Background: Early safety outcomes following transcatheter aortic valve implantation (TAVI) for severe aortic stenosis are critical for patient prognosis. Accurate prediction of adverse events can enhance patient management and improve outcomes. Aim: This study aimed to develop a machine learning model to predict early safety outcomes in patients with severe aortic stenosis undergoing TAVI. Methods: We conducted a retrospective single-centre study involving 224 patients with severe aortic stenosis who underwent TAVI. Seventy-seven clinical and biochemical variables were collected for analysis. To handle unbalanced classification problems, an adaptive synthetic (ADASYN) sampling approach was used. A fined-tuned random forest (RF) machine learning model was developed to predict early safety outcomes, defined as all-cause mortality, stroke, life-threatening bleeding, acute kidney injury (stage 2 or 3), coronary artery obstruction requiring intervention, major vascular complications, and valve-related dysfunction requiring repeat procedures. Shapley Additive Explanations (SHAPs) were used to explain the output of the machine learning model by attributing each variable’s contribution to the final prediction of early safety outcomes. Results: The random forest model identified left femoral artery diameter and aortic valve calcification volume as the most influential predictors of early safety outcomes. SHAPs analysis demonstrated that smaller left femoral artery diameter and higher aortic valve calcification volume were associated with poorer early safety prognoses. Conclusions: The machine learning model highlights of early safety outcomes after TAVI. These findings suggest that incorporating these variables into pre-procedural assessments may improve risk stratification and inform clinical decision-making to enhance patient care.

Background: Transcatheter aortic valve implantation (TAVI) has emerged as a pivotal intervention for managing severe aortic stenosis in high-risk surgical patients. Objective: This study aimed to evaluate the impacts of procedural factors and patient characteristics on TAVI outcomes, with a focus on survival rates, cardiac mortality, and associated complications. Methods: A retrospective, single-center study involving 224 patients who underwent TAVI at the Lithuanian University of Health Sciences from September 2021 to April 2023 was conducted. Data encompassing demographic characteristics, medical history, procedural specifics, and follow-up outcomes were analyzed. Survival and adverse events were assessed at 30 days, 6 months, and 12 months post-TAVI. Results: The study included 224 patients. The mean age in the non-death group was 80 ± 6.17 years (range, 49–91), while that in the cardiac death group was 81.5 ± 6.14 years (range, 70–94; p = 0.079). Males accounted for 37.7% of the non-death group and 50% of the cardiac death group (p = 0.304). Statistical analyses identified factors significantly associated with mortality and complications. The overall survival rate was 88.8%, with cardiac-related mortality observed in 8% of patients. Increased fluoroscopy time (p < 0.001), a higher contrast volume (p = 0.005), and less improvement in aortic valve velocity post-TAVI (p = 0.031) were significantly associated with cardiac mortality. Advanced age and a reduced left ventricular ejection fraction (<50%) were prominent predictors of adverse outcomes. Patients with non-coronary cusp calcification exhibited lower cardiac mortality (p = 0.005), while mitral valve regurgitation was linked to poorer outcomes (p = 0.015). Logistic regression analysis underscored the incremental risks posed by procedural complexities and comorbidities. Conclusions: Procedural factors such as fluoroscopy duration and contrast volume, along with patient-specific attributes including age, left ventricular function, and valve calcification patterns, critically influence TAVI outcomes. These findings emphasize the need for tailored procedural strategies and patient management protocols to mitigate risks and enhance the efficacy of TAVI interventions.

This study aims to evaluate patients with limited state of changes in coronary arteries detected by coronary angiography, the dynamics of these changes over the two years, identify the relevant diagnostic criteria, and assess the efficacy of applied treatment by using speckle tracking echocardiography. Peak radial and circumferential strain and SR (systolic, early, and late diastolic strains) were measured based on the short-axis view; peak longitudinal strain and SR were measured from the apical side of four- two- and three-chamber views. Radial, longitudinal (GLS), circumferential global and regional strains were calculated as an average of measurements. All patients (n - 146) were assigned to normal (control) and CAD groups according to cardiac angiography results. 128 of them were evaluated repeatedly after two years. Depending on angiography findings, LAD (85.83%) stenosis predominate, when subsequently fewer instances of RCA (52.5%) or LCX (40.83%) were observed. Most (about 80%) of the patients had one or two-vessel disease and only 20% had systemic all three-vessel disease. Analysis of STE data in groups during a two-year study period showed statistically reliable differences associated with a particular coronary artery. In the control group: RCA-myocardial circumferential strain (p - 0.037); LAD-no changes; LCX-early (p - 0.013) and late diastolic longitudinal (p - 0.033) strains. Subsequently, in the CAD group: RCA-diastolic circumferential strain rate (p - 0.007); LAD-myocardial longitudinal strain (p - 0.006), systolic longitudinal (p - 0.038) and circumferential strain (p - 0.012) rates, early diastolic circumferential (p - 0.008) and late diastolic longitudinal (p - 0.037) strain rates; LCX-myocardial longitudinal (p - 0.049) strain. Between groups, we detected significant changes in such circumferential strain rates, respectively: RCA- systolic (p=0.037), early diastolic (p=0.019), and late diastolic (p=0.024) strain rates; LAD-no changes; LCX-early diastolic longitudinal strain (p - 0.004). The clinical condition of our patients over the two years has improved both in control and CAD groups, according to GLS. We hold the opinion that microvascular angina (MVA) may be responsible for such an improvement because the main diagnostic criteria and common treatment with ACE inhibitors, statins, beta-blockers, antithrombotic, and nitrates was typical and effective for MVA treatment.