Background: Thrombocytopenia is common in patients with malignancy and frequently encountered prior to surgery. Perioperative guidelines generally recommend prophylactic platelet transfusion before major surgery when platelet counts are <50 × 10⁹/L. In practice, however, transfusions are often administered above this threshold based on perceived bleeding risk, surgeon preference, or procedural complexity. While intended to reduce bleeding risk, platelet transfusion carries potential hazards including thrombotic, infectious, and inflammatory complications. The balance of risk and benefit in surgical oncology patients remains poorly defined.

Methods: We performed a retrospective, multicenter cohort study of adults with active malignancy undergoing surgery under anesthesia from January 1, 2010, to January 1, 2025. Patients with prior splenectomy, bone marrow transplant, or disseminated intravascular coagulation within 1 month were excluded. Two preoperative platelet strata were analyzed: <50 × 10⁹/L (generally meets guideline threshold) and 50–100 × 10⁹/L (generally above threshold). Within each stratum, patients were grouped by receipt of platelet transfusion within 1 week preoperatively versus no transfusion within 1 month. Propensity score matching (1:1) balanced demographic, comorbidity, cancer type, surgical category, and anesthesia variables. The primary outcome was 30-day mortality; secondary outcomes included ICU admission, major postoperative complications, acute kidney failure, postoperative platelet transfusion, and hospital length of stay.

Results: After matching, 3,192 patients in each <50 × 10⁹/L group and 5,613 in each 50–100 × 10⁹/L group were analyzed. In the <50 × 10⁹/L stratum, preoperative platelet transfusion was associated with higher 30-day mortality compared with no transfusion (16.4% vs 11.7%, p<0.001). Transfused patients had greater ICU admission rates (12.0% vs 8.7%, p<0.001), more cerebral infarctions (3.9% vs 2.6%) and myocardial infarctions (3.3% vs 2.1%), and substantially more acute kidney failure (24.9% vs 16.8%, p<0.001). The likelihood of receiving additional platelet transfusions within 1–3 days postoperatively was markedly higher among transfused patients (27.0% vs 5.1%).

In the 50–100 × 10⁹/L stratum, mortality remained higher among transfused patients (3.3% vs 1.7%, p<0.001). These patients also had higher ICU utilization (5.2% vs 3.5%), more cerebral infarctions (3.8% vs 2.1%) and myocardial infarctions (3.4% vs 2.4%), and significantly more acute kidney failure (24.5% vs 11.5%, p<0.001). Postoperative platelet transfusion within 1–3 days occurred in 21.8% of preoperatively transfused patients versus 2.5% of those not transfused. Across both strata, hospital length of stay was longer in transfused patients, and adverse outcome patterns were consistent across cancer types and surgical categories.

Conclusions: In this large, propensity-matched surgical oncology cohort, preoperative platelet transfusion was associated with increased postoperative morbidity and mortality both below and above current guideline thresholds. The persistence of harm in the above-threshold group raises concern for liberal, non-guideline-based prophylactic transfusion, while the continued association with poor outcomes even in the <50 × 10⁹/L group warrants closer evaluation of transfusion timing, dosing, and patient selection. These findings are directly relevant to perioperative oncology care, where transfusion decisions are often made rapidly and with substantial practice variation. They support the need for more restrictive, individualized, and evidence-based transfusion strategies, and underscore the importance of prospective studies to determine which patients truly benefit from perioperative platelet transfusion.

This content is only available as a PDF.
2025
Sign in via your Institution