Deep Learning to Classify AL versus ATTR Cardiac Amyloidosis MR Images.

The aim of this work was to compare the classification of cardiac MR-images of AL versus ATTR amyloidosis by neural networks and by experienced human readers. Cine-MR images and late gadolinium enhancement (LGE) images of 120 patients were studied (70 AL and 50 TTR). A VGG16 convolutional neural network (CNN) was trained with a 5-fold cross validation process, taking care to strictly distribute images of a given patient in either the training group or the test group. The analysis was performed at the patient level by averaging the predictions obtained for each image. The classification accuracy obtained between AL and ATTR amyloidosis was 0.750 for cine-CNN, 0.611 for Gado-CNN and between 0.617 and 0.675 for human readers. The corresponding AUC of the ROC curve was 0.839 for cine-CNN, 0.679 for gado-CNN (p < 0.004 vs. cine) and 0.714 for the best human reader (p < 0.007 vs. cine). Logistic regression with cine-CNN and gado-CNN, as well as analysis focused on the specific orientation plane, did not change the overall results. We conclude that cine-CNN leads to significantly better discrimination between AL and ATTR amyloidosis as compared to gado-CNN or human readers, but with lower performance than reported in studies where visual diagnosis is easy, and is currently suboptimal for clinical practice.

Cutting-Edge Imaging of Cardiac Metastases from Neuroendocrine Tumors: Lesson from a Case Series.

With the increasing availability of high-performance medical imaging for the management of patients with neuroendocrine tumors (NETs), a progressive growth of asymptomatic and incidentally detected cardiac metastases (CMs) has been observed in the recent years. In clinical practice, CMs of NENs are often incidentally detected by whole-body 68Ga-labeled somatostatin analogs or 18F-fluorodihydroxyphenylalanine positron emission tomography/computed tomography, and afterwards accurately characterized by cardiac magnetic resonance (CMR) and/or gated cardiac computed tomography when CMR is contraindicated or not available. The interpreting physician should familiarize with the main imaging features of CM, a finding that may be encountered in NETs patients more than previously thought. Herein, we present a case series of four patients with CMs from small-intestine NETs highlighting strengths and weaknesses of a multimodality imaging approach in clinical practice.

Deep Learning Supplants Visual Analysis by Experienced Operators for the Diagnosis of Cardiac Amyloidosis by Cine-CMR.

BACKGROUND: Diagnosing cardiac amyloidosis (CA) from cine-CMR (cardiac magnetic resonance) alone is not reliable. In this study, we tested if a convolutional neural network (CNN) could outperform the visual diagnosis of experienced operators. METHOD: 119 patients with cardiac amyloidosis and 122 patients with left ventricular hypertrophy (LVH) of other origins were retrospectively selected. Diastolic and systolic cine-CMR images were preprocessed and labeled. A dual-input visual geometry group (VGG ) model was used for binary image classification. All images belonging to the same patient were distributed in the same set. Accuracy and area under the curve (AUC) were calculated per frame and per patient from a 40% held-out test set. Results were compared to a visual analysis assessed by three experienced operators. RESULTS: frame-based comparisons between humans and a CNN provided an accuracy of 0.605 vs. 0.746 (p < 0.0008) and an AUC of 0.630 vs. 0.824 (p < 0.0001). Patient-based comparisons provided an accuracy of 0.660 vs. 0.825 (p < 0.008) and an AUC of 0.727 vs. 0.895 (p < 0.002). CONCLUSION: based on cine-CMR images alone, a CNN is able to discriminate cardiac amyloidosis from LVH of other origins better than experienced human operators (15 to 20 points more in absolute value for accuracy and AUC), demonstrating a unique capability to identify what the eyes cannot see through classical radiological analysis.

68Ga-DOTATOC PET for Treatment Efficacy Evaluation of Cardiac Sarcoidosis.

Cardiac sarcoidosis (CS) has a poor prognosis related to life-threating arrhythmias and heart failure. Treatment includes anti-inflammatory therapies and implantable pacemaker and/or cardioverter defibrillator. The presence of cardiac devices and physiologic myocardial glucose uptake are major limitations of both cardiac magnetic resonance and F-FDG PET/CT, reducing their diagnostic value. Somatostatin-based PET/CT has been proposed to detect active CS. Contrarily to F-FDG uptake, which reflects nonspecific leukocyte infiltration, Ga-DOTATOC may identify active granulomatosis. Herein, we underline the specificity of Ga-DOTATOC PET in challeging clinical situations including refractory CS, and chronic CS in patients with cardiac device, or false-positive F-FDG PET/CT results.

320-row CT transcatheter aortic valve replacement planning with a single reduced contrast media bolus injection.

OBJECTIVE: To reduce the iodine load required for CT Transcatheter Aortic Valve Replacement (TAVR) planning on a 320-row scanner by acquiring the two CT TAVR steps (ECG-gated aortic root CTA and non-gated aorto-ilio-femoral CTA) within a single contrast media bolus injection. METHODS: 50 consecutive patients (82.6±6.9 years; 56% female) were prospectively enrolled and underwent a TAVR planning using a 320-row CT, with ECG-gated aortic root CTA immediately followed by a non-gated aorto-iliac acquisition, all within a single bolus of 40-70mL of Iohexol 350mgI/mL. The Iodine load, image quality, SNR, CNR and radiation dose were compared using a Mann-Whitney test to that of 24 consecutive patients (84.3±4.8 years, 58% female) previously imaged on a 64-row scanner with a conventional two-step protocol. RESULTS: Iodine load was reduced by 44%. All examinations were of diagnostic quality, with improvement of the aortic root CTA image quality (4.9±0.3 versus 4.6±0.5, p<0.01) and a non-significant decrease of the aorto-iliac CTA image quality (4.7±0.6 versus 4.9±0.3, p = 0.07). SNR and CNR were significantly improved in the aortic root CTA (14.0±5.3 and 10.4±4.5 versus 10.3±4.2 and 6.8±3.3, p<0.01 for both) and non-significantly higher in the aorto-iliac CTA (16.5±8.0 and 14.1±7.9 versus 14.7±5.5 and 12.5±5.0, p = 0.42 and p = 0.66). Total radiation dose was reduced by 32%. CONCLUSION: 320-row CT scanner enables a 44% reduction of iodine load in TAVR planning, while maintaining excellent aorto-ilio-femoral arterial enhancement and lowering radiation dose.

Magnetic resonance evaluation of cardiac thrombi and masses by T1 and T2 mapping: an observational study.

The purpose of this work was to evaluate CMR T1 and T2 mapping sequences in patients with intracardiac thrombi and masses in order to assess T1 and T2 relaxometry usefulness and to allow better etiological diagnosis. This observational study of patients scheduled for routine CMR was performed from September 2014 to August 2015. All patients referred to our department for a 1.5 T CMR were screened to participate. T1 mapping were acquired before and after Gadolinium injection; T2 mapping images were obtained before injection. 41 patients were included. 22 presented with cardiac thrombi and 19 with cardiac masses. The native T1 of thrombi was 1037 ± 152 ms (vs 1032 ± 39 ms for myocardium, p = 0.88; vs 1565 ± 88 ms for blood pool, p < 0.0001). T2 were 74 ± 13 ms (vs 51 ± 3 ms for myocardium, p < 0.0001; vs 170 ± 32 ms for blood pool, p < 0.0001). Recent thrombi had a native T1 shorter than old thrombi (911 ± 177 vs 1169 ± 107 ms, p = 0.01). The masses having a shorter T1 than the myocardium were lipomas (278 ± 29 ms), calcifications (621 ± 218 ms), and melanoma (736 ms). All other masses showed T1 values higher than myocardial T1, with T2 consistently >70 ms. T1 and T2 mapping CMR sequences can be useful and represent a new approach for the evaluation of cardiac thrombi and masses.

Multimodality Imaging Diagnostic Approach of Systemic-to-Pulmonary Vein Fistulae.

A 26-year-old man with a history of bilateral lung transplantation for pulmonary cystic fibrosis 6 months before was admitted in our institution for acute heart failure. Cardiac magnetic resonance imaging (CMR) showed an increased aortic output, as aortic flow assessed by velocity mapping was twofold the pulmonary flow, an occluded superior vena cava (SVC), and enlarged azygos vein. A systemic-to-pulmonary vein fistula (SAPVF) was suspected. The selective angiography showed numerous fistulae between intercostals, thyro-cervical, internal mammary arteries and pulmonary veins. The thoracic CT performed before the CMR, which was initially considered as normal, showed well these arteriovenous fistulae after 3D MIP reconstruction. This particular observation highlights the great value of multimodality imaging for the diagnosis of this rare pathology. The MR velocity mapping is a noninvasive imaging technique of great interest to guide the diagnosis of arteriovenous fistulae, and further indicating more invasive complementary imaging modalities like selective arterial angiography.

Native T1 mapping of the heart - a pictorial review.

T1 mapping is now a clinically feasible method, providing pixel-wise quantification of the cardiac structure's T1 values. Beyond focal lesions, well depicted by late gadolinium enhancement sequences, it has become possible to discriminate diffuse myocardial alterations, previously not assessable by noninvasive means. The strength of this method includes the high reproducibility and immediate clinical applicability, even without the use of contrast media injection (native or pre-contrast T1). The two most important determinants of native T1 augmentation are (1) edema related to tissue water increase (recent infarction or inflammation) and (2) interstitial space increase related to fibrosis (infarction scar, cardiomyopathy) or to amyloidosis. Conversely, lipid (Anderson-Fabry) or iron overload diseases are responsible for T1 reduction. In this pictorial review, the main features provided by native T1 mapping are discussed and illustrated, with a special focus on the awaited clinical purpose of this unique, promising new method.

Left ventricular involvement in arrhythmogenic right ventricular cardiomyopathy - a cardiac magnetic resonance imaging study.

BACKGROUND: Few studies evaluated left ventricular (LV) involvement in arrhythmogenic right ventricular dysplasia/cardiomyopathy (ARVD/C). The aim of this study is to determine the frequency, clinical presentation, and pattern of LV involvement in ARVD/C (LV-ARVD/C). METHODS: We retrospectively evaluated the cardiac magnetic resonance (CMR) in 202 patients referred between 2008 and 2012 to our institution, and we determined the presence or the absence of CMR criteria in the revised task force criteria 2010 for the diagnosis of ARVD/C. A total of 21 patients were diagnosed with ARVD/C according to the revised task force criteria 2010. All included patients had no previous history of myocarditis, acute coronary syndrome, or any other cardiac disease that could interfere with the interpretations of structural abnormalities. The LV involvement in ARVD/C was defined by the presence of one or more of the following criteria: LV end-diastolic volume (LVEDV; >95 mL/m(2)), LV ejection fraction (LVEF; <55%), LV late enhancement of gadolinium (LVLE) in a non-ischemic pattern, and LV wall motion abnormalities (WMAs). In the follow-up for the occurrence of cardiac death, ventricular tachycardia (VT) was obtained at a mean of 31 ± 20.6 months. RESULTS: A total of 21 patients had ARVD/C. The median age was 48 (33-63) years. In all, 11 patients (52.4%) had LV-ARVD/C. The demographic characteristics of patients with or without LV were similar. There was a higher frequency of left bundle-branch block (LBBB) VT morphology in ARVD/C (P = 0.04). In CMR, regional WMAs of right ventricle (RV) and RV ejection fraction (RVEF; <45%) were strongly correlated with LV-WMAs (r = 0.72, P = 0.02, r = 0.75, P = 0.02, respectively). RV late enhancement of gadolinium (RVLE) was associated with LV-WMs and LVLE (r = 0.7, P = 0.03; r = 0.8, P = 0.006). LVLE was associated with LV-WMAs, LVEF, and LVEDV (r = 0.9, P = 0.001; r = 0.8, P = 0.001; r = 0.8, P = 0.01). CONCLUSION: LV involvement in ARVD/C is common and frequently associated with moderate to severe right ventricular (RV) abnormalities. The impact of LV involvement in ARVD/C on the prognosis needs further investigations.

The extent of P2Y12 inhibition by clopidogrel in diabetes mellitus patients with acute coronary syndrome is not related to glycaemic control: roles of white blood cell count and body weight.

It was the study objective to determine whether glycaemic control affects the extent of platelet inhibition by thienopyridines as assessed by vasodilator-stimulated phosphoprotein flow cytometry (VASP-FCT) in patients with diabetes mellitus (DM) undergoing percutaneous coronary intervention (PCI) during acute coronary syndrome (ACS). Although the proportion of high on-treatment residual platelet reactivity is higher in DM, the contributions of glycaemic control and other factors associated with DM, such as excess body weight and inflammation, to this impaired platelet inhibition by thienopyridines have not yet been fully characterised. In this study, the extent of P2Y12 ADP receptor pathway inhibition was evaluated by the VASP-FCT. Platelet activation was expressed as the platelet reactivity index (PRI). Low response to clopidogrel (LR) was defined as a PRI of >61%. Four hundred forty-five consecutive ACS patients (DM = 160, NDM = 285) were enrolled. The proportion of LR was higher in DM patients (50 vs. 37.5%). In DM, PRI was not correlated with glycosylated haemoglobin (HbA1c) or glycaemia. In a univariate analysis, LR was associated with age, male sex, overweight, and white blood cell count (WBC). In a multivariate analysis, WBC >10,000 and body weight >80 kg were the sole independent predictors of LR to clopidogrel (hazard ratio (HR) 3.02 [1.36-6.68], p=0.006 and HR 2.47 [1.14-5.35], p = 0.021, respectively). Conversely, in non-DM patients, ST-elevation myocardial infarction was the sole independent predictor of LR. In conclusion, in ACS DM patients undergoing PCI, the extent of P2Y12 inhibition by clopidogrel is not related to glycaemic control but is related to body weight and inflammatory status as assessed by the WBC.

Association of estimated GFR with platelet inhibition in patients treated with clopidogrel.

BACKGROUND: The reasons that decreased glomerular filtration rate (GFR) might alter the clinical efficacy of clopidogrel are poorly understood. STUDY DESIGN: In this study, we sought to evaluate whether decreased GFR alters platelet response to clopidogrel in patients receiving a maintenance dose of clopidogrel (75 mg/d for at least 8 days). SETTINGS & PARTICIPANTS: 126 consecutive patients categorized by estimated GFR: stages 1-2 (>60 mL/min/1.73 m(2); n = 29), stage 3a (45-59 mL/min/1.73 m(2); n = 21); stage 3b (30-44 mL/min/1.73 m(2); n = 26), stage 4 (15-29 mL/min/1.73 m(2); n = 14), and stage 5 (<15 mL/min/1.73 m(2); n = 36) were prospectively enrolled. PREDICTOR: Residual platelet reactivity, defined in the VASP (Vasodilator Stimulated Phosphoprotein) flow cytometry test as platelet reactivity index (PRI) ≥61% and in the VerifyNow turbidimetric-based assay as a value >235 PRU (adenosine diphosphate receptor reaction units) or percentage of platelet inhibition <15%. OUTCOMES: We examined factors associated with low response to clopidogrel using logistic regression. RESULTS: A significant relationship between estimated GFR, PRI, PRU, and percentage of inhibition was found. The prevalence of residual platelet reactivity was highest in patients with GFR stage 5. PRI ≥61% occurred in 52.8% of patients with stage 5 versus 30.8% of stage 3b and 24.1% of stages 1-2 (P = 0.1). PRU >235 was found in 63.6% of patients with stage 5 versus 36.8% of stage 3b and 17.2% of stages 1-2 (P = 0.005). Inhibition <15% affected 66.7% of patients with stage 5 versus 21.1% of stage 3b and 17.2% of stages 1-2 (P < 0.001). In the multivariable model, GFR stage 5 (adjusted prevalence ratio [PR], 3.10; 95% CI, 1.23-9.43; P = 0.02), and obesity (adjusted PR, 1.92; 95% CI, 1.34-2.23; P = 0.004) were the sole predictors of residual platelet reactivity. LIMITATIONS: Interference of hemodialysis with the pharmacokinetics of clopidogrel could not be excluded. CONCLUSION: GFR stage 5 is associated with substantial impairment of platelet inhibition independently of diabetes mellitus.

Cardiovascular mortality in chronic kidney disease patients undergoing percutaneous coronary intervention is mainly related to impaired P2Y12 inhibition by clopidogrel.

OBJECTIVES: We sought to determine whether low platelet response to the P2Y(12) receptor antagonist clopidogrel as assessed by vasodilator-stimulated phosphoprotein flow cytometry test (VASP-FCT) differentially affects outcomes in patients with or without chronic kidney disease (CKD) undergoing percutaneous coronary intervention (PCI). BACKGROUND: Although both CKD and impaired platelet responsiveness to clopidogrel are strong predictors of unfavorable outcome after PCI, the impact of their association is unknown. The platelet VASP-FCT assay is specific for the P2Y(12) ADP receptor pathway. In this test, platelet activation is expressed as the platelet reactivity index (PRI). METHODS: Four-hundred forty unselected patients (CKD: 126, estimated glomerular filtration rate [eGFR] <60 ml/min/1.73 m(2)), no-CKD: 314 eGFR >60 ml/min/1.73 m(2)) undergoing urgent (n = 336) or planned (n = 104) PCI were prospectively enrolled. In each subgroup, patients were classified as low-responders (LR: PRI ≥ 61%) or responders (R: PRI <61%) to clopidogrel. RESULTS: At a mean follow-up of 9 ± 2 months, all-cause mortality, cardiac death, and possible stent thrombosis were higher in CKD than in no-CKD patients. Within the CKD group, the LR status was associated with higher rates of all-cause mortality (25.5% vs. 2.8%, p < 0.001), cardiac death (23.5% vs. 2.8%, p < 0.001), all stent thrombosis (19.6% vs. 2.7%, p = 0.003), and MACE (33.3% vs. 12.3%, p = 0.007). Conversely, in no-CKD patients, the LR status did not affect outcomes. Multivariate analysis identified Killip class ≥ 3, drug-eluting stent implantation, and the interaction between LR and CKD (hazard ratio: 11.96, 95% confidence interval: 1.22 to 116.82; p = 0.033) as independent predictors of cardiac death. CONCLUSIONS: In CKD patients, the presence of low platelet response to clopidogrel is associated with worse outcomes after PCI.

Impact of P2Y12 inhibition by clopidogrel on cardiovascular mortality in unselected patients treated by percutaneous coronary angioplasty: a prospective registry.

OBJECTIVES: The aim of this study was to determine whether low platelet response to the P2Y(12) receptor antagonist clopidogrel as assessed by Vasodilator-stimulated phosphoprotein flow cytometry test (VASP- FCT) predicts cardiovascular events in a high-risk population undergoing percutaneous coronary intervention (PCI). BACKGROUND: Impaired platelet responsiveness to clopidogrel is thought to be a determinant of cardiovascular events after PCI. The platelet VASP-FCT is a new assay specific to the P2Y(12) adenosine diphosphate receptor-pathway. In this test, platelet activation is expressed as platelet reactivity index (PRI). METHODS: Four-hundred sixty-one unselected patients undergoing urgent (n = 346) or planned (n = 115) PCI were prospectively enrolled. Patients were classified as low-response (LR) and response (R) to clopidogrel, depending on their PRI. Optimal PRI cutoff was determined by receiver-operator characteristic curve analysis to 61% (LR: PRI > or =61% and R: PRI <61%). Follow-up was obtained at a mean of 9 +/- 2 months in 453 patients (98.3%). RESULTS: At follow-up, total cardiac mortality rates and possible and total stent thrombosis were higher in LR patients. Multivariate analysis identified creatinine clearance (hazard ratio [HR]: 0.95; 95% confidence interval [CI]: 0.93 to 0.98, p < 0.001), drug-eluting stent (HR: 5.73; 95% CI: 1.40 to 23.43, p = 0.015), C-reactive protein (HR: 1.01; 95% CI: 1.001 to 1.019, p = 0.024), and LR to clopidogrel (HR: 4.00; 95% CI: 1.08 to 14.80, p = 0.037) as independent predictors of cardiac death. The deleterious impact of LR to clopidogrel on cardiovascular death was significantly higher in patients implanted with drug-eluting stent. CONCLUSIONS: In patients undergoing PCI, LR to clopidogrel assessed by VASP-FCT is an independent predictor of cardiovascular death at the PRI cutoff value of > or =61%. The LR clinical impact seems to be dependent on the type of stent implanted.