Perioperative anemia and transfusion thresholds in vascular surgery procedures – a systematic review

Abstract

Abstract Background Anemia is a common condition, with an estimated 1.92 billion cases, or 24.3% of the worldwide population as of 2021, and has been shown to be a predictor of adverse surgical outcomes in a range of patient populations and procedures. Anemia is often treated with blood transfusions perioperatively, but potential benefits must be balanced against risks associated with transfusion including but not limited to cost, lack of availability, infectious complications and transfusion reactions. High quality studies in many populations have demonstrated noninferior outcomes in patients transfused with a restrictive threshold (hemoglobin (Hb) level < 7-8g/dL), but these results many not be applicable to vascular surgery patients. These patients tend to have a high rate of severe underlying cardiovascular disease and oftentimes present with critical limb ischemia which may be particularly sensitive to the tissue hypoxia potentially associated with lower hemoglobin thresholds. Objective We performed a systematic review and qualitative analysis to determine the relative risks and benefits of a restrictive (target Hb < 7-8g/dL) versus liberal (target Hb 9-10g/dL or higher) transfusion threshold in patients undergoing major vascular surgical procedures. Methods We searched Ovid Medline, CENTRAL, EMBASE, and the reference lists of eligible publications for relevant studies published from inception until July 25th, 2025. Studies were screened at abstract and full-text level independently and in duplicate. Any cohort study or randomized controlled trial (RCT) examining either restrictive or liberal transfusion strategies in patients undergoing major vascular procedures were included. Baseline characteristics, surgical procedure details, and transfusion strategies were extracted. Efficacy outcomes included major cardiovascular adverse events (MACEs), surgical complications, mortality, and major ischemic events. Secondary outcomes included documented transfusion reactions as defined by primary authors. Results A total of 4343 studies were identified in our initial search strategy, of which 50 had their full texts reviewed for eligibility. 12 eligible studies were included with a total of 29,707 patients. The most common study type was retrospective cohort (n = 7), followed by RCTs (n = 3) and prospective cohort studies (n = 2). The most common procedures included endovascular or open abdominal aortic aneurysm repair (n = 8), infrainguinal bypass (n = 6), and lower limb amputations (n = 5). Included studies had variable definitions for restrictive transfusion threshold ranging from 7-9 g/dL. Efficacy and safety outcomes were largely dependent on procedure type. Studies which included major, high-risk vascular procedures such as open AAA repair and lower limb amputation trended towards higher rates of mortality, procedural complications, and MACE in the restrictive transfusion groups, though these findings were of variable statistical significance. Studies examining lower-risk procedures including infrainguinal bypass, carotid endarterectomy, and endovascular aneurysm repair had variable results. Multiple studies, including one RCT and two cohort studies noted lower rates of MACE and 30-day mortality in patients transfused with a restrictive strategy, albeit with similarly variable levels of statistical significance. Two other cohort studies in this group noted no differences in patient outcomes with a restrictive strategy. Lastly, pRBCs were noted to be transfused at a higher rate in the liberal group across the included studies, though there was insufficient data on the rates of transfusion reactions between the two strategies. Meta-analysis of the data could not be performed due to significant heterogeneity in the design of the included studies. Conclusions Overall, studies examining higher-risk procedure such as open AAA repairs and lower limb amputations noted trend towards lower MACE, surgical complications, and mortality in patients receiving a liberal transfusion threshold (Hb > 9g/dL). In lower-risk vascular surgical procedures, MACE, surgical complications, and mortality appeared non inferior or superior with a restrictive transfusion threshold. Large randomized trials examining a variety of vascular surgery procedures with standardized hemoglobin thresholds are necessary to establish more conclusive benefits or harms of a restrictive or liberal transfusion strategy in patients undergoing vascular surgery.