Search

Background: For heart transplantation, optimal donor-recipient matching is an important factor in the ongoing development of the United Network for Organ Sharing (UNOS) continuous distribution framework. Donor-recipient sex-mismatch has decreased since the 1990s, but this may be related to the risk posed by size mismatching, particularly when donor hearts are undersized. Thus, the impact of sex-mismatching, controlling for other factors including size mismatch, is uncertain. Methods: Adult first-time, isolated heart transplant patients from the UNOS database between October 1, 1987 and December 31, 2022 were analyzed. Cohorts were separated into male and female recipients. Propensity score matching on known preoperative risk factors was performed. Equivalence testing via Two One-Sided Testing (TOST) was performed to assess between-arm equivalence in postoperative outcomes. Survival differences were measured by the between-arm ratio of restricted mean survival time and binary outcome differences by the odds ratio. Results: In the propensity-matched cohort, we found significant equivalence between arms in both male (TOST p

The echocardiogram allows for a comprehensive assessment of the cardiac musculature and valves, but its rich temporally resolved data remain underutilized. Here, the authors develop a video AI system trained to predict post-operative right ventricular failure.

BackgroundPatients with heart failure (HF) are admitted either under observation (OBS) or inpatient stays; however, there is little data on whether this designation reflects the clinical status of a patient, with significant logistical and financial implications. We sought to compare the outcomes of patients with HF admitted OBS versus inpatient stay (≤2 days; INPT). Methods and ResultsFrom January 1, 2008 to September 30, 2015, our multisite health system saw 21 339 unique patients totaling 52 493 hospital admissions with a primary diagnosis of HF. Patients were excluded if they underwent cardiac surgery (n=611), heart transplantation (n=187), or left ventricular assist device insertion (n=198), or if they died during hospitalization (n=1839). Of the remaining 50 654 discharges, 2 groups were identified: INPT group and OBS group. Outcomes were HF readmission, all‐cause readmission, and all‐cause mortality within 1 year of discharge. Hazard ratios were computed using the Andersen‐Gill method in the Cox proportional‐hazards model. A total of 8709 admissions (17%) occurred in the INPT group and 2648 admissions (5%) occurred in the OBS group. HF readmission rate at 1 year was 55.3% in INPT versus 66.5% in OBS (hazard ratio, 0.75; 95% confidence interval, 0.71–0.80; P