Study design
We have conducted a parallel, double-blind, superiority-randomized controlled trial study at Dilla University Referal Hospital that follows the ethical standards stated in the Declaration of Helsinki and the consolidated standards of reporting Trials (CONSORT) reporting guidelines (23). The participants were randomly assigned to either bupivacaine or a mixture of bupivacaine with tramadol group. The primary outcome of the study was the pain severity between the two groups. The secondary outcome was the overall postoperative analgesia consumption (24hrs.) between the two groups, the time to the first analgesic demand in minutes between the two groups, the incidence of postoperative nausea and vomiting between the two groups, and the incidence of intraoperative shivering between the two groups. The trial registration number is PACTR202310525672884 (13/10/2023).
Definitions of the outcome
Numeric Rating Scale(NRS)
is a reliable method for measuring pain intensity that entails requesting a patient to rate her severity of pain on an eleven-point scale from 0 to 10, where 0 represents no pain and 10 represents the greatest possible pain.
Total analgesic consumption
The overall amount of analgesic drugs in milligrams(mg) consumed in the first twenty-four hours following the surgery.
VAS(Visual analog Scale)
This is a pain rating scale; the scores are derived from patient-reported valuations of symptoms that are recorded at 1 point along a 10 cm line that shows a range between the 2 edges of the scale, which are "no pain" and "worst pain."
Time to first analgesic request
is the time, measured in minutes, between the time the end of surgery and the patient needs analgesics.
Shivering
A trembling of the body that is brought on by an involuntary contraction of the muscles.
Study participants
The study was conducted in Dilla University Referal Hospital from Jan 01, 2023, to August 01, 2023, Pregnant mothers between 15 and 49 undergoing spinal anesthesia were included in the study. We excluded parturients using alcohol, parturients diagnosed with chronic pain, pregnant mothers with systemic diseases like renal impairment, chronic hepatic disease, respiratory disease, and heart disease, morbidly obese patients (BMI > 40kg/m2), spinal anesthesia with adjuvants, operation other than the lower uterine segment incision, pregnant mothers taking opioids preoperatively.
Randomization, blinding, and allocation concealment
The parturients were randomly recruited to take bupivacaine or a combination of bupivacaine with tramadol after fulfilling the inclusion criteria. The process of simple randomization involved selecting a level from a closed envelope having either B or BT, where B represents the bupivacaine group and BT denotes a combination of the bupivacaine with tramadol groups. A 1:1 ratio of a randomly generated number by the computer was employed. The patients in the study, as well as the data collector, were blinded to the medication group that the patients received and the anesthetist in charge who was not part of the study prepared the medications. A sealed, sequentially labeled envelope was used to verify allocation concealment. Before assigning and scheduling the operation, the process was explained in detail to the parturients.
Sample size calculation
Based on a prior Indian study, which demonstrated that wound infiltration with a combination of tramadol and bupivacaine extends the painless period, the sample size was determined by selecting analgesic consumption (N1 = 30, µ1 = 386.17, SD1 = 233.84, for a group consisting of bupivacaine and tramadol, and N2 = 30, µ2 = 192.50, SD2 = 134.77 for bupivacaine alone) (24). A further 10% was added to the enrollment to account for possible dropout, assuming the balanced design, by utilizing alpha 0.05, effect size 0.4525, power 90%, and sample size determined using power analysis with G power software version 3.1.9.7 to be 53. With 30 people in both groups, the sample size increased to 60 in total.
Anesthetic management and surgery
During the preoperative period
The night before the procedure, a pre-anesthetic assessment was completed, and informed consent was obtained following a thorough explanation to the parturient who was scheduled for an elective cesarean section regarding the advantages and risks of both surgery and anesthesia. For solid food, the patient fasted for 8 hours, and for clear fluid, for 2 hours. Before spinal anesthesia was administered on the morning of the procedure, parturients were given 200 mg IV cimetidine and 0.1 mg/kg IV dexamethasone 30 minutes beforehand.
Intra operative period
The non-invasive blood pressure, pulse-oximetry, and electrocardiography (ECG) monitors were attached and recorded. Before spinal anesthesia, 10 ml/kg of isotonic fluid was preloaded into the patients. After the patient was seated, the Tuffers line was marked, the area was cleansed using iodine and alcohol aseptic methods, and 3 milliliters of 2% lidocaine solution (60mg) were then injected into the patient's skin. A 25G cutting-type spinal needle was used midline approach for lumbar puncture while the patient was seated in the L3/4 position. At a rate of 0.2 ml/sec isobaric 0.5% bupivacaine, the subarachnoid block was formed with 0.07 mg/height, where height is measured in centimeters.
The parturient was positioned in a flat position straight away following spinal anesthesia, and to prevent aortocaval compression, the bed was tilted 15 degrees to the left, or a pillow was placed under the right hip. The block's level was measured using the Bromage scale for motor function, alcohol drops for autonomic function, and needle prick for sensory function. A surgical incision is allowed once the sensory level of T6 is attained and the Bromage scale is more than two. A modified Bromage scale, with 0 denoting no motor block, 1 representing incapacity to raise extended legs but still being able to move knees and feet, 2 denoting incapacity to raise extended legs and flex knees but still being able to move feet, and 3 denoting complete motor block of the lower limbs, was used to evaluate the motor block.
An incision was made once the motor and sensory blocks were fully established. Following spinal anesthesia, the patient's hemodynamics (SPO2, HR, ECG, and BP) were monitored right away. Every ten minutes, measurements and records of the patient's hemodynamics (SPO2, HR, ECG, and BP) were made and the grade of intraoperative shivering was recorded. The moment the infant was delivered, Uterotonic medications such as misoprostol (400–600µg sublingual) or oxytocin (10 IU IM; infusion 0.04–0.125 IU/min) were administered (25). Blood loss, blood transfusions, intraoperative fluid requirements, and complications such as bradycardia, hypotension, nausea, and vomiting were all documented.
When systolic blood pressure(SBP) falls greater than 20% from the starting point, it is referred to as hypotension. Intravenous fluid is used to treat this condition; if it persists after fluid administration, a vasopressor (adrenaline 0.01µg/kg) is used; and atropine (0.02µg/kg) is used to treat heart rates that are less than 60 beats per minute. The shivering refers to an uncontrolled shaking of the body caused by a muscular contraction. The method used to grade shivering was the Bedside Shivering Assessment Score (BSAS). A score of 0 indicated no shivering; a score of 1 indicates no shivering at all but modest fasciculation of the face, neck, or peripheral vasoconstriction; a score of 2 indicates obvious muscle activity in a single muscle group; a score of 3 indicates more than two but not generalized muscular activity; and a score of 4 indicates generalized muscular activity (26).
Patients received IV tramadol 0.5 mg/kg if they had a shivering grade of three or above (27). Failure of a spinal anesthetic attempt to produce a block was declared when the level of the block was not strong enough for the planned procedure. The three elements of the block were assessed to declare the failed spinal anesthesia; these were duration, quality, or extent of the local anesthetic action (28). A full dose of local anesthetic (0.07 mg/height) was used and the spinal anesthesia was repeated when there was a total failure, meaning there was no sensory or motor block at all. In the event of partial failure due to insufficient level, patchy, or unilateral block, the spinal anesthesia was repeated with a 25–30% dose reduction. General anesthesia was used when all these attempts failed (29).
Depending on the patient's needs, oxygen was given via a nasal prong at a flow rate of two to three liters per minute both during and after surgery. Following the surgery, the last vital sign was noted, and the wound site was infiltrated with 0.7 mg/kg of 0.25% bupivacaine diluted to 20 ml for group B. The BT group received a mixture of 0.7 mg/kg of 0.25% bupivacaine and 2 mg/kg of tramadol diluted to 20 milliliters by the obstetrician (24). Following the recording of vital signs, the patient was moved to the PACU, where the level of pain was measured using an NRS score. This assessment was made available to both the obstetric ward and the PACU for a maximum of 24 hours.
Postoperative period
While the patient was recovering, their vital signs—BP, HR, and SPO2—were continuously followed and recorded. The pregnant mother was transferred to an obstetric ward once all vital signs had stabilized. A 5-point rating system was used to assess postoperative nausea and vomiting, and protocol-compliant care was given to any problems. Using an NRS scale for 24 hours at T2, T6, T12, T18, and T24 hours—where T is the time—the severity of postoperative pain was assessed.
Intervention
The parturients undergoing cesarean section under spinal anesthesia were allocated to Bupivacaine (B) and Bupivacaine and tramadol group (BT). Parturients in Group B received 0.07mg/kg of 0.25% bupivacaine from the obstetrician at the end of the procedure. Group BT received 0.07mg/kg of 0.25% bupivacaine with 2mg tramadol, both of the drugs were diluted into 20ml by using 0.9% normal saline.
Data collection
Data from parturients and clinical charts of individuals who had cesarean sections under intrathecal anesthesia were gathered for the study using standardized questionnaires. The goal of the study, its risks, and its benefits were explained to the participants in Amharic after being prepared in English. The NRS was used to evaluate the severity of pain at five different time points: T2, T6, T12, T18, and T24, where T stands for time. The obtained information was accurate, comprehensive, and understandable, according to the lead investigator. Among the clinical outcomes evaluated were the incidence of PONV, the severity of postoperative pain, the occurrence of shivering, and the first analgesic required time.
Outcomes
The primary outcome was the severity of pain at 2hr, 6hr, 12hr, 18hr, and 24 hours. The secondary outcomes were total analgesia consumption, first analgesia request time, incidence of shivering, and incidence of postoperative nausea and vomiting.
Statistical analysis
The data has been entered into the Epi-data program version 4.6 following its extraction from the data collector and manual inspection, cleaning, and error correction. Once the data entry process was finished, statistical analysis was carried out by exporting the data into SPSS version 26. The Shapiro-Wilk test and histogram were used to verify the data distribution through statistical testing and visual inspection, and Levene's test was used to determine the homogeneity of variance. The plot of Box and Whisker was used to check outliers. Even if they could not be avoided, the outliers were handled using the median value.
For normally distributed variables, analytical statistics were calculated using independent samples. When analyzing data that was not regularly distributed, the T-test was utilized to compare group means; for categorical variables, the chi-square or Fisher's exact test was employed; and for non-normally distributed data, the Mann-Whitney U test. Welch's t-test was performed if Levene's test for equality of variances showed that the homogeneity of variances assumption was violated. If the data were not normally distributed, a visual inspection was performed to assess the distribution of values for both groups. Boxplot analysis and studentized residuals analysis were performed for repeated measures to assess outliers for values ± 3. We assessed covariance using the Box's M test.
When the two-way interaction's sphericity assumption was broken, the Greenhouse-Geisser adjustment was used. Mauchly's test was employed to verify sphericity. To display the differences at each time point, pairwise comparisons were performed. However, because of the strong two-way interaction between group and time, this is only of marginal interest. We achieve this by decomposing the significant interaction into simple effects, comparing the results analytically, and deciding how to handle family-wise error using the adjusted Bonferroni alpha = 0.025 as a point of reference. If there is a significant interaction between group and time, we use a split file to perform the repeated measure analysis (the simple effects) separately for tramadol and bupivacaine. Instead, a straightforward comparison is made between every level of the repeated measurement element and a baseline vital sign. P values less than 0.05 are considered statistically significant. A uniformly distributed set of numbers was shown as mean ± SD. The non-normally distributed numerical data were shown using the median and interquartile range, and the categorical data were represented using percentages (%).
The 2nd, 6th, 12th, 18th, and 24th-hour pain severity was analyzed using a mixed linear model and a Generalized Estimating Equation (GEE). A General Estimating Equation (GEE) was run to compare the difference in pain severity with a numerical rating scale (NRS) within the study groups (Bupivacaine versus a comparison of Bupivacaine and tramadol). The difference was expressed with statistical significance, odds ratio, and confidence interval about the first hour. A mixed linear model was run to analyze the continuous repeated measurements (visual analog scale) to investigate the mean pain severity score between groups across time points during follow-up. We followed the modified intention-to-treat analysis principle; only subjects who were randomized and who received all the interventions were included in the final analysis.