LAUSR

one procedure recorded, the most recent procedure was considered. Patients were separated by inpatient and outpatient setting, with a KaplanMeier curve created for survival probabilities in each group. Results were compared by log-rank and Wilcoxon tests, with significance at P < .05. Demographics between the survival and mortality groups were compared, and those with P < .1 were placed into a Cox proportional hazards model, with significance at P < .05. Results: A total of 15,915 patients with known 2-yearmortality status were included in the study (2674 inpatients and 13,241 outpatients). Inpatients had higher rates of comorbidities. Inpatients also had higher rates of 30day and 2-year postoperative mortality compared with outpatients (4.8% vs 1.2% [P < .001] and 29.3% vs 18.7% [P < .001]). The Fig demonstrates the Kaplan-Meier curves for inpatient and outpatient groups, with lower 2-year survival probability in inpatients (P < .001). The Table demonstrates results of the Cox proportional hazards model, with inpatient status at a hazard ratio of 1.83 (95% confidence interval, 1.69-1.99; P < .001). Conclusions: Patients who receive HD access surgery as inpatients have higher rates of 30-day and 2-year mortality. Inpatient status was an independent risk factor for increased mortality. Additional study into improved methods of patient selection before HD access surgery in the inpatient setting could minimize unnecessary and high-risk surgery.