The potential of the transformer-based survival analysis model, SurvTrace, for predicting recurrent cardiovascular events and stratifying high-risk patients with ischemic heart disease.

INTRODUCTION: Ischemic heart disease is a leading cause of death worldwide, and its importance is increasing with the aging population. The aim of this study was to evaluate the accuracy of SurvTrace, a survival analysis model using the Transformer-a state-of-the-art deep learning method-for predicting recurrent cardiovascular events and stratifying high-risk patients. The model's performance was compared to that of a conventional scoring system utilizing real-world data from cardiovascular patients. METHODS: This study consecutively enrolled patients who underwent percutaneous coronary intervention (PCI) at the Department of Cardiovascular Medicine, University of Tokyo Hospital, between 2005 and 2019. Each patient's initial PCI at our hospital was designated as the index procedure, and a composite of major adverse cardiovascular events (MACE) was monitored for up to two years post-index event. Data regarding patient background, clinical presentation, medical history, medications, and perioperative complications were collected to predict MACE. The performance of two models-a conventional scoring system proposed by Wilson et al. and the Transformer-based model SurvTrace-was evaluated using Harrell's c-index, Kaplan-Meier curves, and log-rank tests. RESULTS: A total of 3938 cases were included in the study, with 394 used as the test dataset and the remaining 3544 used for model training. SurvTrace exhibited a mean c-index of 0.72 (95% confidence intervals (CI): 0.69-0.76), which indicated higher prognostic accuracy compared with the conventional scoring system's 0.64 (95% CI: 0.64-0.64). Moreover, SurvTrace demonstrated superior risk stratification ability, effectively distinguishing between the high-risk group and other risk categories in terms of event occurrence. In contrast, the conventional system only showed a significant difference between the low-risk and high-risk groups. CONCLUSION: This study based on real-world cardiovascular patient data underscores the potential of the Transformer-based survival analysis model, SurvTrace, for predicting recurrent cardiovascular events and stratifying high-risk patients.

Predictors of long-term success after successful explantation of continuous flow left ventricular assist device support.

OBJECTIVES: Predictors and evaluations of continuous flow left ventricular assist device (cf-LVAD) explantation in recovered patients remain under discussion due to lack of evidence on long-term safety and efficacy. This study summarized our experiences regarding cf-LVAD explantation in non-ischaemic dilated cardiomyopathy patients and estimated a predictor for sufficient myocardial recovery allowing left ventricular assist device explant. METHODS: We retrospectively identified 135 adult patients with cf-LVAD therapy as bridge to heart transplant due to non-ischaemic dilated cardiomyopathy. Of those, 13 patients underwent device explantation (recovery group) after myocardial recovery. Twelve (92%) of the explanted patients were evaluated using our weaning protocol and underwent surgical explantation. Meanwhile, the remaining 122 continued with cf-LVAD therapy (non-recovery group). RESULTS: Multivariate logistic regression analysis revealed time interval between the first heart failure event and cf-LVAD implantation as an independent predictor for successful explantation. The optimal time interval cutoff value to predict cf-LVAD explantation was 7 months, with a sensitivity of 91.0% and specificity of 84.6%. Echocardiography in patients with successful cf-LVAD explantation showed significant improvement of left ventricular function and dimensions at 6 months postoperatively. The 13 explanted patients are currently alive at a median of 30 (interquartile range; 18-58) months after explantation. The survival rate free from rehospitalization due to heart failure following explantation was 100%. Left ventricular function and remodelling after explantation were also preserved. CONCLUSIONS: In non-ischaemic dilated cardiomyopathy patients with a short interval between the first heart failure event and cf-LVAD therapy, left ventricular myocardium may recover in an early phase after device implantation.

A leadless pacemaker implantation for a patient with systemic right ventricle under ventricular assist device support.

Patients with congenitally corrected transposition of the great arteries (ccTGA) often develop complete atrioventricular block and heart failure due to the abnormal disposition of atrioventricular node and disadvantage of systemic right ventricle. These issues are managed with a pacing system and a ventricular assist device (VAD), respectively. While technological advances offer new treatment strategies, the simultaneous deployment of a leadless pacemaker and a VAD in cases of ccTGA remains unexplored. Here, we present a case of leadless pacemaker implantation for a VAD-supported ccTGA patient. The safety of a leadless pacemaker for a subpulmonary left ventricle and electromagnetic interference between devices are major concerns when implanting a leadless pacemaker; however, the current case overcomes these obstacles. There were no perioperative complications, and both devices were functioning without problems during a one-year follow up. We expect that, even in patients with cardiac complexity such as systemic right ventricle under VAD support, a leadless pacemaker could become the treatment of choice if the indication is appropriate, although careful and close follow up is needed. Learning objective: Technological advances expand treatment strategies and provide significant benefits to patients with adult congenital heart disease (ACHD). However, discussion of the combination of a leadless pacemaker and a ventricular assist device (VAD) is rare. We demonstrated the efficacy of a leadless pacemaker for a subpulmonary left ventricle in a patient with systemic right ventricle on VAD. This approach could be an option even for ACHD patients.

Comprehensive Risk Assessment in Patients With Pulmonary Arterial Hypertension Referred for Lung Transplantation.

BACKGROUND: Whether comprehensive risk assessment predicts post-referral outcome in patients with pulmonary arterial hypertension (PAH) referred for lung transplantation (LT) in Japan is unknown.Methods and Results: We retrospectively analyzed 52 PAH patients referred for LT. Risk status at referral was assessed using 3- and 4-strata models from the 2022 European Society of Cardiology and European Respiratory Society guidelines. The 3-strata model intermediate-risk group was further divided into 2 groups based on the median proportion of low-risk variables (modified risk assessment [MRA]). The primary outcome was post-referral mortality. During follow-up, 9 patients died and 13 patients underwent LT. There was no survival difference among 3-strata model groups. The 4-strata model classified 33, 16, and 3 patients as low intermediate, high intermediate, and high risk, respectively. The 4-strata model identified high-risk patients with a 1-year survival rate of 33%, but did not discriminate survival between the intermediate-risk groups. The MRA classified 15, 28, 8, and 1 patients as low, low intermediate, high intermediate, and high risk, respectively. High intermediate- or high-risk patients had worse survival (P<0.001), with 1- and 3-year survival rates of 64% and 34%, respectively. MRA high intermediate- or high-risk classification was associated with mortality (hazard ratio 12.780; 95% confidence interval 2.583-63.221; P=0.002). CONCLUSIONS: Patients classified as high intermediate or high risk by the MRA after treatment should be referred for LT.

[Successful cord blood transplantation for refractory gamma-delta hepatosplenic T-cell lymphoma developed during treatment of Crohn's disease].

The patient was a 21-year-old man who had been diagnosed with Crohn's disease and received infliximab and azathioprine six years earlier. He was admitted with fever and fatigue. Peripheral blood examination showed LDH 2,473 U/l and thrombocytopenia, and contrast-enhanced computed tomography (CT) showed hepatosplenomegaly. Bone marrow biopsy and liver biopsy showed CD4+CD56+TCRγδ＋CD8- atypical cells, leading to a diagnosis of hepatosplenic T-cell lymphoma (HSTCL). The patient was refractory to CHOP and DA-EPOCH, and therefore received cord blood transplantation with myeloablative conditioning. CT showed reduced in hepatosplenomegaly and peripheral blood examination showed LDH 165 U/l and plt 180,000/µl, so the patient was discharged on day117. HSTCL is a tumor of immature γδT cells with a Vδ1 mutation in the spleen, and immunodeficiency has been implicated in its pathogenesis. Patients with inflammatory bowel disease treated with azathioprine are known to have an increased risk of lymphoproliferative disease. In this case, use of immunosuppressive drugs for Crohn's disease may have caused malignant transformation of γδ cells in the intestinal epithelium. Although the patient was refractory to chemotherapy, he was able to achieve remission with early cord blood transplantation and long-term survival is expected.

Multimodality Risk Assessment of Patients with Ischemic Heart Disease Using Deep Learning Models Applied to Electrocardiograms and Chest X-rays.

Comprehensive management approaches for patients with ischemic heart disease (IHD) are important aids for prognostication and treatment planning. While single-modality deep neural networks (DNNs) have shown promising performance for detecting cardiac abnormalities, the potential benefits of using DNNs for multimodality risk assessment in patients with IHD have not been reported. The purpose of this study was to investigate the effectiveness of multimodality risk assessment in patients with IHD using a DNN that utilizes 12-lead electrocardiograms (ECGs) and chest X-rays (CXRs), with the prediction of major adverse cardiovascular events (MACEs) being of particular concern.DNN models were applied to detection of left ventricular systolic dysfunction (LVSD) on ECGs and identification of cardiomegaly findings on CXRs. A total of 2107 patients who underwent elective percutaneous coronary intervention were categorized into 4 groups according to the models' outputs: Dual-modality high-risk (n = 105), ECG high-risk (n = 181), CXR high-risk (n = 392), and No-risk (n = 1,429).A total of 342 MACEs were observed. The incidence of a MACE was the highest in the Dual-modality high-risk group (P < 0.001). Multivariate Cox hazards analysis for predicting MACE revealed that the Dual-modality high-risk group had a significantly higher risk of MACE than the No-risk group (hazard ratio (HR): 2.370, P < 0.001), the ECG high-risk group (HR: 1.906, P = 0.010), and the CXR high-risk group (HR: 1.624, P = 0.018), after controlling for confounding factors.The results suggest the usefulness of multimodality risk assessment using DNN models applied to 12-lead ECG and CXR data from patients with IHD.

Impact of psychiatric disorders on the hemodynamic and quality of life outcome of balloon pulmonary angioplasty in patients with chronic thromboembolic pulmonary hypertension: a retrospective study.

BACKGROUND: Balloon pulmonary angioplasty (BPA) has beneficial effects on pulmonary hemodynamics, exercise capacity, and quality of life (QOL) in patients with chronic thromboembolic pulmonary hypertension (CTEPH). Recently, emerging evidence suggests a relationship between CTEPH and psychiatric disorders (PD). However, data on the clinical efficacy of BPA in CTEPH patients with PD are lacking. METHODS: We retrospectively analyzed 75 patients with inoperable/residual CTEPH who underwent BPA and right-sided heart catheterization before the initial BPA and within 1 year after the last procedure. QOL was evaluated using the European Quality of Life Five Dimension (EQ-5D) scale in 27 patients before and after BPA sessions. Baseline and post-procedural hemodynamic, functional, and QOL parameters were compared between the patients with and without PD. RESULTS: Among the 75 participants, 22 (29.3%) patients were categorized in the PD group. Although PD group had a similar mean pulmonary artery pressure level compared with non-PD group (40 ± 7 vs. 41 ± 9 mmHg, p = 0.477), they tended to have unfavorable QOL status (0.63 ± 0.22 vs. 0.77 ± 0.19, p = 0.102). BPA significantly improved pulmonary hemodynamics, laboratory parameters and exercise tolerance in both groups. BPA also significantly improved EQ-5D scores in the non-PD group (from 0.77 ± 0.19 to 0.88 ± 0.13, p < 0.001), but the scores remained unchanged in the PD group (from 0.63 ± 0.22 to 0.67 ± 0.22, p = 0.770). During the long-term period [1,848 (1,055-2,565) days], both groups experienced similar mortality rates (PD 4.6% vs. non-PD 5.7%, p = 1.000). CONCLUSIONS: BPA improved hemodynamic and functional parameters irrespective of PD, but its effect on QOL was limited in patients with PD.

A new assessment method for right ventricular diastolic function using right heart catheterization by pressure-volume loop.

Diastolic stiffness coefficient (ß) and end-diastolic elastance (Eed) are ventricular-specific diastolic parameters. However, the diastolic function of right ventricle had not been investigated sufficiently due to the lack of established evaluation method. We evaluated the validity of these parameters calculated using only data of right heart catheterization (RHC) and assessed it in patients with restrictive cardiomyopathy (RCM) and cardiac amyloidosis. We retrospectively analyzed 46 patients with heart failure who underwent RHC within 10 days of cardiac magnetic resonance (CMR). Right ventricular end-diastolic volume and end-systolic volume were calculated using only RHC data, which were found to be finely correlated with those obtained from CMR. ß and Eed calculated by this method were also significantly correlated with those derived from conventional method using CMR. By this method, ß and Eed were significantly higher in RCM with amyloidosis group than dilated cardiomyopathy group. In addition, the ß and Eed calculated by our method were finely correlated with E/A ratio on echocardiography. We established an easy method to estimate ß and Eed of right ventricle from only RHC. The method finely demonstrated right ventricular diastolic dysfunction in patients with RCM and amyloidosis.

The ratio of TAPSE to PASP predicts prognosis in lung transplant candidates with pulmonary arterial hypertension.

Lung transplantation (LT) is the only option for patients with pulmonary arterial hypertension (PAH) refractory to maximal medical therapy. However, some patients referred for LT could survive without LT, and its determinants remain unclear. This study aimed to elucidate prognostic factors of severe PAH at the referral time. We retrospectively analyzed 34 patients referred for LT evaluation. The primary outcome was a composite of death or LT. Over a median follow-up period of 2.56 years, eight patients received LT and eight died. Compared with LT-free survival group, pulmonary arterial systolic pressure (PASP) was higher (p = 0.042), and the ratio of tricuspid annular plane systolic excursion (TAPSE) to PASP (TAPSE/PASP) was lower (p = 0.01) in LT or death group. In receiver operating characteristic analysis, the area under the curve was 0.759 (95% confidence interval 0.589-0.929) for TAPSE/PASP to predict primary outcome, and the optimal cut-off value was 0.30 mm/mmHg (sensitivity 0.875 and specificity 0.667). In a multivariate analysis, TAPSE/PASP was independently associated with death or LT. Kaplan-Meier analysis showed a better LT-free survival in patients with TAPSE/PASP ≧0.30 mm/mmHg than in those with < 0.30 mm/mmHg (p = 0.001). Low-level TAPSE/PASP could be a poor prognostic factor in PAH patients referred for LT evaluation.

Prognostic value of follow-up vasoreactivity test in pulmonary arterial hypertension.

BACKGROUND: Acute vasoreactivity test with inhaled nitric oxide (NO) is performed during diagnostic right heart catheterization (RHC) to identify patients with pulmonary arterial hypertension (PAH) who respond to calcium channel blockers. Our purpose was to investigate the prognostic importance of follow-up vasoreactivity test after treatment. METHODS: We retrospectively analyzed 36 PAH patients (mean age, 47 years; 61 % treatment-naïve), who underwent diagnostic and follow-up RHC and vasoreactivity tests at our center. The primary outcome was all-cause mortality. RESULTS: The median time between baseline and follow-up RHC was 9.7 months. Absolute change in mean pulmonary arterial pressure (ΔmPAP) during NO challenge was less pronounced after treatment, but there was great variability among patients. Overall cohort was dichotomized into two groups: preserved vasoreactivity (ΔmPAP ≤ -1 mmHg) and less vasoreactivity (ΔmPAP ≥0 mmHg) at follow-up RHC. Less vasoreactivity group had higher usage rate of endothelin receptor antagonists and parenteral prostacyclin analogues. During a median observation period of 6.3 years after follow-up RHC, 7 patients died, of which 6 showed less vasoreactivity at follow-up. Absolute ΔmPAP ≥0 at follow-up RHC was associated with all-cause mortality in univariable Cox regression analysis (hazard ratio, 8.728; 95 % confidence interval, 1.045-72.887; p = 0.045), whereas other hemodynamic parameters were not. Absolute ΔmPAP ≥0 at follow-up RHC was associated with all-cause mortality in multivariable Cox analysis adjusted for age and known PAH prognostic factors (HR, 12.814; 95 % CI, 1.088-150.891; p = 0.043). Kaplan-Meier survival analysis revealed a significantly worse survival of less vasoreactivity group compared to preserved vasoreactivity group (log-rank test, p = 0.016). CONCLUSIONS: Follow-up vasoreactivity test after treatment could contribute to the detection of high-risk subgroups who might need careful monitoring and referral for lung transplantation.

Prophylactic negative pressure wound therapy is not effective for preventing driveline infection following left ventricular assist device implantation.

BACKGROUND: Driveline infection (DLI) following left ventricular assist device (LVAD) implantation remains an unresolved problem. Negative pressure wound therapy (NPWT) promotes wound healing by applying negative pressure on the surface of the wound. Recently, the prophylactic application of NPWT to closed surgical incisions has decreased surgical site infections in various postsurgical settings. Therefore, we evaluated the efficacy and safety of prophylactic NPWT for preventing DLI in patients with LVAD implantation. METHODS: Prophylactic NPWT was provided to 50 patients who received continuous-flow LVADs as bridge-to-transplant therapy at our institution between May 2018 and October 2020 (NPWT group). The negative pressure dressing was applied immediately after surgery and retained on the driveline exit site for 7 days with a continuous application of -125 mm Hg negative pressure. The primary outcome was DLI within 1 year of LVAD implantation. We compared the rate of DLI incidence in the NPWT group with that in the historical control cohort (50 patients) treated with the standard dressing (SD) who received LVAD implantation between July 2015 and April 2018 (SD group). RESULTS: No severe complications were associated with the NPWT. During the follow-up period, DLI was diagnosed in 16 participants (32%) in the NPWT group and 21 participants (42%) in the SD group. The rates of DLI incidence and freedom from DLI did not differ between groups (p = 0.30 and p = 0.63). CONCLUSIONS: Prophylactic NPWT at the driveline exit site was safe following LVAD implantation. However, it did not significantly reduce the risk of DLI.

Age prediction from coronary angiography using a deep neural network: Age as a potential label to extract prognosis-related imaging features.

Coronary angiography (CAG) is still considered the reference standard for coronary artery assessment, especially in the treatment of acute coronary syndrome (ACS). Although aging causes changes in coronary arteries, the age-related imaging features on CAG and their prognostic relevance have not been fully characterized. We hypothesized that a deep neural network (DNN) model could be trained to estimate vascular age only using CAG and that this age prediction from CAG could show significant associations with clinical outcomes of ACS. A DNN was trained to estimate vascular age using ten separate frames from each of 5,923 CAG videos from 572 patients. It was then tested on 1,437 CAG videos from 144 patients. Subsequently, 298 ACS patients who underwent percutaneous coronary intervention (PCI) were analysed to assess whether predicted age by DNN was associated with clinical outcomes. Age predicted as a continuous variable showed mean absolute error of 4 years with R squared of 0.72 (r = 0.856). Among the ACS patients stratified by predicted age from CAG images before PCI, major adverse cardiovascular events (MACE) were more frequently observed in the older vascular age group than in the younger vascular age group (p = 0.017). Furthermore, after controlling for actual age, gender, peak creatine kinase, and history of heart failure, the older vascular age group independently suffered from more MACE (hazard ratio 2.14, 95% CI 1.07 to 4.29, p = 0.032). The vascular age estimated based on CAG imaging by DNN showed high predictive value. The age predicted from CAG images by DNN could have significant associations with clinical outcomes in patients with ACS.

Residual Pulmonary Vascular Resistance Increase Under Left Ventricular Assist Device Support Predicts Long-Term Cardiac Function After Heart Transplantation.

Aims: We compared hemodynamics and clinical events after heart transplantation (HTx) in patients stratified by the severity of residual pulmonary vascular resistance (PVR) after left ventricular assist device (LVAD) implantation for bridge to transplantation. Methods: We retrospectively analyzed patients who had undergone HTx at the University of Tokyo Hospital. We defined the high PVR group as patients with PVR of >3 Wood Units (WU) as measured by right heart catheterization performed 1 month after LVAD implantation. Results: We included 85 consecutive HTx recipients, 20 of whom were classified in the high PVR group and 65 in the low PVR group. The difference in PVR between the two groups became apparent at 2 years after HTx (the high PVR group: 1.77 ± 0.41 WU, the low PVR group: 1.24 ± 0.59 WU, p = 0.0009). The differences in mean pulmonary artery pressure (mPAP), mean right arterial pressure (mRAP), and mean pulmonary capillary wedge pressure (mPCWP) tended to increase from the first year after HTx, and were all significantly higher in the high PVR group at 3 years after HTx (mPAP: 22.7 ± 9.0 mm Hg vs. 15.4 ± 4.3 mm Hg, p = 0.0009, mRAP: 7.2 ± 3.6 mm Hg vs. 4.1 ± 2.1 mm Hg, p = 0.0042, and mPCWP: 13.4 ± 4.5 mm Hg, 8.8 ± 3.3 mm Hg, p = 0.0040). In addition, pulmonary artery pulsatility index was significantly lower in the high PVR group than in the low PVR group at 3 years after HTx (2.51 ± 1.00 vs. 5.21 ± 3.23, p = 0.0033). The composite event including hospitalization for heart failure, diuretic use, and elevated intracardiac pressure (mRAP ≥ 12 mm Hg or mPCWP ≥ 18 mm Hg) between the two groups was significantly more common in the high PVR group. Residual high PVR was still an important predictor (hazard ratio 6.5, 95% confidence interval 2.0-21.6, and p = 0.0023) after multivariate Cox regression analysis. Conclusion: Our study demonstrates that patients with residual high PVR under LVAD implantation showed the increase of right and left atrial pressure in the chronic phase after HTx.

Renal replacement therapy as a therapeutic option for right heart failure in severe pulmonary arterial hypertension.

Pulmonary arterial hypertension (PAH) is a progressively life-threatening disease that causes right heart failure (RHF). Renal dysfunction frequently complicates PAH with RHF and is associated with a worse prognosis. Renal replacement therapy (RRT) may be a therapeutic option, although its efficacy and safety are unclear. We describe a 30-year-old male with severe PAH who developed renal insufficiency and diuretic-refractory volume overload complicated with RHF but was successfully managed with intermittent RRT via a subcutaneously fixed superficial artery for 4 years. RRT led to haemodynamic stability, which enabled us to carefully de-titrate parenteral PAH drugs without worsening RHF. This case highlights that RRT may be a potential alternative for haemodynamic and volume control of refractory fluid retention complicated with RHF in severe PAH cases. Further studies are warranted to gain more insight into patient selection and the optimal timing of RRT in PAH patients with deteriorating RHF.

Clinical Outcome of Closure of a Small Atrial Septal Defect in a Patient with Pulmonary Arterial Hypertension.

The closure of small/coincidental atrial septal defects (ASDs) in patients with pulmonary arterial hypertension (PAH) has been described in recent major guidelines as useless or even contraindicated. We confirm the effectiveness of "Treat and Repair" for ASD closure through one patient diagnosed with idiopathic PAH with small ASD, under careful observation with right heart catheterization and cardiac magnetic resonance imaging. The clinical decision concerning the closure of ASD with PAH should be made not only by referring to the guidelines but also by evaluating the benefits and risks specific to that case.

Potential value of saline-induced Pd/Pa ratio in patients with coronary artery stenosis.

Background: Fractional flow reserve (FFR) is the current gold standard for identifying myocardial ischemia in individuals with coronary artery stenosis. However, FFR is not penetrated as much worldwide due to time consumption, costs associated with adenosine, FFR-related discomfort, and complications. Resting physiological indexes may be widely accepted alternatives to FFR, while the discrepancies with FFR were found in up to 20% of lesions. The saline-induced Pd/Pa ratio (SPR) is a new simplified option for evaluating coronary stenosis. However, the clinical implication of SPR remains unclear. Objectives: In the present study, we aimed to compare the accuracies of SPR and resting full-cycle ratio (RFR) and to investigate the incremental value of SPR in clinical practice. Methods: In this multicenter prospective study, 112 coronary lesions (105 patients) were evaluated by SPR, RFR, and FFR. Results: The overall median age was 71 years, and 84.8% were men. SPR was correlated more strongly with FFR than with RFR (r = 0.874 vs. 0.713, respectively; p < 0.001). Using FFR < 0.80 as the reference standard variable, the area under the receiver-operating characteristic (ROC) curve for SPR was superior to that of RFR (0.932 vs. 0.840, respectively; p = 0.009). Conclusion: Saline-induced Pd/Pa ratio predicted FFR more accurately than RFR. SPR could be an alternative method for evaluating coronary artery stenosis and further investigation including elucidation of the mechanism of SPR is needed (225 words).

Long-Term renal function after implantation of continuous-flow left ventricular assist devices: A single center study.

BACKGROUND: Implantable continuous-flow left ventricular assist device (LVAD) improve renal function in advanced heart failure. However, the long-term effects of LVAD on renal function have not been investigated thoroughly. We aimed to assess long-term renal function in patients with LVAD support and to identify predictors for late deterioration in renal function (LDRF). METHODS: One hundred patients underwent LVAD implantation as a bridge to transplant at the University of Tokyo Hospital between May 2011 and December 2018. We assessed renal function at intervals (preoperative, 1, 6, 12, 18, 24 and 30 months after LVAD implantation). We divided patients into two groups: "with LDRF," whose renal function at 30 months had decreased by >25% compared with preoperatively (n = 14), and "without LDRF" (n = 55). RESULTS: Renal function improved at 1 month, returned to preoperative levels at 6 months, and remained there up to 30 months after LVAD implantation. However, renal function impairment became evident in patients with LDRF 18 months after LVAD implantation. A ratio of right atrial pressure/pulmonary artery wedge pressure > 0.57 and left ventricular dimension diastole ≤ 67 mm were preoperative independent risk factors for LDRF. In addition, the incidence of perioperative acute kidney injury, ventricular arrhythmia, aortic insufficiency, and late right ventricular failure was significantly higher in patients with LDRF. CONCLUSION: LDRF after LVAD implantation corresponded to several risk factors, including a small left ventricle and LVAD-related complications, such as right ventricular failure.

Automatic detection of vessel structure by deep learning using intravascular ultrasound images of the coronary arteries.

Intravascular ultrasound (IVUS) is a diagnostic modality used during percutaneous coronary intervention. However, specialist skills are required to interpret IVUS images. To address this issue, we developed a new artificial intelligence (AI) program that categorizes vessel components, including calcification and stents, seen in IVUS images of complex lesions. When developing our AI using U-Net, IVUS images were taken from patients with angina pectoris and were manually segmented into the following categories: lumen area, medial plus plaque area, calcification, and stent. To evaluate our AI's performance, we calculated the classification accuracy of vessel components in IVUS images of vessels with clinically significantly narrowed lumina (< 4 mm2) and those with severe calcification. Additionally, we assessed the correlation between lumen areas in manually-labeled ground truth images and those in AI-predicted images, the mean intersection over union (IoU) of a test set, and the recall score for detecting stent struts in each IVUS image in which a stent was present in the test set. Among 3738 labeled images, 323 were randomly selected for use as a test set. The remaining 3415 images were used for training. The classification accuracies for vessels with significantly narrowed lumina and those with severe calcification were 0.97 and 0.98, respectively. Additionally, there was a significant correlation in the lumen area between the ground truth images and the predicted images (ρ = 0.97, R2 = 0.97, p < 0.001). However, the mean IoU of the test set was 0.66 and the recall score for detecting stent struts was 0.64. Our AI program accurately classified vessels requiring treatment and vessel components, except for stents in IVUS images of complex lesions. AI may be a powerful tool for assisting in the interpretation of IVUS imaging and could promote the popularization of IVUS-guided percutaneous coronary intervention in a clinical setting.