Supraclavicular block anesthetizes the brachial plexus where it is in its most compact form, thus providing a complete and reliable block for upper extremity surgery.1 Although the single injection (SI) ultrasound-guided corner pocket technique has been reported to have the highest success rate, this technique may miss the upper part of the plexus, resulting in an incomplete block.2 Cadaver and patient studies by using dye injections have demonstrated that injection into a single location does not result in the spread of the injected dye into all the compartments.3 Septae or a tight muscular membrane between the scalene muscles were found to separate roots of the plexus.4 Several studies comparing different approaches for brachial plexus block have demonstrated multiple injection techniques to be more successful, resulting in a faster onset of anesthesia and higher success rates with no increased incidence in complications.3,5,6 However, these trials did not show consistent results regarding the onset of block or the nerves blocked.
The aim of this study was to compare the sensory block success rate of SI versus triple injection (TI). We hypothesized that a TI performed at the upper, middle, and lower parts of the plexus would have a higher success rate compared with the SI technique.
Design and Patients
We conducted a prospective randomized double-blind study at King Abdulaziz Medical city in Riyadh, Saudi Arabia, between September 2009 and October 2011. All required approvals, including that of the local IRB, were obtained before the start of the study. Ninety-six patients who were scheduled for creation or superficialization of atrioventricular fistula of the upper limb in preparation for renal dialysis were enrolled after obtaining written informed consent. Patients were randomly allocated to receive either a SI or TI ultrasound-guided supraclavicular brachial plexus block. Randomization was performed by using a computer-generated randomized list. The inclusion criteria were ASA physical status <IV, Body Mass Index <35, and a patient age 18 years or older. Exclusion criteria were severe pulmonary disease, poor respiratory reserve, neuropathy, coagulopathy, patient allergy to local anesthetic (LA) drugs, and inability to consent to participate in the study.
A portable ultrasound machine (Sonosite M-turbo, SonoSite, Inc., Bothell, WA) with a 6 to 13 MHz linear probe was used for all patients. A 22-gauge 70-mm stimulating needle (Stimuplex; B. Braun Medical, Bethlehem, PA) was used, and all blocks were performed by 3 anesthesiologists with substantial experience by using this technique. After standard monitoring, an 18- or 20-gauge IV catheter was placed in the upper limb contralateral to the surgical site. Two liters per minute of oxygen were supplied via nasal cannula, and IV premedication (0.01–0.03 mg/kg midazolam) was administered to all patients. With the use of an aseptic technique, the supraclavicular area was scanned for the best view of the brachial plexus (Fig. 1A), a skin wheal was raised, and the needle introduced from the lateral-to-medial direction. For the SI group, the needletip was positioned in the lower part of the cluster formed by the plexus (Fig. 1B) and 10 mL of 1.5% lidocaine with epinephrine was injected incrementally, followed by another 20 mL of 0.5% ropivacaine. If the first 3 mL LA did not demonstrate good spread relative to the plexus, the needle was repositioned in the lower part of the cluster, and anesthetic was again administered until proper LA spread was visualized. For the TI group, LA was injected in 3 aliquots of 10 mL each, each composed of 3.5 mL of 1.5% lidocaine with epinephrine and 6.5 mL of 0.5% ropivacaine. Aliquots were deposited in the upper, middle, and lower thirds of the cluster (Fig. 1C).
The primary outcome was the combined score of sensory blockade of the 5 nerves (median, ulnar, radial, medial cutaneous nerve of the forearm, and musculocutaneous) measured at 5, 10, 15, and 20 minutes after injection. A blinded investigator tested the block by using ice at 5, 10, 15, and 20 minutes in the distributions of the 5 nerves as follows: median (palmar aspect of the second finger), ulnar (fifth finger), radial (dorsum of the hand between the thumb and second finger), musculocutaneous (lateral forearm), and medial cutaneous nerve of the forearm (medial forearm). A validated 3-point scale was used: 0 = no block (patient has normal sensation), 1 = patient can feel cold, but the sensation is reduced compared with the unblocked side, and 2 = complete anesthesia.7 We considered the patient to have a satisfactory sensory block when a minimal score of 9/10 was achieved. Motor blockade was not evaluated, given the superficial nature of the surgery and the lack of any need for muscular relaxation.
The secondary outcomes were onset time, performance time, success rate of each of the above nerves separately, success rate of surgical anesthesia, and complication rate. The onset of the block was evaluated by calculating the percentages of patients with satisfactory sensory block at 5 and 10 minutes. The performance time included the imaging time in addition to the needle time, defined as the point from initial needle insertion until complete injection of the LA. A successful block was one with a good primary outcome (defined above) and adequate surgical block, defined as one tolerating a simulated surgical stimulus (forceps pinching the prospective surgical area) at 30 minutes after injection. The need for LA infiltration at the surgical site or the need to convert to general anesthesia constituted a failed block.
If the composite sensory score was at least 9/10 after 20 minutes, the arm was scrubbed. The block was tested again at 30 minutes by the surgeon by using a forceps at the site of prospective surgery, evaluating for a surgical block. If the patient did not feel pain, surgery was allowed to start; if the patient had discomfort, IV fentanyl supplementation up to 1 μg/kg was administered. In cases of persistent pain at 30 minutes, the surgeon was asked to inject LA into the wound. Finally, if this supplemental LA did not work, the plan was to convert to general anesthesia. Recorded complications included vascular puncture, inadvertent IV injection, significant eye drooping (Horner syndrome), hematoma formation, LA toxicity, pneumothorax, and dyspnea. Patients were contacted 24 hours after surgery to exclude the presence of complications. The surgeon was asked to report the presence or absence of any complications on follow-up in the surgical clinic 2 to 4 weeks after surgery.
Three cases were excluded. The first patient had symptomatic hemidiaphragmatic paresis. The second patient had a brachial plexus anatomical anomaly, with the divisions located inside the middle scalene muscle. In the third case, a large transverse cervical artery was found crossing among the divisions of the plexus, and the infraclavicular approach was used instead. This artery is found normally in 89% of patients, but it is much smaller in diameter (1–3 mm).8
We conducted a pilot study on 15 patients to estimate the percentage of satisfactory sensory block. The rate was 60% at 20 minutes by using SI. Assuming that multiple injections would be capable of increasing this proportion to 90%, with a 95% level of significance and 80% power, by using a 2-sided hypothesis, the required sample size varied between 76 and 90, depending on the exact calculation method. Categorical variables such as the proportion of patients with successful sensory blocks at 5, 10, 15, and 20 minutes were described as percentages (numbers). Continuous variables such as age and Body Mass Index were described as means ± SDs. The χ2 test was used to evaluate SDs in categorical variables between the 2 groups. The Student t test was used to compare continuous variables. 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 1-Way Analysis of Variance on Ranks for intergroup comparisons, and the P value was calibrated with the Benjamini and Hochberg method. All P-values were 2-tailed. P-values of <0.05 were considered significant. SPSS (release 17.0, SPSS Inc., Chicago) and Open Epi (version 2.2) software was used for all statistical analyses.
All patients suffered from end-stage renal disease. There were no significant differences between the 2 groups with regard to age, sex, or body mass index (Table 1).
Combined Sensory Block Success
Sensory block of all 5 nerves was significantly better in the TI group at 10-, 15-, and 20-minute time points (Fig. 2) (all P < 0.035).
Individual Sensory Block Success
Successful block of the lateral cutaneous nerve of the forearm, which is a branch from the musculocutaneous nerve, was always higher in the TI group (Fig. 3A) (all P < 0.026). Complete sensory block of the median nerve was significantly higher in the TI group at 5 (P = 0.013) and 10 (P = 0.031) minutes only, but it did not differ at 15 and 20 minutes (Fig. 3B). The rate of complete sensory block of the medial cutaneous nerve of the forearm was significantly higher in the TI group at 5 minutes only (P = 0.040) but was not different at 10, 15, and 20 minutes (Fig. 3C). The success of the radial nerve block was significantly better in the TI group at 10 minutes (P = 0.08) but not at 5, 15, and 20 minutes (Fig. 3D). The rate of complete ulnar nerve block was significantly higher in the TI group at 5 (P = 0.031) and 10 (P = 0.048) minutes but did not differ between the TI and SI groups at 15 and 20 minutes (Fig. 3E).
Other Secondary Outcomes
The TI group had a more rapid onset of anesthesia. At 10 minutes, 57% of the patients in the TI group had complete sensory block vs 26% in the SI group, which was a significant difference (P = 0.002). The time needed to perform the block was significantly longer in the TI group (mean 6.5 ± 2.1 vs 4.7± 2.1 minutes in the SI group, P < 0.001). There was no difference in the surgical anesthesia at 30 minutes between the groups (87% in SI vs 96% in the TI group, P = 0.253). Six patients (13%) in the SI group and 2 patients (4%) in the TI group needed LA infiltration (P = 0.253). Complications and side effects were minimal in both groups (Table 2).
The main findings of the present study were that the TI technique had a more rapid onset and more complete block in the first 20 minutes after injection than the SI technique. However, there was no difference in success rate for surgical anesthesia at 30 minutes.
Previous studies have reported conflicting results regarding the success rate of supraclavicular block that is improved with a TI compared with a SI technique. Two other similar studies have compared the double injection technique with SI under ultrasound guidance.9,10 Although our study has many similarities to these studies, some important differences are worth noting. In addition to the differing number and sites of the injections given, our study was unique in that we focused on 1 type of surgical procedure to eliminate any confounders stemming from the surgical stimulus or location of the surgery. We also evaluated the outcome differently and chose sensory block of the 5 nerves as our primary outcome.
Our study showed that the TI group surpassed the SI group in achieving a combined sensory block at 5-minute intervals for the first 20 minutes, whereas the study by Roy et al.10 showed no difference at 15 minutes. Similar to the double injection studies, the rate of successful surgical anesthesia did not differ, but the onset time was shorter with TI and the performance time longer.
The slower onset of action in our SI group paralleled findings in another similar study.11 Performance time in the current study being shorter in the SI group was comparable to that in previous studies.9,12,13 Other studies have reported different surgical anesthesia success rates with SI ultrasound-guided brachial plexus blocks. Tsui et al.14 reported a 94.6% surgical success rate, but they used different criteria; for example, they did not consider those who needed supplemental LA as having failed block. Our results are comparable to those of other studies, such as Williams et al.12 (85%) and Arcand et al.15 (87%). It should be emphasized that when comparing our results with previous reports, several factors should be considered including the numbers and sites of injection, the LA used, and the method of sensory analysis.
Blockade of the musculocutaneous nerve was significantly better with TI at all time points, paralleling De Tran et al.’s findings9 of a better sensory block at 15, 20, and 25 minutes. In the present study, radial nerve sensory block was significantly better with TI only at the 10-minute time point, whereas De Tran et al.9 found this difference to extend to other time points as well.
The first injection of the TI technique used in the present study was deep, potentially close to the pleura, and carried the risk of pneumothorax.16 This deep injection was based on the corner block technique described by Soares et al.17 However, the absence of difference in the success of sensory block of the ulnar nerve at 15 minutes thereafter may eliminate the need for the potentially risky corner pocket injection, because this deep injection usually targets the lower part of the plexus that gives origin to the ulnar nerve. However, the effect of omitting this site of injection on other nerve branches needs to be evaluated in a future study.
In conclusion, this study demonstrated that a TI technique for supraclavicular brachial plexus blockade resulted in improved onset and more complete sensory block at 20 minutes compared with a SI technique.
Name: Samer A. Arab, MD.
Contribution: This author helped design and conduct the study and write the manuscript.
Attestation: Samer A. Arab has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
Name: Mohamad K. Alharbi, MD.
Contribution: This author helped conduct the study.
Attestation: Mohamad K. Alharbi has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Eman MS. Nada, MD.
Contribution: This author helped conduct the study and write the manuscript.
Attestation: Eman MS. Nada has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Derar A. Alrefai, MD.
Contribution: This author helped conduct the study.
Attestation: Derar A. Alrefai has seen the original study data and approved the final manuscript.
Name: Hany A. Mowafi, MD.
Contribution: This author helped analyze the data and write the manuscript.
Attestation: Hany A. Mowafi reviewed the analysis of the data and approved the final manuscript.
This manuscript was handled by: Terese T. Horlocker, MD.
1. Neal JM, Hebl JR, Gerancher JC, Hogan QH. Brachial plexus anesthesia: essentials of our current understanding. Reg Anesth Pain Med. 2002;27:402–28
2. Fredrickson MJ, Patel A, Young S, Chinchanwala S. Speed of onset of ‘corner pocket supraclavicular’ and infraclavicular ultrasound guided brachial plexus block: a randomised observer-blinded comparison. Anaesthesia. 2009;64:738–44
3. Thompson GE, Rorie DK. Functional anatomy of the brachial plexus sheaths. Anesthesiology. 1983;59:117–22
4. Chelly JE. Peripheral Nerve Blocks a Color Atlas. 20093rd ed Philadelphia, PA Lippincott Williams & Wilkins
5. Vester-Andersen T, Broby-Johansen U, Bro-Rasmussen F. Perivascular axillary block VI: the distribution of gelatine solution injected into the axillary neurovascular sheath of cadavers. Acta Anaesthesiol Scand. 1986;30:18–22
6. Partridge BL, Katz J, Benirschke K. Functional anatomy of the brachial plexus sheath: implications for anesthesia. Anesthesiology. 1987;66:743–7
7. Redborg KE, Antonakakis JG, Beach ML, Chinn CD, Sites BD. Ultrasound improves the success rate of a tibial nerve block at the ankle. Reg Anesth Pain Med. 2009;34:256–60
8. Cash CJ, Sardesai AM, Berman LH, Herrick MJ, Treece GM, Prager RW, Gee AH. Spatial mapping of the brachial plexus using three-dimensional ultrasound. Br J Radiol. 2005;78:1086–94
9. Tran de QH, Muñoz L, Zaouter C, Russo G, Finlayson RJ. A prospective, randomized comparison between single- and double-injection, ultrasound-guided supraclavicular brachial plexus block. Reg Anesth Pain Med. 2009;34:420–4
10. Roy M, Nadeau MJ, Côté D, Levesque S, Dion N, Nicole PC, Turgeon AF. Comparison of a single- or double-injection technique for ultrasound-guided supraclavicular block: a prospective, randomized, blinded controlled study. Reg Anesth Pain Med. 2012;37:55–9
11. Subramanyam R, Vaishnav V, Chan VW, Brown-Shreves D, Brull R. Lateral versus medial needle approach for ultrasound-guided supraclavicular block: a randomized controlled trial. Reg Anesth Pain Med. 2011;36:387–92
12. Williams SR, Chouinard P, Arcand G, Harris P, Ruel M, Boudreault D, Girard F. Ultrasound guidance speeds execution and improves the quality of supraclavicular block. Anesth Analg. 2003;97:1518–23
13. Perlas A, Lobo G, Lo N, Brull R, Chan VW, Karkhanis R. Ultrasound-guided supraclavicular block: outcome of 510 consecutive cases. Reg Anesth Pain Med. 2009;34:171–6
14. Tsui BC, Doyle K, Chu K, Pillay J, Dillane D. Case series: ultrasound-guided supraclavicular block using a curvilinear probe in 104 day-case hand surgery patients. Can J Anaesth. 2009;56:46–51
15. Arcand G, Williams SR, Chouinard P, Boudreault D, Harris P, Ruel M, Girard F. Ultrasound-guided infraclavicular versus supraclavicular block. Anesth Analg. 2005;101:886–90
16. Bhatia A, Lai J, Chan VW, Brull R. Case report: pneumothorax as a complication of the ultrasound-guided supraclavicular approach for brachial plexus block. Anesth Analg. 2010;111:817–9
17. Soares LG, Brull R, Lai J, Chan VW. Eight ball, corner pocket: the optimal needle position for ultrasound-guided supraclavicular block. Reg Anesth Pain Med. 2007;32:94–5