Secondary Logo

Journal Logo


Double-injection technique assisted by a nerve stimulator for ultrasound-guided supraclavicular brachial plexus block results in better distal sensory–motor block

A randomised controlled trial

Luo, Quehua; Yao, Weifeng; Shu, HaiHua; Zhong, Ming

Author Information
European Journal of Anaesthesiology: March 2017 - Volume 34 - Issue 3 - p 127-134
doi: 10.1097/EJA.0000000000000542
  • Free



Ultrasound is a popular technique for regional blockade because it aids deposition of local anaesthetic close to the desired nerve. The supraclavicular approach to the brachial plexus block (SBPB) is preferred in upper extremity surgery as the trunks and divisions are arranged in a compact manner at the supraclavicular fossa.1,2 Accurate placement of the needle is one of the key elements in achieving rapid and complete peripheral nerve blockade,2 and a nerve stimulator can help locate peripheral nerves.3 The optimal site for local anaesthetic injection in SBPB remains undefined, and recent research has demonstrated a that multipoint technique improves the quality of the block,4–7 especially with regard to complete ulnar nerve blockade. Because the ulnar nerve is derived entirely from the inferior trunk of the brachial plexus, the region it innervates is often incompletely blocked.2,3 In the traditional double-injection technique,7 to increase the probability of blocking the inferior trunk (mainly the ulnar nerve), the first increment of local anaesthetic is given lateral to the subclavian artery and superior to the first rib (the ‘corner pocket’)5 with ultrasound guidance. Using a nerve stimulator allows the needle tip to be accurately positioned adjacent to a nerve by eliciting a successful motor response or paraesthesia. The advantage that this confers remains unclear because only a small number of studies have compared ultrasound vs double guidance in SBPB.3,7,10 We assumed that modifying the current double-injection technique by combining ultrasound guidance with a nerve stimulator would improve the onset and effectiveness of ulnar nerve blockade.

The aim of this study was to compare the block success rates between the classical double injection (DI) and the modified double-injection (MDI) approaches. The classical technique is the double-injection approach described by Tran.7 The MDI technique differs from the classical in that the first increment of local anaesthetic in the ‘corner pocket’ is guided additionally by identification of the ulnar nerve by eliciting a successful motor response or paraesthesia using a nerve stimulator. We hypothesised that targeting the inferior trunk with a nerve stimulator in combination with ultrasound during SBPB would achieve a higher success rate of complete sensory block of the ulnar nerve compared with the classical double-injection technique 15 min after local anaesthetic injection.

Materials and methods

Design and patients

Local research ethics committee approval was obtained (no. 16 Jichang Road, Guangzhou 510405, China; Ethics committee of the First Affiliated Hospital of Guangzhou University of Chinese Medicine on 18 August 2015). The study was conducted at the First Affiliated Hospital of Guangzhou University of Chinese Medicine from October 2015 to January 2016. After obtaining written informed consent, Patients (American Society of Anesthesiologists physical status I to II, aged 18 to 65 years) who were scheduled for surgery of the elbow, forearm, wrist or hand, were enrolled in this study. Exclusion criteria included refusal of consent, coagulopathy, neuropathy, infection at the needle insertion site, allergy to local anaesthetic and a BMI of more than 26 kg m−2. Patients were randomised by a computer-generated schedule into one of two groups for ultrasound-guided supraclavicular brachial plexus block, a classical double-injection group (DI group) or a modified double-injection combined with nerve stimulation group (MDI group). The study was registered at the (no. ChiCTR-IOR-15007588).


On arrival in the preparation room, standard monitors were applied and supplementary oxygen (nasal cannulae at 2 l min−1) was given. Routine i.v. premedication (2 mg midazolam and 50 μg of fentanyl) were administered to all patients after venous access was established. All blocks were performed by senior anaesthesia residents or attending anaesthesiologists who had performed over 60 ultrasound-guided peripheral nerve blocks. A portable ultrasound machine (Sonosite M-turbo, SonoSite, Inc., Bothell, Washington, USA) with a 6 to 13-MHz linear probe was used for nerve visualisation in all patients. A 22-ga 50-mm stimulating needle (B. Braun Melsungen AG, Melsungen, Germany) was used for nerve location in the MDI group.

The classical double-injection technique was similar to that described by Tran et al.7 After obtaining the best possible short-axis view of the subclavian artery and the neural clusters in the supraclavicular fossa (Fig. 1), the anaesthesia provider initially oriented the needle tip to the ‘corner pocket’ at the position of the intersection of the first rib and subclavian artery using an in-plane technique in a lateral to medial direction, and half the volume (11.5 ml) of a 1 : 1 mixture of 2% lidocaine and 1% ropivacaine was injected after a negative aspiration. Subsequently, the remaining volume (11.5 ml) was given at the main neural cluster formed by the brachial plexus trunks and divisions, which were directly visualised by ultrasound.

Fig. 1
Fig. 1:
Transverse sonogram showing the trunks of the brachial plexus can be visualised as hypoechoic circular structures (arrows) lateral to the subclavian artery, which matches the definition of satisfactory imaging in the supraclavicular fossa. SA, subclavian artery; R, the first rib.

For the MDI group, once the stimulating needle was confirmed to be in a satisfactory position for the targeted nerves (the inferior portion of the plexus) in the ‘corner pocket’ with real-time ultrasound imaging guidance (Fig. 2), the electric current was turned on to 0.4 mA (frequency, 2 Hz; pulse width, 0.1 ms). The purpose of electrical stimulation was to elicit the following muscle twitch responses: flexion or paraesthesia of the fourth and the fifth finger or thumb adduction (ulnar nerve). The same type and volume (11.5 ml) of local anaesthetic as described previously was then injected in the location where the desired responses were elicited. Subsequently, the needle tip was repositioned adjacent to the main neural cluster and local anaesthetic was injected incrementally (Fig. 3).

Fig. 2
Fig. 2:
Transverse sonogram of the needle (arrows) in contact with the brachial plexus (nerves) in the corner pocket. The needle shaft and tip is seen as a linear hyperechoic density. N, nerves; R, the first rib; SA, subclavian artery.
Fig. 3
Fig. 3:
Transverse sonogram showing local anaesthetic spread in the corner pocket pushing the nerves to the periphery, thus the needle (arrows) is repositioned within the nerve cluster. LA, local anaesthetic; N, nerves; R, the first rib; SA, subclavian artery.


The sensory–motor blockade of the four nerves (musculocutaneous, median, radial and ulnar nerves) was evaluated at 3, 6, 9, 12, 15 and 30 min after local anaesthetic injection by a single, blinded observer. The primary outcome was the sensory block success of the ulnar nerve 15 min after local anaesthetic injection. Both sensory and motor blockade of the musculocutaneous, median, radial and ulnar nerves were assessed according to a 3-point scale. An ice bag was used for the sensory testing, this was 0 = normal cold sensation 0 = no block, 1 = partially block of cold sensation and 2 = complete anaesthesia. no cold sensation: For motor this was 0 = normal power, 1 = partial paresis, 3 = paralysis Sensory blockade was evaluated in the innervated area of the four nerves as follows: musculocutaneous (lateral forearm), median (palmar aspect of the second finger), radial (dorsum of the hand between the thumb and second finger) and ulnar (fifth finger). Motor blockade was assessed by elbow flexion (musculocutaneous), wrist flexion (median nerve), wrist extension (radial nerve) and flexion of the fourth and the fifth finger (ulnar nerve). We assumed that sensory–motor blockade was satisfactory when a score of 16 (sensory and motor block score of all four nerves reach point 2 = complete anaesthesia) was obtained.

The secondary outcomes were the performance time, satisfactory imaging, number of needle passes and success rate of surgical anaesthesia. For both techniques, the performance time was defined as the time from the start of initial scanning to the removal of the needle. Imaging was observed in real time and for "satisfactory imaging" it was required that two of the three trunks of the brachial plexus could be visualised as hypoechoic circular structures lateral to the subclavian artery (Fig. 1). An additional needle pass was defined as the need for the needle tip to be retracted at least 10 mm. Surgical anaesthesia was defined as the ability to proceed with surgery without the need for analgesic drugs, general anaesthesia, rescue blocks or local anaesthetic infiltration by the surgeon and was assessed at the end of surgery. Once a complaint of pain was made during surgery, the block was considered a failure and patients were allowed to receive rescue medication at the discretion of the responsible anaesthesiologist.

The blinded observer also recorded the incidence of vascular puncture, paraesthesia and injection pain during ultrasound-guided SBPB, as well as the incidence of Horner's syndrome, dyspnoea and toxicity of local anaesthetics for the entire operation. All patients were followed up for 48 h after operation for the clinical manifestation of nerve injury such as paraesthesia or motor deficit.

Statistical analysis

We performed a pilot study on 15 patients with the double-injection approach to estimate the percentage of complete sensory block of the ulnar nerve at 15 min. The rate was 50% at 15 min by this approach. We assumed that the MDI would be capable of increasing this proportion to 80%, with a 95% level of significance and 80% power. Thus, a calculated sample size of 39 patients was required to accomplish this goal. Therefore, we recruited 45 patients per group, assuming a 15% dropout rate.

SPSS for Windows 16.0 (SPSS Inc, Chicago, Illinois, USA) was used for statistical analysis. Normality of the data was tested using the Kolmogorov–Smirnov test, and then the Student t test was used to compare continuous variables. Categorical data were analysed using the χ2 test. Successful sensory block at different observation times were compared by using Friedman Repeated Measures Analysis of Variance on Ranks for within-group comparisons and Kruskal–Wallis one-way analysis of variance on ranks for intergroup comparisons, and the P value was calibrated with the Benjamini and Hochberg method. We considered P values of less than 0.05 to be statistically significant.


Basic information

There were no significant differences between groups with respect to personal characteristics such as age, sex, BMI and surgical site (Table 1). The MDI technique resulted in a longer performance time (5.08 ± 1.41 vs 4.10 ± 0.64 min, P < 0.001) and more frequent needle passes (4.40 ± 1.14 vs 2.87 ± 0.79, P < 0.001). There were three and 0 surgical failures for the double-injection and MDI groups, respectively. However, there were no statistically significant differences in obtaining surgical anaesthesia between the two techniques. No differences were found in terms of adverse events (Horner's syndrome, paraesthesia, injection pain, dyspnoea, vascular puncture, systemic toxicity to local anaesthetics and nerve injury) (Table 2).

Table 1
Table 1:
Personal and surgical data
Table 2
Table 2:
Block characteristics and outcome for the two study groups

Sensory block success

At 15 min, the rate of complete sensory block of the ulnar nerve was significantly higher (93 vs 67%, P = 0.002) in the MDI group (Fig. 4c). The proportions of patients who had achieved complete sensory blockade of all four nerves, at 6, 9 and 12 min, respectively, were significantly higher in the MDI group (all P < 0.05) (Fig. 4a and c, Fig. 5a and c). However, no differences were found in the success rates of complete sensory blockade between the MDI and double-injection groups at 30 min.

Fig. 4
Fig. 4:
Shows the percentage of complete sensory and motor block of the radial (a and b) and ulnar (c and d) nerves at the assessment times. DI, double injection; MDI, modified double injection.
Fig. 5
Fig. 5:
Shows the percentage of complete sensory and motor block of the musculocutaneous (a and b) and median (c and d) nerves at the assessment times DI, double injection; MDI, modified double injection.

Motor block success

Complete motor block of the musculocutaneous nerve was more pronounced in the MDI group from 6 to 15 min (all P < 0.05) (Fig. 5b). The rate of complete motor block of the median nerve was significantly higher in MDI group at 12 and 15 min but was not significantly different at 3, 6, 9 or 30 min (Fig. 5d). As to the complete motor block of the radial nerve, no differences were found at all measurement intervals (all P > 0.05) (Fig. 4b). The rate of complete ulnar nerve motor block was higher in the MDI group at 12 min only (P = 0.032), whereas rates at other measurement intervals did not differ significantly (Fig. 4d).

Combined sensory–motor block success

The composite scores assessing sensory–motor blockade of all four nerves were all significantly higher in the MDI group at 6, 9, 12 and 15-min time points (all P < 0.036). However, no significant differences in sensory–motor blockade were found between the groups at 30 min (P = 0.064) (Fig. 6).

Fig. 6
Fig. 6:
Percentage of patients with composite sensory–motor score of 16 for the four nerves (median, ulnar, radial and musculocutaneous) at the assessment times intervals. DI, double injection; MDI, modified double injection.


In both groups, we first identified the inferior trunk of the brachial plexus with ultrasound guidance, and in the MDI group we then placed the needle within the brachial plexus sheath around the ‘corner pocket’. A successful block in this study was defined as full sensory–motor blockade of all four nerves, and by this definition we achieved a higher success rate at 15 min in the MDI group than in the double-injection group (80 vs 56%). This success rate was an improvement on the 40 and 60% found in similar studies in which the needle was placed directly in the ‘corner pocket’, a method frequently associated with incomplete blockade of the inferior trunk of the brachial plexus.4,7 Failure to completely block the ulnar nerve often means failure of brachial plexus anaesthesia for surgery. In previous studies, unblocked ulnar nerve territory has commonly been present. Only 60% of participants subjected to a double-injection approach with ultrasound-guided SBPB were ready for surgery at 15 min; this rose to 65% with triple injection, and approximately 90% with a targeted intra-cluster injection technique.4,6 Multipoint local anaesthetic injection is more conducive to diffusion through the various extraneural connective tissue barriers and increases the probability of blocking fibres originating from the inferior trunk of the brachial plexus, mainly the ulnar nerve.8,9 Nevertheless, failure in blocking the region innervated by ulnar nerve usually accounts for the majority of incomplete brachial plexus blocks.2 Data from the American Society of Regional Anesthesia and Pain Medicine Conference in 2009 indicated that the anaesthesia supplemention rate was 13.1%, and unplanned general anaesthesia was required in 1.5% following ultrasound-guided supraclavicular blocks.9 The advantages of the MDI technique lie in its relatively simplicity and effectiveness.

There is discordance between the pattern of evoked motor response and a successful SBPB in that stimulation of a positive muscle response does not necessarily predict a successful block. Beach et al.10 demonstrated that using a nerve stimulator did not increase the success rate, and Haleem et al.3 reported that there is an association between the pattern of stimulated motor response and a successful nerve block. Stimulating flexion of the fourth and fifth digits (the inferior trunk) was associated with only a 61.1% success rate. The main explanation for the difference is likely to be in achieving the most suitable site for needle location and the number of injections made. In the setting of real-time ultrasound imaging with well defined anatomy and needle position, our results suggest that stimulation of the inferior trunk can improve the overall success rate (80%) at 15 min, with, remarkably, a 93% success rate for the complete sensory block of the ulnar nerve. In Beach's study, regardless of motor response, under real-time ultrasound imaging, the local anaesthetic was given when the needle tip contacted the outermost hyperechoic layer of the trunks of the brachial plexus. Compared with Beach, in our MDI group, we first identified the inferior trunk of the brachial plexus with the ultrasound guidance, and then confirmed the position by eliciting motor twitches in the ulnar nerve distribution. Compared with Haleem et al. who located the injection site by palpation of the subclavian artery, we visualised the brachial plexus with ultrasound, which we consider more direct and certain compared with landmark techniques. Ultrasound helps to remove the difficulty of the technique and reduces the risk of accidental intravascular injection. Use of a nerve stimulator may also reduce the probability of nerve injury.

The ideal position of the needle tip during ultrasound-guided peripheral nerve block is subject to debate. When the needle tip is placed within the brachial plexus sheath (subfascial), a faster onset of surgical anaesthesia is obtained compared with superficial injection outside the sheath.11–13 But the safety of needle tip placement inside the neural clusters deserves special attention. According to Franco,13 the needle tip should be placed in the connective tissue matrix between the neural elements before local anaesthetic is injected, but this differs from the view of Bigeleisen et al.14 regarding intraneural injection. Perlas et al.15 reported only two (0.4%) cases of transient postoperative numbness using the double-injection technique for 510 consecutive supraclavicular blocks. The current evidence seems to support the safety and effectiveness of the subfascial injection technique when the outermost fascial layer is identified by the hyperechoic ultrasound image. In our practice of ultrasound-guided supraclavicular block, we traversed the brachial plexus sheath and avoided injection into the nerves (the hypoechoic circular structures shown in ultrasonic images) by observing real-time ultrasound images of the needle tip as it advanced, and applied a stimulating current to exclude intraneural injection while guiding the needle tip close to the inferior trunk.16,17 This technique does not guarantee avoidance of intraneural injection; some incidents were associated with use of a stimulating current of 0.4 mA.14 To further exclude intraneural injection, we did not inject local anaesthetic in the presence of injection pain, paraesthesia or high injection resistance. Nonetheless, further studies are needed to verify the superiority and safety of double guidance.

As the primary outcome of the study was successful sensory blockade of the ulnar nerve at 15 min after local anaesthetic injection, a complete sensory–motor blockade of the inferior trunk (mainly ulnar nerve), accompanied by a satisfactory corresponding muscle twitch, may require needle redirection. Indeed, we found that the MDI supraclavicular block was associated with more needle passes. However, use of the MDI technique did not correlate with higher rates of vascular puncture, paraesthesia or postoperative neurologic deficit but was associated with a higher success rate of complete sensory block of ulnar nerve at 15 min (93%). Taking into account the combined results of three other trials,4,6,7 the MDI technique seems to be the best approach to achieving complete sensory block of ulnar nerve at 15 min (>90%).

The effective volume of local anaesthetic required depends mainly on the precise identification of the site of injection, thus accelerating the onset of its action. Although a volume of 30 to 35 ml of solution is more commonly administered, we chose 23 ml for our study as it represented the documented minimum effective anaesthetic volume in 50% of patients18 and allowed us to compare the outcome under the same conditions. We perform ultrasound-guided SBPB and use a volume of 23 ml of local anaesthetic on a daily basis: this is sufficient to produce reliable sensory–motor block.19,20 The use of ultrasound has been recommended to reduce both the local anaesthetic volume and the risk of complications.21–23 The incidence of Horner's syndrome can be as high as 90% when a high volume of local anaesthetic is used, whereas our incidence was 28% in the MDI group.

One limitation of our study is its lack of assessment for neurological dysfunction beyond 48 h, particularly as there was more need for needle redirection. Nevertheless, the MDI approach resulted in better sensory–motor block of all four nerves and a higher success rate of complete sensory block. In the future, large sample size studies may be needed to confirm these findings, especially on its safety and neurological symptoms and signs that may have gone unrecognised.

In conclusion, ultrasound-guided double-injection and the MDI technique provide comparable success rates at 30 min after injection. There were no complications directly related to the techniques or the local anaesthetic injection. Sensory–motor block of all four nerves was significantly better in the MDI group with a higher success rate of complete sensory block of the ulnar nerve within 15 min after local anaesthetic injection. MDI may provide better surgical anaesthesia than classical double-injection technique, although the time needed to perform the block was about 1 min longer.

Acknowledgements relating to this article

Assistance with the study: the authors appreciate the suggestions and interest from Guangzheng Zhong, MD, Department of Anesthesiology, the Second Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China.

Financial support and sponsorship: this work was supported by Department of Anesthesiology, the First Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, China.

Conflicts of interest: none.

Presentation: none.


1. Neal JM, Hebl JR, Gerancher JC, et al. Brachial plexus anesthesia: essentials of our current understanding. Reg Anesth Pain Med 2002; 27:402–428.
2. Sadowski M, Tulaza B, Lysenko L. Renaissance of supraclavicular brachial plexus block. Anaesthesiol Intensive Ther 2014; 46:37–41.
3. Haleem S, Siddiqui AK, Mowafi HA, et al. Brief reports: nerve stimulator evoked motor response predicting a successful supraclavicular brachial plexus block. Anesth Analg 2010; 110:1745–1746.
4. Techasuk W, Gonzalez AP, Bernucci F, et al. A randomized comparison between double-injection and targeted intracluster-injection ultrasound-guided supraclavicular brachial plexus block. Anesth Analg 2014; 118:1363–1369.
5. Soares LG, Brull R, Lai J, et al. Eight ball, corner pocket: the optimal needle position for ultrasound-guided supraclavicular block. Reg Anesth Pain Med 2007; 32:94–95.
6. Arab SA, Alharbi MK, Nada EM, et al. Ultrasound-guided supraclavicular brachial plexus block: single versus triple injection technique for upper limb arteriovenous access surgery. Anesth Analg 2014; 118:1120–1125.
7. Tran DQ, Munoz L, Zaouter C, et al. A prospective, randomized comparison between single- and double-injection, ultrasound-guided supraclavicular brachial plexus block. Reg Anesth Pain Med 2009; 34:420–424.
8. Moayeri N, Bigeleisen PE, Groen GJ. Quantitative architecture of the brachial plexus and surrounding compartments, and their possible significance for plexus blocks. Anesthesiology 2008; 108:299–304.
9. Fredrickson MJ, Patel A, Young S, et al. Speed of onset of ’corner pocket supraclavicular’ and infraclavicular ultrasound guided brachial plexus block: a randomised observer-blinded comparison. Anaesthesia 2009; 64:738–744.
10. Beach ML, Sites BD, Gallagher JD. Use of a nerve stimulator does not improve the efficacy of ultrasound-guided supraclavicular nerve blocks. J Clin Anesth 2006; 18:580–584.
11. Sivashanmugam T, Ray S, Ravishankar M, et al. Randomized comparison of extrafascial versus subfascial injection of local anesthetic during ultrasound-guided supraclavicular brachial plexus block. Reg Anesth Pain Med 2015; 40:337–343.
12. Yazer MS, Finlayson RJ, Tran DQ. A randomized comparison between infraclavicular block and targeted intracluster injection supraclavicular block. Reg Anesth Pain Med 2015; 40:11–15.
13. Franco CD. Connective tissues associated with peripheral nerves. Reg Anesth Pain Med 2012; 37:363–365.
14. Bigeleisen PE, Moayeri N, Groen GJ. Extraneural versus intraneural stimulation thresholds during ultrasound-guided supraclavicular block. Anesthesiology 2009; 110:1235–1243.
15. Perlas A, Lobo G, Lo N, et al. Ultrasound-guided supraclavicular block: outcome of 510 consecutive cases. Reg Anesth Pain Med 2009; 34:171–176.
16. Casati A, Danelli G, Baciarello M, et al. A prospective, randomized comparison between ultrasound and nerve stimulation guidance for multiple injection axillary brachial plexus block. Anesthesiology 2007; 106:992–996.
17. Barrington MJ, Watts SA, Gledhill SR, et al. Preliminary results of the Australasian Regional Anaesthesia Collaboration: a prospective audit of more than 7000 peripheral nerve and plexus blocks for neurologic and other complications. Reg Anesth Pain Med 2009; 34:534–541.
18. Tran DQ, Dugani S, Correa JA, et al. Minimum effective volume of lidocaine for ultrasound-guided supraclavicular block. Reg Anesth Pain Med 2011; 36:466–469.
19. Song JG, Jeon DG, Kang BJ, et al. Minimum effective volume of mepivacaine for ultrasound-guided supraclavicular block. Korean J Anesthesiol 2013; 65:37–41.
20. Sainz Lopez J, Prat Vallribera A, Segui Pericas M, et al. Ultrasound-guided supraclavicular brachial plexus block with small volumes of local anesthetic: technical description and analysis of results. Rev Esp Anestesiol Reanim 2006; 53:400–407.
21. Kjelstrup T, Hol PK, Courivaud F, et al. MRI of axillary brachial plexus blocks: a randomised controlled study. Eur J Anaesthesiol 2014; 31:611–619.
22. Barrington MJ, Kluger R. Ultrasound guidance reduces the risk of local anesthetic systemic toxicity following peripheral nerve blockade. Reg Anesth Pain Med 2013; 38:289–299.
23. Karmakar MK, Shariat AN, Pangthipampai P, et al. High-definition ultrasound imaging defines the paraneural sheath and the fascial compartments surrounding the sciatic nerve at the popliteal fossa. Reg Anesth Pain Med 2013; 38:447–451.
© 2017 European Society of Anaesthesiology