Artificial intelligence to improve decision making in transcatheter aortic valve implantation

Background

The current state of the art in the diagnosis and treatment of cardiovascular diseases has been based on evidence resulting from traditional trials as well as years of clinical experience. Due to interindividual differences and a huge number of possible cofounders, in interventional cardiology a linear algorithm is usually not able to precisely estimate individual risk, therapy, or outcome. With the technological evolution in deep machine learning (ML) and artificial intelligence (AI), clinicians may now address aspects that might not have been investigated previously, as supercomputers may handle the plethora of data that are generated as part of treatment. Ultimately, treatment recommendations and decisions may be made on a personalized level.

Purpose

The aim of this study was to apply AI to routine clinical practice to improve decision making in transcatheter aortic valve implantation (TAVI) to determine the best type and size of prosthesis personalized for each patient with pre-procedural risk stratification.

Methods

All patients included in the study were undergoing TAVI. To predict the clinical outcomes we applied a random forest classification, a ML method with high interpretability. For baseline data 58 features were chosen, including valve type and size used. After removing highly collinear features with a certain variance inflation factor, feature selection was based on impurity-based feature importance as well as permutation importance. The performance of the estimators was evaluated by a five-fold nested stratified cross-validation. To evaluate the model ROC and mean AUC scores were chosen.

Results

A total of 3882 patient datasets were included in this trial. The baseline characteristics were consistent with a high cardiovascular risk typical of this collective. Device success was achieved in 83.3%, pacemaker implantation was necessary in 12.2%, and aortic valvular insufficiency was observed in 2.5%. The 30-day mortality was 3.4% and one-year mortality was 12.7%. The mean AUC for the outcome parameters device success, aortic valvular insufficiency, any pacemaker operation, and 30-day and one-year mortality after five-fold cross validation were 0.61±0.03, 0.71±0.04, 0.66±0.04, 0.67±0.03, and 0.69±0.01, respectively.

Conclusions

We report preliminary data concerning a promising method to improve decision making in the context of TAVI evaluation and planning using ML algorithm. We showed the feasibility with acceptable AUC values for all outcome parameters. Thus, the integration of AI in TAVI strategy planning process offers a valuable tool providing patient focused personalized therapy.

Funding Acknowledgement

Type of funding sources: None.

P2651Seasonal variations in the pathogenesis of acute coronary syndromes

Background

Seasonal variations in acute coronary syndrome (ACS) has been known with the winter being the peak in incidence and mortality. However, underlying pathophysiology for this variation has not been studied.

Purpose

We sought to compare pathobiology of the culprit lesions assessed by optical coherence tomography (OCT) among the four seasons.

Methods

Patients with ACS who underwent OCT were recruited from 6 countries in the Northern Hemisphere. The prevalence of 3 most common pathologies, plaque rupture, plaque erosion and calcified plaque, and other features of coronary plaques were compared among the four seasons.

Results

In 1113 patients with ACS, 284 (25%) patients were admitted in spring, 243 (22%) patients in summer, 290 (26%) patients in autumn and 296 (27%) patients in winter. The proportion of underlying 3 pathologies was significantly different in each season (prevalence of plaque rupture, plaque erosion, calcified plaque was 50%, 39%, and 11%, respectively in the spring; 44%, 43%, and 13% in the summer; autumn: 49%, 39%, and 12% in the autumn; 57%, 30%, and 13% in the winter; P=0.039). The proportion of plaque rupture was higher in winter but lower in summer, and that of plaque erosion was higher in summer, but lower in winter. Maximum and minimum temperatures on the day of OCT procedure were significantly lower in the plaque rupture group than in the plaque erosion group (P=0.02 and P=0.012, respectively). In the rupture group, the prevalence of hypertension was significantly higher in winter, but in the erosion group, it was not different among the four seasons.

Figure 1. The proportion of culprit lesion characteristics were significantly different among the 4 season groups. (P=0.039) The proportion of plaque rupture was significantly higher in winter but lower in summer. In contrast, the proportion of plaque erosion was higher in summer, but lower in winter.

Conclusions

Seasonal variation of the underlying mechanisms of ACS reflects different pathobiology. The proportion of plaque rupture is highest in winter and the proportion of plaque erosion is highest in summer. A different approach may be needed for the prevention and treatment of ACS depending on the season of its occurrence.

P3710Long term follow-up in a real-world study cohort after patent foramen ovale closure

Background

Patent foramen ovale (PFO) closure is the treatment of choice after cryptogenic stroke according recent evidence. The indication is based on results of several randomized controlled trials; however, the results of these trials may not be extrapolated to a real-world clinical setting. Therefore, the aim of the present study was to evaluate long-term outcome regarding recurrent stroke, migraine, and/or peripheral embolism in patients after PFO closure.

Methods

We retrospectively analyzed outcomes of consecutive patients undergoing PFO closure from 2011 to 2018 at two interventional sites with respect to periprocedural events occurring during hospitalization and long-term follow-up. Follow-up data were collected from outpatient visits or telephone interviews.

Results

The analysis included 214 consecutive patients (mean age 52 years; 58% male). The follow-up rate was 96% and the mean follow-up time was 38 (SD 22) months. The index vascular event leading to PFO closure was stroke (n=190; 89%), including patients with repetitive stroke (n=36), embolic myocardial infarction (n=21), and migraine (n=3). One quarter (24.6%) of the population studied had an atrial septal aneurysm (>15 mm). Procedural success was achieved in 98%. There were no procedure-related strokes or deaths. Periprocedural complications occurred in 16 patients (7%): two cases of pericardial tamponade, seven complications at the access site mainly caused by bleeding, two cases of transient atrial fibrillation, and five other complications. The Amplatz Septal Occluder™ was used in two thirds (64.5%) of the cases and the Gore Cardioform™ device in one third (28.6%). Four (2%) patients died during follow-up. None of these patients experienced a recurrent stroke. Ten (5%) other patients experienced a recurrent stroke. Patients with recurrent stroke events were older than patients without recurrent stroke (mean 62.6 [SD 8.8] years vs. mean 52.2 [SD 13.8] years; p=0.015) and had a higher rate of preexisting cerebrovascular occlusive disease (5 [50%] vs. 10 [6%]; p<0.0001). There was no difference in risk for recurrent stroke between patients with one prior stroke and more than one stroke before PFO closure (p=0.71). Atrial fibrillation occurred in 6.6% of the patients during follow-up, but only one of these patients had recurrent stroke. No other anatomic and vascular risk factors or antithrombotic treatments were identified as being predictive of embolic events after closure.

Conclusion

In this real-world PFO closure cohort the recurrent stroke rate is low, although it is higher than reported in the recent randomized controlled trials. Recurrent strokes after PFO closure may reflect additional comorbid risk factors such as age or cerebrovascular occlusive disease that are unrelated to the potential for paradoxical embolism.

Everolimus- vs. novolimus-eluting bioresorbable scaffolds in patients with acute coronary syndrome

BACKGROUND

Limited data exist on bioresorbable scaffolds (BRS) in patients with acute coronary syndrome (ACS). The aim of the present study was to evaluate novolimus-eluting BRS (DESolve) as interventional treatment for patients with ACS, and to compare its 12-month outcomes with the everolimus-eluting bioresorbable scaffolds (Absorb).

METHODS

In this retrospective study, patients with ACS (including unstable angina pectoris, ST-segment elevation myocardial infarction, or non-ST-segment elevation myocardial infarction) treated with either the Absorb or the DESolve BRS were evaluated in a 1:1 matched-pair analysis. Major adverse cardiac events (MACE), including death, myocardial infarction, and target lesion revascularization, were evaluated as a major endpoint. The occurrence of scaffold thrombosis was also assessed.

RESULTS

A total of 102 patients were eligible for this analysis. The rate of MACE at 12 months was comparable between the Absorb and the DESolve group (8.3% vs. 6.8%, p = 0.738). The occurrence of target lesion revascularization (6.2% vs. 4.7%; p = 0.700) and scaffold thrombosis (4.1% vs. 2.1%; p = 0.580) was comparable as well. All instances of scaffold thrombosis occurred within 30 days of the index procedure.

CONCLUSION

In this study, similar 12-month event rates were observed for both BRS types after implantation for the treatment of ACS.

[Procedural aspects in primary PCI: arterial access, stent selection, thrombectomy and treatment of non-culprit lesions]

Acute myocardial infarction was one of the most common causes of death in Germany in 2011. According to the guidelines of the European Society for Cardiology, systemic fibrinolysis and primary percutaneous coronary intervention (PCI) are the methods of choice for acute treatment. Primary PCI should be given priority due to its superiority. The transradial access should be preferred due to the lower bleeding complication rate. In the selection of stents the new generation of drug-eluting stents (DES) are superior to the first generation of bare metal stents (BMS). It has now been demonstrated that the incident rates of DES (e.g. mortality, target vessel revascularization, early and late stent thrombosis and myocardial infarction) are significantly lower. For bioresorbable scaffolds (BRS) long-term results for the use in treatment of ST-elevation myocardial infarction (STEMI) are not yet available but initial results are very promising. However, the selection of a stent needs to be done on an individual basis in order to do justice to all aspects. Data with respect to thrombectomy in acute treatment are heterogeneous. Currently, a thorough consideration of all aspects is necessary because thrombus aspiration can also be associated with an increased rate of incidents. In a state of hemodynamic stability only so-called culprit lesions should currently be treated with a stent. Elective interventions on further stenoses should be carried out after consideration of individual factors and if necessary evaluation of the hemodynamic relevance.

[Historical and current pathophysiological concepts of stress (Tako-Tsubo) cardiomyopathy]

Tako-Tsubo cardiomyopathy (TTC), also referred to as stress cardiomyopathy (SCM), was first described in the 1990s and is characterized by transient left ventricular dysfunction. Its incidence represents 1-2% of all acute coronary syndromes (ACS). In most cases extreme emotional or physical stress precedes this syndrome. The majority of patients affected are postmenopausal women. Since its first description, various hypotheses regarding the pathophysiology of TTC have been discussed. From a historical perspective, coronary vasospasm has often been proposed as a possible cause of this disorder. However, reviews from numerous registries were able to demonstrate that coronary vasospasm plays only a minor role in the pathogenesis of TTC. Several groups showed disturbances in myocardial microcirculation and energy metabolism in the acute phase of TTC. Nevertheless, with regard to the data currently available, it cannot be differentiated whether these changes are the cause or rather the result of TTC. However, recent concepts include an excessive catecholamine overload and morphological changes which are unequivocally documented in TTC. The relation between elevated catecholamine levels and myocardial dysfunction analogous to TTC could be confirmed in animal experiments.In summary, it can be assumed that TTC is caused by an excessive cardiotoxic release of catecholamines. Ventricular dysfunction can be explained by increased numbers of β-adrenergic receptors in the apex, leading to greater vulnerability to catecholamine overload. Individual anatomical differences in the sympathoadrenergic system and distribution from β-adrenergic receptors are presumably responsible for the interindividual occurrence of wall motion abnormalities in TTC.

Stem cell-mediated natural tissue engineering

Recently, we demonstrated that a fully differentiated tissue developed on a ventricular septal occluder that had been implanted due to infarct-related septum rupture. We suggested that this tissue originated from circulating stem cells. The aim of the present study was to evaluate this hypothesis and to investigate the physiological differentiation and transdifferentiation potential of circulating stem cells. We developed an animal model in which a freely floating membrane was inserted into each the left ventricle and the descending aorta. Membranes were removed after pre-specified intervals of 3 days, and 2, 6 and 12 weeks; the newly developed tissue was evaluated using quantitative RT-PCR, immunohistochemistry and in situ hybridization. The contribution of stem cells was directly evaluated in another group of animals that were by treated with granulocyte macrophage colony-stimulating factor (GM-CSF) early after implantation. We demonstrated the time-dependent generation of a fully differentiated tissue composed of fibroblasts, myofibroblasts, smooth muscle cells, endothelial cells and new blood vessels. Cells differentiated into early cardiomyocytes on membranes implanted in the left ventricles but not on those implanted in the aortas. Stem cell mobilization with GM-CSF led to more rapid tissue growth and differentiation. The GM-CSF effect on cell proliferation outlasted the treat ment period by several weeks. Circulating stem cells contributed to the development of a fully differentiated tissue on membranes placed within the left ventricle or descending aorta under physiological conditions. Early cardiomyocyte generation was identified only on membranes positioned within the left ventricle.

Desynchronization: a novel model to induce heart failure

BACKGROUND

Previous large animal heart failure models led to inhomogeneous results. Therefore, we developed a novel model combining rapid pacing with forced ventricular desynchronization.

METHODS

Heart failure was induced in 20 pigs during a pacing period of 21 days. Group A (n = 10) received one right ventricular lead (220 bpm). In group B (n = 10), two leads were implanted in different right ventricular regions with beat-to-beat alternation of activation sites (each lead 110 bpm). Sham-operated pigs (n = 6) served as controls. Hemodynamics were invasively evaluated and tissue was analyzed by immunohistochemistry and zymography.

RESULTS

Hemodynamics were significantly more impaired in group B with an increase of pulmonary capillary wedge and central venous pressure and a reduction of cardiac index (control 4.3 +/- 0.1 l/min/m (2); A 3.6 +/- 0.2; B 2.9 +/- 0.2, P < 0.05). Heart-to-body weight ratio was significantly higher in group B. Histological analyses showed a significant increase of cell diameters and interstitial fibrosis with significantly higher collagen contents in group B.

CONCLUSION

The new model with a combination of rapid pacing and forced desynchronization of the ventricular contraction is superior to traditional heart failure models induced solely by rapid pacing.

Effects of 2-Deoxy-d-Glucose on Proliferation of Vascular Smooth Muscle Cells and Endothelial Cells

2-Deoxy-d-glucose (2-DG) is a glucose analogue that has been proposed for cancer therapy due to its cytostatic properties. Its effect on the proliferation of smooth muscle cells and endothelial cells has not been fully clarified. The aims of this study were to investigate the effects of 2-DG on the proliferation of porcine aortic endothelial cells (PAEC) and porcine smooth muscle cells (PSMC), to establish an overview of its dose-dependent inhibitory capacity and to examine whether the short-term incubation of cells with 2-DG has an impact on cell proliferation in culture. Our results showed a dose-dependent significant inhibitory effect on proliferation, which was more pronounced in PSMC than in PAEC. Even after short-term incubation of cells with 2-DG, relevant inhibition of proliferation was documented. The clinical application of 2-DG might be a promising concept by inhibiting cells that show a potentially rapid proliferation in response to non-malignant stimuli, such as smooth muscle cells after intracoronary stenting.

Negative Inotropic Effect of Rapamycin on Isolated Human Cardiomyocytes

Rapamycin is an increasingly important immunosuppressive drug and reduces restenosis after coronary stenting, but its effects on cardiac contractility are largely unknown. We investigated the acute inotropic effects of rapamycin on isolated human cardiomyocytes. Cardiomyocytes were enzymatically isolated from right atrial appendages obtained during routine coronary artery bypass surgery. Cell morphology was examined by confocal microscopy. Cell contraction was recorded after electrical stimulation. Rapamycin elicited a concentration-dependent decrease in fractional cell shortening ranging from 14.3 ± 2.6% at 10−8 M rapamycin to 26.4 ± 4.2% at 10−5 M. Rapamycin also caused a concentration-dependent decrease in diastolic cell length. Contractile performance of isolated cardiomyocytes was well preserved, as evidenced by the profound positive inotropic effects of high extracellular calcium concentration and the β-adrenoreceptor agonist isoproterenol. The acute negative inotropic effect of rapamycin on human cardiomyocytes might be due to altered calcium homeostasis through the binding of rapamycin to FKBP12.6 and its regulatory function on the ryanodine receptor, with increased calcium leakage from the sarcoplasmic reticulum.

Left Ventricular Tamponade - A Clinical Chameleon

We present a 45-year-old patient who had to undergo aortocoronary bypass surgery after acute posterior myocardial infarction. On day twelve, the patient suddenly developed the classic signs of cardiogenic shock including angina, tachycardia, and hypotension. ECG displayed significant ST-elevations and troponin T was positive. Echocardiography suspected relevant pericardial effusion. However, the typical clinical signs of acute pericardial tamponade(distension of jugular veins, paradoxical pulse) were absent. Therefore, a computed tomography was carried out, which confirmed an isolated left ventricular tamponade resulting in severe diastolic and systolic dysfunction with profoundly impaired left ventricular filling. Immediate operative drainage was necessary since percutaneous pericardiocentesis was impossible given the untypical localization. This case demonstrates that clearcut signs of myocardial infarction can be misleading and may represent the untypical presentation of left ventricular tamponade,particularly in the setting after open-heart surgery.