New risk model for prognostic prediction after surgical aortic valve replacement in hemodialysis patients

Due to the poor prognosis of dialysis patients, accurately predicting life expectancy after aortic stenosis surgery remains challenging, leading to potential misselection of treatment options. This study aimed to develop a prognostic model specific to dialysis patients to facilitate individualized treatment selection.

A total of 171 dialysis patients with aortic stenosis who underwent initial isolated surgical aortic valve replacement at seven cardiovascular centers in Japan between 2011 and 2021 were enrolled. The cohort was randomly divided into the training and validation cohorts in a 2:1 ratio. Risk factors contributing to mortality were identified from preoperative variables, and a prognostic model was developed using the Cox proportional hazards model.

Among the 171 patients, 88 deaths occurred during the total observation period of 488.9 person-years. The cumulative overall survival rates at 1, 3, and 5 years, estimated using the Kaplan-Meier method, were 74.7%, 59.4%, and 38.7%, respectively. An optimal risk model was developed, incorporating six factors: age, serum albumin, peripheral artery disease, sex, insulin-dependent diabetes mellitus, and atrial fibrillation. The model demonstrated strong predictive accuracy, with a 5-year C-statistic of 0.723 (95% confidence interval: 0.658-0.788) and 0.656 (95% confidence interval: 0.543-0.770) in the training and validation cohorts, respectively. Calibration plots confirmed that actual survival up to 5 years was well predicted (intraclass correlation coefficient = 0.918, 95% confidence interval: 0.703-0.981).

The proposed model is a reliable prognostic tool for dialysis patients who underwent surgical aortic valve replacement.