After receiving the institutional Ethics Committee’s approval (Ref: IR.IUMS.REC.1395.27278) and obtaining informed written consent, 60 parturient patients candidates for undergoing caesarean section in a university hospital were assigned in a randomized double-blind clinical trial. The sample size was estimated by:

Equation 1.
(d = µ1 - µ2 = 1.2), P = 90, α = 0.05
Parturients undergoing elective caesarean delivery under spinal anesthesia were recruited between January and June 2016 by double-blinded block randomization.
The study was registered in an international database (Ref: IRCT201601147984N24). The inclusion criteria comprised full-term pregnancy, the age of 18 - 38 years, ASA physical status I - II, elective caesarean delivery, primary or repeat caesarean section, Pfannenstiel incision, and being under spinal anesthesia. The exclusion criteria consisted of drug abuse, bleeding disorders, severe mental disorders, history of allergy to study drugs, gastrointestinal disease, obesity (BMI above 35), conversion to another method of anesthesia, pre-eclampsia and complications during the surgery.
Spinal anesthesia was established using hyperbaric bupivacaine (2.5 mL bupivacaine 0.5%, AstraZeneca, France). On arrival to the recovery room, parturients were randomly allocated to one of the two groups using a random number table. Participants and anesthesiologist performing pain assessments were blinded to group allocation. For post-operative pain management, an i.v. patient-controlled analgesia (PCA) device (Autofuser, ACE Medical Co., South Korea) was used for all patients in both groups. In the DP group, 3 µg kg-1 of dexmedetomidine (Precedex, Hospira Inc., USA) was added to 35 mg kg-1 of paracetamol (Apotel, Cobel Darou, Iran) up to 2 g, and in the DK group the same dose of dexmedetomidine was added to 1 mg kg-1 of ketorolac (Ketorolac, Exir, Iran). The PCA device was set to deliver a continuous infusion rate of 4 mL hr-1.
Patient assessment was done by a physician not aware of the PCA drugs at rest, 6th, 12th, and 24th hour after surgery. Pain score was obtained using a visual analog scale (VAS), (0 = no pain and 100 = worst pain imaginable). Ramsay sedation score (0- restless, 1- tranquil, 2- sleepy, 3- confused but responsive to verbal commands, 4- unresponsive to verbal commands, and 5- no response to painful stimuli), hemodynamic changes (blood pressure and heart rate), complications, patient’s characteristics, and satisfaction rate (exceeded expectation, matched expectation, and less expectation) were recorded.
When the pain score was greater than 30, meperidine (25 mg) was i.v. administered. Complications such as blood pressure and heart rate changes, nausea, vomiting, respiratory depression, bleeding, and dizziness were evaluated and treated, if identified.
The data collected were analyzed using the SPSS version 18 software. Employing the Kolmogorov-Smirnov test, the data were evaluated for normal distribution and accordingly, the Wilcoxon test was used for non-normal distribution data, t-test for normal distribution data, and the Fisher’s Exact test for variables with absolute values like the presence of special symptoms. Differences between the two groups were analyzed applying the Mann-Whitney test; the Wilcoxon test was used for intra-group analysis statistical comparisons with a Bonferroni correction. The Friedman test was employed to analyze the differences between pain assessment hours in the two groups. The qualitative data analysis was performed using the Chi-Square test and P values of < 0.05 were considered significant.