Patients and study design
This prospective open-label randomised controlled study adheres to CONSORT guideline. This study was approved by the Ethics and Research Committee of Universitas Indonesia (0211/UN2.F1/ETIK/2018) and was retrospectively registered in ClinicalTrials.gov (NCT03520205) on May 9th 2018. Written informed consents were obtained from all adult patients with ASA I–II who underwent laparoscopic living donor nephrectomy in Cipto Mangunkusumo Hospital, DKI Jakarta. The inclusion criteria included age 18–60 year with BMI < 30 kg/m2. The exclusion criteria included inability to communicate, history of allergy or contraindication to local anesthetics, contraindication to epidural or QLB (coagulopathy or infections on the injection site). The research protocol was explained to patient who met the inclusion criteria. All patients were educated about QLB and continuous epidural procedures, how to describe the degree of pain with NRS, and how to use the PCA morphine pump when the pain level (NRS) ≥ 4 after the surgery. After obtaining a written approval, the patients were randomised either into the QLB group (intervention group) or the epidural group (control group). Randomisation was conducted in block sizes of 4 using a computerised randomisation sequence by independent research assistants. Randomisation allocation number for each subject was written on paper and put in a closed envelope. The envelope was opened by the anesthesiologist who was appointed to perform epidural or QLB for this study. An independent pain team collected the data, statisticians and study investigators were blinded throughout the study.
Technique for Epidural, Quadratus Lumborum Block and Pain Management
Standard monitoring was placed such as electro cardiogram, oxygen saturation, non-invasive blood pressure monitors and cardiometry ICON™. General anesthesia induction was conducted with administration of fentanyl 2 µg/kg as co-induction, proceeded with propofol 1–2 mg/kg. Endotracheal tube intubation was facilitated with atracurium 0.5 mg/kg. The ventilator was set to volume control with PEEP 5 cmH2O and FiO2 30–50%, breathing frequency was adjusted with ETCO2 target of 35–45 mmHg. The anaesthesia was maintained using sevoflurane with 2.5% MAC, oxygen compressed air ratio of 40:60, atracurium 0.5 mg/kg/hour, with bispectral index target of 40–60, 0.8 for fraction of inspired sevoflurane, and train of four ratios £ 25%. All subjects received fentanyl 1 µg/kg i.v. if there is an increase of change in systolic blood pressure or pulse rate > 20% from the initial value during surgery and the total intraoperative fentanyl usage was recorded. Ephedrine was given if there is a decrease in mean arterial pressure (MAP) less than 65 during the study observation.
Patients in the epidural group had an epidural catheter placement procedure in the left lateral decubitus position after intubation under general anesthesia. After ensuring skin asepsis and draping the area with sterile cover, an 18G Tuohy epidural needle was inserted in vertebral interspace T10–11 and catheter was advanced 5 cm length within the epidural space.6 We observed vacuum epidural catheter aspiration and a test dose of 3 ml bupivacaine 0.25% with adrenaline 1:200,000 without any change in pulse rate or blood pressure to confirm the position of catheter within the epidural space. Then a continuous epidural infusion of bupivacaine 0.25% 6 ml/h was maintained for intraoperative analgesia. After completion of surgery, the continuous epidural dosage was decreased into bupivacaine 0.125% 6 ml/h for 24 hours postoperatively.
All patients in the QLB group received bilateral ultrasound-guided transmuscular QLB (anterior QLB or QLB3) performed by two anaesthetist consultant (DA, P) after induction and intubation under general anaesthesia. Patients were in supine position with the site to be blocked slightly facing upward by pillow underneath it and table tilting. After ensuring skin asepsis of the area, a 2.0–5.5 MHz convex transducer (C5-1E, DC-70, Mindray, Shenzen China) covered with sterile drapes attached to the inferior lumbar (Petit's) triangle that consisted of iliac crest in the inferior region, the latissimus dorsi muscle in the posterior region and external abdominal oblique muscle in the anterior region). The Shamrock sign appeared on the ultrasound, then a 21G 100-mm peripheral block needle (Stimuplex®, BBraun, Mesulngen Germany) was inserted in-plane with ultrasound probe passing in anterior to posterior direction through the QL muscle and reached the border between the QL and psoas major muscle. After confirming negative for blood aspiration, 1 ml normal saline was injected to obtain hydrodissection sign for verifying the needle tip, then a volume of 0.3–0.4 ml/kg bupivacaine 0.25% with a maximum of 25 ml was injected on each side (Figure 1). After completion of surgery, the bilateral QLB procedures were repeated with the same regimen as described above.
Neostigmine 0.04–0.07 mg/kg was given to reverse residual neuromuscular block and patient was extubated when had reached train of ratio ratio of 0.9–1.0. In the recovery room, patient-controlled morphine i.v. was administered using a portable programmed pump (PerfusorÒ Space PCA Infusion Pump System, BBraun, Germany). The PCA setting was 1 mg bolus, lockout time 10 minutes and maximum dosage 6 mg, without basal opioid infusion. All patients were informed how to use the PCA; first at preoperative anaesthesia visit and once again at the recovery room after the surgery. Ondansetron 4 mg i.v. and omeprazole, setuju 40 mg i.v. were administered every 12 hours to prevent postoperative nausea and vomiting (PONV).
Study parameters and statistical analysis
The aim of this study was to compare the analgesic efficacy of QLB with continuous epidural analgesia on postoperative morphine requirement and pain scores with numerical rating scale (NRS). MAP, heart rate, and cardiometry cardiac index were recorded at baseline, during surgery, and 24 hours after anaesthesia recovery. The primary outcome was cumulative morphine requirement in 24 hours after surgery. The secondary outcomes were pain scores at rest and movement, and motoric block at time points 2, 6, 12 and 24 hours after anaesthesia recovery. Pain scores were assessed using NRS (0 = no pain; 1–3 = mild pain, 4–6 = moderate pain, 7–10 = severe pain). Motoric block was evaluated using Bromage score (grade 1 = free movement of leg and feet; 2 = just able to flex knees with free movement of the feet; 3 = unable to flex knees, free movement of feet; 4 = unable to move the legs or feet). Sensoric block was assessed after anaesthesia recovery using pinprick and cold loss sensation with ice and alcohol. Time to first morphine requirement, urinary catheterisation duration, PONV, and paresthesia that occur within 24 hours postoperatively were recorded. Those outcomes were documented by the acute pain service team and ward nurses.
Sample size calculation was based on primary hypothesis that opioid requirement in 24 hours after surgery. Based on previous study, the mean (SD) cumulative morphine requirement 24 hours after surgery was 4 mg (SD = 4.723).7 From a preliminary study, the mean (SD) cumulative morphine requirement 24 hours after surgery was 4 mg (SD = 6.09). The reduction of 20% in cumulative morphine requirement was considered as clinically relevant. A sample size of 28 patients in each group was determined with a statistical power of 0.8 and type-1 error of 0.05. This study recruited 62 patients to allow 10% dropouts.
All statistical studies were analysed using Statistical Package for the Social Sciences (SPSS) version 20 (IBM Corp, Armonk, NY). Differences between numerical variable were analysed using the unpaired t-test for normal distribution data and Mann-Whitney test for abnormal distribution data. Differences between categorical variable were analysed using Chi square test. Numerical data with normal distribution was displayed in a mean (standard deviation), abnormal distribution was displayed in the median (interquartile range) values, and as percentage for categorical variables. The analysis was statistically significant if the p-value was less than 0.05.