This was a prospective, patient-randomized, noninferiority trial. The protocol was approved by the Institutional Review Board of Seoul St. Mary’s Hospital Ethics Committee (approval no. KC18MESI0583) on December 19, 2018. The study was performed in accordance with the principles of the Declaration of Helsinki. The study protocol was prospectively registered at a publicly accessible clinical registration site that is recognized by the International Committee of Medical Journal Editors (Clinical Research Information Service, Republic of Korea, approval number: KCT0003680) on March 27, 2019. Written informed consent was obtained from all patients at our hospital who were enrolled between March 2019 and May 2020. The study adhered to the Consolidated Standards of Reporting Trials (CONSORT) guidelines (Additional file 1) and a CONSORT flow chart is provided in Figure 1. Additional file 2 presents a summary of our study protocol.
Inclusion and exclusion criteria
The inclusion criteria for this study were as follows: age ≥ 20 years; scheduled for elective bimaxillary orthognathic surgery; and American Society of Anesthesiologists (ASA) physical status I or II . Exclusion criteria were as follows: hemodynamic instability requiring rescue therapy, such as strong vasopressor infusion (i.e., epinephrine or norepinephrine); ASA physical status III–V; history of iron drug-related side effects, such as allergy; medication history with anticoagulants, such as aspirin, clopidogrel, or warfarin; anemia (i.e., hemoglobin level < 11 g/dL) [19, 20]; and refusal to participate in the study.
The patients were randomly divided into two groups: the IV iron group and the AWBT group. Randomization was performed using sealed, opaque envelopes. An independent colleague randomly grouped the envelopes in blocks of 10 with a 1:1 ratio to produce an equal distribution across the whole study period. The envelopes were stacked and stored. When an enrolled patient arrived in the treatment area, the upper envelope was opened by the attending physicians who performed the treatments, i.e., IV iron infusion or collection of whole blood. In the operating room, the attending anesthesiologist and nurses were aware of the group allocations, but they were not involved in future patient care or data collection.
In the iron group, IV 500 mg ferric carboxymaltose (Ferinject; Vifor Pharma, Glattbrugg, Switzerland) mixed with 100 mL normal saline was administered for 30 min 4 weeks before surgery in accordance with the manufacturer’s instructions.
In the AWBT group, two units of autologous blood were collected, 1 unit at a time with an interval of at least 2 weeks between collections; thus, each unit was collected on 4 and 2 weeks before surgery. The blood volume collected at one time was 320 mL, and therefore, the total amount of whole blood was 640 mL and was stored at 4°C until the day of surgery. All collected whole blood was transfused intraoperatively using a warming infusion device.
Surgery and anesthesia
Surgical technique and anesthetic care were as described previously . Briefly, the care of patients was standardized between both groups apart from the treatments applied. Bimaxillary orthognathic surgery, including a bilateral sagittal split osteotomy and a Le Fort I osteotomy, was performed by an experienced surgeon, and balanced general anesthesia was provided without pre-medication by experienced attending anesthesiologists. To reduce surgical bleeding, induced hypotensive anesthesia with a systolic blood pressure (SBP) < 100 mmHg was achieved by intermittent IV boluses of sodium nitroprusside or nicardipine, especially between the beginning of osteotomy and the end of osteosynthesis. Hemodynamic monitoring, including SBP, diastolic blood pressure (DBP), heart rate (HR), electrocardiogram, pulse oximetry, and end-tidal carbon dioxide pressure, were regularly recorded every 5 min during surgery. Blood lost by surgical hemorrhage was replaced to prevent hypovolemia with one bag (500 mL) of colloid product (6% hydroxyethyl starch, volulyte) in the IV iron group and by 2 units (640 mL) of autologous whole blood in the AWBT group.
Based on the practice guideline , allogeneic packed red blood cell (PRBC) transfusion was performed at the discretion of the attending anesthesiologists or physicians, when hemoglobin level reached between 6 and 10 g/dL with consideration of potential or actual ongoing hemorrhage, intravascular volume status, organ ischemic signs, or adequacy of cardiopulmonary reserve. These blood products were administered unit-by-unit.
Preoperative findings included demographic variables, ASA physical status, vital signs (i.e., SBP, DBP, and HR). Intraoperative findings included total operation duration, average vital signs during surgery, total amount of fluid input, including crystalloid and colloid, hemorrhage, and urine output. Laboratory variables (i.e., hemoglobin, hematocrit, red blood cell [RBC] count, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, iron, ferritin, C-reactive protein, white blood cell count, neutrophil, lymphocyte, and platelet counts, international normalized ratio [INR], activated partial thrombin time, antithrombin III, and fibrinogen) were regularly measured four times: 4 weeks before surgery (baseline), immediately before surgery, and on PODs 1 and 2.
The primary outcome was the hemoglobin level on POD 1. The secondary outcomes included the levels of inflammatory and coagulation variables on PODs 1 and 2, and changes in blood cell-related variables during POD 2. The incidences of anemia defined as hemoglobin level < 13 g/dL in males and <12 g/dL in females, low iron level defined as <50 μg/L, and low ferritin level defined as <15 ng/mL in males and <10 ng/mL in females were analyzed in the two groups [9, 19]. The degree of discomfort when iron was intravenously infused or whole blood was collected was analyzed using a visual analog scale (VAS; 0 = no discomfort and 10 = the worst discomfort). Surgical complications, such as allogeneic blood transfusion, were analyzed using Clavien–Dindo classification , and total hospital stay was measured in both groups.
The sample size was calculated based on the primary endpoint according to the noninferiority hypothesis. The predetermined noninferiority limit (δ) was set to a difference in hemoglobin level of –1 g/dL between the two groups (IV iron group – AWBT group) that was considered clinically acceptable by experts at our institution. Based on preliminary data, a standard deviation (SD) of 1.3 g/dL was assumed for the hemoglobin distribution. With α = 0.05 and power of 90%, 30 patients were required in each group. Assuming a 10% dropout rate, we decided to enroll 33 patients per group. For the primary outcome, the noninferiority of the IV iron therapy was considered if the lower boundary of the two-sided 95% confidence interval (CI) lay above the noninferiority margin of –1 g/dL .
Values are expressed as the mean ± SD, median with interquartile range (IQR), or as numbers with percentages. The normality of the distribution of the continuous data was evaluated using the Shapiro–Wilk test. The perioperative findings were compared between the two groups using the unpaired t test or the Mann–Whitney U test, and Pearson’s χ2 test or Fisher’s exact test, as appropriate. Serial changes in RBC-related variables were analyzed using the paired t test or Wilcoxon’s signed rank test. All tests were two-sided, and p < 0.05 was taken to indicate statistical significance. Statistical analyses were performed using SPSS for Windows (ver. 24.0; IBM Corporation, Armonk, NY) and MedCalc for Windows software (ver. 11.0; MedCalc Software, Ostend, Belgium).