2026-10-14T12:24:49Z https://www.scidb.cn/oai

10.57760/sciencedb.45617 2026-08-17T11:20:06Z

2026-08-17 Dual-Endpoint Competing Risk Analysis of Cardiovascular and Cerebrovascular Events in Patients Undergoing Maintenance Hemodialysis: An Analysis Based on Cause-specific Cox and Fine-Gray Models doi:10.57760/sciencedb.45617 en The median follow-up time was 36.0 months (interquartile range: 28.0-36.0 months). During follow-up, 124 CV events (23.3%), 149 CBV events (28.0%), and 139 competing events (26.1%) occurred. The 3-year CIFs were 29.84%, 36.36%, and 38.19%, respectively. Because CV and CBV events were not completely mutually exclusive, these CIFs were estimated separately for different target endpoints and should not be simply summed. In the cause-specific Cox model for CV events, prior cardiovascular disease (csHR=1.817, 95% CI: 1.265-2.610, P=0.001), atrial fibrillation (csHR=1.650, 95% CI: 1.071-2.544, P=0.023), and serum albumin (csHR=0.958, 95% CI: 0.919-0.998, P=0.039) were significantly associated with CV events; the Fine-Gray model showed consistent directions of effect. In the cause-specific Cox model for CBV events, prior cardiovascular disease (csHR=2.146, 95% CI: 1.539-2.991, P<0.001) and serum albumin (csHR=0.931, 95% CI: 0.897-0.967, P<0.001) were significantly associated with CBV events. Age reached statistical significance in the Fine-Gray model (sdHR=1.015, 95% CI: 1.001-1.029, P=0.032), but not in the cause-specific Cox model (csHR=1.011, 95% CI: 0.997-1.024, P=0.134). Inter-endpoint heterogeneity analysis showed that effect differences for all risk factors between CV and CBV events did not reach statistical significance (heterogeneity P values ranged from 0.060 to 0.828). hemodialysis; cardiovascular events; competing risks Jing Chen PUBLIC https://creativecommons.org/licenses/by-nc-nd/4.0/ dataset Science Data Bank