Effect of ulinastatin on post-operative blood loss and allogeneic transfusion in patients receiving cardiac surgery with cardiopulmonary bypass: a prospective randomized controlled study with 10-year follow-up

The study was a prospective, randomized, double-blinded and controlled trial. It was sponsored by National Center for Cardiovascular Diseases and was conducted at Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College. The study was approved by the Ethical Review Board of Fuwai Hospital. (Ethical approval No. 2008–366). And written informed consent was provided by all participants.

The inclusion criteria were patients between 18 and 79 years old undergoing elective heart surgery with cardiopulmonary bypass, including the coronary artery bypass graft, valvular repair or replacement, or repair of congenital heart deformities. The exclusion criteria included previous cardiac surgery, hematocrit level less than 33%, platelet count less than 100,000 × 10³/L, allergy to tranexamic acid, and being recruited in other studies.

The surgical procedures and peri-operative care followed the institutional routine. Aspirin and clopidogrel, if any, were discontinued at least 5 days before the operation. In patients using warfarin, it was required that prothrombin time (PT) was normal before the operation. Patients were randomly assigned into three groups for the use of ulinastatin (group U), tranexamic acid (group T) or placebo (group C). The randomization sequence was generated by computer in permuted blocks by a 1:1:1 ratio and was masked in sealed, sequentially numbered and opaque envelopes. Patient enrollment, randomization, and blinding were conducted and supervised by an independent committee. The participants, medical staff, and investigators were unaware of the treatment allocation until the end of the study.

The primary outcome of this study was the total volume of post-operative blood loss. The secondary outcomes included stroke, post-operative myocardial infarction, renal failure, respiratory failure, in-hospital adverse outcomes and long-term morbidities and mortalities. Stroke was stated as new focal neurologic deficit lasting more than 24 h confirmed by a cerebral computed tomography scan and an attending neurologic consultant. Post-operative myocardial infarction was diagnosed by two of the following: prolonged (> 20 min) chest pain not relieved by rest or nitrates, new pathologic Q waves in more than two contiguous electrocardiograph leads, elevated enzyme levels (creatine kinase-MB > 5% of total creatinine phosphokinase or troponin T > 0.5 ng/mL), new wall motion abnormalities, or the need for revascularization. Renal failure was stated as first-time dependency on renal dialysis, an increase of post-operative creatinine of at least 2 mg/dL, or a difference of at least 0.7 mg/dL between baseline value and the maximal post-operative plasma creatinine concentration. Respiratory failure was defined as prolonged mechanical ventilation (> 48 h), the need for continuous positive airway pressure therapy, reintubation, or tracheostomy. The in-hospital adverse outcomes were evaluated and defined as seizure, sudden cardiac arrest, readmission to intensive care unit (ICU), re-operation for surgical cause, using intra-aortic balloon pulsation (IABP) or extracorporeal membrane oxygenation (ECMO) and deep sternal infection. The long-term morbidities included stroke, myocardial infarction, renal failure, respiratory failure, seizure and sudden cardiac arrest.

Study and placebo medication were prepared by the hospital pharmacy. Identical syringes of 50 mL labeled with the randomization number contained transparent solution, 30 mg/Kg body weight of tranexamic acid (Jie Ning®; Changchun Tiancheng Pharmaceutical Co., Changchun, China), 1,000,000 U ulinastatin (Tian Pu Luo An®; Guangdong Tianpu Biochemistry Pharmaceutical Co., Guangzhou, China) or normal saline. There were two syringes prepared for each patient labeled ‘#1’ and ‘#2’. In the tranexamic acid group, both the syringes contained 15 mg/kg tranexamic acid to fulfill a total dosage of 30 mg/kg. In the ulinastatin group, the syringe #1 contained 1,000,000 U ulinastatin and the syringe #2 contained normal saline. In the control group, both the syringes contained normal saline. The syringe #1 and #2 were pumped intravenously after anaesthetic induction and after the administration of protamine respectively.

Postoperative blood loss was assessed via chest drain tubes every 8 h for the first 24 h after admittance to the ICU, and then was assessed every day beyond the first 24 h until the chest drain tubes were withdrawn. Post-operative blood loss was defined as the total volume of drainage from the end of the operation until the removal of the chest tubes. Chest tube drainage more than 300 ml within the first post-operative hour, more than 5 ml/kg per hour consecutively for 3 h, or any signs of pericardial tamponade justified surgical re-exploration to control bleeding.

The criteria for the transfusion of packed red blood cells (RBC) were:

In all other situations the decision to transfuse was left to the discretion of the treating physician. The criteria for transfusion of platelet concentrates and fresh frozen plasma were:

Additional protamine was administered in cases of prolonged ACT (the preoperatively measured ACT served as reference).

All patients were followed up for 10 years via reviewing outpatient records and questionnaires by mail/telephone 30 days post-operatively and annually.

The sample size was calculated based on the volume of post-operative bleeding using one-way ANOVA at an alpha level of 0.05 and effect size of 0.2 with 95% power. Assuming a dropout rate of 10%, the estimated total sample size was 426 patients (142 patients for each group). For continuous variables, normal distribution assumption was assessed. Equal variance assumption was assessed. The differences of these characteristics between groups were performed using independent two-sample t-tests and one-way ANOVA. Mean difference (MD) and its 95% confidence interval (CI) was calculated. Categorical variables were summarized using frequency and percentage and compared using Chi-square test or Fisher’s exact test. The estimated effect size and its precision were presented by the absolute risk difference (RD) and relative risk (RR) with their associated 95% CIs. The Mantel- Haenszel method was applied in the calculation of RR. Survival analysis was performed using the Kaplan-Meier method and log-rank test. All the analyses were performed using SPSS (Version 18.0, SPSS Inc.) software. All tests were two-sided, and a probability value less than 0.05 was considered to be statistically significant. The authors had full access to the data and take responsibility for its integrity.

From April 2008 to Dec 2008, a total of 481 patients were eligible for access in the present study. Of 481 patients, 55 patients were excluded. Twenty patients didn’t meet the inclusion criteria and 35 patients refused to participate. (Fig. 1) The remained 426 patients were randomized to receiving ulinastatin (Group U, n = 142), tranexamic acid (Group T, n = 143), or placebo (Group C, n = 141). The baseline characteristics of each group are shown in Table 1. (Table 1)There was no statistical significant difference in demographic characteristics, main diagnoses and preoperative comorbidity among the groups. Types of operations included on-pump coronary artery bypass graft (CABG), valvular procedures and congenital deformity repairs. There was no significant difference among groups in terms of the constitution of surgical procedures, CPB time and aortic cross-clamping time as seen in Table 2. (Table 2)No significant difference was found in mechanical ventilation time, chest tube removal time, ICU stay and hospital length of stay.

There were significant differences among three groups regarding to the blood loss within 24 h post-operatively, the total blood loss, major bleeding, reoperations, and the amount of RBC and plasma transfusion. (Table 3) In post hoc analyses (Table 3), postoperative blood loss within 24 h (404.87 ± 253.58 ml vs. 527.73 ± 300.4 ml, MD − 122.86 ml, 95% CI − 195.87 ml to − 49.86 ml, p < 0.001, for the first 8 h; 183.94 ± 151.83 ml vs. 205.57 ± 129.57 ml, MD − 21.63 ml, 95% CI − 55.49 ml to 12.22 ml, p = 0.016, for the second 8 h; 99.58 ± 94.75 ml vs. 121.03 ± 101.62 ml, MD − 21.45 ml, 95% CI − 44.58 ml to 1.68 ml, p = 0.029, for the third 8 h; 688.39 ± 393.55 ml vs 854.33 ± 434.03 ml, MD − 165.95 ml, 95% CI − 262.88 ml to − 69.01 ml, p < 0.001, for the wholly first 24 h) and total blood loss (801.7 ± 460.14 ml vs. 1016.67 ± 529.08, MD − 214.98 ml, 95% CI − 338.60 ml to − 91.36 ml, p < 0.001) were significantly reduced in patients receiving ulinastatin compared with placebo. The major bleeding was comparable between group U and group C. It was a trend that there were fewer reoperations in group U than in group C (0.70% vs. 4.26%; RD -0.0355, 95% CI − 0.0716 to 0.0005; RR, 0.166; 95% CI 0.0202 to 1.36; p = 0.055).

Patients in group U had less allogeneic erythrocyte transfusion compared to patients in group C (2.57 ± 3.15 units vs. 3.73 ± 4.21 units; MD -1.16; 95% CI − 2.06 to − 0.265; p = 0.016). Ulinastatin tended to reduce exposure to RBC in contrast to placebo (58.45% vs. 69.50%; RD -0.111, 95% CI − 0.222 to 0.0006; RR 0.841, 95% CI 0.705 to 1.00; p = 0.053). The volume and the exposure of plasma and platelet transfusion was similar between group U and group C. Between group U and group T, there was no significant difference in all indicators (blood loss, major bleeding, re-operation, allogenic transfusion and exposure to transfusion). (Table 3).

There was one in-hospital death in group U and in group C, respectively, and no in-hospital death in group T (0.70, 0.71%, 0, respectively; p = 0.602). There were significantly fewer patients who had respiratory failure post-operatively in group U (0.70, 4.20 and 6.38% in group U, group T and group C, respectively; p = 0.040). However, no significant difference was found among groups in other major morbidities, including stroke, postoperative myocardial infarction and renal failure. In all groups, duration of intensive care unit stay and hospital stay were similar. And no significant difference in adverse outcomes was found. (Table 4)

All patients were followed up for 10 years. There was no significant difference in long-term survival among three groups. (Fig. 2) In further analyses, no significant difference in survival was found between each two groups. There was no difference in major morbidities between groups.