This is a multicenter, randomized, controlled trial. The study protocol was approved by the Ethics Committee of Peking University People's Hospital (No. 2017PHA019) and all 14 participating centers. The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. All participants provided written informed consent.
Participants
Adult patients (18–65 years of age) scheduled for abdominal surgery (intestinal or lower abdomen) or orthopedic surgery (e.g., knee arthroplasty and shoulder reconstruction) under general anesthesia with endotracheal intubation were eligible. Exclusion criteria included: 1) American Society of Anesthesiologists (ASA) Class III or higher; 2) body mass index (BMI) < 18 or > 30 kg/m2; 3) inability to understand the NRS score; 4) history of allergy or hypersensitivity to ibuprofen, aspirin, NSAIDs, or any cyclooxygenase-2 inhibitors; 5) anemia; 6) pregnancy or lactation; 7) history of head trauma, intracranial surgery or stroke, or organic diseases of the central nervous system within 4 weeks before surgery; 8) history of congenital bleeding diseases (such as hemophilia), thrombocytopenia (platelet count below 80×109/L), coagulation disorders; 9) history of organ transplantation or severe cardiovascular and cerebrovascular diseases, liver dysfunction (alanine aminotransferase or glutamic oxalacetic transaminase > 1.5 times normal upper limit), renal dysfunction (blood urea nitrogen > 1.5 times normal upper limit, creatinine > normal upper limit), endocrine system and mental illness; 10) history of gastrointestinal ulcer or bleeding; 11) receiving warfarin, lithium, methotrexate, clopidogrel or other long-term anticoagulant or anti-platelet drugs, two or more antihypertensive agents; 12) exposure to any analgesic, muscle relaxant, or sedative within 24 hours.
Study Medications
Ibuprofen (400 mg in 4-ml; batch #: 170101/181103) and placebo (saline, 4-ml; batch #:170102/181001, the specification of the placebo, such as appearance, color, and odor, was consistent with ibuprofen injection) were provided by Hainan Herui Pharmaceutical Company (Hainan, China). Each package of medication for each subject contained 9 doses with same combination: 2 ampoules of ibuprofen (800mg ibuprofen), or 1 ampoule of ibuprofen and 1 ampoule of saline (400mg ibuprofen), or 2 ampoules of saline (placebo) for per dose.
Randomization and blinding
The three kinds of study drug packages were randomly assigned in a ratio of 1:1:1 using the computer-generated random number sequence. Based on the allocation sequence, drug packages were labeled serially in pharmaceutical company. Subjects were randomized to be assigned to one of the three groups in the order of recruitment and received intravenous infusion of placebo, ibuprofen 400mg or ibuprofen 800 mg. The researchers, patients and doctors, were all blinded throughout the study. The complete random sequence was kept by the primary researcher in Peking University People’s Hospital. Emergency envelopes containing the random allocation individually were prepared for each center, which could be unsealed only in the case of serious adverse events.
Study Design
Subjects were received intravenous infusion of placebo, or ibuprofen at 400 or 800 mg every 6 hours for total of 9 doses, from the time of wound closure to 48 hours. Each infusion lasted for 30 min. All subjects received 5-mg intravenous morphine at the time of wound closure and PCIA (Fig. 1). For PCIA, 100-mg morphine was dissolved in 200-ml saline, the background infusion rate was 0.5 mg/h, bolus was 1 mg, with a lockout time of 5 min and 20-mg 4-hour limit. Subjects had access to morphine via PCIA throughout the treatment period. Study protocol mandated discontinuation of study medication if the patient required the use of additional pain medication, including a narcotic other than morphine or any other non-narcotic pain medication including NSAIDs due to the inability to manage pain by PCIA. These cases were classified as treatment failures.
Anesthesia was induced using 0.03-mg/kg midazolam, 1.5 ~ 2.5-mg/kg propofol or 0.3-mg/kg etomidate, 3-µg/kg fentanyl, and 0.6-mg/kg rocuronium or 0.1-mg/kg vecuronium. Anesthesia was maintained using propofol and remifentanil, or sevoflurane inhalation and remifentanil. It was allowed to add muscle relaxants according to the need of operation and add fentanyl 1µg/kg every 2 hours after anesthesia induction (not allowed within 1.5 hours before the end of the operation). Neuraxial anesthesia and regional blockade were not permitted. According to the study protocol, dexmedetomidine, sufentanil, tramadol, penehyclidine hydrochloride, and scopolamine were not permitted during the surgery; aspirin, other NSAIDs and other drugs with antipyretic (e.g., glucocorticoids) and analgesic effects or anticoagulants (except low molecular weight heparin, LMWH), thrombolytic drugs (streptokinase, urokinase), angiotensin-converting enzyme inhibitors (ACEI), diuretics, lithium agents, methotrexate, etc., were prohibited during the study period.
Outcomes
The primary endpoint measure was morphine consumption within the first 24 hours after surgery, calculated according to the intent-to-treat principle. The secondary endpoints included: 1) Pain intensity at rest and with movement. Pain intensity was evaluated using patient self-reporting with numerical rating scales (NRS) (0 = no pain to 10 = intense pain) at 1, 2, 3, 6, 12, 18, 24, 30, 36, 42 and 48 hours after the first administration of test drugs. In patients who were asleep at the time of evaluation, the pain score was recorded as 2. 2) The area under the pain intensity-time curve (AUC) both at rest and with movement during 1–24 hours after surgery. 3) The number of total and effective PCIA pushes within the first 24 hours after surgery. 4) Percentage of the treatment failures. 5) Incidence of pyrexia (defined body temperature ≥ 38°C) during the study period. 6) Analgesia-related adverse events and laboratory assessments (included biochemistry, electrolytes, blood routine, blood coagulation function, urinalysis, and electrocardiogram) the third day after surgery.
Statistical Analysis
Sample size calculation was based on the following assumptions: 1) 24-hour morphine consumption at 55.9 mg (SD = 20.6) in the placebo group vs. 47.3 mg (SD = 25.6) in the 800-mg group [13]; 2) 2-sided a at 0.05 and 1-b at 0.8. The calculation yielded 116 subjects in each group. Assuming 20% dropout rate, we planned to enroll 396 patients.
Continuous variables were presented as the mean ± standard deviation (SD) if following normal distribution, and as median (interquartile range; IQR) otherwise. The 24-hour morphine consumption was analyzed using ANCOVA after Box-Cox transformation, with intervention, center, patient age, and weight as covariates. Pain scores were analyzed using a mixed effect model for repeated measurement. Area under the pain intensity-time curve (AUC) was compared among groups using Kruskal-Wallis and Nemenyi tests. Chi-square test was used to analyze categorical variables. P < 0.05 was considered statistically significant. All statistical analyses were conducted using SAS 9.4.