Abstract
Aims
We aimed to investigate the role of aortic valve tissue composition from quantitative cardiac computed tomography angiography (CTA) in patients with severe aortic stenosis (AS) for the ...differentiation of disease subtypes and prognostication after transcatheter aortic valve implantation (TAVI).
Methods and results
Our study included 447 consecutive AS patients from six high-volume centres reporting to a prospective nationwide registry of TAVI procedures (POL-TAVI), who underwent cardiac CTA before TAVI, and 224 matched controls with normal aortic valves. Components of aortic valve tissue were identified using semi-automated software as calcific and non-calcific. Volumes of each tissue component and composition (tissue component volume/total tissue volume) × 100% were quantified. Relationship of aortic valve composition with clinical outcomes post-TAVI was evaluated using Valve Academic Research Consortium (VARC)-2 definitions.
High-gradient (HG) AS patients had significantly higher aortic tissue volume compared to low-flow low-gradient (LFLG)-AS (1672.7 vs. 1395.3 mm3, P < 0.001) as well as controls (509.9 mm3, P < 0.001), but increased non-calcific tissue was observed in LFLG compared to HG patients (1063.6 vs. 860.2 mm3, P < 0.001). Predictive value of aortic valve calcium score area under the curve (AUC) 0.989, 95% confidence interval (CI): 0.981–0.996 for severe AS was improved after addition of non-calcific tissue volume (AUC 0.995, 95% CI: 0.991–0.999, P = 0.011). In the multivariable analysis of clinical and quantitative computed tomography parameters of aortic valve tissue, non-calcific tissue volume odds ratio (OR) 5.2, 95% CI 1.8–15.4, P = 0.003 and history of stroke (OR 2.6, 95% CI 1.1–6.5, P = 0.037) were independent predictors of 30-day major adverse cardiovascular event (MACE).
Conclusion
Quantitative CTA assessment of aortic valve tissue volume and composition can improve detection of severe AS, differentiation between HG and LFLG-AS in patients referred for TAVI as well as prediction of 30-day MACEs post-TAVI, over the current clinical standard.
Atherosclerotic plaque quantification from coronary CT angiography (CCTA) enables accurate assessment of coronary artery disease burden and prognosis. We sought to develop and validate a deep ...learning system for CCTA-derived measures of plaque volume and stenosis severity.
This international, multicentre study included nine cohorts of patients undergoing CCTA at 11 sites, who were assigned into training and test sets. Data were retrospectively collected on patients with a wide range of clinical presentations of coronary artery disease who underwent CCTA between Nov 18, 2010, and Jan 25, 2019. A novel deep learning convolutional neural network was trained to segment coronary plaque in 921 patients (5045 lesions). The deep learning network was then applied to an independent test set, which included an external validation cohort of 175 patients (1081 lesions) and 50 patients (84 lesions) assessed by intravascular ultrasound within 1 month of CCTA. We evaluated the prognostic value of deep learning-based plaque measurements for fatal or non-fatal myocardial infarction (our primary outcome) in 1611 patients from the prospective SCOT-HEART trial, assessed as dichotomous variables using multivariable Cox regression analysis, with adjustment for the ASSIGN clinical risk score.
In the overall test set, there was excellent or good agreement, respectively, between deep learning and expert reader measurements of total plaque volume (intraclass correlation coefficient ICC 0·964) and percent diameter stenosis (ICC 0·879; both p<0·0001). When compared with intravascular ultrasound, there was excellent agreement for deep learning total plaque volume (ICC 0·949) and minimal luminal area (ICC 0·904). The mean per-patient deep learning plaque analysis time was 5·65 s (SD 1·87) versus 25·66 min (6·79) taken by experts. Over a median follow-up of 4·7 years (IQR 4·0–5·7), myocardial infarction occurred in 41 (2·5%) of 1611 patients from the SCOT-HEART trial. A deep learning-based total plaque volume of 238·5 mm3 or higher was associated with an increased risk of myocardial infarction (hazard ratio HR 5·36, 95% CI 1·70–16·86; p=0·0042) after adjustment for the presence of deep learning-based obstructive stenosis (HR 2·49, 1·07–5·50; p=0·0089) and the ASSIGN clinical risk score (HR 1·01, 0·99–1·04; p=0·35).
Our novel, externally validated deep learning system provides rapid measurements of plaque volume and stenosis severity from CCTA that agree closely with expert readers and intravascular ultrasound, and could have prognostic value for future myocardial infarction.
National Heart, Lung, and Blood Institute and the Miriam & Sheldon G Adelson Medical Research Foundation.
Post-pericardiotomy Syndrome Tamarappoo, Balaji K.; Klein, Allan L.
Current cardiology reports,
11/2016, Volume:
18, Issue:
11
Journal Article
Peer reviewed
Post-pericardiotomy syndrome (PPS) occurs in a subgroup of patients who have undergone cardiothoracic surgery and is characterized by fever, pleuritic pain, pleural effusion, and pericardial ...effusion. It is associated with significant morbidity, and the leading complications include tamponade and constrictive pericarditis. Epidemiologic studies have found that PPS often occurs among younger patients; however, there is a lack of comprehensive risk stratification. It is therefore important to be able to identify patients who are at high risk for developing this disease. The diagnosis is made if patients present with 2 out of the following 5 criteria; fever, pericardial or pleuritic chest pain, pericardial or pleural friction rub, pericardial effusion, and pleural effusion with elevated C-reactive protein (CRP). Pericardial effusion associated with PPS is detected by echocardiography, and cardiac MRI is used for evaluation of pericardial thickening as well as inflammation associated with PPS. These imaging modalities have been invaluable for monitoring the efficacy of treatment in PPS. Aspirin, nonsteroidal anti-inflammatory agents (NSAID), and colchicine are the mainstay of the current treatment for PPS. Although steroids are used for refractory cases of PPS, they are associated with significant side effects when used for long-term treatment of this disease. It is important for future research to focus on identification of clinical, serologic, and genetic markers that may predispose patients to PPS. There is also a need for clinical trials to address the use of targeted immunomodulatory treatment for this disease.
Stress-only myocardial perfusion imaging (MPI) markedly reduces radiation dose, scanning time, and cost. We developed an automated clinical algorithm to safely cancel unnecessary rest imaging with ...high sensitivity for obstructive coronary artery disease (CAD).
Patients without known CAD undergoing both MPI and invasive coronary angiography from REFINE SPECT were studied. A machine learning score (MLS) for prediction of obstructive CAD was generated using stress-only MPI and pre-test clinical variables. An MLS threshold with a pre-defined sensitivity of 95% was applied to the automated patient selection algorithm. Obstructive CAD was present in 1309/2079 (63%) patients. MLS had higher area under the receiver operator characteristic curve (AUC) for prediction of CAD than reader diagnosis and TPD (0.84 vs 0.70 vs 0.78, P < .01). An MLS threshold of 0.29 had superior sensitivity than reader diagnosis and TPD for obstructive CAD (95% vs 87% vs 87%, P < .01) and high-risk CAD, defined as stenosis of the left main, proximal left anterior descending, or triple-vessel CAD (sensitivity 96% vs 89% vs 90%, P < .01).
The MLS is highly sensitive for prediction of both obstructive and high-risk CAD from stress-only MPI and can be applied to a stress-first protocol for automatic cancellation of unnecessary rest imaging.
Abstract
Objectives
We evaluated the prognostic value of heart rate reserve (ΔHR) and left ventricular ejection fraction reserve (ΔLVEF) among patients with systolic dysfunction.
Background
...Inadequate ΔHR (maximal stress HR − resting HR) and ΔLVEF (LVEF at stress − LVEF at rest) in response to stress are associated with adverse cardiac events. However, the significance of an abnormal ΔHR and ΔLVEF in patients with systolic dysfunction has not been described.
Methods and results
We performed a retrospective analysis of patients with rest LVEF < 45% who underwent dipyridamole stress-rest gated Rb-82 PET myocardial perfusion imaging (PET–MPI) at the Cleveland Clinic between 2006 and 2009. Stress LVEF and volumes were calculated using commercially available software (4DM). A Cox proportional hazards model (CPH) was used to examine the association between ΔLVEF, ΔHR, and all-cause death (ACD). Among 461 patients (mean age 65.7 ± 11.3 years, 82% men) 167 experienced ACD (median follow-up 1045 days). Survival was reduced among patients with ΔHR < 0 (1090 vs. 1300 days, P = 0.04) and ΔLVEF < 0 (1002 vs. 1057 days, P = 0.03). In a CPH after adjusting for confounding variables, ΔHR ≤ 0 and ΔLVEF ≤ 0 were associated with reduced survival (hazard ratio 0.93, P < 0.01 and 0.84, P = 0.01, respectively) with an interaction between age and ΔHR (χ2 = 8.1, P < 0.01). Our model predicts that the magnitude of ΔHR is associated with improved survival among younger patients. For any given ΔLVEF the magnitude of ΔHR has a greater positive effect on survival among younger patients.
Conclusion
Both ΔHR and ΔLVEF during pharmacologic stress PET–MPI provide incremental value in predicting ACD among patients with systolic dysfunction.
Purpose
Artificial intelligence (AI) has high diagnostic accuracy for coronary artery disease (CAD) from myocardial perfusion imaging (MPI). However, when trained using high-risk populations (such as ...patients with correlating invasive testing), the disease probability can be overestimated due to selection bias. We evaluated different strategies for training AI models to improve the calibration (accurate estimate of disease probability), using external testing.
Methods
Deep learning was trained using 828 patients from 3 sites, with MPI and invasive angiography within 6 months. Perfusion was assessed using upright (U-TPD) and supine total perfusion deficit (S-TPD). AI training without data augmentation (model 1) was compared to training with augmentation (increased sampling) of patients without obstructive CAD (model 2), and patients without CAD and TPD < 2% (model 3). All models were tested in an external population of patients with invasive angiography within 6 months (
n
= 332) or low likelihood of CAD (
n
= 179).
Results
Model 3 achieved the best calibration (Brier score 0.104 vs 0.121,
p
< 0.01). Improvement in calibration was particularly evident in women (Brier score 0.084 vs 0.124,
p
< 0.01). In external testing (
n
= 511), the area under the receiver operating characteristic curve (AUC) was higher for model 3 (0.930), compared to U-TPD (AUC 0.897) and S-TPD (AUC 0.900,
p
< 0.01 for both).
Conclusion
Training AI models with augmentation of low-risk patients can improve calibration of AI models developed to identify patients with CAD, allowing more accurate assignment of disease probability. This is particularly important in lower-risk populations and in women, where overestimation of disease probability could significantly influence down-stream patient management.
Objectives
The machine learning ischemia risk score (ML-IRS) is a machine learning–based algorithm designed to identify hemodynamically significant coronary disease using quantitative coronary ...computed tomography angiography (CCTA). The purpose of this study was to examine whether the ML-IRS can predict revascularization in patients referred for invasive coronary angiography (ICA) after CCTA.
Methods
This study was a post hoc analysis of a prospective dual-center registry of sequential patients undergoing CCTA followed by ICA within 3 months, referred from inpatient, outpatient, and emergency department settings (
n
= 352, age 63 ± 10 years, 68% male). The primary outcome was revascularization by either percutaneous coronary revascularization or coronary artery bypass grafting. Blinded readers performed semi-automated quantitative coronary plaque analysis. The ML-IRS was automatically computed. Relationships between clinical risk factors, coronary plaque features, and ML-IRS with revascularization were examined.
Results
The study cohort consisted of 352 subjects with 1056 analyzable vessels. The ML-IRS ranged between 0 and 81% with a median of 18.7% (6.4–34.8). Revascularization was performed in 26% of vessels. Vessels receiving revascularization had higher ML-IRS (33.6% (21.1–55.0) versus 13.0% (4.5–29.1),
p
< 0.0001), as well as higher contrast density difference, and total, non-calcified, calcified, and low-density plaque burden. ML-IRS, when added to a traditional risk model based on clinical data and stenosis to predict revascularization, resulted in increased area under the curve from 0.69 (95% CI: 0.65–0.72) to 0.78 (95% CI: 0.75–0.81) (
p
< 0.0001), with an overall continuous net reclassification improvement of 0.636 (95% CI: 0.503–0.769;
p
< 0.0001).
Conclusions
ML-IRS from quantitative coronary CT angiography improved the prediction of future revascularization and can potentially identify patients likely to receive revascularization if referred to cardiac catheterization.
Key Points
• Machine learning ischemia risk from quantitative coronary CT angiography was significantly higher in patients who received revascularization versus those who did not receive revascularization.
• The machine learning ischemia risk score was significantly higher in patients with invasive fractional flow ≤ 0.8 versus those with > 0.8.
• The machine learning ischemia risk score improved the prediction of future revascularization significantly when added to a standard prediction model including stenosis.
A recently developed camera system for high-speed SPECT (HS-SPECT) myocardial perfusion imaging shows excellent correlation with conventional SPECT. Our goal was to test the diagnostic accuracy of an ...automated quantification of combined upright and supine myocardial SPECT for detection of coronary artery disease (CAD) (≥ 70% luminal diameter stenosis or, in left main coronary artery, ≥ 50% luminal diameter stenosis) in comparison to invasive coronary angiography (ICA).
We studied 142 patients undergoing upright and supine HS-SPECT, including 56 consecutive patients (63% men; mean age ± SD, 64 ± 13 y; 45% exercise stress) without known CAD who underwent diagnostic ICA within 6 mo of HS-SPECT and 86 consecutive patients with a low likelihood of CAD. Reference limits for upright and supine HS-SPECT were created from studies of patients with a low likelihood of CAD. Automated software adopted from supine-prone analysis was used to quantify the severity and extent of perfusion abnormality and was expressed as total perfusion deficit (TPD). TPD was obtained for upright (U-TPD), supine (S-TPD), and combined upright-supine acquisitions (C-TPD). Stress U-TPD ≥ 5%, S-TPD ≥ 5%, and C-TPD ≥ 3% myocardium were considered abnormal for per-patient analysis, and U-TPD, S-TPD, and C-TPD ≥ 2% in each coronary artery territory were considered abnormal for per-vessel analysis.
On a per-patient basis, the sensitivity was 91%, 88%, and 94% for U-TPD, S-TPD, and C-TPD, respectively, and specificity was 59%, 73%, and 86% for U-TPD, S-TPD, and C-TPD, respectively. C-TPD had a larger area under the receiver-operating-characteristic curve than U-TPD or S-TPD for identification of stenosis ≥ 70% (0.94 vs. 0.88 and 0.89, P < 0.05 and not significant, respectively). On a per-vessel basis, the sensitivity was 67%, 66%, and 69% for U-TPD, S-TPD, and C-TPD, respectively, and specificity was 91%, 94%, and 97% for U-TPD, S-TPD, and C-TPD, respectively (P = 0.02 for specificity U-TPD vs. C-TPD).
In this first comparison of HS-SPECT with ICA, new automated quantification of combined upright and supine HS-SPECT shows high diagnostic accuracy for detecting clinically significant CAD, with findings comparable to those reported using conventional SPECT.
Objectives We compared electrocardiogram-gated computed tomography (CT) myocardial perfusion imaging (MPI) based on quantification of the extent and severity of perfusion abnormalities to that ...measured with single-photon emission computed tomography (SPECT) MPI. Background Contrast-enhanced CT-MPI has been used for the identification of myocardial ischemia. Methods We performed CT-MPI during intravenous adenosine infusion in 30 patients with perfusion abnormalities on rest/adenosine stress SPECT-MPI acquired within 60 days (18 stress-rest CT-MPI and 12 stress CT-MPI only). The extent and severity of perfusion defects on SPECT-MPI were assessed on a 5-point scale in a standard 17-segment model, and total perfusion deficit (TPD) was quantified by automated software. The extent and severity of perfusion defects on CT-MPI was visually assessed by 2 observers using the same grading scale and expressed as summed stress score and summed rest score; visually quantified TPD was given by summed stress score/(maximal score of 68) and summed rest score/68. The magnitude of perfusion abnormality on CT-MPI in regions of the myocardium was defined. Results On a per-segment basis, there was good agreement between CT-MPI and SPECT-MPI with a kappa of 0.71 (p < 0.0001) for detection of stress perfusion abnormalities. Automated TPD on SPECT-MPI was similar to visual TPD from CT-MPI (p = 0.65 stress TPD, and p = 0.12 ischemic TPD stress-rest) with excellent agreement (bias = −0.3 for stress TPD, and bias = 1.2 for ischemic TPD) on Bland-Altman analysis. Software-based quantification of the magnitude of stress perfusion deficit and ischemia on CT-MPI were similar to that for automated TPD measured by SPECT (p = 0.88 stress, and p = 0.48 ischemia), with minimal bias (bias = 0.6, and bias = 1.2). Conclusions Stress and reversible myocardial perfusion deficit measured by CT-MPI using a visual semiquantitative approach and a visually guided software-based approach show strong similarity with SPECT-MPI, suggesting that CT-MPI–based assessment of myocardial perfusion defects may be of clinical and prognostic value.
Shape index and eccentricity index are measures of left ventricular morphology. Although both measures can be quantified with any stress imaging modality, they are not routinely evaluated during ...clinical interpretation. We assessed their independent associations with major adverse cardiovascular events (MACE), including measures of poststress change in shape index and eccentricity index.
Patients undergoing SPECT myocardial perfusion imaging between 2009 and 2014 from the Registry of Fast Myocardial Perfusion Imaging with Next-Generation SPECT (REFINE SPECT) were studied. Shape index (ratio between the maximum left ventricular diameter in short axis and ventricular length) and eccentricity index (calculated from orthogonal diameters in short axis and length) were calculated in end-diastole at stress and rest. Multivariable analysis was performed to assess independent associations with MACE (death, nonfatal myocardial infarction, unstable angina, or late revascularization).
In total, 14,016 patients with a mean age of 64.3 ± 12.2 y (8,469 60.4% male were included. MACE occurred in 2,120 patients during a median follow-up of 4.3 y (interquartile range, 3.4-5.7). Rest, stress, and poststress change in shape and eccentricity indices were associated with MACE in unadjusted analyses (all
< 0.001). However, in multivariable models, only poststress change in shape index (adjusted hazard ratio, 1.38;
< 0.001) and eccentricity index (adjusted hazard ratio, 0.80;
= 0.033) remained associated with MACE.
Two novel measures, poststress change in shape index and eccentricity index, were independently associated with MACE and improved risk estimation. Changes in ventricular morphology have important prognostic utility and should be included in patient risk estimation after SPECT myocardial perfusion imaging.