Anterior Quadratus Lumborum Block Does Not Provide Superior Pain Control after Hip Arthroscopy: A Double-blinded Randomized Controlled Trial

This randomized double-blinded trial received institutional review board approval, was registered at ClinicalTrails.gov (NCT03432650; principal investigator: Stephen C. Haskins, M.D.; February 12, 2018), and adheres to Consolidated Standards of Reporting Trials guidelines. The trial was conducted in accordance with the original protocol. Although no changes to methods and definitions were made during this trial, the clinical trial registration was updated after the completion of a previous trial to correct an original oversight. The full trial protocol can be obtained upon request. Written informed consent was obtained from all participants before surgery.

Patients 18 to 80 yr of age being treated with an ambulatory hip arthroscopy by a participating surgeon coinvestigator between June 2018 and December 2019 were eligible for this study. All patients were enrolled at a single site and were approached by the research study staff on the day of surgery. Exclusion criteria encompassed hepatic or renal insufficiency, age younger than 18 and older than 80 yr, allergy or intolerance to one of the study medications, chronic gabapentin/pregabalin use (regular use for longer than 3 months), chronic opioid use (daily opioids use for longer than 3 months), contraindication for spinal anesthesia, non-English speakers, and any arthroscopy procedures including lipogems, revisions, or repair of the gluteus medius. Study enrollment ceased when the target sample size was obtained.

Patients were randomized in a 1:1 ratio, via a computer-generated randomization schedule, to receive a single shot anterior quadratus lumborum block (30 ml 0.5% bupivacaine with 2 mg preservative-free dexamethasone) or no block. A statistician who was not involved in the analysis of the data prepared the randomization schedule. A research assistant provided an opaque card with the randomization written on it to the coinvestigator anesthesiologist on the day of surgery if a patient enrolled. The cards were prepared by research staff who had access to the randomization schedule but were not otherwise involved in the study. Upon patient arrival to the operating room, the investigator anesthesiologist assigned to the case opened a sequentially numbered sealed opaque envelope containing assignment. Coinvestigator anesthesiologists were unblinded and administered the block, but research assistants enrolling and collecting data were blinded. Patients were also blinded as the anterior quadratus lumborum block was performed under deep sedation.

All patients received either a spinal (4 ml mepivacaine) or combined spinal–epidural anesthetic (5 ml of 2% lidocaine) with deep IV sedation (up to 5 mg of midazolam, IV fentanyl, and IV propofol infusion titrated to effect). Intraoperative antiemetics consisted of IV ondansetron (4 mg) and IV dexamethasone (4 mg). Intraoperative analgesics were IV fentanyl (maximum 100 mcg), IV acetaminophen (maximum 1,000 mg), and IV ketorolac (maximum 30 mg). Up to 2 mg of IV hydromorphone hydrochloride was given at the anesthesiologist’s discretion at the end of the case.

Anterior quadratus lumborum block was performed on the ipsilateral surgical side under deep IV sedation to ensure adequate patient blinding, but before performing neuraxial technique using an X-Porte ultrasound system (FujiFilm SonoSite, USA) with a curvilinear transducer (5–2 MHz, C60xp) and a 22-gauge, 4-in needle (Chiba) in the lateral decubitus position using a sterile technique. The patient was placed in the lateral decubitus position with the operative side facing up. The transducer was placed in the transverse position immediately cephalad to the iliac crest at the level of the posterior axillary line. The needle was inserted in-plane in a posterolateral to anteromedial fashion until the needle tip transversed the quadratus lumborum muscle and pierced the fascia between the quadratus lumborum and the psoas major muscle. Thirty milliliters of 0.5% bupivacaine with 2 mg preservative-free dexamethasone was injected with an endpoint of separation of the quadratus lumborum and psoas major muscles (fig. 1). Experienced regional anesthesiologists performed all block procedures.

To assess intra-abdominal fluid extravasation, the operating room anesthesiologist, trained in the focused assessment with sonography for trauma examination, evaluated the patient preoperatively to assess for fluid within the abdomen or pelvis. Postoperatively, the same anesthesiologist repeated the examination, and patients with new fluid in either compartment were diagnosed to have intra-abdominal fluid extravasation. The amount of fluid present was graded as “small,” “moderate,” or “large.” To increase the ultrasound examination’s sensitivity, the patients were placed in the Trendelenburg position while evaluating the perihepatic and perisplenic space and reverse Trendelenburg to evaluate the pelvic compartment. An anesthesiologist blinded to the case reviewed images independently. In points of disagreement, the images were reviewed by a third independent expert to reach a consensus.

Before surgery and upon resolution of the neuraxial block in the PACU, a blinded research assistant independently assessed each patient for quadriceps weakness. The patient was placed supine with a cushion underneath the knee, resulting in a 45-degree angle at the knee. Quadriceps strength of both legs was assessed with a dynamometer placed on the anterior of the ankle, between the malleoli. Patients were instructed to extend their legs three times each, with a 30-s pause between each attempt. After each attempt, the strength was recorded, and patients rated their pain using a numerical rating scale. A blinded assessor determined analgesia in the L1 to L3 dermatomal distribution by touch and temperature discrimination using an alcohol swab.

Postoperative pain medications included oxycodone 5/10 mg with IV hydromorphone hydrochloride 0.5 mg every 5 min for breakthrough pain in the PACU. Patients were prescribed oxycodone/oxycodone with acetaminophen (unless contraindicated) and an oral nonsteroidal anti-inflammatory drug (naproxen or indomethacin).

The primary outcome, numerical rating scale pain scores at rest and movement, was assessed at 30 min and 1, 2, 3, and 24 h after PACU arrival. Secondary outcomes included the presence of intra-abdominal fluid extravasation after surgery, opioid use, the incidence of antiemetic use, the incidence of hospital admission, time to discharge from PACU, patient satisfaction with pain management, change in quadriceps strength, the incidence of urinary retention, and hypotension. Questionnaires were also administered assessing opioid-related symptoms and adverse effects using the opioid-related symptom distress scale during PACU stay, and on postoperative day 1.¹⁶  The opioid-related symptom distress scale uses 4-point Likert scales to evaluate the frequency, severity, and bothersomeness of 12 symptoms. The composite score ranges from 0 to 4. The Quality of Recovery 40 was also administered in the PACU and on postoperative day 1. The Quality of Recovery 40 is a 40-item questionnaire that assesses five dimensions of recovery after surgery and anesthesia—comfort, emotions, physical independence, patient support, and pain—and has a mean time to completion of 5 min.¹⁷  The self-administered Leeds assessment of neuropathic symptoms and signs of pain scale was performed 6 months after surgery if the patient had persistent hip pain. A post hoc subgroup analysis of the intra-abdominal fluid extravasation groups (anterior quadratus lumborum block vs. control) was performed to assess pain scores at rest and movement in the first 24 h. This analysis was performed post hoc because the incidence of intra-abdominal fluid extravasation is low, and patients could not be prospectively randomized to either have or not have intra-abdominal fluid extravasation. However, given that intra-abdominal fluid extravasation is an independent indicator of postoperative pain, the analysis was performed post hoc to evaluate for a difference.

The balance on demographics and surgical variables was compared by calculating standardized differences, where the difference in means or proportions was divided by the pooled SD. An imbalance was defined as a standardized difference with an absolute value greater than 1.96 × (2/48)^1/2 = 0.4.¹⁸  Continuous variables are summarized as mean ± SD or median with interquartile range. Categorical variables are summarized as counts and percentages. All analyses were performed on an intention-to-treat basis.

The primary outcomes, numerical rating scale pain score at rest and numerical rating scale pain score with movement at 30 min and 1, 2, 3, and 24 h after PACU arrival, were compared between the quadratus lumborum block and no block groups using regression based on a generalized estimating equations approach, with treatment group, time (30 min and 1, 2, 3, and 24 h after PACU arrival), and baseline numerical rating scale pain score as fixed effects, and participant as a random effect. An unstructured correlation model was used. The adjusted marginal means of pain scores aggregated across all time points were reported separately by group, with contrast for the difference in groups and 95% CI. P value is adjusted using Bonferroni adjustment for two primary outcomes. A P value less than 0.025 was considered significant for the primary outcome. As secondary outcomes, we added a time times treatment interaction term to the generalized estimating equations model to examine numerical rating scale difference at each observed time. For other secondary outcomes, continuous secondary outcomes measured at a single time point were analyzed using independent t tests or Wilcoxon rank sum tests. Categorical secondary outcomes measured at a single time point were compared between groups using chi-square and Fisher exact tests. Secondary outcomes measured at multiple time points per patient were analyzed using the generalized estimating equations method with time and treatment interactions. Effect sizes for continuous and binary secondary outcomes are presented as differences in means, and odds ratios, respectively, with corresponding 95% CI and unadjusted P values. A P value less than 0.05 in secondary outcomes should be interpreted as suggestive.

All statistical hypothesis tests were two-sided. Statistical analyses were performed with SAS Version 9.4 (SAS Institute, USA).

Previous literature found the mean ± SD for numerical rating scale pain at rest and with movement 24 h after surgery to be 3.3 ± 3.0 and 5.5 ± 3.0, respectively.²  We determined that a sample size of 40 patients per group would provide 80% power at a two-sided alpha level of 0.025 to detect a 1.3-point difference in numerical rating scale pain score at rest or numerical rating scale pain score with movement 24 h after surgery between the anterior quadratus lumborum block and no anterior quadratus lumborum block groups.¹⁹  We planned to enroll a total of 96 patients to account for attrition and protocol violations (20%).

Ninety-six patients scheduled for a hip arthroscopy were randomized to receive either an anterior quadratus lumborum block or no block. All patients enrolled were included in the analysis (fig. 2). Baseline demographics, preoperative pain scores, and intraoperative variables between the two groups were similar (table 1).

For the primary outcome, quadratus lumborum block with multimodal analgesia was not superior to multimodal analgesia alone. There was no overall treatment difference in pain scores between the two groups over the study time. For secondary outcomes looking at numerical rating scale difference at each observed study time, the numerical rating scale at rest was only significantly lower in the anterior quadratus lumborum block group (3.3 ± 3.4) compared to the control group (4.4 ± 3.3; differences in means, 1.3 [95% CI, 0.2 to 2.4]; P = 0.023) at 30 min after PACU arrival. Of note, the correlation between delivery of intraoperative IV hydromorphone hydrochloride and numerical rating scale pain score at 30 min was weak (0.07 for numerical rating scale rest and 0.14 for numerical rating scale move). There was no significant difference between the anterior quadratus lumborum block versus no block groups for pain scores at rest or with movement at any other time point (table 2).

Intra-abdominal fluid extravasation was present in both groups, 18.8% in the anterior quadratus lumborum block group and 22.9% in the no block group. There was no difference in subgroup numerical rating scale pain scores for patients with intra-abdominal fluid extravasation (table 3).

Opioid consumption for 0 to 24 h, PACU antiemetic use, and PACU patient satisfaction were also not significantly different (table 4). There was no difference between the opioid-related symptom distress scale and the Quality of Recovery 40 scores in the PACU and postoperative day 1. Regarding the anterior quadratus lumborum block mitigating side effects of hypotension and urinary retention, there was no statistical difference between the anterior quadratus lumborum block group and the no block group. There was no difference in the incidence of nausea and vomiting between the groups both in the PACU and on postoperative day 1 (table 4). There was no difference in the change of quadriceps strength on the operative side between groups (table 4; differences in means, 1.9 [95% CI, –1.5 to 5.3]; P = 0.268). At the 6-month follow-up, there was no difference in the self-administered Leeds assessment of neuropathic symptoms and signs of pain scale scores between the groups.

This prospective double-blinded randomized controlled trial assessed the clinically analgesic effect of anterior quadratus lumborum block with multimodal analgesia compared to multimodal analgesia alone in hip arthroscopy. The only benefit in the anterior quadratus lumborum block intervention was at the 30-min point after PACU arrival. However, while statistically significant, a 1.1 difference in pain score may not represent a clinically meaningful difference. There was no difference found in secondary outcomes regarding opioid use, nausea/vomiting, antiemetic use, the incidence of hospital admission, time to discharge from PACU, patient satisfaction with pain management, change in quadriceps strength, or incidence of urinary retention and hypotension. In sum, these findings suggest that anterior quadratus lumborum block, in addition to multimodal analgesia, may have limited value in the setting of hip arthroscopy.

This study demonstrated that the anterior quadratus lumborum block did not cause a significant motor deficit despite previous reports suggesting up to 90% of quadriceps weakness after the anterior quadratus lumborum block.²⁰  A retrospective study by Ueshima and Hiroshi reported that 65 out of 81 patients receiving anterior quadratus lumborum block experienced quadriceps weakness after performing bilateral anterior quadratus lumborum block with a total of 40 cc of 0.375% levobupivacaine.²⁰  Quadriceps weakness caused by blockade of the lumbar plexus can impede early mobilization and physical therapy and cause falls, all of which delay PACU discharge.²  Despite the anterior quadratus lumborum block’s proximity to the lumbar plexus, the local anesthetic is injected posterior to the psoas major muscle. In contrast, the lumbar plexus is encased within the psoas major muscle. The cadaveric study by Dam et al. demonstrated that out of 10 anterior quadratus lumborum block injections, no dye surrounded the lumbar plexus block, suggesting that the lumbar plexus is not affected.²¹  Changes in quadriceps strength were common in both of our groups, likely due to pain or surgical swelling limiting quadricep firing. Without a control group, it is possible that other retrospective reports misinterpreted standard surgery-related postoperative weakness as the direct result of the anterior quadratus lumborum block. Finally, although our study was not powered to assess this outcome, the addition of anterior quadratus lumborum block to multimodal analgesia did not provide superior pain control for patients with intra-abdominal fluid extravasation.

Contrary to the retrospective study by McCrum et al.,⁸  our analysis suggests that for hip arthroscopy, the addition of anterior quadratus lumborum block to multimodal analgesia is not superior to multimodal analgesia alone for postoperative pain control, reducing opioid or antiemetic consumption, and patient satisfaction. Similar to the study by Brixel et al. evaluating posterior quadratus lumborum block for elective total hip arthroplasty, we also found that this block is not a good clinically analgesic option.⁷  These findings are meaningful as the anterior quadratus lumborum block is relatively new. There are many hypotheses regarding the surgical procedures that might benefit from the performance of this block. However, the lack of treatment effect with performing anterior quadratus lumborum block for hip arthroscopy in addition to multimodal analgesia helps to identify the surgical procedures that do not benefit clinically from these novel blocks, particularly given that many of these blocks are deep, challenging to perform, and associated with potential complications. Additionally, the publication of prospective randomized studies with a lack of treatment effect ensures increased accuracy of future meta-analyses investigating the block’s clinical analgesic effect.

The limitations of this study include the single-institutional nature, which limits external validity. We did not control for the type of hip arthroscopy performed. A sham block was not performed; however, we felt a sham block was not appropriate. The quadratus lumborum block is deep, and we did not want to risk potential violation of the peritoneum or kidney without the benefit of the postoperative analgesia. Further, we excluded patients with chronic pain or opioid dependence, who may have benefitted from additional analgesia from a regional technique.

There may be an analgesic role for the quadratus lumborum block as a rescue block for patients in severe postoperative pain. The quadratus lumborum block is performed far from the surgical site and is motor sparing; therefore, it can be performed easily without removing the surgical dressing. Also, the postprocedural analgesia might decrease the need for hospital admission for pain control while simultaneously enabling ambulation, unlike the lumbar plexus block. Additionally, patients with a history of chronic pain or high opioid dependence may benefit, no matter how marginally, from any supplemental analgesic to decrease the need for high-dose opioids postoperatively. However, despite these potential indications, our findings conclude that it does not seem prudent to routinely perform the anterior quadratus lumborum block on patients undergoing hip arthroscopy if multimodal analgesia is provided in the perioperative setting.

The authors acknowledge George Birch, B.S., and Nicole Brunetti, B.S., Department of Anesthesiology, Critical Care, and Pain Management at Hospital for Special Surgery, New York, New York, for serving as backup research assistants who played a significant role in data collection. The authors thank Patricia Pang, M.D., for support with study design, Kara Fields, M.S., for statistical design, and David H. Kim, M.D., for help with recruitment (all from the Department of Anesthesiology, Critical Care, and Pain Management, Hospital for Special Surgery). In-kind support was provided by FujiFilm SonoSite (USA) in the form of a dedicated study probe (curvilinear transducer, 5–2 MHz, C60xp).

This study was funded by East River Medical Associates, New York, New York, and the Hospital for Special Surgery Anesthesiology Department Research and Education Fund, New York, New York. REDCap (powered by Vanderbilt University, Nashville, Tennessee) use was supported by the National Center for Advancing Translational Science of the National Institutes of Health (Bethesda, Maryland; grant No. UL1TR000457).

Dr. Memtsoudis declares a financial relationship with Teikoku (San Jose, California), Sandoz (Princeton, New Jersey), HATH (Bedford Hills, New York), Parvizi Surgical Innovation (Philadelphia, Pennsylvania), Centauros Healthcare Analytics (Scottsdale, Arizona), and SGM Consulting LLC (Holmdel, New Jersey). Dr. Kelly declares a financial relationship with Arthrex (Naples, Florida), Smith & Nephew (Watford, United Kingdom), and Organicell (Miami, Florida). Dr. Ranawat declares a financial relationship with Enhatch (Hoboken, New Jersey) and Conformis (Billerica, Massachusetts); serves as a consultant for Stryker (Kalamazoo, Michigan), Smith & Nephew (London, United Kingdom), Anika (Bedford, Massachusetts), Bodycad (Quebec, Quebec, Canada), Xiros (Leeds, United Kingdom), NewClip (Haute-Goulaine, France), Ranfac (Avon, Massachusetts), and Marrow Cellulation (Avon, Massachusetts); and is on the speakers’ bureau for Pfizer (New York, New York). The other authors declare no competing interests.

Full protocol available at: haskinss@hss.edu. Raw data available at: haskinss@hss.edu.