Duke University Medical Center, Durham, NC, North Carolina, United States
Background/Case Studies: The Red Cell Pathogen Inactivation (ReCePI) trial was a Phase III, randomized non-inferiority study comparing amustaline/glutathione pathogen-reduced (PR) and conventional red blood cells (RBCs) for the support of acute anemia during and for 7 days after complex cardiac or thoracic-aorta surgery.
Study
Design/Methods: The primary endpoint was the incidence of acute kidney injury (AKI) defined as a change in serum creatinine (≥0.3 mg/dL) from baseline within 48 hours of surgery. Blood utilization (components and total Hb transfused) was assessed. Patient hemoglobin (Hb) levels were recorded at baseline, daily for 7 days and at 28 days after surgery. Surgical and post-operative blood loss were estimated from medical records.
Results/Findings: Subjects (581) were randomized in 18 US hospitals and 321 (159 Test and 162 Control) RBC-transfused recipients comprised the modified intent-to-treat (mITT) population. Test and Control subjects had similar histories, baseline characteristics, surgeries and 7-day total blood loss (median [IQR] Test 1500 [940-2475] mL, Control 1733 [1060-2880] mL, p=0.31) and had comparable Hb levels at baseline, post-surgery (median (IQR] 9.8 [8.9-10.9] g/dL Test, 9.6 [8.6-10.6] g/dL Control, p=0.16) and for 28 days. PR RBC units contained ~5% less total Hb (median [IQR] Test 58.0 (53.0-62.0) g Hb vs. Control 61.0 (57.0-66.0) g Hb, p< 0.001) and were stored longer prior to transfusion (median [IQR] Test 23.8 (16.9-29.4) days vs. Control 21.8 (15.0-28.3) days, p< 0.001). The Test arm received ~10% less total study plus non-study RBC Hb over 7-days (median Test 169.0g Hb vs. Control 188.0g Hb, p=0.01). Both groups received a median of 3 RBC components, However, Test subjects 32/159 (20.1%) required 5 or more RBC units within 7 days compared to Control 48/162 (29.6%, p=0.05), and more Control subjects required non-study RBC transfusions due to exceeding the available RBC supply [Test 35/159 (22.0%) subjects vs. Control 45/162 (27.8%) subjects]. Plasma utilization was significantly less in the Test group (median Test 2 units vs. Control 2 units, P=0.02) while platelet and cryoprecipitate use were not different. The incidence of AKI was 29.3% (46/157) for Test and 28.0% (45/161) for Control subjects in the mITT group. Non-inferiority for the incidence of AKI was achieved in both the mITT (p< 0.001) and the per protocol analyses (P=0.03). Adverse events, serious adverse events and deaths on study were not different.
Conclusions: PR RBCs and conventional RBCs demonstrated equivalent support for patients undergoing cardiac or thoracic-aorta surgery while using ~10% less transfused total RBC Hb and fewer plasma units. The incidence of AKI was non-inferior to that with conventional RBCs.
