This randomized double-blinded controlled trial was conducted at a University Hospital between November 2022 and May 2023, after the institutional ethics committee approval (MD-246-2022). Clinical trial registration was done before patients’ enrolment at ClinicalTrials.gov (NCT05610384, Date of registration: 09/11/2022). Informed consent was obtained from the patient before enrolment in the study.

Participants were adult (40–65 years) women with American Society of Anesthesiologist (ASA) classification of I-II, scheduled for open elective abdominal hysterectomy with or without salpingo-oophorectomy.

Exclusion criteria were renal impairment (history of renal impairment or kidney function tests above normal reference range), history of gastrointestinal bleeding or ulceration, inflammatory bowel disease, allergy to any of the study’s drugs, significant cardiac morbidity (impaired contractility, ischemic heart disease). Patients undergoing surgery for suspected gynecological cancer, patients on chronic analgesic medication, and patients deemed unable to understand the visual analogue scale (VAS) were also excluded from the study.

Randomization was done using an online randomizer (https://www.graphpad.com/quickcalcs/randomize1/) in 1:1 ratio. The group assignment and drug preparation instructions were put inside sequentially-numbered opaque envelopes. An independent researcher handled the envelope-opening and drug-preparation. The three scheduled doses were prepared at once; and the unused preparations were stored in the refrigerator and were marked with the patient’s name, hospital number and the time for administration. Ketorolac (Ketolac 30 mg/ 2 mL, AMRIYA PHARM. IND, Alexandria, Egypt) was prepared by diluting 30 mg in 200 mL normal saline; and ibuprofen (Ibuprofen-Arabcomed 100 mg/mL, ARABCOMED, Cairo, Egypt) was prepared by diluting 800 mg in 200 mL normal saline. The attending anesthetist, nurse, data collector and the patient were blinded to the group assignment.

Thirty minutes before surgery, the participants received the first dose of the study drug according to the group assignment in addition to 1 gm intravenous paracetamol (Medalgesic 10 mg/mL, ARABCOMED, Cairo, Egypt). The study drug was then given every 8 h and paracetamol was given every 6 h for 24 h.

In the operating room, electrocardiogram, pulse oximetry, and non-invasive blood pressure monitor were applied, and a prophylactic antiemetic was administered (4 mg dexamethasone).

Induction of general anesthesia was achieved by 2 mg/kg propofol and 1 mcg/kg fentanyl; tracheal intubation was facilitated by 0.5 mg/kg atracurium after loss of consciousness. Anesthesia was maintained with isoflurane 1-1.2% in oxygen/air admixture and 0.1 mg/kg atracurium every 20 min. Additional analgesic boluses (1 mcg/kg fentanyl) were given when needed according to the attending anesthetist’s discretion.

Postoperatively, the VAS was assessed at rest (static) and during hip and knee flexion (dynamic; by asking the patient to bend her knee while in the supine position) at 0.5, 2, 4, 6, 10, 18, and 24 h after leaving the operating room. An intravenous morphine bolus (2 mg) was given when the VAS was > 3 (or at any time upon patient request) and can be repeated if the pain persisted for 30 min after the initial bolus. If postoperative nausea and vomiting occurred, intravenous 4-mg ondansetron was given.

The primary outcome was the mean postoperative dynamic VAS during the first 24 h. The secondary outcomes were static and dynamic VAS, time to first analgesic requirement (defined as the time from extubation until first analgesic requirement), intraoperative fentanyl requirements, postoperative morphine requirements, time to independent movement (defined as time from extubation until being able to move independently), intra- and postoperative hemodynamic measurements (heart rate and systolic blood pressure were recorded every 15 min intraoperatively, and at 0.5, 2, 4, 6, 10, 18 and 24 h postoperatively). The occurrence of opioid-related complications was recorded including nausea and vomiting (incidence and No. of episodes), itching, urine retention, sedation level using the Modified Ramsay Sedation Score, respiratory depression (defined as respiratory rate less than 8 breath per minute). At the end of the study, the participant was asked to evaluate her satisfaction with pain management on a scale of 0–10 (in which a score of 10 means strongly satisfied and a score of zero means strongly unsatisfied); We recorded the value at which the patient expressed her level of satisfaction. Patients’ demographic data (age, weight, body mass index, ASA), surgical characteristics (duration of surgery, need for blood transfusion, the use of vasopressors, amount blood loss [via visual assessment of surgical gauze and recording the amount of blood suctioned from the surgical field]), pre- and 24-h postoperative blood hemoglobin concentration were also recorded.

In a previous study, the difference in the mean dynamic VAS between the two groups was 1.45 with standard deviations of 2.29 and 2.43 for the ibuprofen and ketorolac groups, respectively [8]. We calculated the sample size using the mentioned standard deviations (2.29 and 2.43) and to detect a difference of 1.4 in the mean dynamic VAS. Having a study power of 80% and an alpha error of 0.05, the minimum number of patients would be 92. The number of envelopes was increased to 100 to compensate for possible dropouts. The sample size was calculated using the MedCalc (14.10.2) software.

Statistical package for social science (SPSS) software, version 26 for Microsoft Windows (IBM. Corp., NY, USA) was used for data analysis. Categorical data are presented as frequency (%) and were analyzed by the Chi squared test or Fisher exact test as fitting. Continuous data were checked for normality using the Shapiro-Wilk test and are presented as mean ± standard deviation or median (quartiles) according to the data distribution. Unpaired continuous data were analyzed using the unpaired t test or Mann Whitney test according to the distribution of the data. Repeated measures were analyzed using the analysis of variance for repeated measures for normally distributed data (heart rate and systolic blood pressure). Area under the curve for the systolic blood pressure and heart rate readings over time was calculated and compared between the two groups. To compare the dynamic and static VAS values adjusted for the effect of time and morphine dose, we used a generalized estimating equation model that included the main effect of the group, time, and total morphine dose. The Bonferroni test was used to adjust for multiple testing. A P-value less than 0.05 was considered statistically significant.