Analgesic efficacy of ultrasound-guided adductor canal blockade after arthroscopic anterior cruciate ligament reconstruction

A randomised controlled trial

Abstract

CONTEXT 

Anterior cruciate ligament (ACL) reconstruction surgery is associated with moderate to severe postoperative pain, which may be ameliorated by peripheral nerve blocks. The adductor canal block (ACB) is an almost exclusively sensory nerve block that has been demonstrated to reduce pain and opioid consumption after major knee surgery.

OBJECTIVES 

To investigate the analgesic effect of ACB in patients receiving a basic analgesic regimen of paracetamol and ibuprofen after arthroscopic ACL reconstruction under general anaesthesia.

DESIGN 

Randomised, double-blind, placebo-controlled, parallel groups.

SETTING 

Day Case Surgery, University of Copenhagen, Glostrup Hospital, Denmark, June 2010 to March 2012.

PATIENTS 

Fifty patients, aged 18 to 70 years, scheduled for arthroscopic ACL reconstruction.

INTERVENTIONS 

Patients were randomised to receive ACB with either 30 ml ropivacaine 7.5 mg ml⁻¹ (n = 25) or 30 ml 0.9% saline (n = 24).

MAIN OUTCOME MEASURES 

Primary outcome was pain score (0 to 100 mm) during standing at 2 h after surgery. Secondary outcomes were pain at rest, during standing and after walking 5 m, opioid consumption and opioid-related side effects for 24 h after surgery.

RESULTS 

Median (interquartile range) pain scores for the primary outcome were 20 (12 to 36) mm in the ropivacaine and 20 (10 to 44) mm in the control group (P = 0.84, 95% confidence interval for difference of −9 to 12 mm). No significant differences were observed in any of the secondary outcomes.

CONCLUSION 

An analgesic regimen with paracetamol and ibuprofen provides acceptable postoperative pain control after arthroscopic ACL reconstruction. ACB did not confer further benefit in our patients.

TRIAL REGISTRATION 

Clinicaltrials.gov Identifier: NCT01212666.

Introduction

Arthroscopic anterior cruciate ligament (ACL) reconstruction is most often performed as a day case procedure. It is associated with moderate to severe postoperative pain that may prevent same day discharge.^1–9 Various different analgesic methods have been investigated, including systemic and intra-articular analgesics,^1,10–12 neuroaxial and peripheral nerve blocks.^{2–3,6–8,13–18} At the present time, there is no gold standard for the management of postoperative pain after this procedure.

Femoral nerve block (FNB) has been shown to provide effective analgesia after ACL reconstruction in a number of randomised clinical trials.^7–9,15,19 However, concern has been raised about this block, as it reduces the strength of the quadriceps muscle, which may result in delayed mobilisation and a risk of falling.^20–23 Some authors have questioned the clinical relevance of the effect of the FNB.^2,24,25 Consequently, alternative regional anaesthetic techniques that preserve muscle function have been investigated, including blockade of various portions of the saphenous nerve.^26,27 Recently, the almost exclusively sensory adductor canal block (ACB) of the saphenous and the obturator nerves was found to reduce pain and morphine consumption after total knee arthroplasty.^28,29 The aim of this study was to investigate the effect of ACB on postoperative pain and opioid requirements after arthroscopic ACL reconstruction under general anaesthesia.

Materials and methods

This research was conducted at Glostrup University Hospital, the Capital Region of Denmark and enrolled patients from June 2010 to March 2012. The study was approved by the local Regional Ethics Committee, the Capital Region of Denmark, Kongens Vænge 2, Hillerød (Chairperson Catharina Madsen, Protocol no.: H-4-2010-032) on 7 June 2010, by the Danish Medicines Agency, Axel Heides Gade 1, Copenhagen (Chairperson Anne M. Holm, journal no.: 2612-4263) on 15 March 2010 and by the Danish Data Protection Agency. The trial was conducted in accordance with the Helsinki Declarations and monitored by the Good Clinical Practice (GCP) unit at Copenhagen University Hospital. The design and the description of the trial adhere to the Consolidated Standards of Reporting Clinical Trials (CONSORT) statement. The trial was registered at http://www.clinicaltrials.gov (NCT01212666).

All patients received written information about the trial at their first consultation at the Department of Orthopaedics. Patients were enrolled after the preoperative anaesthetic evaluation. Written informed consent was obtained from all patients included in the study. The majority of the patients had their surgery performed in the day case surgery unit. All patients planned for arthroscopic reconstruction of ACL were screened for inclusion except for a period of 4 weeks (during the summer of 2011) wherein screening and inclusion were not possible because of logistical problems. Patients were eligible if scheduled for arthroscopic ACL under general anaesthesia, aged 18 to 70 years, with BMI 19 to 35 kg m⁻² and classified as American Society of Anesthesiologists (ASA) physical status I–II. Patients with a history of allergic reactions to any of the study medications, drug or alcohol abuse, or with a daily intake of analgesics were excluded from the trial.

Interventions

All patients received 1000 mg paracetamol and 400 mg ibuprofen as premedication 30 min before surgery. General anaesthesia was performed with propofol (variable rate) and remifentanil 30 μg kg⁻¹ h⁻¹ (fixed rate). Sufentanil 0.2 μg kg⁻¹ was administered 30 min before expected awakening. Hypotension was treated with an infusion of 0.9% saline and if necessary, with intravenous ephedrine 5 to 10 mg or phenylephrine 0.1 to 0.2 mg at the discretion of the attending anaesthetist. Two surgeons performed the surgical procedures. The autologous tendon graft was harvested from the semitendinosus muscle with or without accessory tendon graft from the semimembranosus or gracilis muscle. The incision site for the graft was on the medial surface of the lower leg, 1 to 2 cm above the tibial tuberosity. All knees were explored for meniscal and articular cartilage damage.

The ACB was performed immediately after surgery before awakening the patient. The block was performed under ultrasound guidance with the use of a LogiQe (GE Healthcare, Waukesha, Wisconsin, USA), high-frequency linear probe. The position for the block was identified halfway between the base of the patella and the superior anterior iliac spine (Fig. 1). With the transducer placed in a transverse position across the thigh, the sartorius muscle, the femoral artery and vein were identified, with the saphenous nerve usually lying laterally to the artery. A 70 mm, 23-gauge Braun Stimuplex needle (B. Braun, Melsungen, Germany) was inserted ‘in-plane’ through the sartorius muscle until the tip of the needle was in a position just lateral to the saphenous nerve. The blinded study medication (30 ml ropivacaine 7.5 mg ml⁻¹ or 0.9% saline) was injected with aspiration initially, and then after each 5 ml of injection, whilst the expansion of the adductor canal was visualised on ultrasound. Postoperative analgesia consisted of 1000 mg paracetamol and 400 mg ibuprofen administered orally at 2, 10 and 18 h postoperatively. In addition, ketobemidone 2.5 mg, with a minimum dosing interval of 10 min, was given at patient request (as an intravenous injection during the first two postoperative hours and thereafter as oral tablets). Nausea was treated with intravenous ondansetron 4 mg and, if necessary, repeated with 1 mg supplemental doses. No other analgesic, antiemetic or sedative drugs were administered during the 24-h follow-up period.

Figure

Outcomes

The primary outcome measure was pain during standing, assessed at 2 h postoperatively.

Secondary outcome measures were pain at rest [area under the curve (AUC) for assessments performed at 0, 1, 2, 4, 6, 8 and 24 h postoperatively], during standing (at 1, 2, 4, 6, 8 and 24 h postoperatively) and after walking 5 m (at 2, 4, 6, 8 and 24 h postoperatively). In addition, cumulative ketobemidone consumption, nausea, vomiting and sedation were assessed at 0, 1, 2, 4, 6, 8 and 24 h postoperatively.

Assessment of outcomes

Data were collected by an investigator consulting the patients directly in hospital and afterwards by telephone. All results obtained by the telephonic interview were cross-checked with a questionnaire completed by the patients and returned to the investigator. The patients were instructed in the use of the 0 to 100 mm visual analogue scale (VAS) with 0 and 100 mm referring to ‘no pain’ and ‘worst pain imaginable’, respectively, at inclusion to the study. Nausea and sedation were assessed on a four-point verbal scale: none, 0; slight, 1; moderate, 2; severe, 3 and the number of vomiting episodes were counted. Total consumption of ketobemidone and ondansetron was also recorded.

Sample size estimation

On the basis of previous studies, we expected a VAS score of 50 mm (SD 25) in the control group at 2 h postoperatively when patients were mobilised to a standing position.^5,8 We considered a 50% reduction (25 mm) to be clinically relevant. With a type I error (α) of 5% and a power (1-β) of 90%, a sample size of 44 patients, 22 in each group, would be required. In order to take account for potential dropouts, we included 25 patients in each group.

Blinding and randomisation

The study was randomised, double-blinded and placebo-controlled. Study medication was prepared by the hospital pharmacy. Fifty identical packages containing either ropivacaine 7.5 mg ml⁻¹ or 0.9% saline (control group) were labelled with name of the project and numbered according to a computer-generated block randomisation list prepared by the pharmacy in five blocks, each containing 10 numbers. Data from the patients were registered according to the randomisation number. Each package was opened and the medicine prepared in a syringe by a nurse not involved in the study or postoperative care of the patient. All medications administered ‘in hospital’ were given to the patient and registered by one of the investigators. No investigator, person treating or nursing the patients was aware of group assignment until all patients had been included and data collection was completed.

Statistical analysis

Data are presented as median [interquartile range (IQR)] or mean (SD) and were analysed using the Mann–Whitney U-test for unpaired data with Hodges–Lehmann estimation of 95% confidence interval (CI) for median difference and the unpaired Student's t-test, respectively. For nausea and sedation, the arithmetic mean scores for each patient were calculated and compared. Categorical data (number of patients vomiting) were analysed using the chi-square test. The Bonferroni correction was used to correct for multiple comparisons. A P value of less than 0.05 was considered statistically significant. Calculations were performed using SPSS 17.0 for Windows (SPSS, Chicago, Illinois, USA). The investigators performed all the statistical analyses.

Results

Ninety patients were considered for inclusion in the study (see flow diagram, supplemental digital content, https://links.lww.com/EJA/A34). Fifty patients were recruited and randomly assigned to their allocated groups. One patient in the control group withdrew his consent 1 h postoperatively and was excluded, as he preferred a different analgesic treatment to that prescribed in the protocol. Data from 49 patients were analysed. One patient in the intervention group had to stop mobilising at 8 h postoperatively due to intra-articular bleeding and pain. Demographic and perioperative data are summarised in Table 1. The groups were similar regarding demographic and perioperative data. The distributions of data for pain, opioid consumption and side effects followed skewed distributions.

Table 1: Patient demographics and perioperative data

Primary outcome

The difference in median (IQR) pain scores (Fig. 2) during standing at 2 h postoperatively was not statistically significant between groups: 20 (12 to 37) versus 20 (10 to 44) mm in the ropivacaine versus the control group, respectively (P = 0.84, 95% CI for difference −9 to 12 mm).

Figure

Secondary outcomes

There were no statistically significant differences (Figs 2–4 and Table 2) between the ropivacaine and control groups using weighted average AUC (mm h⁻¹) with regard to pain scores: at rest (0 to 24 h) 13 versus 11 mm (−5 to 7) (P = 0.67); during standing (1 to 24 h) 16 versus 14 mm (−4 to 11) (P = 0.74); and after walking 5 m (2 to 24 h), 22 versus 24 mm (7 to 15) (P = 0.59). All values are median (95% CI for difference).

Figure

Figure

Table 2: Weighted average level area under the curve visual analogue score pain

Postoperative (0 to 24 h) median (IQR) ketobemidone consumptions were 7.5 (3.1 to 14.4) and 5.0 mg (2.5 to 8.8) in the ropivacaine and control groups, respectively (P = 0.13). Regarding sedation, levels of nausea, number of vomits, number of patients vomiting and consumption of ondansetron, there were no significant differences at any time point between groups (see supplemental digital content: ‘Side effects’, https://links.lww.com/EJA/A34). All blocks were successfully performed and no complications were recorded.

Discussion

The results of this study demonstrated no significant additional analgesic effects of ACB after arthroscopic ACL reconstruction, neither at 2 h postoperatively when patients were standing (primary outcome), nor at rest, during standing or after walking 5 m from 0 to 24 h postoperatively. Further, the need for additional ketobemidone from 0 to 24 h postoperatively was not significantly different in groups.

These results were not expected. Our sample size estimation was based upon the assumption that patients in the control group would have a VAS score of 50 mm at 2 h postoperatively when mobilised to a standing position. The observed pain scores were considerably and surprisingly lower at a median VAS of 20 mm in both groups. We must then consider if our trial had sufficient sensitivity to detect an effect of any analgesic intervention. Adequate sensitivity in acute pain trials is usually considered to require ‘moderate pain’, which has been demonstrated to correspond to at least 30 mm on a VAS scale.^30–32 In contrast to our primary outcome, pain scores at rest and during standing at 1 h postoperatively, and during walking at 2 h postoperatively, were above 30 mm in the control group, but still no significant effects of the ACB could be demonstrated at these time points. Further, the median differences between groups were low, with rather narrow 95% CIs. Consequently, we believe that our study has sufficient power to exclude major, clinically relevant effects of ACB after this surgical procedure.

All patients in our study received a basic analgesic regimen with paracetamol, ibuprofen and opioids on request, which may explain the rather low pain scores in both groups. Consequently, a possible conclusion of our study may be that this basic analgesic regimen is sufficient for pain control after arthroscopic ACL reconstruction and that the addition of regional nerve blocks is not warranted as a routine practice. However, a possible additional benefit from the ACB in populations with higher pain scores cannot be ruled out. Consequently, further studies are needed before final conclusions about the efficacy of ACB can be drawn.

Although disputable,^25,33 FNB has been shown to provide significant analgesia after ACL reconstruction in some randomised clinical trials.^7–9,15,19 As FNB reduces the strength of the quadriceps muscle, which may result in delayed mobilisation and a risk of falling, alternative regional anaesthetic techniques with preserved muscle function are warranted.^20–23 The saphenous nerve is the largest cutaneous branch of the femoral nerve. It supplies the cutaneous area of the anterior and medial part of the leg, ankle and foot. It joins the patellar plexus supplying the area above the patella and the sub-sartorial plexus supplying the medial part of the thigh. It sends branches to the knee joint, supplying the antero-inferior part of the knee capsule.^34,35 Furthermore, the terminal end of the posterior branch of the obturator nerve enters the distal part of the canal, and the administration of high volumes of local anaesthetic into the canal will theoretically affect these fibres, and add to the potential analgesic effect of the ACB. Akkaya et al.²⁶ showed significant reductions of VAS pain scores and opioid consumption with a saphenous nerve block during the first 24 h after arthroscopic resection of the medial meniscus. Two studies of ACB after total knee arthroplasty both demonstrated reduced pain scores compared with placebo.^28,29 Only one study has investigated the analgesic effect of a saphenous block after ACL repair. This study demonstrated reduced pain scores with an infra-patellar nerve block compared with placebo.²⁷ We were not able to confirm these findings. It is worth noting though that all patients in this trial were capable of walking 5 m, 2 h after surgery.

In order to avoid unblinding of the study, no sensory testing of the lower extremity was performed in our patients, and therefore, it was not possible to evaluate the success rate of the block. As a consequence, we have to consider the possibility that block failure may have been the cause of the lack of a significantly different analgesic effect of the ACB. However, the technique used to perform the ultrasound-guided ACB is simple, using the femoral artery as a consistent landmark. Jenstrup et al.²⁸ reported an ACB success rate of 94%, which is comparable with other studies blocking the saphenous nerve in the adductor canal.

In conclusion, we found that a basic analgesic regimen with paracetamol, ibuprofen and opioid, as required, provided acceptable pain control in our surgical population after arthroscopic ACL reconstruction and that the addition of ACB is not warranted. However, further studies in populations with higher postoperative pain scores are needed before firm conclusions about the efficacy of ACB after arthroscopic ACL reconstruction can be made.

Acknowledgements

Assistance with the study: the authors gratefully acknowledge the invaluable assistance from the nurses and colleagues at the Day Case Surgery Unit and Department of Anaesthesiology, University of Copenhagen, Glostrup Hospital, Capital Region of Denmark. In addition, we would like to thank the GCP-Unit, Copenhagen University Hospital, Capital Region of Denmark, for their support and for monitoring the study.

Sources of funding: this trial was supported by the Department of Anaesthesiology, University of Copenhagen, Glostrup Hospital, Capital Region of Denmark.

Conflicts of interest: none declared.