Yosuke Ishii

Objective To evaluate whether implementation of chest physiotherapy [CPT] focused on postural drainage was associated with a reduced incidence of postoperative atelectasis following pediatric cardiac surgery. Methods This single-center retrospective cohort study included 102 consecutive patients aged ≤2 yr who underwent cardiac surgery via median sternotomy between September 2014 and August 2023. Patients were divided into a non-intervention group (n = 46; before January 2019) and an intervention group (n = 56; after implementation of an intensive care unit [ICU]–based CPT program delivered by dedicated physical therapists). The primary outcome was postoperative atelectasis. Multivariable logistic regression was performed to identify independent predictors of postoperative atelectasis. Results Postoperative atelectasis occurred less frequently in the intervention group than in the non-intervention group (17.9% vs. 37.0%, p = 0.030). Severe (Grade 3) atelectasis was observed in one patient in the intervention group and six patients in the non-intervention group. Multivariable analysis identified CPT intervention (odds ratio [OR], 0.295; 95% confidence interval [95% CI], 0.116–0.782; p = 0.011) and longer duration of intubation (OR, 1.008; 95% CI, 1.001–1.015; p = 0.017) as independent predictors of postoperative atelectasis. No patient in the intervention group required reintubation, whereas three patients (6.5%) in the non-intervention group underwent reintubation. Conclusions Implementation of an early postoperative CPT program centered on postural drainage was associated with a reduced incidence of postoperative atelectasis following pediatric cardiac surgery. These findings support further prospective studies to confirm the effectiveness of structured respiratory rehabilitation in reducing postoperative pulmonary complications.

Background: After coronary artery bypass grafting (CABG), graft and native-vessel anatomy complicate perfusion SPECT, and preoperative thresholds transfer poorly. We evaluated whether perfusion-anatomy concordance on hybrid SPECT/CT was associated with major adverse cardiac events (MACE) and compared it with perfusion severity and 123I-BMIPP mismatch.Methods: This retrospective study included 131 consecutive post-CABG patients who underwent stress/rest perfusion SPECT, 123I-BMIPP SPECT, and coronary CT angiography. Hybrid status was matched when a reversible defect corresponded to significant stenosis or occlusion in the supplying native artery or graft. Firth-penalized Cox regression was prespecified (5.4 events per parameter).Results: Over a median follow-up of 99.7 months (interquartile range 47.9–115.7), MACE occurred in 27 patients (20.6%). Event proportions were 38.8% versus 9.8% by matched status (log-rank P < 0.001), 30.4% versus 13.3% for a summed difference score ≥ 4 (P = 0.009), and 22.8% versus 18.9% for a summed mismatch score ≥ 7 (P = 0.269). After clinical adjustment, matched status remained associated with MACE (Firth-penalized hazard ratio 3.9970; 95% confidence interval 1.7666–9.7395; P < 0.001). This model showed the lowest unpenalized Akaike information criterion (226.32) and the highest ridge-stabilized optimism-corrected concordance index (0.704), with no proportional-hazards violation detected (P = 0.731). Perfusion severity was associated with MACE (unpenalized hazard ratio 2.9219; 1.2039–7.0916; P = 0.018) but showed evidence of nonproportional hazards (P = 0.036).Conclusions: Perfusion-anatomy concordance identifies a high-risk post-CABG subgroup. It provided the best model fit and, unlike perfusion severity, no evidence of nonproportional hazards. BMIPP mismatch was not associated with MACE.

Background and Clinical Significance: Hemoptysis caused by acute type B aortic dissection rupturing into the lung is rare but life-threatening; Case Presentation: An 81-year-old man receiving maintenance hemodialysis developed hemoptysis due to an aortobronchial fistula involving the left upper lobe. The ulcer-like projection was located on the greater curvature of the distal aortic arch and enlarged during the clinical course. Immediate thoracic endovascular aortic repair (TEVAR) and open repair were initially considered unsuitable because of hostile access through a previous abdominal aortic graft, a shaggy aorta, and severe cardiac comorbidities. Left upper lobe bronchial ligation was performed as a temporizing airway-isolation procedure and initially controlled airway bleeding. However, recurrent hemoptysis occurred after further enlargement of the aortic lesion and development of a new fistula involving the left lower lobe. Emergency TEVAR was then performed through the previous abdominal aortic prosthetic graft, and no further hemoptysis occurred; Conclusions: Bronchial ligation may temporarily control life-threatening airway bleeding when immediate aortic repair is not feasible, but strict surveillance and readiness for subsequent aortic intervention remain essential.

Background: To describe an aortic arch incision and closure technique (AICT) for proximal fixation of a frozen elephant trunk (FET) and to report early outcomes. Methods: We retrospectively reviewed 15 consecutive patients who underwent distal arch repair with an FET using AICT (mean age 77 ± 7 years; 14 men). Indications were distal arch aneurysm (n = 12), acute Stanford type B dissection (n = 2), and distal arch enlargement after thoracic endovascular aortic repair (n = 1). Under circulatory arrest, an oblique arch aortotomy was created, the FET was deployed antegrade, trimmed, and sutured to the native aortic wall during simultaneous closure, allowing extended posterior fixation. Clinical outcomes and postoperative computed tomography were assessed. Results: No ischemic complications related to graft kinking or thrombosis, reoperation for bleeding, stroke, spinal cord ischemia, or organ failure occurred. One patient died of pneumonia on postoperative day 47 (6.7%). Cervical branch reconstruction was required in 12 patients (80%), whereas two patients with type III arch morphology and acute angulation were treated without debranching via a Zone 3 aortotomy. At a median follow-up of 29 months, no proximal endoleak was observed; one distal endoleak occurred without reintervention. Coronary bypass grafts remained patent in all patients with concomitant or prior CABG. Conclusions: AICT provided secure proximal FET fixation and arch closure while preserving the ascending aorta, offering an alternative to total arch replacement in selected distal arch pathologies.

Background: Falls among hospitalized patients are a critical issue that often leads to prolonged hospital stays and increased health care costs. Traditional fall risk assessments typically rely on standardized scoring systems; however, these may fail to capture the complex and multifactorial nature of fall risk factors. Objective: This retrospective observational multicenter study aimed to develop and validate a machine learning–based model to predict in-hospital falls and to evaluate its performance in terms of discrimination and calibration. Methods: We analyzed the data of 83,917 inpatients aged 65 years and older with a hospital stay of at least 3 days. Using Diagnosis Procedure Combination data and laboratory results, we extracted demographic, clinical, functional, and pharmacological variables. Following the selection of 30 key features, 4 predictive models were constructed: logistic regression, extreme gradient boosting, light gradient boosting machine (LGBM), and categorical boosting (CatBoost). The synthetic minority oversampling technique and isotonic regression calibration were applied to improve the prediction quality and address class imbalance. Results: Falls occurred in 2173 (2.6%) patients. CatBoost achieved the highest F1-score (0.189, 95% CI 0.162-0.215) and area under the precision-recall curve (0.112, 95% CI 0.091-0.136), whereas LGBM had the best calibration slope (0.964, 95% CI 0.858-1.070) with good discrimination (F1-score 0.182, 95% CI 0.156-0.209; area under the precision-recall curve 0.094, 95% CI 0.078-0.113). Logistic regression had the lowest discrimination (F1-score 0.120, 95% CI 0.100-0.143). Shapley Additive Explanations analysis consistently identified low albumin, impaired transfer ability, and the use of sedative-hypnotics or diabetes medications as major contributors to fall risk. In incident report analysis (n=435), 49.2% of falls were toileting-related, peaking between 4 and 6 AM, with bedside falls predominating in high or very high risk groups. Conclusions: CatBoost and LGBM offer clinically valuable prediction performance, with CatBoost favored for high-risk patient identification and LGBM for probability-based intervention thresholds. Integrating such models into electronic health records could enable real-time risk scoring and trigger targeted interventions (eg, toileting assistance and mobility support). Future work should incorporate dynamic, time-varying patient data to improve real-time risk prediction.

BACKGROUND Falls among hospitalized patients are a critical issue that often leads to prolonged hospital stays and increased health care costs. Traditional fall risk assessments typically rely on standardized scoring systems; however, these may fail to capture the complex and multifactorial nature of fall risk factors. OBJECTIVE This retrospective observational multicenter study aimed to develop and validate a machine learning–based model to predict in-hospital falls and to evaluate its performance in terms of discrimination and calibration. METHODS We analyzed the data of 83,917 inpatients aged 65 years and older with a hospital stay of at least 3 days. Using Diagnosis Procedure Combination data and laboratory results, we extracted demographic, clinical, functional, and pharmacological variables. Following the selection of 30 key features, 4 predictive models were constructed: logistic regression, extreme gradient boosting, light gradient boosting machine (LGBM), and categorical boosting (CatBoost). The synthetic minority oversampling technique and isotonic regression calibration were applied to improve the prediction quality and address class imbalance. RESULTS Falls occurred in 2173 (2.6%) patients. CatBoost achieved the highest F1-score (0.189, 95% CI 0.162-0.215) and area under the precision-recall curve (0.112, 95% CI 0.091-0.136), whereas LGBM had the best calibration slope (0.964, 95% CI 0.858-1.070) with good discrimination (F1-score 0.182, 95% CI 0.156-0.209; area under the precision-recall curve 0.094, 95% CI 0.078-0.113). Logistic regression had the lowest discrimination (F1-score 0.120, 95% CI 0.100-0.143). Shapley Additive Explanations analysis consistently identified low albumin, impaired transfer ability, and the use of sedative-hypnotics or diabetes medications as major contributors to fall risk. In incident report analysis (n=435), 49.2% of falls were toileting-related, peaking between 4 and 6 AM, with bedside falls predominating in high or very high risk groups. CONCLUSIONS CatBoost and LGBM offer clinically valuable prediction performance, with CatBoost favored for high-risk patient identification and LGBM for probability-based intervention thresholds. Integrating such models into electronic health records could enable real-time risk scoring and trigger targeted interventions (eg, toileting assistance and mobility support). Future work should incorporate dynamic, time-varying patient data to improve real-time risk prediction.

Surgical repair of thoracoabdominal aortic aneurysms (TAAA) is a complex procedure that is associated with marked morbidity and mortality. Endovascular TAAA repair, initially reserved for patients at high risk, has yielded promising results and is performed on an increasing range of patients. This case report describes a hybrid approach—a combination of endovascular and open repair—for a patient with high-risk TAAA after aortic dissection. A 79-year-old man had undergone aortic replacement for chronic aortic dissection 20 years previously. His saccular TAAA, extending from the Th12 level to the renal artery origin, demonstrated progressive enlargement (56–60 mm) on computed tomography (CT) scans and compressed the celiac artery. In light of the patient' s age and thoracotomy history, a hybrid repair strategy involving total debranching thoracic endovascular aortic repair (TEVAR) with four-branch abdominal reconstruction was selected to minimize surgical risk. The surgery was performed via a midline abdominal incision. Bypass surgery was first performed using a quadrant artificial graft to restore left renal, right renal, superior mesenteric, and common hepatic artery blood flow. The celiac artery was coiled, and TEVAR was completed. Intraoperative contrast and postoperative CT showed no endoleak and a patent debranching graft. The patient recovered without complications and was discharged 10 days postoperatively. The findings for the present hybrid procedure indicate that debranching TEVAR is less invasive and yields better outcomes than open aortic replacement, particularly for patients at high risk.

Background: Cardiac neoplasms may cause life-threatening symptoms associated with cerebral infarction, ventricular arrhythmias, and heart failure. Emergency surgery or preoperative treatment may be required for these patients. However, no study has reported the surgical outcomes in cases involving cardiac neoplasms with life-threatening complications. The current study investigated the mid- to long-term outcomes of surgery in patients with cardiac neoplasms in life-threatening conditions. Methods: This study retrospectively analyzed 36 consecutive patients who underwent resection for cardiac neoplasms with life-threatening cardiovascular, respiratory, and cerebral nervous system complications from January 2000 to December 2022. Their mean age at surgery was 54.9 years. In terms of fatal events, one patient who experienced a ventricular tachycardia storm caused by a left ventricular neoplasm was placed under deep sedation and managed with a ventilator preoperatively. Seven patients who presented with limb motor paralysis and visual defects had cerebral infarction. Two of the seven patients with cerebral infarction received cerebrovascular treatment before cardiac surgery. Results: During the follow-up period, cerebral- and cardiovascular-related deaths were not recorded. All postoperative cerebral and cardiovascular complications were new-onset cerebral infarction (n = 2) (with symptoms that improved during the long term). The mean follow-up period was 6.2 years. The 5- and 10-year survival rates of all patients were 89.8% and 78.7%, respectively. There were no significant differences in postoperative prognosis between patients with preoperative cerebral infarctions and those without. Conclusions: The long-term surgical outcome of patients with life-threatening symptomatic cardiac neoplasm was good. Thus, preoperative complications did not affect prognosis.