Abstract
Background
The aim of this study was to investigate the effects of ulinastatin, a protease inhibitor, and blood transfusion on perioperative surgical complications, changes of systemic inflammatory response syndrome (SIRS) scores, and levels of interleukin-6 (IL-6), interleukin-8 (IL-8) and tumour necrosis factor-α (TNF-α) in patients undergoing liver resection.
Materials and methods
Patients aged 18–65 years were enrolled and divided into four groups (12 patients in each group): a control group, a group given ulinastatin (UTI group), a group given blood transfusion (BT group), and a group given both blood transfusion and ulinastatin (BT+UTI group). Patients were randomised to receive ulinastatin or not, whereas blood transfusion was administered based on a transfusion trigger. Ulinastatin was given at a dose of 100,000 units/10 kg, infused 15 min before allogeneic blood transfusion or after completion of the liver resection. The patients were followed up for 3 days to record surgical complications, SIRS scores and levels of IL-6, IL-8 and TNF-α.
Results
Forty-four patients were included in the data analysis. The SIRS rate (SIRS scores ≥2) was significantly higher in the BT groups than in the control group at 6 hours and on day 3 after surgery and was significantly lower in the BT+UTI group than in the BT group on day 3 after surgery. Allogeneic blood transfusion significantly increased and ulinastatin significantly decreased postoperative levels of IL-6, IL-8, and TNF-α. The length of stay in hospital was significantly longer in the BT groups than in the control group but was not significantly different between the BT+UTI and BT groups.
Conclusion
A single dose of ulinastatin before allogeneic blood transfusion may lower the rate of postoperative SIRS and levels of IL-6, IL-8 and TNF-α associated with allogeneic blood transfusion and improve patients’ postoperative recovery.
Keywords: blood transfusion, pro-inflammatory cytokines, systemic inflammatory response syndrome, protease inhibitor, hepatic resection
Introduction
Surgical procedures may cause substantial bleeding and, therefore, necessitate allogeneic blood transfusion. However, allogeneic blood transfusion has been reported to cause immune suppression and be associated with an increase of postoperative complications, including infections and multiple organ failure, and even decrease both short- and long-term survival after cardiac surgery1–4. Although the issue is still controversial5, various studies showed that the increase of complications was associated with the volume and age of transfused blood1,2,6. During the storage of red blood cells, neutrophil enzymes, pro-inflammatory cytokines and other immunomodulatory substances accumulate in the stored units and contribute to the increase of complications and/or mortality associated with blood transfusion3,7–9. Therefore, a viable therapeutic option to reduce the adverse effects associated with blood transfusion could be pharmacological prevention of the activation of pro-inflammatory cytokines. Ulinastatin (UTI), one of the Kunitz-type human protease inhibitors found in urine, has the physiological effects of inhibiting neutrophil elastase and activation of various pro-inflammatory cytokines10–12. It could, therefore, be expected to reduce the blood transfusion-associated increase of postoperative complications.
Previous studies demonstrated that allogeneic blood transfusion induced an increase of polymorphonuclear leucocyte elastase (PMNE) and interleukin 6 (IL-6), while a single, intraoperative dose of ulinastatin inhibited the increase of PMNE but not that of IL-6 in patients undergoing gastrectomy13,14. These studies were, however, focused on a single pro-inflammatory cytokine and did not provide sufficient information on the association between the cytokine and postoperative outcomes after blood transfusion. It has been reported that not only IL-6 but also interleukin-8 (IL-8) and tumour necrosis factor-α (TNF-α) are important early biomarkers for predicting the development of multiple organ failure and mortality after cardiac surgery9,15,16.
In this prospective, clinical controlled study, we investigated the perioperative changes of IL-6, IL-8 and TNF-α, systemic inflammatory response syndrome (SIRS) scores and postoperative surgical complications in adult patients undergoing liver resection and the effects of allogeneic blood transfusion and ulinastatin on the changes of these parameters.
Materials and methods
Patients
The study was performed between June 1, 2010 and July 31, 2011 in the First Affiliated Hospital of Sun Yat-sen University in Guangzhou, China. The study protocol was approved by the institutional ethics committee of the hospital (2010 No.29). After informed consent to potential participation in the study had been obtained from each subject, adult patients (18–65 years old; American Society of Anesthesiologist [ASA] grade 1–2) undergoing liver resection were initially recruited. Patients planned to undergo liver resection were chosen as the study population because they were likely to bleed during surgery. The patients were divided into four groups: the first group received neither blood transfusion nor ulinastatin (control group), the second group received only ulinastatin (UTI group), the third group received only blood transfusion (BT group), and the fourth group received both blood transfusion and ulinastatin (BT+UTI group). Forty-eight patients were enrolled in this study until the four study groups had 12 patients each (Table I and Figure 1). The patients were not randomly assigned to receive or not blood transfusion, because this treatment was determined by the change in haemoglobin during surgery to protect the patients. However, all the enrolled patients were randomised, using the sealed envelope method, to receive ulinastatin or placebo (normal saline). The study subjects were divided into four groups because we tried to clarify whether blood transfusion results in increases of plasma cytokines (e.g. IL-6, IL-8 and TNF-α) and deterioration of postoperative clinical outcomes in patients undergoing liver resection and subsequently to investigate whether ulinastatin could improve any adverse effects caused by blood transfusion.
Table I.
Demographics and clinical outcome of the patients.
| Group | |||||
|---|---|---|---|---|---|
|
|
|||||
| Control | UTI | BT | BT+UTI | P value | |
| Age (years, mean±SD) | 49±15 | 55±10 | 54±11 | 49±11 | 0.639 |
| Gender (male/female) | 7/5 | 6/5 | 6/4 | 7/4 | 0.978 |
| Body weight (kg, mean±SD) | 60±10 | 62±5 | 58±12 | 63±13 | 0.851 |
| ASA grade (I/II) | 4/8 | 5/6 | 3/7 | 3/8 | 0.820 |
| Duration of surgery (min) | 214.2±118.3 | 226.8±41.1 | 247.8±78.1 | 248.4±94.2 | 0.813 |
| Duration of anaesthesia (min) | 268.5±120.0 | 265.7±40.3 | 300.0±80.2 | 283.6±99.8 | 0.876 |
| Blood loss, intraoperative (mL) | 422±381‡§ | 493± 511‡§ | 1,913±493*† | 1,636±369*† | 0.000 |
| Transfused crystalloid fluid volume (mL) | 2,290±692 | 2,457±605 | 2,250±342 | 2,490±1,052 | 0.877 |
| Transfused colloid fluid volume (mL) | 925±577 | 1,142±377 | 1,250±463 | 1,136±636 | 0.631 |
| Transfused blood volume, intraoperative (mL) | 0 ‡§ | 0 ‡§ | 525±237*† | 418± 60*† | 0.000 |
| Transfused fresh frozen plasma (mL) | 0 ‡§ | 0 ‡§ | 175±198*† | 136±201*† | 0.000 |
| Storage period of the transfused blood (mean, days) | NA | NA | 20.8±5.2 | 22.7±4.4 | 0.933 |
| Transfused blood volume, postoperative (mL) | 0 | 0 | 0 | 0 | – |
| Length of stay in hospital | 10.4±2.1‡§ | 11.6± 2.4‡ | 20.1±6.7*† | 15.6±3.7* | 0.000 |
| Surgery complications (including liver/kidney/cardiac/lung/nervous/vascular/acute serious complications) | 0 | 0 | 0 | 0 | – |
| Postoperative infection | 0 | 0 | 0 | 0 | – |
ASA: American Society of Anesthesiologists; control group: no blood transfusion or ulinastatin administration; UTI group: ulinastatin only; BT group: blood transfusion only; BT+UTI group: both blood transfusion and ulinastatin administration;
P <0.05 compared to control group;
P <0.05 compared to the UTI group;
P <0.05 compared to the BT group;
P <0.05 compared to the BT+UTI group.
Figure 1.
Flow chart of screened, included, excluded and eliminated patients. BT: blood transfusion; UTI: ulinastatin; NS: normal saline.
Patients with any of the following criteria before surgery were excluded: haemoglobin less than 110 g/L; a previous history of transfusion; fever or any symptom of infection; coagulation disorder; cardiac, respiratory, renal or nervous system disorder and complications; significant increases of liver enzymes and bilirubin; and use of aprotinin and/or gabexate mesylate (these drugs may affect inflammatory responses in patients) during the patients’ stay in the hospital. Patients were withdrawn (“eliminated”) from the study if they refused to be followed up after surgery or had an intraoperative blood loss of more than 3,000 mL. According to these criteria, four patients were eliminated from the final statistical analysis. The flow chart of screened, included, excluded and eliminated patients is shown in Figure 1.
Intervention
Atropine 0.01 mg/kg (maximum 0.5 mg) and luminal sodium 2 mg/kg (maximum 100 mg) were administered as premedication intramuscularly 30 min before the patient entered the operating theatre. After monitors had been connected to the patient, an epidural catheter was inserted into an intervertebral space at T9/T10 or T10/T11 for maintenance anaesthesia and postoperative analgesia. Anaesthesia was induced with propofol 2 mg/kg followed by fentanyl 3–5 μg/kg. Endotracheal intubation was facilitated by cis-atracurium at a dose of 0.2 mg/kg. An arterial line and right internal jugular vein line were set up routinely. Anaesthesia was maintained with sevoflurane 1–2%, propofol 3–5 mg/kg/h, remifentanil 0.2 μg/kg/min, and an epidural block using 0.5% ropivacaine. The antibiotic cefuroxime was given as prophylaxis against infections and repeated after 24 hours in the ward. Omeprazole 40 mg, for gastric protection, was administered routinely after induction of anaesthesia. Oesophageal temperature was monitored and maintained between 35.5 °C and 37.0 °C during surgery. Packed red blood cells (China Red Cross Society) were transfused if the haemoglobin decreased below 70 g/L and maintained between 70 g/L and 90 g/L during surgery. Ulinastatin (Techpool Biopharma Co. Ltd, Guangzhou, Guangdong, China) 100,000 units/10 kg was infused 15 min before the allogeneic blood transfusion or at the time the liver resection was completed if the patient did not reach the transfusion trigger. Because previous studies13,14,17 indicated that ulinastatin 300,000 units did not inhibit the increase of plasma IL-6 associated with blood transfusion in patients undergoing abdominal surgery, we increased the dose up to 100,000 units/10 kg expecting clearer and stronger effects in the present study, in accordance with our clinical experience and that of others18. A dose of up to one shot of 1,000,000 units was considered to be safe in a clinical trial18.
Blood sampling
Blood samples were obtained before surgery, and immediately, 6 hours, 1 day and 3 days after surgery. Blood sampling was terminated if the patient refused to give samples at any of the above time points (one patient in the UTI group refused to give blood samples after surgery). The sample (4 mL) was immediately placed into a test-tube containing EDTA-K2 anticoagulant and centrifuged at 3,000 g for 20 min at a temperature of 4 °C. The supernatants (about 2 mL) were carefully collected, aliquoted into Eppendorf cups and stored at a temperature of −70 °C until measurements were performed.
Cytokine measurements
Plasma levels of IL-6, IL-8, and TNF-α were determined from the isolated supernatant using enzyme-linked immunosorbent assays (ELISA) according to the manufacturer’s instructions (R&D systems, Shanghai, China). All assays were standardised by titrating the appropriate purified recombinant mediator of known concentration. The concentrations of cytokine biomarkers in the samples were determined by interpolation from the corresponding standard curve. All samples were assayed in duplicate. The sensitivity of the ELISA for IL-6 was less than 0.7 pg/mL, that for IL-8 was less than 3.5 pg/mL, and that for TNF-α was less than 1.6 pg/mL.
Systemic inflammatory response syndrome
A SIRS score (0–4) was calculated before surgery and at 6 hours, day 1 and day 3 after surgery and was set as the primary clinical outcome in the present study. Each patient received one point for each clinical manifestation present (hyperthermia or hypothermia, tachycardia, tachypnoea, and leucocytosis or leucopenia) according to previously consensus panel guidelines published by the American College of Chest Physicians/Society of Critical Care Medicine (ACCP/SCCM)19,20. The above-listed clinical manifestations were defined as: (i) a body temperature higher than 38 °C or lower than 36 °C; (ii) a heart rate faster than 90 beats/min; (iii) a respiratory rate faster than 20 breaths/min or hyperventilation, as indicated by a PaCO2 of less than 4.3 kPa; and (iv) a white blood cell count of more than 12,000 cells/mm3 or less than 4,000 cell/mm3, or the presence of greater than 10% immature neutrophils (“bands”). A patient was defined as having SIRS when the score was ≥2.
The follow-up period was established as 3 days because in the pilot trial we found that SIRS was manifested within 3 days after surgery, but significantly alleviated and almost disappeared on day 4 after surgery. Immediate postoperative data were not considered because the anaesthetics may still have been having effects on the patients.
Data collection
The preoperative and intraoperative clinical data were collected by an anaesthesiologist blinded to the ulinastatin treatment. The postoperative clinical outcomes were independently determined by an investigator blinded to the patients’ grouping. All these data were recorded in a well-design form, which included the following parameters: the patient’s demographic data (age, gender, body weight, preoperative ASA grade), duration of surgery, duration of anaesthesia, intraoperative blood loss, intraoperative urine volume, intraoperative input (crystalloid and colloid fluid volume), intraoperative and postoperative transfusion of red blood cells and other blood products, intraoperative vital signs, length of stay (LOS) in hospital, postoperative infection (central venous catheter infection, wound infection, urinary tract infection, intraperitoneal infections, pressure sores or skin ulcers, anastomotic fistula, bacteraemia, sepsis, etc.), postoperative complications (including respiratory, cardiac, vascular, hepatic, renal and nervous system complications), and in-hospital survival.
Statistical analysis
All analyses were performed using SPSS for Windows Version 16.0.1 (SPSS Inc., Chicago, IL, USA). Categorical variables are expressed as either direct data or constituent ratios (percentages). Continuous variables are expressed as means and standard deviations (SD). Categorical variables were compared using the chi-square likelihood ratio test. Continuous variables were compared using one-way factorial analysis of variance (ANOVA) followed by Bonferroni’s post hoc test. Repeated measures with analysis of variance followed by Bonferroni’s post hoc test was employed to determine within-group differences of concentrations of plasma cytokines, such as IL-6, IL-8 and TNF-α, over time. The effects of blood transfusion, ulinastatin and their interaction on plasma cytokines were analysed by ANOVA two-way factorial analysis. The effect of ulinastatin administration on SIRS scores after adjusting for the blood transfusion factor was analysed by the chi-square likelihood ratio analysis. Differences are considered statistically significant when the P value is <0.05.
Results
Background
As shown in Table I, there were no significant differences in age, gender, body weight, preoperative ASA grade, duration of surgery, duration of anaesthesia, transfused crystalloid and colloid fluid volume and postoperative blood transfusion among the four groups (P >0.05). The intraoperative blood loss was significantly greater in the BT and BT+UTI groups than in the control and UTI groups (P =0.000). Thus, the patients in the BT and BT+UTI groups were transfused with blood. However, the volume and storage period of the transfused blood were not significantly different between these two groups (P >0.05). Simultaneously, some patients in these groups were transfused with fresh-frozen plasma. There were no significant differences in the volume of fresh-frozen plasma between these two groups (P >0.05). None of the patients in any of the four groups receive postoperative blood transfusions and none received any platelets or other blood products during the perioperative period.
Clinical outcomes
As shown in Table I, patients in the BT and BT+UTI groups stayed in hospital significantly longer than patients in the control group (P <0.05). The time spent in hospital by patients in the BT group was significantly longer than that of the patient in the UTI group (P <0.05). The length of stay in hospital did not differ between the control group and the UTI group (P >0.05) or between the BT group and the BT+UTI group (P >0.05). No postoperative infections or surgical complications occurred in any of the four groups of patients.
Systemic inflammatory response syndrome scores
No patients in any group had SIRS (SIRS score ≥2) prior to surgery (Table II). However, SIRS scores were ≥2 in some patients in all four groups at various times after surgery, indicating that the surgical procedure resulted in cases of SIRS (Table II). The percentage of SIRS (SIRS scores ≥2) in patients in the BT and BT+UTI groups was significantly higher than that in the control and UTI groups at 6 hours after surgery (P <0.05) (Table II), and the percentage of SIRS in patients in the BT group was significantly higher than that in the control group on day 3 after surgery (P =0.005) (Table II), indicating that blood transfusion increased the postoperative SIRS rate. The percentage of patients who developed SIRS on day 3 after surgery was significantly lower in the BT+UTI group than in the BT group (P =0.024) (Table II). There was no significant difference in the percentage of patients who developed SIRS between the BT+UTI group and the control group (P >0.05) (Table II). After adjusting for the blood transfusion factor, ulinastatin administration was also found to have a significant inhibitory effect on SIRS rate on day 3 after surgery (P =0.024) (Table III): a single intraoperative dose of ulinastatin inhibited the increase of SIRS rate caused by blood transfusion.
Table II.
Systemic inflammatory response syndrome (SIRS) scores during surgery.
| Observed time points | Group | SIRS scores (n (%)) | ||||||
|---|---|---|---|---|---|---|---|---|
|
|
||||||||
| No (SIRS) | Yes (SIRS) | |||||||
|
|
|
|||||||
| 0 | 1 | Total | 2 | 3 | 4 | Total | ||
| Preoperative | Control | 11 | 1 | 12 (100%) | 0 | 0 | 0 | 0 |
| UTI | 10 | 1 | 11 (100%) | 0 | 0 | 0 | 0 | |
| BT | 10 | 0 | 10 (100%) | 0 | 0 | 0 | 0 | |
| BT+UTI | 8 | 3 | 11 (100%) | 0 | 0 | 0 | 0 | |
|
| ||||||||
| 6 h after surgery | Control | 3 | 4 | 7 (58.3%) | 3 | 2 | 0 | 5 (41.7%)ठ|
| UTI | 1 | 5 | 6 (54.5%) | 5 | 0 | 0 | 5 (45.5%)‡ | |
| BT | 0 | 0 | 0*† | 2 | 6 | 2 | 10 (100%)*† | |
| BT+UTI | 1 | 1 | 2 (18.2%) | 7 | 2 | 0 | 9 (81.8%)* | |
|
| ||||||||
| Day 1 after surgery | Control | 1 | 3 | 4 (33.3%) | 4 | 4 | 0 | 8 (66.7%) |
| UTI | 1 | 5 | 6 (54.5%) | 4 | 1 | 0 | 5 (45.5%)‡ | |
| BT | 0 | 1 | 1 (10%) | 1 | 5 | 3 | 9 (90%)† | |
| BT+UTI | 0 | 2 | 2 (18.2%) | 8 | 1 | 0 | 9 (81.8%) | |
|
| ||||||||
| Day 3 after surgery | Control | 4 | 4 | 8 (66.7%) | 3 | 1 | 0 | 4 (33.3%)‡ |
| UTI | 3 | 4 | 7 (63.6%) | 2 | 2 | 0 | 4 (36.4%)‡ | |
| BT | 0 | 1 | 1 (10%) | 2 | 6 | 1 | 9 (90%)*†§ | |
| BT+UTI | 1 | 5 | 6 (54.5%) | 4 | 1 | 0 | 5 (45.5%)‡ | |
Control group: no blood transfusion, no ulinastatin administration; UTI group: ulinastatin administration only; BT group: blood transfusion only; BT+UTI group, both blood transfusion and ulinastatin administration;
P <0.05 compared to Control group;
P <0.05 compared to UTI group;
P <0.05 compared to BT group;
P<0.05 compared to BT+UTI group.
Table III.
Chi-square analysis of effects of ulinastatin administration on systemic inflammatory response syndrome (SIRS) scores after adjusting for blood transfusion.
| Observed time points | Ulinastatin | Chi-square value | P value | ||
|---|---|---|---|---|---|
|
|
|||||
| Yes | No | ||||
| Preoperative | |||||
| Blood transfusion | yes | 0 (0%) | 0 (0%) | - | - |
| no | 0 (0%) | 0 (0%) | - | - | |
| 6 h after surgery | |||||
| Blood transfusion | yes | 9/11 (81.8%) | 10/10 (100.0%) | 2.778 | 0.096* |
| no | 5/11 (45.5%) | 5/12 (41.7%) | 0.034 | 0.855* | |
| Day 1 after surgery | |||||
| Blood transfusion | yes | 9/11 (81.8%) | 9/10 (90.0%) | 0.292 | 0.589* |
| no | 5/11 (45.5%) | 8/12 (66.7%) | 1.058 | 0.304* | |
| Day 3 after surgery | |||||
| Blood transfusion | yes | 5/11 (45.5%) | 9/10 (90.0%) | 5.074 | 0.024* |
| no | 4/11 (36.4%) | 4/12 (33.3%) | 0.023 | 0.879* | |
Likelihood ratio; P: statistical significance.
Plasma cytokines
The plasma IL-6 levels in the control group increased post-operatively, reaching a peak on day 1 after surgery (Figure 2). The levels 6 hours and 1 day after surgery were significantly higher than those prior to surgery (P <0.05), while there were no significant differences between IL-6 levels before surgery, immediately after surgery and day 3 after surgery (P >0.05), indicating that the surgical procedure resulted in an increase of IL-6 level for about 1 day (Figure 2). However, if blood was transfused, the IL-6 levels (i.e. those in the BT group) increased further and came to a peak on day 3 after surgery (Figure 2). The IL-6 levels in the BT group 6 hours, 1 day and 3 days after surgery were significantly higher than those before surgery (P <0.05) (Figure 2) and significantly higher than those in the control group on day 3 after surgery (P <0.05) (Figure 2). The two-way factorial analysis indicated that blood transfusion induced significant changes of IL-6 immediately, at 6 hours and 3 days after surgery when adjusting for ulinastatin administration and the interaction of blood transfusion and ulinastatin administration (Table IV). These findings indicate that blood transfusion further increased the postoperative IL-6 level and maintained it raised for at least 3 days. However, when ulinastatin was administered (BT + UTI group), the peak levels of IL-6 occurred on day 1 after surgery rather than on day 3 (Figure 2). The IL-6 levels in the BT+UTI group were significantly lower than those in the BT group on day 3 after surgery (P <0.05) (Figure 2). The two-way factorial analysis indicated that ulinastatin administration resulted in significant changes of IL-6 on day 3 after surgery when adjusting for blood transfusion and the interaction of blood transfusion and ulinastatin administration (Table IV). These data indicate that ulinastatin inhibited the increase of postoperative IL-6 level caused by blood transfusion. Table IV also shows that there was an interaction between blood transfusion and ulinastatin administration on day 3 after surgery. These data together with those in Figure 2 imply that, on day 3 after surgery, ulinastatin had significantly more inhibitory effects than blood transfusion.
Figure 2.
Effects of blood transfusion and liver resection on perioperative levels of interleukin 6 (IL-6). Data are expressed as mean ± S.E.M. T0: before surgery, T1: immediately after surgery, T2: 6 hours after surgery, T3: day 1 after surgery and T4: day 3 after surgery. Control group, no blood transfusion or ulinastatin administration; UTI group, ulinastatin only; BT group, blood transfusion only; BT+UTI group, both blood transfusion and ulinastatin administration.
Table IV.
ANOVA two-way factorial analysis for plasma cytokines.
| Blood transfusion | Ulinastatin | Blood transfusion × Ulinastatin | |||||
|---|---|---|---|---|---|---|---|
|
|
|||||||
| Cytokine | Time Points | F value | P value | F value | P value | F value | P value |
| IL-6 | Preoperative | 0.002 | 0.964 | 0.535 | 0.470 | 0.046 | 0.832 |
| Immediately postoperative | 5.955 | 0.021 | 0.226 | 0.638 | 1.938 | 0.174 | |
| 6 h postoperative | 4.935 | 0.034 | 1.410 | 0.245 | 0.033 | 0.857 | |
| 1 day postoperative | 0.268 | 0.608 | 2.401 | 0.132 | 0.004 | 0.951 | |
| 3 days postoperative | 6.949 | 0.013 | 7.118 | 0.012 | 5.384 | 0.028 | |
| IL-8 | Preoperative | 1.053 | 0.313 | 0.324 | 0.574 | 0.447 | 0.509 |
| Immediately postoperative | 0.563 | 0.459 | 0.250 | 0.621 | 0.528 | 0.473 | |
| 6 h postoperative | 2.564 | 0.120 | 0.331 | 0.570 | 1.148 | 0.293 | |
| 1 day postoperative | 4.817 | 0.036 | 3.575 | 0.069 | 4.220 | 0.049 | |
| 3 days postoperative | 17.154 | 0.000 | 5.156 | 0.031 | 7.439 | 0.011 | |
| TNF-α | Preoperative | 0.044 | 0.835 | 1.441 | 0.239 | 0.085 | 0.772 |
| Immediately postoperative | 0.058 | 0.811 | 0.369 | 0.548 | 0.666 | 0.421 | |
| 6 h postoperative | 2.916 | 0.098 | 2.386 | 0.133 | 0.017 | 0.898 | |
| 1 day postoperative | 2.525 | 0.123 | 1.104 | 0.302 | 0.004 | 0.953 | |
| 3 days postoperative | 5.124 | 0.031 | 5.657 | 0.024 | 0.006 | 0.940 | |
F: F-test statistics; P: significance.
The plasma levels of IL-8 levels in the control, UTI and BT+UTI groups did not change over time after surgery (P >0.05), indicating that the surgical procedure did not result in any postoperative increase of IL-8 level (Figure 3). However, in the BT group the IL-8 levels at 6 hours and on days 1 and 3 after surgery were significantly higher than those before surgery (P <0.05) (Figure 3). The levels of IL-8 in the BT group were significantly higher than those in the control group on days 1 and 3 after surgery (P <0.05) (Figure 3). The two-way factorial analysis showed that blood transfusion induced significant changes of IL-8 on days 1 and 3 after surgery when adjusting for ulinastatin administration and the interaction of blood transfusion and ulinastatin administration (Table IV). These findings indicate that blood transfusion resulted in an increase of postoperative IL-8 levels. However, the levels of IL-8 in the BT+UTI group were significantly lower than those in the BT group on day 3 after surgery (P <0.05) (Figure 3). The two-way factorial analysis showed that ulinastatin administration resulted in significant changes of IL-8 on day 3 after surgery when adjusting for blood transfusion and the interaction of blood transfusion and ulinastatin administration (Table IV). These findings indicate that ulinastatin inhibited the increase of postoperative IL-8 level caused by blood transfusion. Table IV also shows that there was an interaction between blood transfusion and ulinastatin administration with regards to IL-8 levels on days 1 and 3 after surgery. These data, together with those in Figure 2, imply that ulinastatin had more significant inhibitory effects on IL-8 on days 1 and 3 after surgery when given to patients who had undergone blood transfusion.
Figure 3.
Effects of blood transfusion and liver resection on perioperative levels of interleukin 8 (IL-8). Data are expressed as mean ± S.E.M. T0: before surgery, T1: immediately after surgery, T2: 6 hours after surgery, T3: day 1 after surgery and T4: day 3 after surgery. Control group, no blood transfusion or ulinastatin administration; UTI group, ulinastatin only; BT group, blood transfusion only; BT+UTI group, both blood transfusion and ulinastatin administration.
The plasma TNF-α levels in the control and UTI groups did not change according to time after surgery (P >0.05), indicating that the surgical procedure did not result in increases of TNF-α level (Figure 4). However, in the BT group the TNF-α levels on day 3 after surgery were significantly higher than those before surgery (P <0.05) (Figure 4). In the BT+UTI group the TNF-α levels on day 1 after surgery were significantly higher than those before surgery (P <0.05). The two-way factorial analysis revealed that blood transfusion induced significant changes of TNF-α on day 3 after surgery when adjusting for the factor of ulinastatin administration (Table IV). These findings indicate that blood transfusion induced a postoperative increase of TNF-α. The two-way factorial analysis revealed that ulinastatin administration resulted in significant changes of TNF-α on day 3 after surgery when adjusting for the factor of blood transfusion (Table IV), although the TNF-α levels in the BT group did not differ from those in the BT+UTI group (P >0.05) (Figure 4). These findings indicate that ulinastatin had an inhibitory effect on postoperative TNF-α levels in both control and blood transfused patients, independent of blood transfusion.
Figure 4.
Effects of blood transfusion and liver resection on perioperative levels of tumor necrosis factor-α (TNF-α). Data are expressed as mean ± S.E.M. T0: before surgery, T1: immediately after surgery, T2: 6 hours after surgery, T3: day 1 after surgery and T4: day 3 after surgery. Control group, no blood transfusion or ulinastatin administration; UTI group, ulinastatin only; BT group, blood transfusion only; BT+UTI group, both blood transfusion and ulinastatin administration.
Discussion
SIRS is a systemic response to a wide variety of insults and is regarded as an important factor responsible for morbidity and mortality following major surgery19–21. The syndrome was considered to be present when the SIRS score was ≥2, including, but not limited to, more than one of a series of clinical manifestations and physiological changes of the body19,21. It has been reported that the incidence of post-operative SIRS varies greatly from 10% to more than 70% depending on the type of surgery22–24. Furthermore, it has been suggested that allogeneic blood transfusion is capable of triggering SIRS and is an independent risk factor for the syndrome8,21,25. The results of our study indicate that a transfusion of a mean of 400–600 mL (Table I) of allogeneic packed red blood cells increases the risk of SIRS after liver resection, a finding consistent with those of previous studies8,21–25.
As reported, IL-6 is related directly to the appearance of endotoxins, which activate many complement factors and cytokines and induce neutrophils to release PMNE26,27. TNF-α, which is secreted by activated macrophages, can stimulate other inflammatory cytokines, including IL-8 in endothelial cells and monocytes in the blood28,29. Secreted IL-8 promotes tissue injury by neutrophil infiltration and the secretion of lysosomal enzymes30. The activation of all of these pro-inflammatory factors has been suggested to be involved in mediating the post-traumatic sequelae of SIRS and even multiple organ failure9,16,21,29. Transfused allogeneic blood is emerging as an inflammatory agent capable of priming neutrophils, inducing numerous pro-inflammatory substances such as IL-6, IL-8, and TNF-α and triggering the release of PMNE1,2,6,8,14,31. Allogeneic blood transfusion may, therefore, increase the risk of SIRS. To the best of our knowledge, there are few studies providing direct evidence on an association between pro-inflammatory cytokines and risk of SIRS in surgical patients receiving allogeneic blood transfusion, although the phenomenon has been reported in trauma patients21. In the present study, we examined the perioperative levels of IL-6, IL-8 and TNF-α and found that the surgical procedure of liver resection resulted in an increase of plasma IL-6 level in the early postoperative period (for about 1 day). Moreover, intraoperative allogeneic blood transfusion further increased levels of IL-6, IL-8 and TNF-α in the 3 days after surgery (Figures 2, 3, 4), consistent with the results of previous studies 13,14,16,25. Interestingly, the incidence of SIRS in the control and BT groups seemed to parallel the plasma levels of IL-6, IL-8 and TNF-α over time (Figures 2, 3, 4 and Tables II–IV). These results could be interpreted as meaning that the increases of IL-6, IL-8 and TNF-α levels may contribute to the increased incidence of SIRS in patients receiving allogeneic blood transfusion.
However, the time to peak IL-6 level (on day 3 after surgery in the BT group, see Figure 2) was not consistent with the findings of the study by Nishiyama et al., in which the IL-6 level peaked at the end of the surgery after blood transfusion14. This discrepancy in results might be because the storage time of the transfused blood products used in our study (20.8±5.2 days) was longer than that of the products used in the study by Nishiyama et al. (13.5±3.2 days); the older products could have had the effect of inducing the release of more pro-inflammatory substances even on day 3 after surgery19,21.
As a protease inhibitor, ulinastatin has the physiological characteristics of inhibiting neutrophil elastase, trypsin, α-chymotrypsin, plasmin, and cathepsin G 10–12, and also inhibits the processing of pro-interleukin 1β (a 31 kDa inactive peptide) to IL-1β (a 17 kDa active peptide)32,33. It has been demonstrated that the transcription of IL-6, IL-8 and TNF-α is a secondary event induced by the bioactive IL-1β 32,34. Ulinastatin, therefore, has the capacity to inhibit activation of pro-inflammatory cytokines such as IL-6, IL-8 and TNF-α associated with blood transfusion, as shown in Figures 2, 3, 4. In fact, the effect of ulinastatin on inhibition of cytokines could have been underestimated because we did not treat the blood samples with a protease inhibitor cocktail, which may increase the values of cytokines in protease inhibitor (e.g. ulinastatin) sample35. Based on the above-mentioned physiological characteristics, ulinastatin is used in the management of acute inflammatory disorders, such as acute pancreatitis, circulatory insufficiency, shock, disseminated intravascular coagulation, SIRS and multiple organ dysfunction syndrome36–39. However, there are no reports on the effect of ulinastatin on inhibiting the risk of SIRS associated with blood transfusion in patients undergoing surgery. In the present study we found that a single dose of ulinastatin (100,000/10 kg) given 15 minutes before a blood transfusion decreased the risk of SIRS associated with the intraoperative transfusion. The increases of IL-6, IL-8 and TNF-α associated with blood transfusion were also inhibited by ulinastatin, indicating that the ulinastatin-related decrease in the risk of SIRS may be associated with the inhibition of pro-inflammatory cytokines.
With regards to ulinastatin’s effects on pro-inflammatory cytokines associated with blood transfusion, there are some divergences between our study and others. Karasawa et al.17 found that ulinastatin inhibited the increase of IL-6 and PMNE caused by upper abdominal surgical stimuli but not by blood transfusion. Nishiyama et al.13,14 found that ulinastatin inhibited the increase of PMNE induced by blood transfusion but did not inhibit the increase of IL-6 caused by surgery or blood transfusion in patients undergoing gastrectomy when the ulinastatin, at a dose of 300,000 units, was given prior to the blood transfusion. In contrast, in our study ulinastatin markedly inhibited the increase of IL-6 on day 3 after intraoperative blood transfusion. The reason for these differences is unclear. Karasawa et al.17 administered the ulinastatin after induction of anaesthesia, which may be too early for the effect of ulinastatin to be maintained because the plasma half-life of this protease inhibitor is about 33 minutes41. We and Nishiyama et al.13,14 administered the ulinastatin around the start of blood transfusion and so it could have inhibited some pro-inflammatory cytokines (PMNE in the study by Nishiyama et al.; IL-6, IL-8 and TNF-α in our study). However, the dose of ulinastatin we used was larger (almost double) than that in the study by Nishiyama et al., which might be another cause for the differences between the results of the two studies. We did not measure the changes of plasma concentration of PMNE in our study since the inhibitory effect of ulinastatin on PMNE has already been well clarified7,8,13,14,25.
In the present study, we found that the time spent in hospital was significantly longer for patients who received allogeneic blood transfusion than those who did not (Table I). The reason for this is unclear. It may be because of the allogeneic blood transfusion-induced increase of cytokines (Figures 2, 3, 4) or because the intraoperative and postoperative conditions of patients receiving allogeneic blood transfusion were more serious. It has been reported that pro-inflammatory cytokines, such as IL-6 and IL-8, may be used as early prognostic factors in patients with organ dysfunction after cardiac surgery15; ulinastatin’s inhibition of pro-inflammatory cytokines could, thereby, be beneficial by attenuating the inflammatory over-response and promoting the recovery of patients receiving blood transfusion. Unfortunately, the time spent in hospital by patients in the BT+UTI group (15.6±3.7 days) was not statistically different from that of patients in the BT group (20.1±6.7 days) (P =0.141; Table I), although it seemed to be slightly shorter. Furthermore, we did not demonstrate any differences in other postoperative complications between patients who did or did not receive ulinastatin (Table I). Thus the association between blood transfusion-induced increases of plasma cytokines and postoperative outcomes and morbidities is still uncertain although it may contribute to the increased incidence of SIRS.
There are some limitations to this study. Although the drug administration was randomised (ulinastatin versus control), the assignment to blood transfusion or not could not be randomised and introduced some bias. However, the general condition of the patients in this study was comparable among the four groups, and the confounding factor could be adjusted for by using two-way ANOVA and stratified chi-square testing during the statistical analysis. The follow-up investigator who recorded the SIRS scores was independent and blinded to the patients’ grouping and observation bias could be eliminated. Another limitation is that the sample size was relatively small to investigate changes in postoperative infections and other surgical complications (indeed no postoperative infections or surgery complications were observed among the four groups in this study). Future randomised, blinded studies with greater statistical power will be conducted to demonstrate the clinical benefits of ulinastatin in decreasing surgical complications and optimising other outcomes.
In conclusion, in this prospective clinical controlled setting, a single dose of ulinastatin given before allogeneic blood transfusion reduced the risk of postoperative SIRS and lowered the levels of plasma IL-6, IL-8 and TNF-α associated with intraoperative allogeneic blood transfusion in patients undergoing liver resection and might, therefore, improve patients’ postoperative recovery.
Footnotes
Conflicts of interest disclosure
This work was supported by a research fund offered to Drs Wenqi Huang and Haihua Shu by Techpool Biopharma Co. Ltd, Guangzhou, Guangdong, China (No. 0120095).
References
- 1.Koch CG, Li L, Sessler DI, et al. Duration of red-cell storage and complications after cardiac surgery. N Engl J Med. 2008;358:1229–39. doi: 10.1056/NEJMoa070403. [DOI] [PubMed] [Google Scholar]
- 2.Shander A. Emerging risks and outcomes of blood transfusion in surgery. Semin Hematol. 2004;41:117–24. doi: 10.1053/j.seminhematol.2003.11.023. [DOI] [PubMed] [Google Scholar]
- 3.Vamvakas EC, Blajchman MA. Transfusion-related immunomodulation (TRIM): an update. Blood Rev. 2007;21:327–48. doi: 10.1016/j.blre.2007.07.003. [DOI] [PubMed] [Google Scholar]
- 4.Napolitano LM. Current status of blood component therapy in surgical critical care. Curr Opin Crit Care. 2004;10:311–7. doi: 10.1097/01.ccx.0000140948.98019.8a. [DOI] [PubMed] [Google Scholar]
- 5.Edgren G, Kamper-Jorgensen M, Eloranta S, et al. Duration of red blood cell storage and survival of transfused patients (CME) Transfusion. 2010;50:1185–95. doi: 10.1111/j.1537-2995.2010.02583.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.van de Watering L, Lorinser J, Versteegh M, et al. Effects of storage time of red blood cell transfusions on the prognosis of coronary artery bypass graft patients. Transfusion. 2006;46:1712–8. doi: 10.1111/j.1537-2995.2006.00958.x. [DOI] [PubMed] [Google Scholar]
- 7.Sparrow RL, Patton KA. Supernatant from stored red blood cell primes inflammatory cells: influence of prestorage white cell reduction. Transfusion. 2004;44:722–30. doi: 10.1111/j.1537-2995.2004.03113.x. [DOI] [PubMed] [Google Scholar]
- 8.Zallen G, Moore EE, Ciesla DJ, et al. Stored red blood cells selectively activate human neutrophils to release IL-8 and secretory PLA2. Shock. 2000;13:29–33. doi: 10.1097/00024382-200013010-00006. [DOI] [PubMed] [Google Scholar]
- 9.Maier B, Lefering R, Lehnert M, et al. Early versus late onset of multiple organ failure is associated with differing patterns of plasma cytokine biomarker expression and outcome after severe trauma. Shock. 2007;28:668–74. [PubMed] [Google Scholar]
- 10.Jonsson-Berling BM, Ohlsson K, Rosengren M. Radioimmunological quantitation of the urinary trypsin inhibitor in normal blood and urine. Biol Chem Hoppe Seyler. 1989;370:1157–61. doi: 10.1515/bchm3.1989.370.2.1157. [DOI] [PubMed] [Google Scholar]
- 11.Hirose J, Ozawa T, Miura T, et al. Human neutrophil elastase degrades inter-alpha-trypsin inhibitor to liberate urinary trypsin inhibitor related proteins. Biol Pharm Bull. 1998;21:651–6. doi: 10.1248/bpb.21.651. [DOI] [PubMed] [Google Scholar]
- 12.Jin X, Deng L, Li H, et al. Identification and characterization of a serine protease inhibitor with two trypsin inhibitor-like domains from the human hookworm Ancylostoma duodenale. Parasitol Res. 2011;108:287–95. doi: 10.1007/s00436-010-2055-z. [DOI] [PubMed] [Google Scholar]
- 13.Nishiyama T, Aibiki M, Hanaoka K. The effect of ulinastatin, a human protease inhibitor, on the transfusion-induced increase of plasma polymorphonuclear granulocyte elastase. Anesth Analg. 1996;82:108–12. doi: 10.1097/00000539-199601000-00019. [DOI] [PubMed] [Google Scholar]
- 14.Nishiyama T, Hanaoka K. Do the effects of a protease inhibitor, ulinastatin, on elastase release by blood transfusion depend on interleukin 6? Crit Care Med. 2001;29:2106–10. doi: 10.1097/00003246-200111000-00010. [DOI] [PubMed] [Google Scholar]
- 15.Sablotzki A, Dehne MG, Friedrich I, et al. Different expression of cytokines in survivors and non-survivors from MODS following cardiovascular surgery. Eur J Med Res. 2003;8:71–6. [PubMed] [Google Scholar]
- 16.Bilgin YM, Brand A. Transfusion-related immunomodulation: a second hit in an inflammatory cascade? Vox Sang. 2008;95:261–71. doi: 10.1111/j.1423-0410.2008.01100.x. [DOI] [PubMed] [Google Scholar]
- 17.Karasawa S, Uehara K, Yokota T, Nomoto Y. Changes in serum concentrations of cytokines by ulinastatin and blood transfusion in upper abdominal surgery. (in Japanese) J Clin Anesthes. 1996;20:1169–75. [Google Scholar]
- 18.Bingyang J, Jinping L, Mingzheng L, et al. Effects of urinary protease inhibitor on inflammatory response during on-pump coronary revascularisation. Effect of ulinastatin on inflammatory response. J Cardiovasc Surg (Torino) 2007;48:497–503. [PubMed] [Google Scholar]
- 19.Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest. 1992;101:1644–55. doi: 10.1378/chest.101.6.1644. [DOI] [PubMed] [Google Scholar]
- 20.Bone RC, Sprung CL, Sibbald WJ. Definitions for sepsis and organ failure. Crit Care Med. 1992;20:724–6. doi: 10.1097/00003246-199206000-00002. [DOI] [PubMed] [Google Scholar]
- 21.Dunne JR, Malone DL, Tracy JK, Napolitano LM. Allogenic blood transfusion in the first 24 hours after trauma is associated with increased systemic inflammatory response syndrome (SIRS) and death. Surg Infect (Larchmt) 2004;5:395–404. doi: 10.1089/sur.2004.5.395. [DOI] [PubMed] [Google Scholar]
- 22.Wang XY, Niu CL, Zhang L, et al. Effect of enteral nutrition on liver function and inflammatory response after abdominal operation in patients complicated with liver dysfunction. Zhonghua Wei Chang Wai Ke Za Zhi. 2011;14:336–9. (In Chinese) [PubMed] [Google Scholar]
- 23.Mommsen P, Andruszkow H, Fromke C, et al. Effects of accidental hypothermia on posttraumatic complications and outcome in multiple trauma patients. Injury. 2013;44:86–90. doi: 10.1016/j.injury.2011.10.013. [DOI] [PubMed] [Google Scholar]
- 24.Tsujimoto H, Ono S, Takahata R, et al. Systemic inflammatory response syndrome as a predictor of anastomotic leakage after esophagectomy. Surg Today. 2011;42:141–6. doi: 10.1007/s00595-011-0049-9. [DOI] [PubMed] [Google Scholar]
- 25.Fransen E, Maessen J, Dentener M, et al. Impact of blood transfusions on inflammatory mediator release in patients undergoing cardiac surgery. Chest. 1999;116:1233–9. doi: 10.1378/chest.116.5.1233. [DOI] [PubMed] [Google Scholar]
- 26.Wen D, Zhou XL, Li JJ, et al. Plasma concentrations of interleukin-6, C-reactive protein, tumor necrosis factor-alpha and matrix metalloproteinase-9 in aortic dissection. Clin Chim Acta. 2012;413:198–202. doi: 10.1016/j.cca.2011.09.029. [DOI] [PubMed] [Google Scholar]
- 27.Muller M, Scheel O, Lindner B, et al. The role of membrane-bound LBP, endotoxin aggregates, and the MaxiK channel in LPS-induced cell activation. J Endotoxin Res. 2003;9:181–6. doi: 10.1179/096805103125001595. [DOI] [PubMed] [Google Scholar]
- 28.Hanazaki K, Monma T, Hiraguri M, et al. Cytokine response to human liver ischemia-reperfusion injury during hepatectomy: marker of injury or surgical stress? Hepatogastroenterology. 2001;48:188–92. [PubMed] [Google Scholar]
- 29.Cohen J. The immunopathogenesis of sepsis. Nature. 2002;420:885–91. doi: 10.1038/nature01326. [DOI] [PubMed] [Google Scholar]
- 30.Mukaida N, Matsumoto T, Yokoi K, et al. Inhibition of neutrophil-mediated acute inflammation injury by an antibody against interleukin-8 (IL-8) Inflamm Res. 1998;47(Suppl 3):S151–7. doi: 10.1007/s000110050308. [DOI] [PubMed] [Google Scholar]
- 31.Ho J, Sibbald WJ, Chin-Yee IH. Effects of storage on efficacy of red cell transfusion: when is it not safe? Crit Care Med. 2003;31:S687–97. doi: 10.1097/01.CCM.0000099349.17094.A3. [DOI] [PubMed] [Google Scholar]
- 32.Dinarello CA. Biologic basis for interleukin-1 in disease. Blood. 1996;87:2095–147. [PubMed] [Google Scholar]
- 33.Inoue K, Takano H, Shimada A, et al. Urinary trypsin inhibitor protects against systemic inflammation induced by lipopolysaccharide. Mol Pharmacol. 2005;67:673–80. doi: 10.1124/mol.104.005967. [DOI] [PubMed] [Google Scholar]
- 34.Kelk P, Claesson R, Hanstrom L, et al. Abundant secretion of bioactive interleukin-1beta by human macrophages induced by Actinobacillus actinomycetemcomitans leukotoxin. Infect Immun. 2005;73:453–8. doi: 10.1128/IAI.73.1.453-458.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Simpson JL, Gibson PG, Wark PA. Optimization of sputum-processing methods for the measurement of interleukin-5: effects of protease inhibition. Respirology. 2002;7:111–6. doi: 10.1046/j.1440-1843.2002.00380.x. [DOI] [PubMed] [Google Scholar]
- 36.Inamo Y, Okubo T, Wada M, et al. Intravenous ulinastatin therapy for Stevens-Johnson syndrome and toxic epidermal necrolysis in pediatric patients. Three case reports. Int Arch Allergy Immunol. 2002;127:89–94. doi: 10.1159/000048174. [DOI] [PubMed] [Google Scholar]
- 37.Inoue K, Takano H, Yanagisawa R, Yoshikawa T. Protective effects of urinary trypsin inhibitor on systemic inflammatory response induced by lipopolysaccharide. J Clin Biochem Nutr. 2008;43:139–42. doi: 10.3164/jcbn.2008059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Inoue K, Takano H. Urinary trypsin inhibitor as a therapeutic option for endotoxin-related inflammatory disorders. Expert Opin Investig Drugs. 2010;19:513–20. doi: 10.1517/13543781003649533. [DOI] [PubMed] [Google Scholar]
- 39.Ribeiro JK, Cunha DD, Fook JM, Sales MP. New properties of the soybean trypsin inhibitor: Inhibition of human neutrophil elastase and its effect on acute pulmonary injury. Eur J Pharmacol. 2010;644:238–44. doi: 10.1016/j.ejphar.2010.06.067. [DOI] [PubMed] [Google Scholar]
- 40.Jonsson-Berling BM, Ohlsson K. Distribution and elimination of intravenously injected urinary trypsin inhibitor. Scand J Clin Lab Invest. 1991;51:549–57. doi: 10.3109/00365519109104564. [DOI] [PubMed] [Google Scholar]