Article Text
PDF
Abstract
Introduction Effective analgesia after total hip arthroplasty must minimize pain and optimize early ambulation. Lumbar plexus blocks (LPBs) provide analgesia but may cause motor weakness. Quadratus lumborum blocks (QLBs) may provide analgesia with preserved motor strength.
Methods This trial randomized subjects scheduled for elective hip arthroplasty to receive an LPB or lateral QLB for postoperative analgesia. The primary outcome was opioid consumption at 12-hour postoperative. Non-inferiority of lateral QLBs compared with LPBs was conducted using a one-sided two-sample t-test. Secondary outcomes included pain scores, cumulative opioid consumption, quadriceps strength, time to ambulation, and distance ambulated. Differences in pain scores and opioid consumption over time between groups were evaluated using a linear mixed model.
Results The trial consented and randomized 111 subjects and 103 completed the study: LPB (n=50) and lateral QLB (n=53). Mean (95% CI) cumulative opioid consumption (mg) at 12-hour postoperative was not found to be non-inferior in the lateral QLB (15.9 (12.7 to 19.2)) vs the LPB (12.7 (10.2 to 15.1)) group (p=0.625). Pain scores in postoperative anesthetic care unit (PACU) and 24-hour postoperative did not differ. The maximum distance ambulated did not differ, but lateral QLB patients were 2.4 times more likely to ambulate in the first 12 hours (p=0.024) and had significantly greater quadriceps strength in PACU (p<0.001).
Discussion Although we were unable to demonstrate non-inferiority for opioid consumption at 12-hour postoperative, strength and mobilization were improved in lateral QLB subjects.
Trial registration number NCT04402437.
- Nerve Block
- analgesia
- Pain, Postoperative
- Lower Extremity
- Acute Pain
Data availability statement
Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as online supplemental information.
Statistics from Altmetric.com
Key message
Quadratus lumborum blocks (QLBs) may improve analgesia after hip arthroplasty, but most studies have focused on the anterior approach. This study evaluated postoperative opioid consumption in patients randomized to lumbar plexus blocks or lateral QLBs.
The study was ceased early as blinded therapists and surgeons perceived that one group had greater strength and improved mobilization. While the study was underpowered to demonstrate non-inferiority for opioid consumption at 12-hour postoperative, the lateral QLB group was 2.4 times more likely to ambulate at less than 12-hour postoperative.
This study supports lateral QLBs as an analgesic option after hip arthroplasty.
Introduction
Total hip arthroplasty (THA) improves quality of life and increased demand is projected over the coming years.1 While lumbar plexus blocks (LPBs) reduce postoperative pain and opioid consumption after THA,2 3 risks include the potential for quadriceps weakness, bleeding, and neuraxial local anesthetic spread.
Quadratus lumborum blocks (QLBs) may be an alternative regional technique for analgesia after THA.4–6 Compared with LPBs, lateral QLBs are more superficial, which may reduce the risk of bleeding, and placed further from the neuraxis. However, lateral QLBs have not been compared with LPBs for postoperative analgesia following THA.
The purpose of this study was to examine postoperative analgesia in patients undergoing THA and randomized to LPBs or lateral QLBs (previously called QLB1). We hypothesized that lateral QLBs would provide non-inferior postoperative analgesia compared with LPBs as measured by opioid consumption at 12 hours after surgery.
Methods
The trial was conducted in accordance with the original protocol, and written informed consent was obtained from all subjects. This manuscript adheres to applicable Consolidated Standards of Reporting Trials guidelines.
On the day of surgery, patients were invited to participate, provided with informed consent, and enrolled if eligible by study staff. Inclusion criteria consisted of age >18 years and undergoing elective THA. Exclusion criteria included allergy to study medications, weight under 40 kg, unable or unwilling to provide informed consent, and substance abuse.
Enrollment occurred from July 2020 to June 2021. Consenting subjects were consecutively assigned a three-digit number (001–184) and randomized to either QLB or LPB using a computer-generated list created by a statistician before study initiation using simple randomization. Randomization lists were kept in sealed envelopes and opened by the regional anesthesia team prior to block positioning. Other than the regional anesthesia team, all patients, care team members, and research staff were blinded to randomization.
Protocol
In preoperative holding, subjects were placed in the lateral decubitus position (operative side superior) and administered intravenous sedation (midazolam (0–2 mg), fentanyl (0–100 µg), dexmedetomidine (0–20 µg)). To maintain blinding, surface landmarks were marked with a marking pen (LPB) and ultrasound anatomy scanned and identified (QLB) for all subjects regardless of randomization. Skin was aseptically prepped, and lidocaine skin wheals placed at both appropriate block needle insertion sites. For both blocks, a 10 cm, 21-gage, echogenic stimulating needle (B Braun Medical, Bethlehem, Pennsylvania, USA) was inserted and ropivacaine (20 mL, 0.5%) slowly injected (3–5 mL aliquots) with intermittent aspiration.
Lumbar plexus block
At the L4 level, 4 cm lateral to midline, the needle was inserted perpendicular to the skin. Nerve stimulation (initial 1.0mA, 2 Hz, 0.1 ms) was used to produce a quadriceps muscle response (0.5–0.6 mA) as previously described.7
Lateral QLB
A high frequency (13–6 MHz), linear or low frequency (5–2 MHz) ultrasound probe was used to visualize the external oblique, internal oblique, and transversus abdominus muscles and moved laterally to identify the lateral aponeurosis of the transversus abdominus muscle and the lateral aspect of the QL muscle. An echogenic needle was then inserted using an in-plane technique and advanced (anterior to posterior) until the needle tip was deep to the aponeurosis of the transversalis abdominus muscle and lateral to the QL muscle as previously described (online supplemental file 1).6 8–11 Ropivacaine was injected and local anesthetic spread observed deep to the aponeurosis and lateral to the QL muscle on ultrasound imaging.
Supplemental material
Anesthetic care
Anesthetic care was standardized. Unless contraindicated, patients were prescribed a standardized multimodal protocol including preoperative oral acetaminophen (1000 mg) and celecoxib (200–400 mg based on renal function). If celecoxib was not given, ketorolac (15–30 mg intravenous based on renal function) was administered during surgical closure. Intraoperatively, spinal anesthesia (bupivacaine 10–12 mg) was supplemented with intravenous sedation. General anesthesia was utilized in the event of the inability to place or inadequate subarachnoid block. Surgeons performed periarticular injection immediately before closure (ropivacaine 0.2% with ketorolac 30 mg and clonidine 100mcg) for all patients. Postoperative anesthetic care unit (PACU) orders were standardized and included hydromorphone (0.2 mg intravenously every 10 min for severe pain). Postoperative surgical orders were standardized: oral acetaminophen (1000 mg every 8 hours), methocarbamol (750–1000 mg every 8 hours), celecoxib (200 mg two times per day), and oxycodone as needed for moderate and severe pain (5–10 mg).
Outcomes
Collected data included demographics, opioid consumption, pain rating using the Visual Analog Scale (VAS) and Numeric Rating Scale (NRS), quadriceps muscle strength, and ambulation times and distances. Demographic data collected included patient age, sex, race, weight (kg), body mass index (kg/m2), and use of preoperative opioids (yes or no). Block placement time was defined as time out to block needle removal. Block duration was defined as the time from block placement to the time that the patient reported resolution of numbness in the blocked extremity. Surgical duration was captured as incision to wound closure. Opioid consumption was collected as intraoperative (including block placement) or postoperative (after anesthesia end). VAS measurements were taken throughout by having patients mark on a 100 mm line (0 mm: no pain to 100 mm: worst pain). NRS data was collected by asking the patient to rate their pain on a scale of 0 to 10 (0: no pain, 10: worst pain). Motor strength was assessed in preoperative holding, on PACU arrival, 24-hour postoperative, and on visits by physical therapy (0: none; 1: muscle flicker without movement; 2: movement, but not against gravity; 3: movement against gravity; 4: movement against some resistance; 5: normal strength). Sensory assessments were made to cold stimulus and evaluated numb (no sensation) or not (any perception of cold). Ambulation distances were measured by physical therapists.
Power
The primary outcome for assessing non-inferiority was 12-hour postoperative opioid consumption assuming a non-inferiority margin of no more than a 20% increase. All opioids were converted to intravenous morphine milligram equivalents (MME) for comparison (1 mg morphine intravenous equal to fentanyl 10 µg intravenous, hydromorphone 0.15 mg intravenous, oxycodone 2 mg oral, and meperidine 7.5 mg intravenous).3 A previous study found mean intravenous MME in patients undergoing THA with a LPB was 20.4±13.1 MME at 24 hours.2 A prior power calculation found a sample size of 92 subjects per group (184 total) would provide 80% power to detect non-inferiority using a one-sided two-sample t-test (significance level α=0.05) and assuming a mean of 20.4±13.1 MME in the LPB group, the margin of equivalence is 20% of the mean opioid consumption in the QL group, and a true difference between the means equals 0.
Statistics
Descriptive statistics were calculated for all patient and procedural characteristics. The primary outcome of interest was 12-hour postoperative opioid consumption. Non-inferiority of QLBs compared with LPBs was conducted using a one-sided two-sample t-test approach. Secondary outcomes include pain scores, opioid consumption, quadriceps strength, time to first ambulation, and maximum distance ambulated. Differences in pain scores and opioid consumption over time between block groups were evaluated using a linear mixed model approach. The models for opioid consumption and pain scores over time included fixed effects for block type, postoperative time point, and block type by time interaction and a random patient effect to account measures on the same patient over time, and p values for pairwise comparisons at each timepoint were Bonferroni adjusted to account for multiple comparisons between block types over time. Differences between block types in postoperative quadricep strength was evaluated using the Cochran-Armitage trend test. Differences between groups in time to first ambulation were evaluated using a Cox proportional hazards model and maximum distance ambulated at 24- hour postoperative was evaluated using a Wilcoxon rank sum test. Model assumptions for all outcomes were assessed graphically and transformations were considered as needed. All analyses were conducted in SAS V.9.4.
Results
The study enrolled 111 patients (July 23, 2020–June 30, 2021): 57 LPB and 54 QLB (figure 1). In seven subjects randomized to LPB, quadriceps muscle stimulation at 0.5–0.6mA was not achieved and they were withdrawn from the study. One patient in the QLB group chose to not participate after consenting and withdrew. Of note, the study was stopped early due to concerns from blinded care providers that half of the patients were mobilizing faster, promoting day of surgery physical therapy and discharge. This resulted in a clinical decision by surgical team to cease allowing patient enrollment in the study and exclusively requesting lateral QLB for patients undergoing THA. The final study population included 103 patients (50 LPB and 53 QLB). A post hoc power calculation based on the observed data found the study only had 34% power to demonstrate non-inferiority, assuming a non-inferiority margin of 20%. Patient and procedure characteristics did not differ between groups (table 1). No block-related harm or adverse effects, including falls, were noted in either group.
Patientand procedural characteristics
Consolidated Standards of Reporting Trials flow diagram. LP, lumbar plexus; QL, quadratus lumborum.
Postoperative opioid consumption
Opioid consumption is presented in table 2. Mean (SD) cumulative opioid consumption (intravenous MME) at 12-hour postoperative was 12.7 (8.89) in the LPB group and 15.9 (12.1) in the QLB group. This equates to at 25% increase in cumulative opioids at 12 hours in the QLB group and was not found to be non-inferior (p=0.625). However, there were not clinically meaningful differences in opioid consumption in the first 24-hour postoperative, and the difference between groups over time was relatively consistent (online supplemental file 2).
Supplemental material
Cumulative postoperative opioid consumption presented as mean (95% CI) in intravenous morphine milligram equivalents
Pain scores
There was not a significant difference in VAS or NRS pain scores at all time point assessed (table 3). Mean (95% CI) 24-hour postoperative pain scores on a VAS scale were 39.1 (31.9, 46.3) in the LPB group and 40.1 (33.2, 47.0) in the QLB group. Postoperative pain scores were associated with preoperative pain scores and increased postoperative time (p<0.001 for both). Specifically, patients reporting higher preoperative VAS scores reported higher VAS pain scores in PACU and at 24-hour postoperative on average, controlling for block type.
Secondary outcomes
Quadricep strength
Patient quadricep strength did not differ significantly between the QLB and LPB groups prior to surgery (p=0.949). QLB patients had higher quadricep strength scores on PACU arrival compared with LPB patients (p<0.001, table 3). At 24-hour postoperative, the difference between the QLB and LPB groups in quadricep strength scores no longer differed (p=0.548).
Ambulation
There was not a difference in the maximum distance ambulated by block type (p=0.744; table 3). A change point at 12-hour postoperative was introduced in the Cox proportional hazards model evaluating the association between block type and the probability of ambulating to address the proportional hazards assumption. Patients randomized to QLBs were 2.4 times as likely to ambulate in the first 12-hour postoperative relative to patients randomized to LPBs (p=0.026; HR (95% CI): 2.44 (1.11 to 5.73)). However, the likelihood of ambulating did not differ between the groups after 12-hour postoperative (p=0.505). The greater likelihood of patients in the QLB group ambulating before 12-hour postoperative remained significant controlling for whether the surgery ended before 12:00 hour. The cumulative incidence of ambulation by block type is shown in figure 2.
Cumulative incidence plot for the probability of ambulating. The dashed vertical line at 12-hour postoperative represents the changepoint in the Cox model of time to first ambulation regressed on block type. The HRs (95% CI) comparing QL to LP blocks in the first 12-hour postoperative and for times greater than 12 hours are shown to the left and right (respectively) of the dashed line. LP, lumbar plexus; QL, quadratus lumborum.
Discussion
This randomized, prospective study did not find a lateral QLB to be non-inferior to an LPB for reducing postoperative opioid consumption in intravenous MME after THA. Although early study cessation resulted in inadequate power to demonstrate non-inferiority, both postoperative opioid consumption and pain scores did not differ clinically between block groups at all examined times.
Two recent studies have also compared anterior QLBs and LPBs following THA with similar findings. In a retrospective, propensity matched cohort design, opioid consumption and pain scores did not differ between patients who received anterior QLBs (n=30) or LPBs (n=30) for analgesia after THA.12 Further, a recent randomized controlled trial (RCT) examined patients receiving an anterior QLB (n=23) or LPB (n=23) with ropivacaine (0.5%, 20 mL) for analgesia following THA.13 They found the QLB to provide non-inferior pain control as measured with NRS pain scores. They also noted that opioid consumption did not differ between groups, while quadriceps weakness was less common in the QLB group. While our findings support the results of these prior studies and support the QLB as an alternative regional procedure for analgesia after THA, our study design focused on the lateral QLB to promote muscle strength while maintaining analgesia.
Prior publications support the utilization of the lateral QLB for hip surgery. La Colla et al reported successful analgesia in two patients undergoing hip surgery with lateral QLBs without quadriceps weakness and documented loss of sensation over T6-L3 dermatomes.5 In two retrospective cohort studies of subjects undergoing outpatient hip arthroscopy and receiving a QLB or no block, QLB patients received less opioids10 11 and reported less pain.10 Similarly, in a blinded RCT design in patients with femoral neck fractures and randomized to femoral nerve block or lateral QLB, QLB patients had less postoperative pain and opioid consumption.6 Likewise, in a prospective RCT of patients undergoing outpatient hip arthroscopy with lateral QLB or sham, opioid consumption was reduced 28.3% in patients receiving a lateral QLB.8 Our study adds more information regarding the utility of the lateral QLB approach, which has been less studied for analgesia for THA.
While publications have noted improved pain or decreased opioid consumption with the anterior12–17 and lateral5 6 8 QLB approaches, the mechanism of analgesia for the lateral approach is not well understood. However, in an imaging study of lateral, posterior, and anterior QLB approaches, there was wide variability in injectate spread within each approach with only the anterior approach reaching the lumbar plexus (25%).18 As our goal was to examine if we could provide hip analgesia without blocking the lumbar plexus, a lateral QLB was selected for our study. While further studies our needed to elucidate the analgesic mechanism, our findings support the utility of the lateral QLB approach.
As early ambulation after lower extremity arthroplasty may reduce morbidity19 and hospital length of stay,20 21 minimization of motor weakness and promotion of early ambulation are essential. Patients randomized to lateral QLBs had greater quadricep strength in PACU and were almost 2.4 times more likely to ambulate in the first 12 hours after surgery. While the few RCTs examining motor function have not noted quadricep strength to be decreased by QLBs compared with placebo after hip surgery8 22 or have shown improved strength when compared with LPB13 or fascia iliac block23 24 weakness was reported in one publication in 30% of subjects following an anterior QLB.25 This weakness may result from injectate spread to the lumbar paravertebral space and lumbar plexus,15 and this was another reason we examined the lateral QLB over the anterior QLB approach. Despite improved quadriceps muscle strength and earlier times to first ambulation in the lateral QLB group, we did not note a difference in the maximum distance ambulated in the first 24 hours.
Finally, our results are generalizable to most patients undergoing elective THA with few patient exclusions. The lateral QLB is a straightforward technique with ultrasound guidance and has less concerns for bleeding or deep plexus block complications compared with the LPB. Additionally, while we did not complete our desired enrollment goal, it is notable that our achieved participant number exceeds or matches the number of subjects enrolled in prior publications.
Limitations
This study has some clear limitations. Our greatest limitation is that our study was stopped prior to completing enrollment. We are, therefore, underpowered to evaluate for non-inferiority. A greater number of LPBs were not able to be placed compared with QLBs. This may be due to use of a stimulated vs ultrasound guided LPB technique. Another limitation is that we collected quadriceps strength on arrival to PACU. Although this was done for patients in both groups, we are unable to assess if the weakness was due to residual spinal anesthetic or the randomized block. Further, while there has been a major movement to have physical therapy assist patients with ambulation in the PACU, the timeliness of this can vary. Indeed, one prospective observational cohort study of 215 patients found waiting for physiotherapy or radiographs to delay postoperative recovery in up to 20% of patients following lower extremity arthroplasty.26 Finally, our normal practice is to include perineural dexamethasone in our blocks to prolong analgesia. This was not done in our study as it is an off-label use.
Conclusions
Although we were unable to demonstrate non-inferiority for opioid consumption at 12-hour postoperative, there were not clinically meaningful difference in opioid consumption and pain control over time between groups. Further, strength and mobilization were improved in the lateral QLB group.
Data availability statement
Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as online supplemental information.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by Medical University of South CarolinaOffice of Research IntegrityCharleston, SC, USAIRB Protocol ID: Pro00098482. Participants gave informed consent to participate in the study before taking part.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Presented at This work has been accepted for presentation at The American Society of Regional Anesthesia and Pain Medicine 2022 Annual Meeting (Las Vegas, NV; Abstract #3127).
Contributors TK: This author helped with data collection, interpretation of the results, and manuscript writing and editing. CDW: This author helped with data collection, interpretation of the results, and manuscript writing and editing. ORCID 0000-0002-1309-5710. BJW: Conflicts of interest: none. This author helped with study design, data interpretation, statistical analysis, and manuscript writing and editing. ORCID 0000-0002-7124-5158. EH: This author helped with data collection, interpretation of the results, and manuscript writing and editing. SB: This author helped with data collection and manuscript writing and editing. SHW: This author helped with study conception, study procedures, interpretation of the results, manuscript writing and editing, and is the guarantor of the study.
Funding This work was supported by internal departmental support (Department of Anesthesia and Perioperative Medicine, Medical University of South Carolina). This project was also supported by the South Carolina Clinical & Translational Research Institute, Medical University of South Carolina’s CTSA, NIH/NCRR Grant Number 1UL1TR001450.
Disclaimer The contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH or NCRR.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.