Does Proximal Adductor Canal Block Provide Better Analgesic Efficacy than Distal Adductor Canal Block in Patients Undergoing Knee Arthroplasty: A Systematic Review and Meta‐Analysis of Randomized Controlled Trials

Abstract

To compare the analgesic efficacy and adverse events of proximal versus distal ACB for adults undergoing knee arthroplasty, we searched PubMed, Cochrane, Web of Science, and Embase to identify all eligible randomized controlled trials (RCTs). The study quality of the RCTs was evaluated using the Cochrane risk of bias assessment tool. Heterogeneity among studies was examined by Cochrane Q test. Our primary outcomes were pain intensity at rest/during movement and morphine consumption. Statistical analyses were conducted by RevMan Software. Seven eligible studies involving 400 subjects were included in this meta‐analysis with 202 participants in the proximal ACB group and 198 participants in the distal ACB group. The results demonstrated that proximal ACB provided significantly better pain relief at rest at 2 h (SMD −0.27, 95% CI −0.54 to −0.01, four trials, 222 participants, I ² = 0, p = 0.04) and 24 h (SMD −0.28, 95% CI −0.48 to −0.08, seven trials, 400 participants, I ² = 0, p = 0.006) following the surgery. We found no evidence of a difference in postoperative pain intensity at other timepoints. Furthermore, we noted no evidence of a difference in cumulative morphine consumption and occurrence of adverse events. Proximal ACB provides better pain relief and comparable adverse effects profile compared with distal ACB. The analgesic benefit offered by proximal ACB, however, did not appear to extend beyond the first 24 h. The overall evidence level was mostly low or very low, which requires more well‐organized multicenter randomized trials in the future.

Proximal ACB provides better pain relief compared with distal ACB, which did not appear to extend beyond the first 24 h.

Background

Knee arthroplasty is among the most crucial surgical treatments for knee osteoarthritis (OA). ¹ The demand for primary total knee arthroplasty (TKA) is expected to increase by 673% to 3.48 million in the United States by 2030. ² Major goals following knee arthroplasty include sufficient postoperative pain relief to assist early physical therapy, which aids patients in returning to their physical capacity and being discharged early from the hospital. The use of regional analgesia in joint arthroplasty is associated with superior postoperative outcomes and may reduce the risk of chronic post‐surgical pain. ³ Moreover, as part of a multimodal approach to analgesia, peripheral nerve blocks can effectively offer postoperative pain relief after knee arthroplasty. ⁴ , ⁵

Traditionally, femoral nerve block has been the gold standard regional analgesic technique for postoperative pain treatment following knee arthroplasty. ⁴ However, there have been concerns regarding its risk of quadricep muscle strength loss. ⁶ The adductor canal runs distally from the apex of the femoral triangle to the adductor hiatus, which comprises femoral vessels, the saphenous nerve, a nerve branch to the vastus medialis muscle, and occasionally the posterior branch of the obturator nerve. ⁷ Adductor canal block (ACB), which targets the sensory branches of the saphenous nerve and spares the motor branches of the femoral nerve, has recently been demonstrated to provide better conditions for early rehabilitation, quicker return to mobility, and less risk of accidental falls during hospital care than femoral nerve block. ⁸ , ⁹ Currently, ACB is usually performed with ultrasonography and can be used as a single‐shot or continuous nerve block provided through a catheter. However, the ideal location for ACB placement within the canal remains unclear. The proximal block is positioned at the entrance just distal to the apex of the femoral triangle and is expected to provide improved analgesia and a more remote insertion site from the surgical field. ¹⁰ However, there is a theoretical risk of the proximal block causing quadriceps weakness by targeting several afferent branches of the femoral nerve. ¹¹ Placement of the distal block between the inguinal crease and the top of the patella may decrease the risk of motor blockade and provide improved analgesia for the posterior aspect of the knee owing to its distribution to the popliteal fossa and blockade of the popliteal plexus. ¹² , ¹³ However, it may involve an additional risk of contaminating the sterile surgical field. This controversy could be attributed to the lack of elucidation of the neural contents of the adductor canal and their relative sensory contributions to the knee. Initially, the saphenous nerve was considered the relevant femoral nerve branch for TKA analgesia and was the primary target of ACB. ¹⁴ However, recent evidence suggests that the nerve to the vastus medialis in the canal may also significantly contribute to knee innervation. ¹⁵ , ¹⁶

Several recent randomized controlled trials (RCTs) have reported that proximal ACB allows significantly less pain than distal ACB. ¹⁷ , ¹⁸ However, Lee et al. recently showed that patients in the distal ACB group experienced significantly less pain at rest and during movement than those in the proximal ACB group. ¹⁹ Moreover, other studies have reported no significant between‐group difference in terms of postoperative pain and total opioid consumption. ²⁰ , ²¹ Accordingly, we lack consensus on the optimal location for ACB placement within the canal. Thus, we conducted a systematic review and meta‐analysis to assess the analgesic efficacy and adverse events associated with proximal versus distal ACB in adults undergoing knee arthroplasty.

Methods

For this review, we followed the preferred reporting items for systematic reviews and meta‐analyses (PRISMA) statement. ²² This review was registered with PROSPERO, the international prospective register of systematic review of the National Institute for Health Research (accessed on April 9, 2023, CRD42023412491).

Search Strategy

PubMed, Cochrane library, Web of science, and Embase were searched to retrieve relevant papers dating up to May 25, 2023 with no language restriction. Our search strategy was based on a PICOS methodology and both Medical Subject Headings (MeSh) and text words were used (Supplementary Table S1). Literature search strategies were developed using terms which were related to adductor canal and knee arthroplasty. Adductor canal related terms included “adduct*” or “saphenous*” or “adductor canal” or “adductor canal block” or ACB or “saphenous nerve” or “saphenous nerve block” or SNB. Knee arthroplasty related terms included “knee arthroplas*” or “knee replac*” or “knee surgery” or “knee surgical” or “arthroplasties, knee replacement”. The reference lists of available studies were manually searched to identify additional articles for potential inclusions. The selection process was conducted by two individual investigators (QQL and ZKZ) independently and disagreements were resolved through discussions with the third investigator (DYC). A flow diagram of our search strategy is depicted in Figure 1.

FIGURE 1.

Study flow diagram.

Literature Inclusion and Exclusion Criteria

Two reviewers (QQL and ZKZ), working independently and in duplicate, identified and evaluated potentially eligible trials according to predefined inclusion criteria. Inclusion criteria were as follows: (1) Participants: adult participants (≥18 years old) undergoing primary unilateral knee arthroplasty regardless of their sex or pathological conditions. (2) Intervention and comparator: We included all RCTs comparing proximal versus distal ACB. Only studies using ultrasound technique were included without restrictions with the administration method (single‐shot or continuous blockade). (3) Outcomes: The primary outcome measures included the mean difference in postoperative pain at rest or during movement [2 h (within postoperative care unit), 24 h, 48 h] and cumulative morphine consumption (2 h, 24 h, 48 h). The secondary outcomes included postoperative quadriceps strength (24 h, 48 h), range of motion (ROM), rates of block‐related adverse events (accidental vascular puncture, paresthesia, motor blockade, failed block, neurological impairment), catheter insertion time, operative time, length of stay, successful rate, satisfactory level. (4) Study design: We included RCTs performed on humans. Exclusion criteria were as follows: (1) Studies on pregnant or lactating females. (2) Studies using landmark guidance technique due to the potential risk of inappropriate injection site. Disagreements regarding the study selection process were resolved by discussion with the third researcher (DYC).

Data Extraction

QQL extracted data from each eligible trial according to prepared data extraction form (Table 1). The extracted data were checked by another investigator (ZKZ) to reduce reviewer errors. If there were discrepancies, group consensus and a third reviewer was consulted to ensure accuracy of data (DYC). Data extracted from the eligible studies include first author, year of publication, country, characteristics of participants (age), number of participants in each group, type of anesthesia, block timing and type, composition of anesthesia, location of proximal and distal ACB, perioperative pain control method, and primary outcomes. If the data were only reported as graph, we extracted the values using GetData Graph Digitizer 2.24 software. If studies reported the median, range, and sample size, then the mean and standard deviation (SD) values were estimated. ²³ , ²⁴ If the data were incomplete or missing, we tried to contact the leading authors of relevant articles.

TABLE 1.

Demographic characteristics of the included studies.

Author, Year	Country	Mean age (years)	Sample size	Anesthesia	Block timing and type	Composition of anesthesia	Location of proximal ACB	Location of distal ACB	Perioperative pain control	Primary outcomes
Mariano 2014	USA	66 vs. 65	25 vs. 24	General anesthesia	Pre‐op; continuous	15 mL of 2% mepivacaine with epinephrine, 2.5 μm/mL	Beyond the apex of the femoral triangle	Halfway between the ASIS and the patella	Periarticular anesthetic injection and postoperative opioid	Time to success
Sztain 2015	USA	69 vs. 69	24 vs. 26	General or spinal anesthesia	Pre‐op; continuous	0.2% ropivacaine at 8 mL/with total of 30 min	Mid‐thigh	2–3 cm proximal to the adductor hiatus in the distal AC	Periarticular anesthetic and postoperative opioid	Pain intensity
Romano 2015	USA	60.8 vs. 62.9	28 vs. 28	Spinal anesthesia	Pre‐op; continuous	0.2% ropivacaine at a rate of 6–8 mL/h for 24 h	Superficial femoral artery passed beneath the medial border of the sartorius muscle	Half the distance between the inguinal crease and top of the patella	Tibial nerve block and postoperative opioid	Post‐op opioids consumption
Fei 2020	China	68.6 vs. 67.47	30 vs. 30	General anesthesia	Pre‐op; continuous	0.2% ropivacaine at a rate of 6 mL/h	At the proximal end of the AC	3–5 cm caudal to the proximal end of the canal	Postoperative PCA analgesia for 48 h	Cumulative sufentanil consumption
Greenky 2020	USA	68.4 vs. 66.1	34 vs. 29	Spinal anesthesia	Pre‐op (proximal) and post‐op (distal); single shot	15 mL of ropivacaine 0.5%	At the junction of the middle and distal third of the thigh	Deep to the vastus medialis obliquus at the level of the adductor tubercle	Pre‐ and postoperative opioid	Pain score, ROM, opioid consumption
Lee 2022	Korea	71.5 vs. 71.2	31 vs. 32	Spinal anesthesia	Post‐op; continuous	0.2% ropivacaine at a rate of 6 mL/h with a lockout time of 30 min	Medial borders of ALM and the sartorius muscle intersect	1/15th of the femur length from the proximal end of the AC	Pre‐ and post‐op oral opioid, peri‐articular anesthetic	Pain intensity
Chaiperm 2023	Thailand	68.7 vs. 70.8	32 vs. 24	Spinal anesthesia	Post‐op; continuous	NR	Conventional anesthesiologist assisted ACB catheter insertion	5‐cm distance from the level of the superior pole of the patella	Pre‐ and post‐op oral opioid, periarticular anesthetic	Pain intensity

Abbreviations: AC, adductor canal; ACB, adductor canal block; ASIS, anterior superior spine; ALM, adductor longus muscle; NR, not being reported; PCA, patient‐controlled analgesia; pre‐op, preoperative; post‐op, postoperative; ROM, range of motion.

Risk of Bias Assessment

Two review authors (QQL and ZKZ), working independently and in duplicate, assessed risk of bias of the eligible studies by using the Cochrane risk of bias assessment tool of seven points based on the following criteria: (1) Random sequence generation, (2) allocation concealment, (3) blinding of participants and personnel, (4) blinding of outcome assessment, (5) incomplete outcome data, (6) selective reporting, (7) other sources of biases such as baseline imbalance. ²⁵ Based on the recommendations of the Cochrane Handbook, risk of bias was judged to be L, H, and U, which is interpreted as low risk, high risk, and unknown risk of bias, respectively. Both responsible review authors resolved disagreements by discussion with a third review author (DYC).

Measures of Treatment Effect

For continuous outcomes, we estimated the mean difference (MD) or standardized mean difference (SMD) with 95% confidence interval (CI), and for proportions (dichotomous outcomes), we calculated the relative risk (RR) with 95% CI. For the continuous outcomes “postoperative pain” and “cumulative postoperative morphine consumption”, we used the SMD as a summary statistic in meta‐analysis because we did not transform the results. For other continuous outcomes including insertion time, operative time, and length of stay, we used the MD as a summary statistic in meta‐analysis.

Summary of Finding Table

We used the principles of the GRADE system to assess the quality of the body of evidence associated with the primary outcomes (postoperative pain intensity and cumulative opioid consumption) in our review (Guyatt 2008), and constructed a summary of findings (SoF) table using the GRADE software regarding the risk of bias (methodologic quality), the directness of the evidence, heterogeneity of the data, precision of effect estimates, and risk of publication bias (Table 2). ²⁶

TABLE 2.

Quality of evidence based on GRADE criteria.

Outcome	SMD (95% CI)	No. of subjects (Studies)	Comments	Quality of evidence
Postoperative pain at rest (VAS 0 to 100 mm, NRS 0 to 10) (2 h)	−0.27 (−0.54; −0.01)	222 (4 RCT)	Downgraded by one level for imprecision (failed required population)	⨁⨁⨁◯ MODERATE
Postoperative pain at rest (VAS 0 to 100 mm, NRS 0 to 10) (24 h)	−0.28 (−0.48; −0.08)	400 (7 RCT)	NA	⨁⨁⨁⨁ HIGH
Postoperative pain at rest (VAS 0 to 100 mm, NRS 0 to 10) (48 h)	−0.03 (−0.71; 0.64)	300 (5 RCT)	Downgraded by one level for inconsistency (unexplained high heterogeneity) (I 2 = 88%). Downgraded by one level for imprecision (large confidence interval). Downgraded by one level for imprecision (failed required population).	⨁◯◯◯ VERY LOW
Postoperative pain during movement (VAS 0 to 100 mm, NRS 0 to 10) (2 h)	−0.25 (−0.55; 0.06)	169 (3 RCT)	Downgraded by one level for imprecision (large confidence interval). Downgraded by one level for imprecision (failed required population).	⨁⨁◯◯ LOW
Postoperative pain during movement (VAS 0 to 100 mm, NRS 0 to 10) (24 h)	−0.15 (−0.38; 0.09)	281 (5 RCT)	Downgraded by one level for imprecision (failed required population).	⨁⨁⨁◯ MODERATE
Postoperative pain during movement (VAS 0 to 100 mm, NRS 0 to 10) (48 h)	0.08 (−0.71; 0.87)	229 (4 RCT)	Downgraded by one level for inconsistency (unexplained high heterogeneity) (I 2 = 88%). Downgraded by one level for imprecision (large confidence interval). Downgraded by one level for imprecision (failed required population).	⨁◯◯◯ VERY LOW
Cumulative morphine consumption (mg) (24 h)	−0.14 (−0.36; 0.07)	337 (6 RCT)	Downgraded by one level for imprecision (failed required population).	⨁⨁⨁◯ MODERATE

Abbreviations: CI, Confidence interval; MD, Mean difference; NA, not applicable; NRS, numerical rating scale; RCT, randomized controlled trial; VAS, visual analogue scale.

Evidence downgraded for imprecision for wide confidence interval >0.6 and/or sample size <100.

GRADE criteria.

High certainty: We are very confident that the true effect lies close to that of the estimate of the effect.

Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect.

Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect.

Statistical Analysis

Meta‐analysis was conducted by combining studies which were clinically similar in participants, intervention, comparator, and outcome (PICO). All analyses were carried out using Review Manager 5.1 (Cochrane Collaboration, UK). Heterogeneity among the included studies was quantitatively assessed with the chi‐squared test (p‐value <0.1) and I ² test, with I ² > 50% indicating significant heterogeneity. ²⁷ We used a random‐effects model if a significant heterogeneity was detected; otherwise, a fixed‐effects model was applied. Funnel plots were not included in this study as tests for funnel plot asymmetry is not recommended when a meta‐analysis contains fewer than 10 studies, due to the low power for detecting true effects not ascribed to chances. ²⁵ We did not perform subgroup analyses because less than 10 trials were included for the outcomes in this study. ²⁶ We performed sensitivity analyses focused on the influence of study quality by excluding trials assessed as having high risk of bias and low methodological quality. We also performed a sensitivity analysis to test the influence of using the fixed‐effect or random‐effect model.

Results

Selection and Identification of Studies

Figure 1 shows the flow diagram of paper inclusion and selection process. In summary, a total of 6731 publications were identified from the following databases including PubMed (538), Cochrane library (1140), Web of science (1890), and Embase (3163), which yielded 3232 papers after removing duplicates (3499 articles). After screening the retrieved manuscripts based on titles and abstracts, we retrieved 20 full texts, ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ 13 were excluded and seven were finally included in the review and meta‐analysis according to inclusion and exclusion criteria. ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²⁸ , ²⁹ Trials were excluded for the following reasons: not randomized controlled trials (n = 1), ³² unsuitable control group (n = 2), ³⁴ , ³⁵ other comparison (n = 8), ³⁰ , ³¹ , ³³ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ cross‐over design (n = 1), ³⁸ and not true ACB (n = 1). ⁴¹

Study Characteristics

Table 1 describes the baseline characteristics of the included trials. Overall, 14 treatment arms were extracted from seven RCTs that included a total of 400 participants, of which 202 participants were in the proximal ACB group and 198 were in the distal ACB group. The year of publication of the included trials ranged from 2014 to 2023. Four trials were conducted in the United States, ¹⁷ , ¹⁸ , ²⁸ , ²⁹ one in China, ²¹ one in Korea, ¹⁹ and one in Thailand. ²⁰ The number of participants in these studies varied from 49 to 63. Studies included more female adults than males. The population undergoing knee arthroplasty was similar regarding diagnosis and ranged from 65 to 71.5 years of age.

Five trials performed an ultrasound‐guided injection in both proximal and distal groups, ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²⁸ , ²⁹ whereas the other two studies compared the analgesic effect of anesthesiologist administered proximal ACB with surgeon‐administered distal ACB. ¹⁸ , ²⁰ The included studies used the local anesthetic in different concentrations. Four trials used 0.2% ropivacaine, whether the other studies used 2% mepivacaine with epinephrine ²⁹ and 0.5% ropivacaine. ¹⁸ Five studies performed the nerve block preoperatively, ¹⁷ , ¹⁸ , ²¹ , ²⁸ , ²⁹ and the other two studies conducted the block procedure postoperatively. ¹⁹ , ²⁰ Most trial authors selected continuous postoperative administration of local anesthetic except for the study by Greenky et al., who performed a single injection procedure. ¹⁸ For surgical procedures, patients in four studies received spinal anesthesia, ¹⁸ , ¹⁹ , ²⁰ , ²⁸ two trials performed general anesthesia, ²¹ , ²⁹ and the remaining one used either general or spinal anesthesia. ¹⁷ Three of the included trials reported that an additional multi‐modal analgesic regimen was started preoperatively and was continued postoperatively. ¹⁸ , ¹⁹ , ²⁰ Opioids were given as rescue analgesics in all studies. One of the included studies reported the use of a prophylactic drug against postoperative nausea and vomiting. ²⁰

Risk of Bias Assessment

The risk of bias assessment was displayed in Figures 2 and 3. Generally, the level of evidence of the included studies was high, with all studies considered to have a low risk of bias for random sequence generation due to adequate randomization. Five of the included studies described allocation concealment and were considered to have low risk of bias, ¹⁷ , ¹⁸ , ¹⁹ , ²¹ , ²⁹ while the other two studies were judged as having unclear risk of bias as they did not provide clear information on allocation concealment. ²⁰ , ²⁸ Four of the seven studies were double‐blind studies with the participant and the provider of the intervention blinded to therapy, ¹⁷ , ²¹ , ²⁸ , ²⁹ while we assessed the other three studies as having unclear risk of bias for blinding (performance bias) because blinding was not mentioned. ¹⁸ , ¹⁹ , ²⁰ We rated five trials as having low risk of bias for blinding of outcome assessment (detection bias), ¹⁹ , ²⁰ , ²¹ , ²⁸ , ²⁹ one trial as having unclear risk of bias, ¹⁸ and one trial as having high risk of bias due to non‐blinding of the outcome assessor. ¹⁷ Six trials reported that all participants were included in the analysis; we assessed them as having low risk of bias for incomplete outcome data. ¹⁷ , ¹⁹ , ²⁰ , ²¹ , ²⁸ , ²⁹ Due to insufficient data sources, we rated all studies as having unclear risk of bias for selective reporting and other potential sources of bias.

FIGURE 2.

Risk of bias summary: review authors' judgment about each risk of bias item for each included study. +, low risk of bias; −, high risk of bias; ?, unclear risk of bias.

FIGURE 3.

Risk of bias graph: review authors' judgment about each risk of bias item presented as percentages across all included studies.

Primary Outcome: Mean Differences in Postoperative Pain at Rest/During Movement at 2 h (within the Postoperative Unit), 24 h, 48 h

A total of seven studies compared postoperative pain intensity at rest and during movement at three different time points. As shown in Figure 4, the results showed that proximal ACB provided significantly better pain relief at rest at 2 h following the surgery (SMD −0.27, 95% CI −0.54 to −0.01, four trials, 222 participants, I ²  = 0%, p = 0.04), however, there is no difference in pain intensity during movement at 2 h following the surgery (SMD −0.25, 95% CI −0.55 to 0.06, three trials, 169 participants, I ² = 0%, p = 0.11). The pain intensity at rest at 24 h following surgery was also significantly lower in proximal ACB group than distal ACB group (SMD −0.28, 95% CI −0.48 to −0.08, seven trials, 400 participants, I ² = 0%, p = 0.006); similarly, there is no difference in pain intensity during movement at 24 h after surgery (SMD −0.15, 95% CI: −0.38 to 0.09, five trials, 281 participants, I ² = 0%, p = 0.23) (Figure 5). The results in Figure 6 showed no significant differences in pain intensity between proximal and distal ACB at rest (SMD −0.03, 95% CI −0.71 to 0.64, five trials, 300 participants, I ² = 88%, p = 0.92) and during movement (SMD 0.08, 95% CI −0.71 to 0.87, four trials, 229 participants, I ² = 88%, p = 0.84) 48 h after surgery.

FIGURE 4.

Forest plot of the postoperative pain score at rest (a) and during movement (b) 2 h after surgery.

FIGURE 5.

Forest plot of the postoperative pain score at rest (a) and during movement (b) 24 h after surgery.

FIGURE 6.

Forest plot of the postoperative pain score at rest (a) and during movement (b) 48 h after surgery.

We did not perform subgroup analysis because less than 10 studies were included for the above outcomes. Sensitivity analyses focusing on the influence of study quality excluding trials with high risk of bias showed again no significant differences (pain during movement: 2 h: SMD −0.20, 95% CI −0.56 to 0.16, p = 0.28, Sztain; pain at rest: 24 h: SMD −0.26, 95% CI −0.47 to −0.05, p = 0.02, Sztain; pain during movement: 24 h: SMD −0.16, 95% CI −0.42 to 0.10, p = 0.22, Sztain; pain at rest: 48 h: SMD 0.16, 95% CI −0.56 to 0.88, p = 0.67, Sztain; pain during movement: 48 h: SMD 0.24, 95% CI −0.78 to 1.25, p = 0.65, Sztain). However, the results of pain at rest 2 h postop showed significant difference after excluding the trial with high risk of bias (pain at rest: 2 h: SMD −0.23, 95% CI −0.53 to 0.07, p = 0.14, Sztain). We did not perform an analysis focusing on the influence of missing data because all data were reported within the trials. The sensitivity analysis focusing on the influence of using the fixed‐effect model showed no significant differences between random‐ and fixed‐effect models for the outcome pain at rest and during movement at all timepoints postop. We did not prepare a funnel plot due to the small size of included studies (less than 10 trials). Finally, we calculated the optimal information size (OIS) for the outcomes of pain at rest and during movement at 2 h, 24 h, and 48 h, respectively; results showed that the number of necessary participants was not reached except for the outcome of pain at rest at 24 h postop.

Using the GRADE approach, we downgraded postoperative pain at rest (2 h) and pain during movement (24 h) to moderate‐quality due to imprecision (failed required population). We downgraded the level of evidence for the outcomes of postoperative pain at rest and during movement (48 h) to very low‐quality evidence due to inconsistency (unexplained high heterogeneity) and imprecision (large confidence interval, failed required population). We downgraded postoperative pain during movement (2 h) to low‐quality evidence due to imprecision (large confidence interval, failed required population). Postoperative pain at rest 24 h postop was rated as high‐quality evidence.

Primary Outcome: Cumulative Mean Morphine Consumption [2 h (within the Postoperative Care Unit), 24 h, 48 h]

Six trials reported the cumulative morphine requirement at three different assessments. ¹⁷ , ¹⁹ , ²⁰ , ²¹ , ²⁸ , ²⁹ At all time points, there were no significant differences between groups (2 h: SMD −0.08, 95% CI −0.51 to 0.34, three trials, 159 participants, I ² = 45%, p = 0.70; 24 h: SMD −0.14, 95% CI −0.36 to 0.07, six trials, 337 participants, I ² = 0%, p = 0.20; 48 h: SMD −0.13, 95% CI −0.43 to 0.17, three trials, 173 participants, I ² = 33%, p = 0.40) (Figure 7).

FIGURE 7.

Forest plot of the postoperative cumulative morphine consumption (a) 2 h, (b) 24 h, and (c) 48 h after surgery.

Sensitivity analyses focusing on the influence of study quality excluding trials with high risk of bias showed again no significant difference (2 h: MD −0.21, 95% CI −0.59 to 0.17, p = 0.28; 24 h: MD −0.13, 95% CI −0.36 to 0.10, p = 0.28; 48 h: MD −0.09, 95% CI −0.45 to 0.26, p = 0.61). The sensitivity analysis focusing on the influence of using the fixed‐effect model did not show significant difference for this outcome compared with random‐effect model. Then, we calculated the OIS and the number of necessary participants was not reached for the outcome of cumulative morphine consumption (24 h). We did not perform an OIS for the outcome of cumulative morphine consumption (2 h and 48 h) because included participants were two few (<200 participants).

Using the GRADE approach, we rated the evidence for cumulative morphine consumption (2 h and 48 h) as low quality due to imprecision (unexplained heterogeneity, failed required population). We downrated the level of evidence for cumulative morphine consumption (24 h) from high to moderate due to imprecision (failed required population).

Secondary Outcome: Degree of Quadriceps Muscle Strength (24 h and 48 h)

Five trials reported data focusing on the degree of quadriceps muscle strength at two time points. ¹⁹ , ²⁰ , ²¹ , ²⁸ , ²⁹ Due to large heterogeneity in terms of different muscle tests, we decided to analyze this outcome qualitatively. All included trials showed no significant difference in quadriceps muscle strength following proximal ACB compared with distal ACB at 24 h after surgery. Similarly, at 48 h postop, four trials reported no significant difference in quadriceps muscle strength between groups.

Secondary Outcome: Postoperative ROM

Two trials reported the postoperative ROM between proximal and distal ACB group. ¹⁸ , ²⁰ Chaiperm et al. reported that there were no significant differences at all evaluated times in knee extension and flexion between two groups. ²⁰ However, Greenky et al. showed a significant increase in active flexion 24 h postop in proximal ACB group than distal ACB group, whereas there was no difference in active flexion on 6 weeks postop. ¹⁸

Secondary Outcome: Rates of Block‐Related Adverse Events (Accidental Vascular Puncture, Paresthesia, Motor Blockade, Failed Block, Neurological Impairment)

Only two included trials (123 participants) reported the number of participants with postoperative fall, ¹⁹ , ²¹ and no accidental falls occurred postoperatively. One included trial reported the number of patients suffering from vascular puncture and no vascular puncture occurred. ²⁹ Three of the included studies reported no catheter dislodgment occurred postoperatively. ²¹ , ²⁸ , ²⁹ Three trials reported the rate of fluid leakage of the catheter between groups. ¹⁷ , ²¹ , ²⁹ As shown in Figure 8, the results revealed no significant difference in the rate of fluid leakage between groups (RR 2.69, 95% CI 0.65 to 11.12, three trials, 158 participants, I ² = 0%, p = 0.17). None of the patients in two included studies suffered from puncture point infection. ¹⁹ , ²¹ Lee et al. mentioned that there was no participant suffering from local anesthetic toxicity. ¹⁹ Greenky et al. reported that one participant in the proximal and one participant in the distal ACB group had a hematoma, which was unrelated to the infection. ¹⁸ They also reported that in the anesthesiologist performed group, two subjects underwent manipulation under anesthesia. Mariano et al. mentioned that there is no difference in the numbness score between groups. ²⁹ Generally, Chaiperm et al. reported that there were no injection‐related complications in both groups. ²⁰

FIGURE 8.

Forest plot of the rate of fluid leakage of the catheter.

Secondary Outcome: Rates of Opioid‐Related Adverse Events (Postoperative Nausea and Vomiting, Pruritus, Sedation at 24 h and 48 h)

Only three included trials reported opioid‐related adverse events. ²⁰ , ²¹ , ²⁸ All of them reported that the two treatment groups did not differ significantly in terms of episodes of PONV after surgery. Chaiperm et al. reported that there were seven (23.3%) and 4 (13.8%) patients suffering from dizziness 24 h after surgery in the proximal and distal ACB group, respectively, and the difference did not reach significant difference. ²⁰ For all other opioid‐related adverse event, no data were available.

Secondary Outcome: Insertion Time

Three trials reported the catheter insertion time between proximal and distal ACB group. ¹⁷ , ²¹ , ²⁹ As shown in Figure 9, there is no significant difference in insertion time between groups (MD 0.09, 95% CI −0.48 to 0.66, three trials, 159 participants, I ² = 0%, p = 0.75).

FIGURE 9.

Forest plot of the insertion time.

Secondary Outcome: Operative Time

Five trials reported the operative time between proximal and distal ACB group, ¹⁷ , ¹⁹ , ²⁰ , ²¹ , ²⁸ As shown in Figure 10, there is no significant difference between groups (MD −0.53, 95% CI −4.70 to 3.63, five trials, 288 participants, I ² = 18%, p = 0.80).

FIGURE 10.

Forest plot of the operative time.

Secondary Outcome: Length of Stay

Four trials reported the length of stay between proximal and distal ACB group. ¹⁸ , ²⁰ , ²¹ , ²⁸ As shown in Figure 11, there is no significant difference between groups in terms of LOS (MD −0.03, 95% CI −0.21 to 0.15, four trials, 238 participants, I ² = 0%, p = 0.74).

FIGURE 11.

Forest plot of the length of stay.

Secondary Outcome: Successful Rate

Three trials reported the success rate of blockade of adductor canal between proximal and distal ACB group. ¹⁷ , ²⁸ , ²⁹ All of them reported that there is no significant difference in the rate of successful blocking between groups.

Secondary Outcome: Satisfaction Level

Three studies reported the overall satisfaction level between groups, ²⁰ , ²¹ , ²⁹ and no significant difference was reported regarding satisfaction level between groups.

Discussions

Summary of Main Findings

This meta‐analysis included seven RCTs (400 participants) comparing proximal and distal ACB in adults undergoing knee arthroplasty. Our findings indicated that compared with distal ACB, proximal ACB provided better pain relief during rest at 2 h and 24 h after surgery; however, there were no significant between‐group differences in postoperative pain intensity at the other time points. Moreover, quantitative analysis revealed no significant between‐group differences in cumulative morphine requirement and postoperative quadriceps strength. Contrastingly, qualitative analysis revealed a significant increase in active flexion at 24 h postoperative in the proximal ACB group compared with that in the distal ACB group; however, this difference did not reach significance at 48 h after surgery. There was no significant between‐group difference in the opioid‐related adverse events of PONV. Only two trials considered accidental falls during postoperative care, with none reporting such cases. Moreover, block‐related adverse events were poorly reported, and meta‐analyses could only be performed for the rate of fluid leakage, which did not show a significant between‐group difference. Qualitative analysis revealed no significant between‐group differences in the rates of successful blocks. Finally, there were no between‐group differences in the insertion time, operative time, LOS, or satisfaction level.

Overall Completeness and Applicability of Evidence

In our systematic review, we included seven trials involving 400 participants. However, the sample size of each trial was small, which increased the risk of heterogeneity and limited external validation. Furthermore, we could not achieve the calculated minimum number of patients required for all primary outcomes. Additionally, the included RCTs had variable reporting or inconsistent definitions of some outcomes. For example, although most studies reported the degree of quadricep muscle strength, they used different muscle tests; further, there remains no international standard for reporting functional recovery after knee arthroplasty. Therefore, this outcome was qualitatively described instead. Moreover, although six of the seven trials reported cumulative opioid consumption, none reported all relevant opioid‐related adverse events, which are clinically more relevant for the patient than cumulative opioid consumption. Additionally, three trials reported the PONV, but not for all investigated time points. The incidence of accidental falls during postoperative care was also poorly reported, and thus meta‐analysis could not be performed. Further, none of the studies reported chronic post‐surgical pain, which is another clinically relevant outcome following knee surgery, or block‐relevant adverse events other than the rate of fluid leakage.

Quality of the Evidence

Generally, most RCTs had a low risk of bias for all aspects rated in our risk assessment tool, except for blinding. One RCT was rated to have a high risk for assessor blinding, which was challenging to effectively implement given the nature of the block location. The most important limitation was the small number of trials reporting each outcome, which impeded many planned subgroup analyses exploring heterogeneity. Accordingly, the evidence for many primary outcomes was downgraded by two levels due to inconsistency (unexplained heterogeneity). Imprecision (failed required population and large confidence intervals) was another major limitation of our findings. Collectively, we rated only two primary outcomes (postoperative pain at rest and during movement [24 h]) as high‐quality evidence; contrastingly, many other outcomes were rated as low‐ or very low‐quality evidence with a high risk that future studies might reach different conclusions.

Potential Biases in the Review Process

Regarding publication bias, we may have missed some unpublished trials. Nonetheless, we used a comprehensive and systematic procedure to retrieve all RCTs; furthermore, the review authors independently selected studies from the initially identified articles, which led to a low risk of publication bias. Since all included RCTs had <100 participants per arm, small‐study effects could have biased the meta‐analysis results towards an inflated treatment effect. ⁴² Furthermore, we only performed subgroup analyses to explore heterogeneity when >10 trials were included. Accordingly, we considered these aspects in the GRADE ratings and downgraded the relevant evidence, as required. Since we analyzed numerous primary clinical outcomes at different time points, elucidating the evidence obtained from the included trials was challenging. Most trials only focused on specific topics (e.g., cumulative morphine consumption) and missed several other important postoperative pain‐ and block‐related outcomes. Therefore, the relevant core outcome domains for acute postoperative pain that should be reported in future trials on interventions for postoperative pain treatment must be promptly defined. The clinical heterogeneities of this study include age/sex difference, administration method of the blockade, compliance, different concentrations and types of local anesthetic (0.2% ropivacaine in four studies vs. 2% mepivacaine with epinephrine and 0.5% ropivacaine), sample size ranging from 24 to 34 and different perioperative pain control methods, potentially introducing bias to our analysis.

Agreements and Disagreements with Other Studies or Reviews

The conclusions of this review were consistent with those of a previously published meta‐analysis. ⁴³ Specifically, both reviews revealed no significant between‐group differences in total opioid consumption, catheter insertion time, or block success rate. However, our review found that proximal ACB could provide less pain during rest at 2 h and 24 h after surgery, which contrasts with their results. In the previous review, two of the five included trials did not meet their inclusion criteria, which impeded the interpretation of their results. Comparatively, our review had a higher probability of detecting statistical significance, given the inclusion of recently published RCTs.

Implications for Practice

Our findings indicated that compared with distal ACB, proximal ACB allowed better pain relief and a comparable adverse effect profile. However, the analgesic benefit offered by proximal ACB did not appear to extend beyond the first 24 h. Nevertheless, given the poor reporting of most primary outcomes, the quality of evidence for many of the outcomes was low or very low. Additionally, given the lack of data, we could not perform subgroup analyses, especially with respect to the influence of continuous versus single‐shot regional analgesia, which may have influenced the results.

Implications for Research

The findings of this review are based on the results of a few small‐scale RCTs. Accordingly, more RCTs comparing proximal and distal ACB are warranted. Moreover, future related studies should ensure proper reporting of relevant outcomes (opioid‐related adverse events, block‐related adverse events, and accidental falls during postoperative care). Although many trials used different tests to assess the degree of quadriceps muscle strength, only a few studies reported the number of patients who suffered from accidental falls or motor weakness during postoperative care. Furthermore, we could not perform subgroup analyses (e.g., single shot vs. continuous adductor canal blocks) owing to limited data. Future trials are warranted to clarify this comparison. Emerging evidence indicates the possibility of intraoperatively accessing the adductor canal ⁴⁴ , ⁴⁵ ; therefore, future studies are warranted to investigate the efficacy of intraoperative intra‐articular ACB administered by an ultrasound‐guided anesthesiologist. Additionally, the anatomical block location must be standardized when comparing the analgesic effects of different approaches in future studies.

Conclusions

Our findings indicated that compared with distal ACB, proximal ACB could provide better pain relief at 2 h and 24 h after surgery. However, there was no between‐group difference in pain intensity at rest and during movement at 48 h, total opioid consumption, catheter insertion time, block success rate, and postoperative quadriceps muscle strength. Further well‐designed trials are warranted to confirm the efficacy and safety of proximal versus distal ACB, especially with a focus on patient‐relevant outcomes, including opioid‐related and block‐related adverse events, rather than just opioid consumption or the degree of quadriceps muscle strength.

Author Contributions

Conceptualization, QQL and QJ; Data curation, QQL, DYC, and ZKZ; Formal analysis, QQL; Investigation, QQL, DYC, and ZKZ; Methodology, QQL and SWT; Software, QQL; Supervision, SWT and QJ; Writing—original draft, QQL; Writing—review and editing, QJ and SWT. All authors have read and agreed to the published version of the manuscript. QQL and ZKZ contributed equally to this work.

Conflict of Interest Statement

All authors declare that they have no conflict of interest.

Ethics Statement

Not applicable.

Supporting information

TABLE S1. Summary of the searching strategy.

Data Availability Statement

All data relevant to the study are included in the article or uploaded as supplementary information.