The study was performed according to ICH-GCP (International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use – Good Clinical Practice) guidelines. We also observed the tenets of the Declaration of Helsinki, had the study authorized by the relevant national authority, and had it approved by the appropriate Research and Ethics Committee prior to enrollment. The study was performed under EudraCT No. 2006-000451-17 between June 2006 and June 2009 at the Marienkrankenhaus, Soest, Germany. The trial was also registered at www.clinicaltrials.gov
under NCT00738608, protocol number 2006-000451-17.
We recruited male and female patients between 19 and 70 years of age who were scheduled to undergo therapeutic arthroscopy (e.g., meniscal surgery, joint shaving) of the knee under general anesthesia. Only patients undergoing therapeutic arthroscopy (as opposed to diagnostic arthroscopy) were selected, because patients undergoing diagnostic procedures may not suffer enough postoperative pain to justify treatment with etoricoxib. All patients who undergo therapeutic knee arthroscopy are routinely treated as inpatients for 2 days after surgery. Before they were enrolled in the study, the patients had to provide written informed consent.
Patients with a creatinine clearance of ≤ 50 mL/min (calculated using the Cockcroft-Gault formula); opiate addiction; a known hypersensitivity to the active ingredient in etoricoxib or to any of the excipients in the film-coated placebo tablet, to NSAIDs, or to aspirin; patients with active peptic ulcer or active gastrointestinal bleeding, inflammatory bowel disease, liver cirrhosis, cholestasis, elevated liver function tests (i.e. ALT or AST levels more than 3 times the upper limit of normal); or severe hepatic dysfunction (i.e. patients with a serum albumin concentration of < 25 g/L or with a Child-Pugh score of ≥ 10) were excluded for safety reasons. Other exclusion criteria were pregnancy; breast feeding; congestive heart failure (i.e. patients with New York Heart Association (NYHA) class II–IV heart failure); hypertension with inadequately controlled blood pressure (BP) (i.e. systolic BP > 160 mmHg or diastolic BP > 100 mmHg); a known history of ischemic heart disease; peripheral arterial disease; and/or cerebrovascular disease.
Interventions, randomization, blinding, and concealment of allocation
Patients received either a single oral tablet of etoricoxib (120 mg) or one look-alike film-coated placebo tablet 1 h before induction of anesthesia for arthroscopy. A randomization code was prepared at the hospital using the Microsoft Excel 2007 function RANDBETWEEN (1;2) by a person not involved in data collection. Generation of the list was repeated until a list with equal distribution of 1 and 2 resulted. Etoricoxib and placebo were prepared according to this randomization list and labeled with consecutive numbers from 1 to 66, and patients received the medication with the lowest available number. The patients, nurses, surgeons, and anesthesiologists directly involved in patient care were blind regarding the content of the oral study medication for the duration of the study (i.e. until all queries were resolved and the database was closed). For every patient, a sealed envelope with the treatment allocation was kept at the intensive care unit of the hospital so that immediate decoding of the study medication would have been possible for individual patients in the event of a serious adverse reaction.
Study schedule, outcomes, and objectives
Before surgery, patients underwent standard preoperative examinations, including electrocardiography; physical examination; determination of vital signs; urine pregnancy test (in women of childbearing age); and measurement of hematological, blood coagulation, and standard clinical chemistry parameters. In all cases, arthroscopy was performed under general anesthesia (i.e. total intravenous anesthesia (TIVA) with propofol and fentanyl).
After the end of surgery and anesthesia, the patient was transferred to the recovery room, where he was immediately connected to a PCA pump. There, 30 min after the end of anesthesia (t = 0), all patients were able to answer questions and to rate their pain on a 10-cm visual analog scale (VAS) at rest and during movement (range of the scales: 0–10, with 0 = no pain and 10 = worst imaginable pain). After checking of stable vital signs and full recovery, the patient was transferred from the recovery room to an orthopedic ward. At the ward, 2 h, 4 h, 6 h, and 24 h after the first measurement, the patients were visited by a physician (HL) and the measurements were repeated. Analgesia was administered according to usual hospital standards using a PCA pump (which was set to deliver a single bolus of morphine at a dose of 0.02 mg/kg body weight up to 6 times per hour). The amount of morphine that had been used by the time points specified above and the time of the administration of the first PCA pump dose were documented. The primary endpoint of the study was to determine the extent to which a single preoperative dose of 120 mg etoricoxib reduced postoperative opioid consumption during the 24-hour period after therapeutic knee arthroscopy.
Patient alertness was scored on a numerical rating scale (0 = alert and oriented, 1 = slightly drowsy, 2 = mildly sedated but arousable by shaking, and 3 = deeply sedated, not arousable) by the anesthesiologist responsible for the recovery room. Heart rate (defined normal range: 40–120/min.), blood pressure (defined normal range: systolic 95–195 mmHg) and respiratory rate (defined normal range: 10–25/min.) were measured at the times specified above. The patients were also asked if they were satisfied with the level of analgesia that had been achieved at those times: (1 = very satisfied, 2 = satisfied, 3 = dissatisfied, and 4 = very dissatisfied; 1 and 2 were regarded as successful pain management and 3 and 4 were regarded as unsuccessful pain management).
Sample size and statistical analysis
Because there were insufficient data available in the literature on the preoperative use of etoricoxib at the time this study was planned, the sample size estimate was based on several assumptions. We assumed that patients in the placebo group would require a mean morphine dose of 55 mg during the first 24 h after surgery and those patients who received etoricoxib would require a mean morphine dose of 36 mg (35% less) over the same time period. We assumed a common standard deviation of 22 mg (i.e. 40% of the mean morphine dose in the placebo group). We also assumed an alpha of 0.05, a beta of 0.10, and a dropout rate of 10%. These assumptions resulted in a sample size of 33 patients per group.
The data are presented either as mean (SD) or as median (range). They were tested for normal distribution using the Kolmogorov-Smirnov test. Where there was normal distribution, Student’s t-test was used to compare the two groups at each time point. Otherwise, non-parametric tests such as the Mann-Whitney test were used. Patients’ overall morphine requirements, VAS scores, and vital signs over the first 24-h period postoperatively were examined by repeated-measures analysis of variance (ANOVA). Differences in the rates of side effects between the groups were calculated using the chi-2 test or Fisher’s exact test, as appropriate. Analyses were performed using SPSS software version 16. All tests were two-tailed, and differences were considered statistically significant for p < 0.05.