This multi-centre double-blinded randomized-controlled trial was conducted at two teaching hospitals in Adelaide, Australia; Noarlunga Health Services (NHS) and Flinders Medical Centre (FMC). Institutional ethics approval was obtained (SALHN/HREC/292.20) and written informed consent was acquired from all participants. The trial was registered prior to commencement (NTR; NL9147; principal investigator: D-Y.L; date of registration: 25th of December 2020, URL: https://www.trialregister.nl/trial/9147). This study conforms to the Consolidated Standards of Reporting Trials (CONSORT) and the CONSORT extension for trials reporting patient-related outcomes [17, 18]. The study ran from June 28 to November 8 2021.
The inclusion criteria were adult patients presenting for primary elective THA under spinal anesthesia, without contraindications for regional analgesia, who were able to provide informed consent and reliably report symptoms to the research team. Exclusion criteria were an inability to provide first party consent (e.g. due to cognitive impairment or language barrier) and contraindications for or patient refusal of spinal anesthesia and/or regional analgesia.
Randomization, blinding and study intervention
Patients were randomized to either PENG block (intervention) or sham block (control). Randomization was performed by the principal investigator only via an online randomization computer generator (www.sealedenvelope.com) on a 1:1 basis. Members of the surgical team, members of the Acute Pain Service (APS), nursing staff and patients were all blinded to the intervention. To ensure blinding, the anaesthesiologist performing the preoperative block was different from the anaesthesiologist managing the patient intraoperatively and conducting the postoperative assessments.
Following the administration of spinal anaesthesia, the allocated block was placed using ultrasound guidance with a curvilinear probe (2.5-5 MHz).
PENG: 20 mL of ropivacaine 0.5% (100 mg) prepared by the anaesthesiologist performing the block was used. The area was aseptically prepped and draped. The curvilinear probe was placed transversely, medial to the anterior inferior iliac spine with the medial end of the probe rotated in a caudad direction to align to the superior pubic ramus. A 100 mm sonoplex needle was inserted in-plane under ultrasound guidance. 20mLs of local anaesthetic was injected as a plane block between the psoas fascia and superior pubic rami.
This block was simulated by the anaesthesiologist by prepping, scanning and draping as per PENG block protocol. The probe and a blunt needle, with a 20 mL syringe filled with saline attached, were held against the skin similar to the PENG block and a sufficient pause to simulate the block being performed was conducted, without actual administration of any medicine.
Following placement of either block, a small cross was drawn with a surgical marker to cover the puncture site or absence thereof.
The study was designed to represent daily practice and to achieve high external validity. Anaesthetic technique was standardized to a spinal anaesthesia with 0.5% Isobaric bupivacaine (range 10-14 mg) without use of intrathecal opioids. A single 8 mg intravenous dose of dexamethasone was administered at the time of the block. Surgical technique was performed at the discretion of the treating orthopaedic surgeon, including routine use of LIA in all patients at a dose of 100 mL of 0.1% ropivacaine with 1 mg epinephrine. Postoperative analgesia regime was standardized with round-the-clock acetaminophen and NSAIDs if no contraindication, and if needed tramadol, oxycodone, and/or fentanyl on a nurse administered basis.
The rationale for using isobaric bupivacaine is to reflect usual practice at our institution, where the longer duration is suited to the surgery .
Preoperatively, individual patient pain experience was evaluated using the Pain Catastrophizing Scale . Pain scores were obtained preoperatively (baseline), 3-h postoperatively in the Recovery Unit (Day 0), and on postoperative Day 1 (16 to 22 h postoperatively, standardized), marking the maximum pain score during active movement (quadriceps muscle strength test) at each time point. Pain scores were recorded using a numeric rating scale (NRS) ranging from 0 (absence of pain) to 10 (worst pain imaginable) and grouped as no (NRS 0), mild (NRS 1–4), moderate (NRS 5–7) or severe pain (NRS 8–10).
Perioperative opiate doses were recorded preoperatively, intraoperatively, on day 0 and each postoperative day for three days with quantities converted to oral morphine equivalents. Chronic opioid use and chronic preoperative pain were defined as daily opioid use or pain interfering with activities of daily living for a duration of greater than three months.
Mobilization: Postoperatively at Day 0 once the spinal had recessed, and Day 1, a blinded anaesthesiologist assessed quadriceps muscle strength using the Oxford muscle strength grading with grouping of results into intact (5/5), reduced (1–4/5) and absent (0/5). If a patient reported reduced or absent quadriceps muscle strength, the test was carried out on the non-operative side to ensure it was not due to residual spinal effect. Day 0 measurements of dynamic pain and quadriceps strength were standardised to three hours from end time of surgery. A Timed Up-and-Go test was conducted preoperatively and on Day 1 postoperatively by physiotherapists. In this test, the patient starts in a seat at standard height, stands, walks ten feet, turns around, walks back, and sits back down .
Patient-reported outcome measures (PROMs)
Baseline preoperative anxiety and depression were noted using the validated Patient-reported outcomes measurement information system (PROMIS) anxiety and depression item banks . These PROMs, along with the Pain Catastrophizing Scale, assess factors that have previously shown to influence pain experience and function . Preoperatively and on Day 1, quality of recovery was evaluated using the Quality of Recovery (QoR-15) questionnaire .
The APS assessed patient satisfaction and pain management on Day 1 in a blinded fashion. Pain scores as a maximum on movement, quadriceps muscle strength, patient satisfaction and PROMs were collected using a scripted format. Complications throughout hospital admission, according to Clavien-Dindo classification grade, time to first mobilization and time to discharge were also recorded . First mobilisation was accompanied by physiotherapy and assessment for suitability was twice a day.
The primary outcome was the NRS pain score at Day 0. Secondary outcomes were: NRS pain score (at Day 1), Day 0 and 1 quadriceps muscle strength, perioperative opiate use, postoperative complications, length of hospital stay, patient satisfaction and PROMs.
Sample size calculation and statistical analyses
A priori power calculation was carried out using PASS 14 Power Analysis and Sample Size Software (Kaysville, Utah, USA) based on pain scores from a pilot study and a previous PENG randomized-controlled trial. This showed a mean pain score of 4 out of 10 points after THA on Day 0 without placement of PENG block. This was reduced to a score of 2 out of 10 points with placement of PENG block, with a standard deviation (SD) of 2 [13, 14]. A two-tailed independent-samples t-test for the difference between the two unpaired means with an alpha-error of 0.05 and power of 0.80 showed that 18 patients in each arm were required to detect a difference, 36 total. Given the high attrition rate in the pilot study, we accounted for a 40% dropout which brought numbers to 26. This was rounded up to 30.
Data collection and entry, and statistical analyses were conducted in a blinded fashion. The analysis was performed on an intention-to-treat basis using SPSS version 27 (IBM Corp., Armonk, NY, USA) and GraphPad Prism version 9 (GraphPad Software, La Jolla, Calif, USA). Parametricity of continuous variables was determined using the Shapiro–Wilk test. Normally distributed continuous variables are expressed as mean (SD), and nonparametric variables as median (range). Univariate analysis was carried out using the chi2 test or Fisher’s exact test (for n < 10) for categorical variables, and the Mann–Whitney U-test for nonparametric continuous variables or the Student’s t-test for parametric continuous variables. A p-value of < 0.05 was considered statistically significant.