In our previous pilot trial comparing the effect of individualized transfusion strategy and liberal transfusion strategy (8), we found 36.5% in the individualized group and 89.4% in the liberal group of patients received red cells transfusion. Besides, we found that 76.7% in 400 patients undergoing major spine surgeries received red cells transfusion under guidance of current Chinese transfusion guidelines. Based on these findings, this trial is designed to test the hypothesis that individualized transfusion strategy by application of West China Liu's Score, as compared with a restrictive transfusion strategy guided by current transfusion guidelines and a liberal transfusion strategy of 10 grams per deciliter, the widely accepted “safe” threshold, red cells transfusion requirement can be reduced without increase of mortality, post-operative complications, or the cost of hospitalization.
This trial is a prospective, multicenter, randomized controlled trial that will test the superiority of individualized transfusion strategy (red cells transfusion guided by West China Liu's Score) to restrictive or liberal transfusion strategy in terms of reducing red cells transfusion demands safely and effectively. The protocol of the trial has been registered at http://www.clinicaltrials.gov (NCT01597232) and a brief flowchart of the whole study is summarized in the figure. This study is conducted according to the principles outlined in the Declaration of Helsinki. This trial is investigator-initiated with grant support from Sichuan University Education Foundation (SCUEF). All investigators in participating centers are qualified through training to conduct the trial. All patients have to sign the informed consent prior to study entry.
This study protocol has been first approved by the Biological-Medical Ethical Committee of West China Hospital, Sichuan University, Chengdu, Sichuan, China on 10 January 2012, and then approved by all the participating centers successively. Details of the study will be explained thoroughly to the potential subjects and(or) their legal guardian by the investigators. The signed informed consent forms must be obtained from all eligible patients before enrollment. Patients will be given at least 24 hours to consider before enrollment. Participation to the study is entirely voluntary and patients can withdraw from the study anytime. The privacy of all participants will be protected. Personal medical records will be checked only by designated investigators and inspectors, and they will not export any confidential information. Data anonymity will be applied in the whole process of data management, and all collected data will be kept and analyzed centrally at West China Hospital.
Patients with intra-operative estimated blood loss more than 1000 ml or 20% of total blood volume, or anticipated hemoglobin level less than 10 grams per deciliter will be screened for inclusion and the signed informed consent form will be collected. During surgery or in postoperative phase, and after all autologous or cell-saver blood has been transfused back to the patient, patients with hemoglobin level less than 10 grams per deciliter will be formally enrolled.
The detailed inclusion and exclusion criteria are summarized in table 2.
In the individualized-strategy group, decision for red cell transfusion trigger and target will be made in accordance with the West China Liu's score and hemoglobin level. That is, whenever the red cells transfusion is considered, Liu's score will be calculated and compared with the instant hemoglobin level. If the score is less than the instant hemoglobin level, red cells transfusion will not be necessary. If the score is more than the instant hemoglobin level, red cells should be transfused, and the volume of red cells is calculated as double of the difference of Liu's score minus the instant hemoglobin level.
In the restrictive-strategy group, patients will be managed according to Chinese or European guidelines (3-5) for red cells transfusion. That is, red cells transfusion is always not recommended in patients with hemoglobin level above 10 grams per deciliter, required when hemoglobin level decrease to 7 grams per deciliter, and depends on the attending physician's judgment with evaluation of patients' cardiopulmonary reserve and oxygen consumption. Besides, the red cells transfusion target is not stated in the guidelines and also depends on the physician's judgment.
In the liberal-strategy group, patients will be transfused if the hemoglobin level decreases to less than 10 grams per deciliter, in order to maintain hemoglobin not less than 10 grams per deciliter during perioperative period.
All patients recruited to the study will be collectively randomized using SAS programming (SAS Windows 9.1) by Department of Anesthesia, West China Hospital, Chengdu, Sichuan, China. The randomization will be stratified by the participating centers, and each center is allocated with 120 subjects with the ratio of 1: 1: 1. Study IDs are 5-digit numbers that are randomly generated and assigned into one of the three groups by SAS. Results of the group allocation will be concealed in a non-transparent envelop with only a 5-digit number on the surface. Envelops will be distributed to the centers. At each center, eligible patients enrolled in the study will be randomly assigned a study ID in the sequence according to the order of the patient recruited.
The investigators, responsible anesthesiologists, surgeons, and the research nurses will be aware of the treatment allocation. The patients, staff responsible for follow-up, and statisticians will be blinded to the treatment assignment. After discharge, all patients will be followed up by telephone calls for five years.
There are two primary outcomes, the incidences of red cells transfusion requirement, and combined in-hospital death for any reason and in-hospital serious complications.
In-hospital serious complications are listed as follows:
1) Cardiac events including cardiac arrest, myocardial infarction, definite diagnosis of angina pectoris, definite diagnosis of arrhythmia, heart block needing temporary or permanent pacemaker, definite diagnosis of heart failure, acute coronary syndrome, and cardiac infection needing antibiotics or other treatment;
2）Central nervous system events including stroke, definite diagnosis of carotid or intracranial thrombosis, definite diagnosis of intracranial hemorrhage, and intracranial infection needing antibiotics or other treatment;
3） Pulmonary events including definite diagnosis of atelectasis needing treatment, tension pneumothorax, hemothorax, or hemopneumothorax needing thoracic close drainage, pulmonary edema needing treatment, bronchopleural fistula needing surgical intervention, respiratory failure needing mechanical ventilation, pulmonary embolism with definite diagnosis, and pulmonary infection needing antibiotics or other treatment;
4） Digestive system events including anastomotic fistula, bile leakage, or biloma needing surgical intervention, definite diagnosis of hepatic dysfunction or failure, and diarrhea, gastrointestinal infection, or intra-abdominal infection needing antibiotics treatment;
5） Urinary/reproductive system events including urinary tract obstruction needing surgical intervention, acute renal insufficiency needing dialysis, and urinary/reproductive infection needing antibiotics treatment;
6） Other complications including re-operation for postoperative bleeding, definite diagnosis of disseminated intravascular coagulation, disruption of wound needing surgical intervention, infection of incisional wound needing antibiotics and other treatment, and definite diagnosis of systemic inflammatory response syndrome or sepsis.
The secondary outcomes are listed as follows:
1） Incidences of in-hospital infectious complications;
2） Intensive care unit (ICU) admission rate and ICU length of stay;
3） Length of hospital stay;
4） Hemoglobin level at different time points;
5） Red cells transfusion cost and hospital admission cost;
6） Healing status of surgical incision, which is divided into grade I, II, and Ⅲ. Grade Ⅰ was defined as the wound healing nicely without any adverse reaction, grade Ⅱ was defined as the inflammatory wound without the need of re-incision, and grade Ⅲ was defined as the wound suppurated with the need of re-incision for clearance;
7） Post-operative quality of life, which is evaluated by the 8-item Short-Form Health Survey (SF-8);
8） Death for any reason within one and five year postoperatively.
All patients' demographic information, diagnosises, interventions, and in hospital serious complications and mortality will be captured from electronic hospital medical records and recorded in the paper version of case report forms. The collected data will be uploaded to electronic case report forms at www.pottstrial.com within seven days after discharge. The follow-up information will be uploaded to this website within three days after each follow-up.
Sample Size and Statistical Analysis
Our primary hypothesis is that the individualized transfusion strategy will reduce the red cells transfusion demand (incidences of red cells transfusion requirement) while do not increase the adverse events (in hospital serious complications and death rate for any reason) compared to restrictive transfusion strategy. To control the overall Type I error rate under 0.05, we set the Type I error rate to be 0.025 for sample size calculation for each of the two primary hypotheses two-sided testing. We also set the power to be 90% for each outcome.
In our pilot study comparing transfusion guided by individualized and liberal transfusion strategy, we found that among 191 of patients, 36.5% in the individualized group and 89.4% in the liberal group (P<0.001) received red cells transfusion in perioperative period, and a median of zero units of red cells were transfused in the individualized group and 3.5 units in the liberal group (P<0.001) (8). Besides, we sampled 400 records of blood transfusion in spine surgeries with blood loss more than 800 ml from 2006 to 2009 in West China Hospital, Sichuan University, and found that 76.7% patients received red cells transfusion and the number of transfused Red cells units was 2.7 U on average. Therefore, it is reasonable to assume a 36.5% and 76.7% rate (difference of proportions 40.2%) for red cells transfusion requirements for the individualized strategy and the restrictive strategy group. We conservatively assume a 20% difference between the two groups (36.5% vs 56.5%).Using Z-test, we will need 153 patients in each group to detect a significant between group difference in the primary outcome of the proportion of patients who received red cells transfusion with a 90% power. Considering an estimated 20% dropout rate, 192 patients in each group are required for evaluate this primary outcome in this study. We plan to recruit 400 patients in each group for the study which will achieve a higher than 99% power. For the combined in-hospital death for any reason and in-hospital serious complications outcome, in a pilot study conducted at West China hospital, we observed rates of 6.6% and 7.9% in the individualized strategy and the restrictive strategy groups, respectively. Assuming a similar difference in this study, with 360 patients in each group to finish study, we will have a greater than 95% power to detect a significant difference between the two groups in the combined in-hospital death and serious complication rates. The power analysis is based on a two-sample Chi-square test of proportions. The analysis using Mantel-Haenszel test accommodating the center difference will be more powerful. In addition, according to the incidences of secondary outcomes and death rates for any reason within one year post-operatively, we consider a difference of 2% (e.g. 4% vs. 6%) within groups to be clinically important, with an estimated 20% dropout rate, and 4050 patients are required.
A total of 4050 patients will be enrolled at 35 centers in China, adding 150 patients for potential subjects loss due to various reasons. Two formal interim analysis will be performed when the number of enrolled patients reaches approximately 1080 for the primary outcomes of red cells transfusion requirement and combined in-hospital mortality and serious complications. Quality and validity check on all collected data, data merge across the 35 centers, and data analysis will be performed by professional statisticians at West China Hospital. All primary and secondary outcomes will be analyzed on an intent-to-treat basis using SAS 9.4 (Cary, NC). Data distribution will be checked using Kolmogorov--Smirnov goodness of fit test and homogeneity will be checked using C-variances test. Summary variables will be expressed as mean ± standard deviation (with normal distribution) or median and interquartile range (for skewed distribution). Quantitative data will be compared using an analysis of variance (ANOVA) among three groups, and t-test for comparison between two groups. For primary outcomes, transfusion rates and the in-hospital combined complications and death rates between the individualized strategy and the restrictive strategy groups, we will use Cochran-Mantel-Haenszel test for comparison, and data with skewed distribution will be analyzed using Mann-Whitney U test. A P value<0.025 is considered significant.
The principal investigator, a study coordinator, and the Office of Scientific Research at West China Hospital are jointly responsible for all aspects of the study protocol and amendments. Dr. Ren Liao, associate professor of Department of Anesthesiology, West China Hospital, will be responsible for site monitoring. Data collection and follow-up will be performed by three dedicated affiliated research nurses. Designated trial monitors will review all investigational data for accuracy and completeness to ensure protocol compliance periodically.
The study is currently in the process of recruiting participants in all the trial centers.