.65-0.78 vs. 0.59, CI: 0.57-0.62; P < 0.001], thin-cap fibroatheroma by OCT (fractal box counting dimension of high attenuation voxels vs. presence of low attenuation voxels, AUC: 0.80, CI: 0.72-0.88 vs. 0.66, CI: 0.58-0.73; P < 0.001), and NaF18-positivity (surface of high attenuation voxels vs. presence of two high-risk features, AUC: 0.87, CI: 0.82-0.91 vs. 0.65, CI: 0.64-0.66; P < 0.001). CONCLUSION: Coronary CTA radiomics identified invasive and radionuclide imaging markers of plaque vulnerability with good to excellent diagnostic accuracy, significantly outperforming conventional quantitative and qualitative high-risk plaque features. Coronary CTA radiomics may provide a more accurate tool to identify vulnerable plaques compared with conventional methods. Further larger population studies are warranted.

The Effects of Preoperative Aspirin on Coronary Artery Bypass Surgery: a Systematic Meta-Analysis.

BACKGROUND AND OBJECTIVES: Aspirin plays an important role in the maintenance of graft patency and the prevention of thrombotic event after coronary artery bypass graft surgery (CABG). However, the use of preoperative aspirin is still under debate due to the risk of bleeding. METHODS: From PubMed, EMBASE, and Cochrane Central Register of Controlled Trials, data were extracted by 2 independent reviewers. Meta-analysis using random effect model was performed. RESULTS: We performed a systemic meta-analysis of 17 studies (12 randomized controlled studies and 5 non-randomized registries) which compared clinical outcomes of 9,101 patients who underwent CABG with or without preoperative aspirin administration. Preoperative aspirin increased chest tube drainage (weighted mean difference 177.4 mL, 95% confidence interval [CI], 41.3-313.4; p=0.011). However, the risk of re-operation for bleeding was not different between the preoperative aspirin group and the control group (3.2% vs. 2.4%; odds ratio [OR], 1.23; 95% CI, 0.94-1.60; p=0.102). There was no difference in the rates of all-cause mortality (1.6% vs. 1.5%; OR, 0.98; 95% CI, 0.64-1.49; p=0.920) and myocardial infarction (MI) (8.7% vs. 10.4%; OR, 0.83; 95% CI, 0.66-1.04; p=0.102) between patients with and without preoperative aspirin administration. CONCLUSIONS: Although aspirin increased the amount of chest tube drainage, it was not associated with increased risk of re-operation for bleeding. In addition, the risks of early postoperative all-cause mortality and MI were not reduced by using preoperative aspirin.

Influence of Sex on Relationship Between Total Anatomical and Physiologic Disease Burdens and Their Prognostic Implications in Patients With Coronary Artery Disease.

Background Total atherosclerosis disease burden is associated with clinical outcomes in patients with coronary artery disease. However, the influence of sex on the relationship between total anatomical and physiologic disease burdens and their prognostic implications have not been well defined. Methods and Results A total of 1136 patients who underwent fractional flow reserve (FFR) measurement in all 3 major coronary arteries were included in this study. Anatomical and physiologic total disease burden was assessed by SYNTAX (Synergy Between Percutaneous Coronary Intervention With Taxus and Cardiac Surgery) score, residual SYNTAX score, a total sum of FFR in 3 vessels (3-vessel FFR), and functional SYNTAX score. The primary end point was major adverse cardiac events, a composite of cardiac death, myocardial infarction, and ischemia-driven revascularization at 2 years. There were no differences in angiographic diameter stenosis, SYNTAX score, or residual SYNTAX score between women and men. However, both per-vessel FFR (0.89±0.10 versus 0.87±0.11, P<0.001) and 3-vessel FFR (2.72±0.13 versus 2.69±0.15, P<0.001) were higher in women. Multivariable Cox regression analyses showed that total anatomical and physiologic disease burdens were significantly associated with 2-year major adverse cardiac events, and there was no significant interaction between sex and total disease burden for clinical outcomes. Conclusions Despite similar angiographic disease severity, both per-vessel and per-patient physiologic disease severity was less in women than in men. There was no influence of sex on prognostic implications of total anatomical and physiologic disease burdens in patients with coronary artery disease. Clinical Trial Registration URL: https://www.clinicaltrials.gov . Unique identifier: NCT01621438.

Identification of High-Risk Plaques Destined to Cause Acute Coronary Syndrome Using Coronary Computed Tomographic Angiography and Computational Fluid Dynamics.

OBJECTIVES: The authors investigated the utility of noninvasive hemodynamic assessment in the identification of high-risk plaques that caused subsequent acute coronary syndrome (ACS). BACKGROUND: ACS is a critical event that impacts the prognosis of patients with coronary artery disease. However, the role of hemodynamic factors in the development of ACS is not well-known. METHODS: Seventy-two patients with clearly documented ACS and available coronary computed tomographic angiography (CTA) acquired between 1 month and 2 years before the development of ACS were included. In 66 culprit and 150 nonculprit lesions as a case-control design, the presence of adverse plaque characteristics (APC) was assessed and hemodynamic parameters (fractional flow reserve derived by coronary computed tomographic angiography [FFRCT], change in FFRCT across the lesion [△FFRCT], wall shear stress [WSS], and axial plaque stress) were analyzed using computational fluid dynamics. The best cut-off values for FFRCT, △FFRCT, WSS, and axial plaque stress were used to define the presence of adverse hemodynamic characteristics (AHC). The incremental discriminant and reclassification abilities for ACS prediction were compared among 3 models (model 1: percent diameter stenosis [%DS] and lesion length, model 2: model 1 + APC, and model 3: model 2 + AHC). RESULTS: The culprit lesions showed higher %DS (55.5 ± 15.4% vs. 43.1 ± 15.0%; p < 0.001) and higher prevalence of APC (80.3% vs. 42.0%; p < 0.001) than nonculprit lesions. Regarding hemodynamic parameters, culprit lesions showed lower FFRCT and higher △FFRCT, WSS, and axial plaque stress than nonculprit lesions (all p values <0.01). Among the 3 models, model 3, which included hemodynamic parameters, showed the highest c-index, and better discrimination (concordance statistic [c-index] 0.789 vs. 0.747; p = 0.014) and reclassification abilities (category-free net reclassification index 0.287; p = 0.047; relative integrated discrimination improvement 0.368; p < 0.001) than model 2. Lesions with both APC and AHC showed significantly higher risk of the culprit for subsequent ACS than those with no APC/AHC (hazard ratio: 11.75; 95% confidence interval: 2.85 to 48.51; p = 0.001) and with either APC or AHC (hazard ratio: 3.22; 95% confidence interval: 1.86 to 5.55; p < 0.001). CONCLUSIONS: Noninvasive hemodynamic assessment enhanced the identification of high-risk plaques that subsequently caused ACS. The integration of noninvasive hemodynamic assessments may improve the identification of culprit lesions for future ACS. (Exploring the Mechanism of Plaque Rupture in Acute Coronary Syndrome Using Coronary CT Angiography and Computational Fluid Dynamic [EMERALD]; NCT02374775).