Introduction
The ultrasound-guided supraclavicular block (SCB) is a commonly used approach for brachial plexus block in upper extremity surgery [1]. Despite needle technique innovations in SCB (e.g., single or multiple injections [≥ 2 sites] with or without peripheral nerve stimulation), incomplete blockade of the ulnar nerve and high risk of intraneural injection remain clinically concerning [2,3]. A recent scanning technique known as the sequential ultrasound imaging technique has been found to help in accurately identifying individual neural elements of the brachial plexus at different trajectory levels above the clavicle [4]. Therefore, several studies have highlighted the feasibility of “selective or targeted blockade” or “entire blockade” of the brachial plexus with multiple SCB approaches and/or needle techniques [1,5]. Moreover, balancing between satisfying the block dynamics and avoiding intraneural injection has been evolving in clinical practice because of the existing fascial structure [4,6].
Recently, Siddiqui et al. [7] described an intertruncal approach (IA) to SCB for a complete blockade of the entire brachial plexus while avoiding intraneural injections. Compared to the classical approach (CA), which targets the compact plexus at the level of the trunks and divisions, the IA-SCB targets three independent trunks between the interscalene groove and the level of the first rib. After identifying the epineurium of each trunk, a double-injection technique with local anesthetic (LA) deposited in the investing adipose layers between the three trunks is performed. Theoretically, this is an extrafascial injection technique that allows for reliably confirming a subepineurial injection. Based on the experience of more than 200 cases, this study reported ideal motor and sensory blockade of the entire brachial plexus in general terms. However, at present, no detailed data on the characteristics of sensory-motor blockade and block-related variables for IA-SCB are known.
Therefore, the primary aim of this study was to systematically evaluate the effects of the IA-SCB on the complete sensory blockade of all four terminal nerves (musculocutaneous, median, radial, and ulnar nerves) compared to the CA-SCB using a double-injection technique. Based on our clinical experience, we hypothesized that the IA-SCB would provide complete sensory blockade 20 min after LA injection non-inferior to that of the CA-SCB. Therefore, the present study was designed as a non-inferiority trial.
Materials and Methods
Study participants
After receiving approval from the Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University (no. 2020-02-118-01), the trial was prospectively registered in the China Clinical Trial Registry (no. ChiCTR2000040199) on November 25, 2020, and conducted according to the Consolidated Standards of Reporting Trials (CONSORT) guidelines. The study protocol has been reported previously, and no changes were made throughout the trial [8]. Written informed consent was obtained from all the participants. A total of 122 patients scheduled for elective elbow, forearm, wrist, or hand surgery with brachial plexus block between November 26, 2020, and August 31, 2021, were enrolled. Eligible patients were identified at the preoperative visit one day before surgery and were informed of the study protocol. We enrolled adults (aged 18–75 years) with an American Society of Anesthesiologists (ASA) physical status score I–III. Exclusion criteria were as follows: 1) patient refusal of the brachial plexus block; 2) inability to perform the nerve block because of coagulopathy (defined as any coagulation disorders contraindicated for a peripheral nerve block), pre-existing neuropathy, infection at the supraclavicular fossa, hypersensitivity, or allergy to LA; 3) body mass index > 35 kg/m2; 4) pregnancy; and 5) severe mental illness or cognitive dysfunction (inability to communicate or cooperate).
Randomization and blinding
A research assistant who was not otherwise involved in other stages of the study used computer-generated simple randomization to assign patients to either the CA-SCB (n = 61) or IA-SCB (n = 61) groups at a 1:1 ratio. The random number was concealed in an opaque envelope that was opened only by the block practitioner immediately before performing the procedure. To eliminate performance bias, all blocks were performed by fellow or staff anesthesiologists (WF. Y, JQ. G, JY. G, and HB. X) with an extensive experience (> 60 attempts) in each technique. The nerve block was performed by one anesthesiologist, while another anesthesiologist provided intraoperative care. All treating anesthesiologists, outcome assessors, and research assistants were blinded to the group allocation.
Supraclavicular brachial plexus block performance
No medication was administered to the patients before arrival in the dedicated room for the SCB. After applying standard ASA monitoring and supplemental oxygen, intravenous midazolam 0.05 mg/kg or midazolam 0.05 mg/kg combined with fentanyl 0.5 µg/kg was administered for anxiolysis, according to the attending anesthesiologist’s discretion. Before performing the block, a pre-scan was performed using a 6–15 MHz high-frequency linear array transducer (Sonosite M-turbo, SonoSite, Inc.) to identify the trajectory of the brachial plexus above the level of the first rib. After sterilizing and infiltrating the skin with 2% lidocaine (1–2 ml), an ultrasound-guided IA-SCB or CA-SCB was performed according to group allocation with a 22-gauge, 80-mm insulated stimulating needle (B. Braun Melsungen AG).
For the CA-SCB, the double-injection technique was performed as described by Techasuk et al. [9], with the neural stack of elliptical hypoechoic trunks and divisions at the level of the first rib as the target ultrasound window. Briefly, the needle tip was oriented to the “corner pocket” between the subclavian artery and brachial plexus using the in-plane technique in a lateral-to-medial direction. Most of the LA mixture (15 ml, 1% lidocaine [Shandong Hualu Pharmaceutical Co., Ltd.] and 0.5% ropivacaine [Astrazeneca Pharmaceutical Co., Ltd.]) was deposited into the “corner pocket.” Subsequently, the needle was withdrawn and targeted at the center of the main neural cluster. The remaining volume (10 ml) of the LA mixture was then carefully injected into that central position (Fig. 1A).
For the IA-SCB, the nerve procedure was performed according to the method described by Siddiqui et al. [7], with the optimal order of injections following Endersby et al. [10]’s suggestion. The intention of the target ultrasound image was to visualize each trunk separately and simultaneously (superior, middle, and inferior trunks), each surrounded by the fascia in the plexus compartment. After skin infiltration, the block needle was advanced in-plane to the ultrasound beam in a lateral-to-medial direction until the needle tip was inserted immediately into the intertruncal plane between the middle and inferior trunks, where 15 ml of the LA mixture was injected. The LA was injected into small aliquots, while the needle was slowly advanced into the middle of the intertruncal plane. Subsequently, the remaining volume (10 ml) of the LA mixture was injected into the intertruncal plane between the superior and middle trunks (Fig. 1B).
Outcome measures
The primary outcome of this study was to determine the effects of the IA-SCB on the ipsilateral sensory blockade of all four terminal nerves of the brachial plexus, defined as the percentage of patients with complete sensory blockade 20 min after LA injection. In both groups, the sensory-motor blockade was assessed every 5 min for 30 min post-block. Sensory blockade was graded on a 3-point scale using a cold test (0 = no block, 1 = partial anesthesia, 2 = complete anesthesia). Sensory blockade of the musculocutaneous, median, radial, and ulnar nerves was assessed on the lateral aspect of the forearm, volar aspect of the thumb, lateral aspect of the dorsum of the hand, and volar aspect of the fifth finger, respectively. Similarly, motor blockade was graded on a 3-point scale (0 = no block, 1 = paresis, 2 = paralysis). Motor blockade of the musculocutaneous, radial, median, and ulnar nerves was evaluated by elbow flexion, thumb abduction, thumb opposition, and thumb adduction, respectively. A total sum ≥ 7 (of 8 points possible) for the sensory or motor scores of all four terminal nerves was considered complete sensory or motor blockade, respectively [9,11].
Secondary outcomes included the following: (1) nerve injury, defined as persistent paresthesia or motor weakness of the operative upper limb on follow-up 7 days after surgery; (2) block-related complications (intravascular injection, local anesthetic systemic toxicity, pneumothorax, and Horner syndrome); (3) partial or complete hemidiaphragmatic paresis (HDP), defined as a reduction in diaphragmatic excursion measured using M-mode ultrasonography before (baseline) and 35 min after block completion (complete = > 75% reduction from baseline, no hemidiaphragmatic movement, or paradoxical movement; partial = 25%–75% reduction; absent = < 25% reduction) [12]; (4) surgical anesthesia, defined as completion of scheduled surgery without a rescue nerve block, rescue analgesia (e.g., intravenous sufentanil 5 μg bolus), infiltrate LA (lidocaine 1%, maximum 10 ml), or conversion to general anesthesia; (5) level of difficulty of scanning, needle tip/shaft visualization, and identification of the injection site based on a comprehensive evaluation completed by the practitioner who performed the block (numeric rating scale: 0 = no difficulty, 10 = extremely difficult); (6) block-related variables, including performance time, needle pass, and procedure-related pain (numeric rating scale: 0 = no pain, 10 = worst possible pain); and (7) duration of surgery, defined as the interval from the beginning of the skin incision to skin suture completion. Definitions and evaluations of the aforementioned variables are described in detail in our previously defined study protocol [8].
Sample size calculation and statistical analysis
Based on a pilot study of 15 patients in each group, the proportion of patients with complete sensory blockade 20 min after LA injection was 73% (11/15) in the CA-SCB group and 87% (13/15) in the IA-SCB group. The study was designed as a non-inferiority trial to exclude the possibility that the IA-SCB was more than 5% inferior to the CA-SCB in terms of the percentage of patients with complete sensory blockade 20 min after LA injection. A non-inferiority margin of 5% was approved by the protocol committee based on the reasoning that this low threshold would be acceptable to patients with sensory blockade failure [13]. The required sample size per group was calculated to be 54 with a statistical power of 80% and a one-sided 95% CI using an online computing software (https://www2.ccrb.cuhk.edu.hk/stat/proportion/tspp_sup.htm). To account for a possible dropout rate of 10%–15%, the total sample size was inflated to 120–127 participants. A total of 122 participants were thus included in the study (n = 61 per group).
Statistical analyses were performed using SPSS for Windows version 18.0 (IBM, SPSS Inc.). Continuous variables are presented as mean±SD, normality was first assessed using the Kolmogorov-Smirnov test and then analyzed using an independent-samples t test. Categorical variables, such as the proportion in each group with complete sensory or motor blockade, success rates, and adverse events, are presented as the frequency, n (%). The Pearson 2 test, Fisher’s exact test (expected count < 5), or Mann-Whitney U test was used for categorical variables and other non-normally distributed data, as appropriate. A P < 0.05 was considered statistically significant for all results.
Results
A total of 122 patients were recruited and 113 completed the protocol for the primary outcome. Among 122 patients, six patients in the IA-SCB group and three in the CA-SCB group could not undergo the planned nerve block because of atypical imaging. For the former, distinguishing between the middle and inferior trunk or identifying the inferior trunk was difficult, while for the latter, nerve clusters were not concentrated on the lateral side of the subclavian artery but were distributed above or below the artery. These nine patients received an interscalene block, axillary block, or general anesthesia. Thus, 113 patients were included in the final analysis (Fig. 2). The baseline characteristics of the patients and surgeries were comparable (Table 1).
Regarding the primary outcome, 79.3% of the patients in the CA-SCB group and 72.7% in the IA-SCB group had complete sensory blockade of all four nerves at 20 min post-block (Fig. 3A). The difference in incidence between the two groups was –6.6% (95% CI [–22.3% to 9.1%]; P value for non-inferiority = 0.206), which exceeded the predefined non-inferiority margin of –5% (Fig. 3B). Thus, we could not confirm that the IA-SCB was non-inferior to the CA-SCB in achieving a complete sensory blockade 20 min post-block. No differences in the percentages of complete sensory and motor blockade were observed between the two groups within 30 min (Figs. 3C and D, all P > 0.05, these exact P values have been annotated at the corresponding time points). Details regarding the characteristics of complete sensory or motor blockade of the four nerves of the brachial plexus are presented in Fig. 4. Compared with the CA-SCB group, significantly more patients had complete sensory blockade of the median nerve at 10, 15, and 20 min and of the ulnar nerve at 15, 25, and 30 min in the IA-SCB group, and complete motor blockade of the radial nerve only at 20 min (all P < 0.05). However, the IA-SCB group had a lower proportion of patients with complete sensory blockade of the musculocutaneous nerve at 10, 15, 20, 25, and 30 min and complete motor blockade at 10, 20, 25, and 30 min (all P < 0.05).
None of the patients had developed persistent paresthesia or weakness in the upper extremities by the 7-d follow-up. Additionally, no intergroup differences were found in the success rates of surgical anesthesia (90.9% vs. 96.6%; P = 0.393) or block-related complications (Table 2). However, the IA-SCB group had a higher incidence of HDP 35 min after injection (47.3% vs. 25.9%; P = 0.018). In addition, the IA-SCB technique required more needle passes than the CA-SCB technique and longer imaging (49.0 s ± 16.4 vs. 21.3 s ± 4.4), needling (95.7 s ± 20.8 vs. 77.4 s ± 24.2), and performance (145 s ± 28.3 vs. 98.7 s ± 24.1) (all P < 0.001) times (Table 2).
Discussion
In this randomized trial, as the lower limit of the 95% CI was below the –5% margin (–22.3% to 9.1%) [14], we failed to demonstrate the non-inferiority of the IA-SCB compared to the CA-SCB for complete sensory blockade 20 min after LA injection. Indeed, the group with the higher proportion of complete sensory blockade was reversed compared with the pilot study; that is, the CA-SCB group had a higher proportion with complete sensory blockade than the IA-SCB group, which may be explained by the sample size. A sufficient sample size can reflect the true proportion with complete sensory blockade and reduce bias. However, the proportion of patients with complete sensory-motor blockade within 30 min was comparable between the two groups. Interestingly, the IA-SCB technique resulted in better sensory blockade of the ulnar nerve and worse sensory-motor blockade of the musculocutaneous nerve compared to the CA-SCB technique. However, the scanning window for IA imaging is narrow, and anatomical variations may exist. Thus, more cases of atypical imaging were found with the IA than with the CA. Importantly, this can make imaging or LA injection difficult. In addition, the parameters associated with the blocking operation (e.g., performance time, number of needle passes, and difficulty level) for the IA-SCB were inferior to those for the CA-SCB.
These discrepancies between the IA-SCB and CA-SCB could be attributed to the following: (i) subfascial double injection with the IA-SCB vs. extrafascial + intracluster injection with the CA-SCB, (ii) the large span and fascia barrier for the superior trunk [7], and (iii) the different volumes of LA injected in the superior and middle trunks. The “corner pocket” technique used in the CA-SCB is a kind of extrafascial injection [15] that may interfere with LA spread to the inferior trunk. In contrast, the needle traverses the prevertebral fascia (also known as the paraneural sheath) to enter the plexus compartment, and the LA is injected into the two intertruncal planes of the IA-SCB. Moreover, the inferior trunk can be clearly visualized with the “hydrodissection” effect of the LA injection. However, in the IA-SCB, the superior trunk is the most superficial and has the widest plexus compartment [4,7]. LA deposited between the superior and middle trunks may therefore not anesthetize this area, especially with a small volume of LA, because the musculocutaneous nerve generally emerges from the lateral cord (distal to the superior trunk) [16]. In addition, a septum (an additional layer of hyperechoic connective tissue) usually exists between the anterior compartment containing the divisions of the superior trunk and the posterior compartment containing the middle and inferior trunks [6], which may act as a physical barrier to the diffusion of the LA. In addition, a large portion of LA (15 ml) was injected between the superior and middle trunks in a previous study [17], whereas only 10 ml of LA was deposited in our study. Nonetheless, the IA-SCB has considerable potential to achieve good block dynamics of all four nerves of the brachial plexus and may serve as a promising alternative approach to the CA-SCB; however, the block characteristics of the superior trunk need to be monitored.
Given that the superior trunk is an independent compartment in this area, adding a single LA injection to the anterior or lateral side of the superior trunk or increasing the volume of the LA may improve the sensory-motor blockade of the musculocutaneous nerve [18]. However, our results show that the IA-SCB provides reliable surgical anesthesia and similar proportions of block-related complications (e.g., nerve injury, vascular puncture, paresthesia, and Horner syndrome). These findings suggest that both techniques have similar efficacy and relatively high safety. However, the performance time and number of needle passes were relatively poor for the IA-SCB. We attributed this discrepancy to some special anatomical characteristics of the IA-SCB. First, its scanning window is narrow, and all three trunks are in the same plane [7,19]. Thus, a slight repositioning of the probe causes a change in the target ultrasound window, particularly in the imaging of the inferior trunk. Performing a pre-scan using the sequential ultrasound imaging technique may help to clarify the trajectory of the brachial plexus and locate the optimal site for performing the IA-SCB [4,20]. Second, the two intertruncal planes between the three trunks are difficult to identify because of the tightly compressed, homogeneous fascia. Therefore, we recommend using the pre-injection technique with 2–3 ml of LA or saline for superior identification of the epineurium before advancing the needle tip to the target site [19]. In terms of nerve block performance, the intertruncal approach is suitable for distinguishing between the three trunks in the same plane, not only to perform the anatomy-based subfascial injection technique, but also to allow for a selective trunk block.
In the present study, the incidence of ipsilateral HDP was significantly higher after the IA-SCB (47.3%) than after the CA-SCB (25.9%). Previous studies have shown that a higher incidence of HDP is mainly associated with the site and volume of LA injections and the related injection techniques [21,22]. An injection site further away from the interscalene groove may limit the cephalic spread of LA to the cervical nerve roots (C3–C5) and ipsilateral phrenic nerve when the same concentration and volume of LA is used. Recent studies have found that subfascial LA injection can shorten the onset time but can also lead to a higher risk of HDP owing to the longitudinal spread of LA [23–25]. This is not surprising given that the total LA was injected under the paraneural sheath but outside of the epineurium of the three trunks for the IA-SCB, while the larger portion of LA was injected extrafascially (“corner pocket” technique) for the CA-SCB. The difference in HDP between the IA-SCB and CA-SCB could be partly attributed to the fact that the injection point of the CA-SCB is farther from the phrenic nerve. Consequently, future anatomical and clinical investigations are needed to determine the optimal LA volume and multipoint injection technique (> 2 sites) for the IA-SCB.
Our study has several limitations. First, we used a total LA volume of 25 ml in both groups even though a smaller volume would have been sufficient for a supraclavicular or infraclavicular block according to our previous studies [13,26]. However, this volume is relatively small compared to that used in other previous studies (30–40 ml) [1,7]. Larger volumes may enhance block dynamics but increase block-related complications (e.g., HDP). The imaging and injection techniques for the IA-SCB need to be further refined to simplify the nerve block procedure and reduce complications. Second, both the IA-SCB and CA-SCB were performed by experienced anesthesiologists. This was helpful for reducing bias; however, the clinical value of these techniques is likely significantly reduced for novices. Third, although we were under the assumption that the original techniques were implemented correctly, the paraneural sheath and fascial compartments surrounding the brachial plexus are relatively difficult to image and distinguish using real-time ultrasound in clinical anesthesia practice [4,6,27]. For example, ensuring that the needle tip reaches the target position in all cases is difficult as the needle tip may accidentally be positioned beneath the epineurium of the trunks in the IA-SCB.
In conclusion, our study was unable to show that the IA-SCB was non-inferior to the CA-SCB in terms of complete sensory blockade of all four nerves at 20 min post-block. In terms of the blocking effect, the IA-SCB achieved similar sensory-motor blockade characteristics within 30 min of LA injection and resulted in better sensory blockade of the ulnar nerve but unsatisfactory sensory-motor blockade of the musculocutaneous nerve, with a relatively high risk of HDP. Further trials are required to determine whether a triple-injection or sub-epineurium injection technique could further improve block dynamics.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
