Exclusion criteria
Patients aged <20 years, or >82 years, and those with renal dysfunction, hepatic failure, peptic ulcer, aspirin-induced asthma, or arrhythmia were excluded. Patients who were taking steroids or nonsteroidal antiinflammatory drugs including aspirin were also excluded from the study. In addition, the cases that began with not colorectal surgery but the other operation such as liver resection will be excluded from the study because in such cases surgeons usually do not pull mesentery.
Study protocol
Forty-five patients, who were scheduled to undergo elective surgery for colorectal cancer via laparotomy from October 2012 to January 2013 were enrolled. Using a random number table, patients were randomly divided into 3 groups as follows: a preoperative group receiving flurbiprofen axetil (Ropion; Kaken Pharmaceutical, Tokyo, Japan) directly before surgery, a post-MTS group receiving flurbiprofen axetil following MTS onset, and a control group who were not administered flurbiprofen axetil. Patients in the preoperative group (
n = 16) were administered flurbiprofen axetil (1 mg/kg; maximum dose = 50 mg) intravenously after induction of anesthesia, just prior to the initial incision. Patients in the post-MTS group (
n = 14) received intravenous flurbiprofen axetil (1 mg/kg; maximum dose = 50 mg) after MTS onset. In this study, MTS was defined as a flushing of the face during an abdominal surgical procedure within 1 h after the initiation of surgery, and was diagnosed by two anesthesiologists [
13]. If we could not confirm MTS onset, flurbiprofen axetil was not administered. Patients in the control group (
n = 15) were not administered flurbiprofen axetil.
When hypotension requiring treatment were observed, ephedrine (Ephedrine; Nichi-Iko Pharmaceutical, Toyama, Japan), phenylephrine (Neosinejin; Kowa Company, Nagoya, Japan), or fluid (crystalloid solution or colloid solution), or all three were used at the discretion of the anesthesiologist in charge.
Throughout the perioperative care, all the patients were treated by our enhanced recovery after surgery (ERAS) protocols [
14,
15]. No pre- and postoperative fasting (provision of oral nutrition) as well as intensive pre-admission counselling, avoidance of sodium/fluid overload, intraoperative warm-air body heating, enforced postoperative mobilization, and multimodal team care were among the main changes brought about by the introduction of ERAS protocols to our hospital [
14,
15]. The course of the surgical procedure included peritoneal incision, opening of the abdomen, liver palpation (abdominal examination), unfolding of the colon, resection, and anastomosis; the same surgical team performed all the operations.
The subjects in each study arm were assigned in blinded fashion to the surgeons; however, the study design remained single-blind for the study team’s anesthesiologist.
All patients in all groups received general anesthesia in combination with epidural anesthesia. After admission into the operating room, an epidural catheter was inserted. Anesthesia was induced with propofol (1.5–2 mg·kg−1), rocuronium (0.6 mg·kg−1), sevoflurane (1–3%), remifentanyl (0.25–0.5 μg·kg−1·min−1), and ephedrine and dopamine as needed. After intubation, anesthesia was maintained with oxygen, air, sevoflurane (1.2–1.4%), and remifentanyl (0.1–0.25 μg·kg−1·min−1). Then, the radial artery was cannulated for continuous measurement of arterial pulse pressure. Following administration of 0.375% ropivacaine (4–6 mL) through the epidural catheter, the surgery was started. The tidal volume setting of the mechanical ventilator was 8–10 mL/kg.
To confirm the synthesis of prostacyclin, which triggers MTS, we measured the levels of 6-keto-PGF1α, a stable metabolite of prostacyclin. Because a previous report described the mean onset time of MTS to be 16 ± 5 min after skin incision with symptoms persisting over approximately 30 min [
16], blood was collected from an arterial line at the initiation of surgery (T0) as well as at 15, 30, and 60 min after the initiation of surgery (T15, T30, and T60, respectively). The samples were refrigerated and centrifugal separation was rapidly performed. Then, blood plasma was subjected to cryopreservation. 6-Keto-PGF1α levels within the samples were measured using enzyme-linked immunosorbent assay (Enzo Life Sciences, Inc.) at the Department of Health Chemistry, Graduate School of Pharmaceutical Sciences, University of Tokyo.
Statistical analyses
Results were displayed as mean value ± standard deviation (mean ± SD), and statistical processing was performed as follows. The statistical methods used for comparison between the groups were Chi-squared test for frequency data. Comparison of frequency data between 2 groups was performed using the independent t-test. As for 6-keto-PGF1α levels, an analysis of covariance (ANCOVA) comparing each pair out of the three intervention groups were used for analyses at each of time points (e.g. T15, T30, T60 for 6-keto-PGFIα level), each time taking the baseline measurement (e.g. 6-keto-PGFIα level at T0) as a covariate (both 6-keto-PGF1α levels and mean blood pressure). Since multiple tests were performed, a correction for multiple testing, a Bonferroni, was employed (In these cases, P < 0.0056 (0.05/3 × 3) was considered statistically significant). As for mean blood pressure, the Student’s t-test was used for comparing two time points (e.g. T0 as the baseline measurement, and T15) in each intervention group, and since multiple tests were performed, a correction for multiple testing, a Bonferroni, was employed (In these cases, P < 0.017 (0.05/3) was considered statistically significant). Because mean blood pressure was greatly influenced by the use of vasopressors and phenylephrine with various strength, analyses of its changes at T30, T45, and T60 were not performed, which were beyond our concern. Statistical analysis was performed using Dr. SPSS II for Windows (Ver. 11), and P < 0.05 was considered statistically significant unless otherwise mentioned.