Impact of Preoperative Anemia on Hospitalization, Death, and Overall Survival in Patients With Peripheral Artery Disease Undergoing Endovascular Therapy: A Retrospective Cohort Study in the United States and Canada

Purpose: Preoperative anemia is associated with adverse outcomes after cardiac and noncardiac surgeries, but outcomes after an endovascular peripheral vascular intervention (PVI) are not well established. We aimed to assess the association of preoperative anemia with 30 day death, hospital length of stay (LOS), and overall (long term) survival in patients undergoing an endovascular PVI for peripheral artery disease. Materials and Methods: In this retrospective, cohort study in the United States and Canada, we queried the national Vascular Quality Initiative database for all endovascular PVIs performed between 2010 and 2019, and outcomes were correlated with patients' hemoglobin (Hb) levels. Anemia was classified as mild (Hb=10-13 g/dL for men and 10-12 g/dL for women), moderate (Hb=8-9.9 g/dL), and severe (Hb<8 g/dL). Results: A total of 79 707 adult patients who met study criteria underwent endovascular PVI. The mean age was 68 years, and 59% of patients were male. Anemia was documented in 38 543 patients (48%) and was mild in 27 435 (71%), moderate in 9783 (25%), and severe in 1325 (4%). The median follow-up duration was 4 years (range, 1.25-5.78 years). On univariate analysis, 30 day mortality, total LOS, and overall survival were significantly associated with the level of preoperative anemia. These associations persisted in the multivariate models. Kaplan-Meier survival analysis demonstrated an association of death with degree of anemia (p<0.001). Conclusion: The presence and degree of preoperative anemia were independently associated with increased 30 day mortality and LOS and decreased overall survival for patients with peripheral artery disease who had undergone endovascular PVI. Clinical Impact: The findings from this study have many implications for how to approach vascular surgery in patients with variable hemoglobin levels. Our findings will strengthen our ability to conduct accurate preoperative risk stratification for patients undergoing peripheral vascular interventions. This may also mitigate healthcare expenditures if findings are applied in a way that can lower patient length of postoperative stay while also maintaining quality of care and patient safety. Our results will also serve as guidance for clinical trials, and future prospective trials should evaluate the effect of preoperative optimization of hemoglobin as a potentially modifiable risk factor for outcomes.