Phase II Comparison Study of Intraoperative Autotransfusion for Major Oncologic Procedures

Background. Intraoperative autotransfusion (10AT) has been avoided in oncologic surgery because of possible tumor cell dissemination. Through a prior Phase I study, we demonstrated that malignant cells are not present in blood filtered for IOAT. We hypothesized that autotransfusion could be safely used for patients undergoing major oncologic procedures and reduce the need for allogeneic blood. Materials and Methods. A Phase II, IRB-approved, prospective evaluation was conducted of patients undergoing gastrointestinal oncologic procedures. All procedures were conducted with blood salvaged for IOAT, and the collected volume was autotransfused if it was >100 ml. Quality of life (QoL) was assessed by questionnaire at regular intervals. Results. A total of 92 patients were enrolled with median age of 56 years. The most commonly performed procedures were hepatectomy (47%) and pancreaticoduodenectomy (26%). The median preoperative hemoglobin (Hgb) was 13.1 (range, 9-16), and the median estimated blood loss was 350 ml (range, 20-4000 ml). Of the 92 total patients, 32 (35%) received IOAT with a median volume of 255 ml (range, 117-1499 ml). Multivariate analysis identified that patients with preoperative Hgb >11 g/dl (P = .02), and blood loss of 400-900 ml (P = .03) benefited from IOAT with a reduction in postoperative blood transfusion rate. Patients with discharge Hgb >10 g/dl showed higher mean QoL scores throughout their recovery. At a median followup of 18 months, the rates of recurrence in the IOAT and the non-IOAT groups were equivalent (38 vs. 39%, P = .9). Conclusions. Intraoperative autotransfusion can be used safely and effectively for major oncologic procedures. Furthermore, degree of discharge anemia is associated with lower quality of life in patients undergoing oncologic gastrointestinal surgery.