E Brief Report
A Comparison of Posterior and Medial Cord Stimulation for Neurostimulation-Guided Vertical Infraclavicular Block: A Randomized Noninferiority Clinical Trial Chun Woo Yang, MD,* Sung Mee Jung, MD,† Hee Uk Kwon, MD, PhD,‡ Po Soon Kang, MD, PhD,‡ Choon Kyu Cho, MD,‡ Jin Young Oh, MD,‡ Younsuk Lee, MD, PhD,§ and Junghee Choi, MD║ BACKGROUND: We investigated whether medial cord stimulation is inferior to posterior cord stimulation for vertical infraclavicular block with respect to block success. METHODS: Ninety-six patients scheduled for upper limb surgery were randomly elicited a medial or posterior cord response for infraclavicular block using 40 mL of 0.5% ropivacaine. We assessed block success (complete sensory block of the 5 nerves in the forearm at 50 minutes) as the primary end point and block procedure characteristics and adverse events as secondary end points. RESULTS: The block success rates did not differ significantly between medial and posterior cord stimulation (95.7% [44/46] vs 91.7% [44/48], 95% CI of difference, –7.4% to 15.6%), while the secondary end points were comparable in both groups. CONCLUSIONS: Needle manipulation to elicit medial cord response is noninferior to posterior cord response of block success during neurostimulation-guided vertical infraclavicular block. (Anesth Analg 2014;118:874–8)
T
he type of distal motor response evoked is an important factor that influences the overall success of neurostimulation-guided infraclavicular block (ICB). Although previous studies1,2 have reported improved success with posterior cord stimulation over either medial or lateral cord stimulation in ICB using a lateral or pericoracoid approach, these findings might not apply to more medial vertical ICB because of the different topography of the cords between the 2 approaches.3,4 We recently showed that a median or ulnar nerve response elicited with medial cord stimulation also yielded a high success rate in vertical ICB.5 We investigated whether needle manipulation to elicit a medial cord response is inferior to elicitation of a posterior cord response in terms of block success for vertical ICB. We hypothesized that the closer anatomic association of the brachial plexus cords at the vertical infraclavicular site compared with the paracoracoid site for ICB would result in equivalent block efficacy from stimulation of either the medial or posterior cords. From the *Department of Anesthesiology and Pain Medicine, Cheju Halla General Hospital, Jeju-si, Jeju special self-governing province; †Department of Anesthesiology and Pain Medicine, Yeungnam University School of Medicine, Daegu; ‡Department of Anesthesiology and Pain Medicine, Konyang university hospital, Daejeon; §Department of Anesthesiology and Pain Medicine, Dongguk University Ilsan Hospital, Goyang; and ║Department of Anesthesiology and Pain Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea. Accepted for publication January 8, 2014. Funding: No funding. The authors declare no conflicts of interest. Reprints will not be available from the authors. Address correspondence to Sung Mee Jung, MD, Department of Anesthesiology and Pain Medicine, 170, Hyeonchung-ro, Nam gu, Daegu, 705-717, South Korea. Address e-mail to [email protected]. Copyright © 2014 International Anesthesia Research Society DOI: 10.1213/ANE.0000000000000143
874 www.anesthesia-analgesia.org
METHODS
With IRB approval and written informed consent, 96 patients scheduled for upper limb surgery were enrolled in this study. Exclusion criteria included any contraindication to regional anesthesia, pregnancy, allergy to local anesthetics, or neurological disorders. With the use of sealed opaque envelopes, patients were randomly assigned to receive a single injection ICB after needle manipulation to stimulate either the posterior (group P) or medial cord (group M). Standard noninvasive monitoring was used, and no premedication or sedation was administered before block. All blocks were performed with 40 mL of 0.5% ropivacaine by the first author before surgery in the recovery room. Patients were placed supine, with the arm adducted and the palm facing up. After identification of the landmarks, the needle insertion site was marked halfway between the jugular notch and the ventral process of the acromion. After skin disinfection, a local anesthetic skin wheal was raised at the needle entry point. Using a nerve stimulator (Stimuplex® HNS 12, B. Braun, Melsungen, Germany) initially set at 1.5 mA, 0.1 millisecond, and 1 Hz, a 22-gauge, 50-mm insulated needle (Stimuplex® A; B.Braun) was advanced directly in a posterior direction until elicitation of the desired distal motor response. To elicit the desired motor response, we used the following protocol based on a previous anatomical study:3 1. If no motor response was elicited on initial needle insertion, the needle was repositioned, with a slight deviation in the caudal, cranial, or lateral (not medial) direction until the maximum depth of 5 and 4 cm (man and woman, respectively) was reached. 2. If the first response was flexion of the elbow, pronation of the forearm, or radial movement of the hand, the needle had to be advanced deeper without angulation until stimulation of either the posterior or medial cord was obtained. April 2014 • Volume 118 • Number 4
Effect of Distal Motor Response on Infraclavicular Block
3. If the first elicited response corresponded to posterior cord stimulation in group M, the needle was repositioned in the caudal direction to elicit the medial cord response. If the first elicited response corresponded to medial cord stimulation in group P, the needle was repositioned in the cranial direction to elicit the posterior cord response. A satisfactory response to nerve stimulation was considered to occur when the desired motor response was elicited at a stimulation current between 0.2 and 0.5 mA. Finger flexion (with or without wrist flexion) was accepted as stimulation of the medial cord,6 and finger and/or wrist extension as stimulation of the posterior cord. If a satisfactory response was not obtained within 20 minutes of needle insertion, the patient was excluded from the study. Block performance time, number of needle redirections, and patient discomfort were also recorded by the anesthesia nurse assisting the anesthesiologist performing the block. Block performance time was defined as the time between the block needle insertion and needle withdrawal. The number of needle redirections was defined as either forward or backward movement of needle at least 1 cm or more. Degree of patient discomfort related to the block procedure was reported on a numerical rating scale (0–10) after completion of the block. Sensory and motor block assessment was performed by a blinded investigator after local anesthetic administration (time zero). Sensory block was assessed in the distribution of the radial, median, ulnar, musculocutaneous, and medial antebrachial cutaneous nerve by a cold test with the following scale: 0 = normal sensation, 1 = less cold, and 2 = not cold when compared with the contralateral arm every 5 minutes until 50 minutes or complete block. A complete sensory block was defined as a score of 2 in each nerve distribution. Since the complete sensory block with ICB using long-acting local anesthetics such as ropivacaine might have delayed onset and continued to evolve for as long as 1 hour,7,8 we defined the successful block as complete sensory block of the 5 nerves in the forearm within 50 minutes of injection. Failure of the block was considered if one (or more) of these sensory distributions was not blocked. The block was deemed apparently sufficient for surgery if sensory block was sufficient to tolerate surgical incision without any supplementation, despite not achieving complete sensory block of all nerves within 50 minutes. The time between the end of the injection and development of successful block was defined as block onset time. Motor block was evaluated every 10 minutes until 50 minutes or complete block using the forearm flexion, thumb abduction, thumb opposition, and thumb adduction (for the musculocutaneous, radial, median, and ulnar nerve, respectively) and scored as follows: 0 = no block, 1 = paresis, and 2 = paralysis. After a study period of 50 minutes or occurrence of a complete sensory block, the patients were taken to the operation room. If the block was adequate, supplemental sedation with propofol infusion was administered intraoperatively, according to patient preference. In the case of intraoperative pain, rescue analgesia was provided with 50 to 100 μg IV fentanyl. In case of persistent pain, general anesthesia was administered.
April 2014 • Volume 118 • Number 4
The surgical procedures, surgery time, tourniquet time, and tourniquet pain were recorded. Adverse events (vascular puncture, local anesthetic toxicity, Horner syndrome, dyspnea, and pneumothorax) were also noted.
Data Analysis
This study was conducted to determine whether needle manipulation to elicit medial cord response is inferior to that to elicit posterior cord based on successful neurostimulation-guided ICB. For noninferiority of medial cord stimulation versus posterior cord stimulation, a maximum difference of 10% (noninferiority margin) in success rate was considered acceptable. The choice of the noninferiority margin was based on clinical judgment. Assuming a success rate of 97% in both groups based on a previous study,5 the sample size for α = 0.05 and 80% power was calculated at 36 patients per group.9 To increase the power, we included 48 patients per group. The primary end point was analyzed by calculating 2-sided 95% confidence interval (CI) for the difference using Wilson method with continuity correction.10 Secondary end points were analyzed using the Mann-Whitney test
Table 1. Patient, Surgical, and Anesthetic Characteristics Age (y) Gender (M/F) Height (cm) Weight (kg) ASA physical status (I/II/III) Site of surgery Hand/forearm/elbow (n) Duration of surgery (min) Tourniquet time (min) Tourniquet pain (n)
Group P, (n = 48) 46 ± 15 28/20 164 ± 9 65 ± 11 36/11/1
Group M, (n = 46) 46 ± 15 28/18 167 ± 10 67 ± 12 34/10/1
P 0.964 0.836 0.146 0.287 0.820
25/18/5 59 ± 46 44 ± 37 3 (6.3)
22/19/5 62 ± 45 53 ± 36 3 (6.5)
0.980 0.746 0.203 1
Data are presented as the mean (SD) or numbers (%). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord; ASA = American Society of Anesthesiologists.
Figure 1. Percentage of patients with successful block over time (with 95% confidence interval). *Group M was inferior to Group P for success rate at 10 minutes: difference −16.2% (95% CI, −33.5% to 2.4%). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord. Bracketed values are the number of patients.
www.anesthesia-analgesia.org
875
E Brief Report for continuous variables and the χ2 test or Fisher exact test for categorical variables. The 95% CI of the median treatment difference in block onset time was calculated using the “Hodges-Lehman Estimator” as proposed by Altman et al.11 A 2-sided P < 0.05 was considered statistically significant. Statistical analysis was performed using the SPSS version 12 statistical software (SPSS, Chicago, IL)
Table 2. Procedural Characteristics Block performance time (min) Number of needle redirections (n) Patient discomfort (NRS 0–10) Desired motor response at first trial (n) Performance time (min) Patient discomfort (NRS 0–10)
Group P, (n = 48) 5.5 ± 2.5
Group M, (n = 46) 5.7 ± 2.8
P 0.843
3 (1–7)
3 (1–6)
0.949
2 (1–2)
2 (1–2)
0.860
16 (33.1)
18 (39.1)
0.705
4.2 ± 1.7 1 (1–3)
4.8 ± 2.9 1.5 (1–2)
0.703 1
Data are presented as the mean (SD), median (interquartile range), or numbers (%). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord; NRS = numerical rating scale.
RESULTS
Of 96 patients enrolled, 94 completed the study. Two patients in group M were excluded because of the inability to elicit the desired motor response. Patient characteristics were comparable between groups (Table 1). The success rate of group M was not inferior to that of group P after ICB (95.7% (44/46) vs 91.7% (44/48), 95% CI of difference −7.4% to 15.6%; Fig. 1). The number of patients with block sufficient for surgery and supplementation was comparable between groups (Table 3). Although there were significant differences in motor block of the radial (at 10 minutes), median (at 20 and 30 minutes) and ulnar nerve (at 20 minutes) (Fig. 2), comparable results were found for onset time (median difference 0 minute, 95% CI, −5 to 0, P = 0.71; Table 3), block performance time, number of needle redirections, patient discomfort (Table 2), efficacy of the sensory block (Fig. 3), and adverse events (Table 4) in both groups.
DISCUSSION
This study showed that needle manipulation to elicit either medial or posterior cord response provided a similar success rate block without differences in the block efficacy, performance time, and adverse events for vertical ICB. This is in contrast to previous studies that suggested posterior cord
Figure 2. Percentage of patients with complete motor block for each nerve over time (with 95% confidence interval). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord.
876 www.anesthesia-analgesia.org
anesthesia & analgesia
Effect of Distal Motor Response on Infraclavicular Block
Table 3. Block Characteristics Onset time (min) Blocks sufficient for surgery (n) Supplementation (n) Failure (n)
Group P, (n = 48) 10 (5–10) 46 (95.8) 2 (4.2) 0 (0)
Table 4. Adverse Events Group M, (n = 46) 10 (5–15) 45 (97.8) 1 (2.2) 0 (0)
P 0.71 1
Data are presented as the number of patients (%) or median (interquartile range). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord.
stimulation as the appropriate end point for the paracoracoid approach to ICB.1,2 This discrepancy might have resulted from the difference in topographical anatomy of cords between vertical and paracoracoid approaches of ICB. Since both the posterior and medial cord are close to each other and directly posterior (deep) to the lateral cord in the midinfraclavicular area,3 injection of local anesthetic may result in homogeneous distribution around all 3 cords with either medial or posterior stimulation elicited for vertical ICB. Another possible explanation might be the differences in the definition of medial cord stimulation. We have defined the distal motor response from medial cord stimulation as finger flexion (with or without wrist flexion). The median nerve has a dual-cord contribution from the lateral and medial cords. With the common plexus anatomy, finger flexion most likely identifies medial cord or medial root to median nerve stimulation, but wrist flexion may result from either medial or lateral cord stimulation or stimulation of the medial or lateral median nerve roots.6 Finally, the relatively large volume of local anesthetics used might have been responsible for our results. Our study has some limitations. First, it is possible that needle manipulations to search for specific distal motor response
Vascular puncture Horner syndrome Dyspnea
Group P, (n = 48) 0 (0) 0 (0) 0 (0)
Group M, (n = 46) 3 (6.5) 1 (2.2) 1 (2.2)
P 0.113 0.489 0.489
Data are presented as the number of patients (%). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord.
might influence the spread of local anesthetics. Second, the assessment time of 50 minutes may account for the favorable results in both groups, at least in part. This long period of assessment may not be acceptable in clinical practice. Finally, we did not evaluate the influence of specific cord stimulation on postoperative outcomes such as block duration, pain score, time to first analgesia, and opioid consumption. In conclusion, this study demonstrated that needle manipulation to elicit either medial or posterior cord response provided similar block-related characteristics during neurostimulation-guided vertical ICB. E DISCLOSURES
Name: Chun Woo Yang, MD. Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript. Attestation: Chun Woo Yang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Sung Mee Jung, MD. Contribution: This author helped design the study, analyze the data, and write the manuscript. Attestation: Sung Mee Jung has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Figure 3. Percentage of patients with complete sensory block for each nerve over time (with 95% confidence interval). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord; MAC = medial antebrachial cutaneous nerve.
April 2014 • Volume 118 • Number 4
www.anesthesia-analgesia.org
877
E Brief Report Name: Hee Uk Kwon, MD, PhD. Contribution: This author helped analyze the data and write the manuscript. Attestation: Hee Uk Kwon has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Po Soon Kang, MD, PhD. Contribution: This author helped analyze the data and write the manuscript. Attestation: Po Soon Kang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Choon Kyu Cho, MD. Contribution: This author helped analyze the data and write the manuscript. Attestation: Choon Kyu Cho has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Jin Young Oh, MD. Contribution: This author helped collect the data. Attestation: Younsuk Lee has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Younsuk Lee, MD, PhD. Contribution: This author helped analyze the data. Attestation: Younsuk Lee has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Junghee Choi, MD. Contribution: This author helped analyze the data. Attestation: Junghee Choi has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. This manuscript was handled by: Terese T. Horlocker, MD.
878 www.anesthesia-analgesia.org
REFERENCES 1. Bloc S, Garnier T, Komly B, Leclerc P, Mercadal L, Morel B, Dhonneur G. Single-stimulation, low-volume infraclavicular plexus block: influence of the evoked distal motor response on success rate. Reg Anesth Pain Med 2006;31:433–7 2. Lecamwasam H, Mayfield J, Rosow L, Chang Y, Carter C, Rosow C. Stimulation of the posterior cord predicts successful infraclavicular block. Anesth Analg 2006;102:1564–8 3. Moayeri N, Renes S, van Geffen GJ, Groen GJ. Vertical infraclavicular brachial plexus block: needle redirection after elicitation of elbow flexion. Reg Anesth Pain Med 2009;34:236–41 4. Sauter AR, Smith HJ, Stubhaug A, Dodgson MS, Klaastad Ø. Use of magnetic resonance imaging to define the anatomical location closest to all three cords of the infraclavicular brachial plexus. Anesth Analg 2006;103:1574–6 5. Yang CW, Kang PS, Kwon HU, Lee KC, Lee MJ, Kim HY, Choi EK, Lim HK, Kim CW. Effects of increasing the dose of ropivacaine on vertical infraclavicular block using neurostimulation. Korean J Anesthesiol 2012;63:36–42 6. Weller RS, Gerancher JC. Brachial plexus block: “best” approach and “best” evoked response–where are we? Reg Anesth Pain Med 2004;29:520–3 7. Liisanantti O, Luukkonen J, Rosenberg PH. High-dose bupivacaine, levobupivacaine and ropivacaine in axillary brachial plexus block. Acta Anaesthesiol Scand 2004;48:601–6 8. Rettig HC, Gielen MJ, Boersma E, Klein J. A comparison of the vertical infraclavicular and axillary approaches for brachial plexus anaesthesia. Acta Anaesthesiol Scand 2005;49: 1501–8 9. Blackwelder WC. “Proving the null hypothesis” in clinical trials. Control Clin Trials 1982;3:345–53 10. Newcombe RG. Interval estimation for the difference between independent proportions: comparison of eleven methods. Stat Med 1998;17:873–90 11. Altman D, Machin D, Bryant T, Gardner M. Statistics with Confidence. Confidence Intervals and Statistical Guidelines, 2nd ed. London: BMJ Books, 2000
anesthesia & analgesia
A Comparison of Posterior and Medial Cord Stimulation for Neurostimulation-Guided Vertical Infraclavicular Block: A Randomized Noninferiority Clinical Trial Chun Woo Yang, MD,* Sung Mee Jung, MD,† Hee Uk Kwon, MD, PhD,‡ Po Soon Kang, MD, PhD,‡ Choon Kyu Cho, MD,‡ Jin Young Oh, MD,‡ Younsuk Lee, MD, PhD,§ and Junghee Choi, MD║ BACKGROUND: We investigated whether medial cord stimulation is inferior to posterior cord stimulation for vertical infraclavicular block with respect to block success. METHODS: Ninety-six patients scheduled for upper limb surgery were randomly elicited a medial or posterior cord response for infraclavicular block using 40 mL of 0.5% ropivacaine. We assessed block success (complete sensory block of the 5 nerves in the forearm at 50 minutes) as the primary end point and block procedure characteristics and adverse events as secondary end points. RESULTS: The block success rates did not differ significantly between medial and posterior cord stimulation (95.7% [44/46] vs 91.7% [44/48], 95% CI of difference, –7.4% to 15.6%), while the secondary end points were comparable in both groups. CONCLUSIONS: Needle manipulation to elicit medial cord response is noninferior to posterior cord response of block success during neurostimulation-guided vertical infraclavicular block. (Anesth Analg 2014;118:874–8)
T
he type of distal motor response evoked is an important factor that influences the overall success of neurostimulation-guided infraclavicular block (ICB). Although previous studies1,2 have reported improved success with posterior cord stimulation over either medial or lateral cord stimulation in ICB using a lateral or pericoracoid approach, these findings might not apply to more medial vertical ICB because of the different topography of the cords between the 2 approaches.3,4 We recently showed that a median or ulnar nerve response elicited with medial cord stimulation also yielded a high success rate in vertical ICB.5 We investigated whether needle manipulation to elicit a medial cord response is inferior to elicitation of a posterior cord response in terms of block success for vertical ICB. We hypothesized that the closer anatomic association of the brachial plexus cords at the vertical infraclavicular site compared with the paracoracoid site for ICB would result in equivalent block efficacy from stimulation of either the medial or posterior cords. From the *Department of Anesthesiology and Pain Medicine, Cheju Halla General Hospital, Jeju-si, Jeju special self-governing province; †Department of Anesthesiology and Pain Medicine, Yeungnam University School of Medicine, Daegu; ‡Department of Anesthesiology and Pain Medicine, Konyang university hospital, Daejeon; §Department of Anesthesiology and Pain Medicine, Dongguk University Ilsan Hospital, Goyang; and ║Department of Anesthesiology and Pain Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea. Accepted for publication January 8, 2014. Funding: No funding. The authors declare no conflicts of interest. Reprints will not be available from the authors. Address correspondence to Sung Mee Jung, MD, Department of Anesthesiology and Pain Medicine, 170, Hyeonchung-ro, Nam gu, Daegu, 705-717, South Korea. Address e-mail to [email protected]. Copyright © 2014 International Anesthesia Research Society DOI: 10.1213/ANE.0000000000000143
874 www.anesthesia-analgesia.org
METHODS
With IRB approval and written informed consent, 96 patients scheduled for upper limb surgery were enrolled in this study. Exclusion criteria included any contraindication to regional anesthesia, pregnancy, allergy to local anesthetics, or neurological disorders. With the use of sealed opaque envelopes, patients were randomly assigned to receive a single injection ICB after needle manipulation to stimulate either the posterior (group P) or medial cord (group M). Standard noninvasive monitoring was used, and no premedication or sedation was administered before block. All blocks were performed with 40 mL of 0.5% ropivacaine by the first author before surgery in the recovery room. Patients were placed supine, with the arm adducted and the palm facing up. After identification of the landmarks, the needle insertion site was marked halfway between the jugular notch and the ventral process of the acromion. After skin disinfection, a local anesthetic skin wheal was raised at the needle entry point. Using a nerve stimulator (Stimuplex® HNS 12, B. Braun, Melsungen, Germany) initially set at 1.5 mA, 0.1 millisecond, and 1 Hz, a 22-gauge, 50-mm insulated needle (Stimuplex® A; B.Braun) was advanced directly in a posterior direction until elicitation of the desired distal motor response. To elicit the desired motor response, we used the following protocol based on a previous anatomical study:3 1. If no motor response was elicited on initial needle insertion, the needle was repositioned, with a slight deviation in the caudal, cranial, or lateral (not medial) direction until the maximum depth of 5 and 4 cm (man and woman, respectively) was reached. 2. If the first response was flexion of the elbow, pronation of the forearm, or radial movement of the hand, the needle had to be advanced deeper without angulation until stimulation of either the posterior or medial cord was obtained. April 2014 • Volume 118 • Number 4
Effect of Distal Motor Response on Infraclavicular Block
3. If the first elicited response corresponded to posterior cord stimulation in group M, the needle was repositioned in the caudal direction to elicit the medial cord response. If the first elicited response corresponded to medial cord stimulation in group P, the needle was repositioned in the cranial direction to elicit the posterior cord response. A satisfactory response to nerve stimulation was considered to occur when the desired motor response was elicited at a stimulation current between 0.2 and 0.5 mA. Finger flexion (with or without wrist flexion) was accepted as stimulation of the medial cord,6 and finger and/or wrist extension as stimulation of the posterior cord. If a satisfactory response was not obtained within 20 minutes of needle insertion, the patient was excluded from the study. Block performance time, number of needle redirections, and patient discomfort were also recorded by the anesthesia nurse assisting the anesthesiologist performing the block. Block performance time was defined as the time between the block needle insertion and needle withdrawal. The number of needle redirections was defined as either forward or backward movement of needle at least 1 cm or more. Degree of patient discomfort related to the block procedure was reported on a numerical rating scale (0–10) after completion of the block. Sensory and motor block assessment was performed by a blinded investigator after local anesthetic administration (time zero). Sensory block was assessed in the distribution of the radial, median, ulnar, musculocutaneous, and medial antebrachial cutaneous nerve by a cold test with the following scale: 0 = normal sensation, 1 = less cold, and 2 = not cold when compared with the contralateral arm every 5 minutes until 50 minutes or complete block. A complete sensory block was defined as a score of 2 in each nerve distribution. Since the complete sensory block with ICB using long-acting local anesthetics such as ropivacaine might have delayed onset and continued to evolve for as long as 1 hour,7,8 we defined the successful block as complete sensory block of the 5 nerves in the forearm within 50 minutes of injection. Failure of the block was considered if one (or more) of these sensory distributions was not blocked. The block was deemed apparently sufficient for surgery if sensory block was sufficient to tolerate surgical incision without any supplementation, despite not achieving complete sensory block of all nerves within 50 minutes. The time between the end of the injection and development of successful block was defined as block onset time. Motor block was evaluated every 10 minutes until 50 minutes or complete block using the forearm flexion, thumb abduction, thumb opposition, and thumb adduction (for the musculocutaneous, radial, median, and ulnar nerve, respectively) and scored as follows: 0 = no block, 1 = paresis, and 2 = paralysis. After a study period of 50 minutes or occurrence of a complete sensory block, the patients were taken to the operation room. If the block was adequate, supplemental sedation with propofol infusion was administered intraoperatively, according to patient preference. In the case of intraoperative pain, rescue analgesia was provided with 50 to 100 μg IV fentanyl. In case of persistent pain, general anesthesia was administered.
April 2014 • Volume 118 • Number 4
The surgical procedures, surgery time, tourniquet time, and tourniquet pain were recorded. Adverse events (vascular puncture, local anesthetic toxicity, Horner syndrome, dyspnea, and pneumothorax) were also noted.
Data Analysis
This study was conducted to determine whether needle manipulation to elicit medial cord response is inferior to that to elicit posterior cord based on successful neurostimulation-guided ICB. For noninferiority of medial cord stimulation versus posterior cord stimulation, a maximum difference of 10% (noninferiority margin) in success rate was considered acceptable. The choice of the noninferiority margin was based on clinical judgment. Assuming a success rate of 97% in both groups based on a previous study,5 the sample size for α = 0.05 and 80% power was calculated at 36 patients per group.9 To increase the power, we included 48 patients per group. The primary end point was analyzed by calculating 2-sided 95% confidence interval (CI) for the difference using Wilson method with continuity correction.10 Secondary end points were analyzed using the Mann-Whitney test
Table 1. Patient, Surgical, and Anesthetic Characteristics Age (y) Gender (M/F) Height (cm) Weight (kg) ASA physical status (I/II/III) Site of surgery Hand/forearm/elbow (n) Duration of surgery (min) Tourniquet time (min) Tourniquet pain (n)
Group P, (n = 48) 46 ± 15 28/20 164 ± 9 65 ± 11 36/11/1
Group M, (n = 46) 46 ± 15 28/18 167 ± 10 67 ± 12 34/10/1
P 0.964 0.836 0.146 0.287 0.820
25/18/5 59 ± 46 44 ± 37 3 (6.3)
22/19/5 62 ± 45 53 ± 36 3 (6.5)
0.980 0.746 0.203 1
Data are presented as the mean (SD) or numbers (%). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord; ASA = American Society of Anesthesiologists.
Figure 1. Percentage of patients with successful block over time (with 95% confidence interval). *Group M was inferior to Group P for success rate at 10 minutes: difference −16.2% (95% CI, −33.5% to 2.4%). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord. Bracketed values are the number of patients.
www.anesthesia-analgesia.org
875
E Brief Report for continuous variables and the χ2 test or Fisher exact test for categorical variables. The 95% CI of the median treatment difference in block onset time was calculated using the “Hodges-Lehman Estimator” as proposed by Altman et al.11 A 2-sided P < 0.05 was considered statistically significant. Statistical analysis was performed using the SPSS version 12 statistical software (SPSS, Chicago, IL)
Table 2. Procedural Characteristics Block performance time (min) Number of needle redirections (n) Patient discomfort (NRS 0–10) Desired motor response at first trial (n) Performance time (min) Patient discomfort (NRS 0–10)
Group P, (n = 48) 5.5 ± 2.5
Group M, (n = 46) 5.7 ± 2.8
P 0.843
3 (1–7)
3 (1–6)
0.949
2 (1–2)
2 (1–2)
0.860
16 (33.1)
18 (39.1)
0.705
4.2 ± 1.7 1 (1–3)
4.8 ± 2.9 1.5 (1–2)
0.703 1
Data are presented as the mean (SD), median (interquartile range), or numbers (%). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord; NRS = numerical rating scale.
RESULTS
Of 96 patients enrolled, 94 completed the study. Two patients in group M were excluded because of the inability to elicit the desired motor response. Patient characteristics were comparable between groups (Table 1). The success rate of group M was not inferior to that of group P after ICB (95.7% (44/46) vs 91.7% (44/48), 95% CI of difference −7.4% to 15.6%; Fig. 1). The number of patients with block sufficient for surgery and supplementation was comparable between groups (Table 3). Although there were significant differences in motor block of the radial (at 10 minutes), median (at 20 and 30 minutes) and ulnar nerve (at 20 minutes) (Fig. 2), comparable results were found for onset time (median difference 0 minute, 95% CI, −5 to 0, P = 0.71; Table 3), block performance time, number of needle redirections, patient discomfort (Table 2), efficacy of the sensory block (Fig. 3), and adverse events (Table 4) in both groups.
DISCUSSION
This study showed that needle manipulation to elicit either medial or posterior cord response provided a similar success rate block without differences in the block efficacy, performance time, and adverse events for vertical ICB. This is in contrast to previous studies that suggested posterior cord
Figure 2. Percentage of patients with complete motor block for each nerve over time (with 95% confidence interval). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord.
876 www.anesthesia-analgesia.org
anesthesia & analgesia
Effect of Distal Motor Response on Infraclavicular Block
Table 3. Block Characteristics Onset time (min) Blocks sufficient for surgery (n) Supplementation (n) Failure (n)
Group P, (n = 48) 10 (5–10) 46 (95.8) 2 (4.2) 0 (0)
Table 4. Adverse Events Group M, (n = 46) 10 (5–15) 45 (97.8) 1 (2.2) 0 (0)
P 0.71 1
Data are presented as the number of patients (%) or median (interquartile range). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord.
stimulation as the appropriate end point for the paracoracoid approach to ICB.1,2 This discrepancy might have resulted from the difference in topographical anatomy of cords between vertical and paracoracoid approaches of ICB. Since both the posterior and medial cord are close to each other and directly posterior (deep) to the lateral cord in the midinfraclavicular area,3 injection of local anesthetic may result in homogeneous distribution around all 3 cords with either medial or posterior stimulation elicited for vertical ICB. Another possible explanation might be the differences in the definition of medial cord stimulation. We have defined the distal motor response from medial cord stimulation as finger flexion (with or without wrist flexion). The median nerve has a dual-cord contribution from the lateral and medial cords. With the common plexus anatomy, finger flexion most likely identifies medial cord or medial root to median nerve stimulation, but wrist flexion may result from either medial or lateral cord stimulation or stimulation of the medial or lateral median nerve roots.6 Finally, the relatively large volume of local anesthetics used might have been responsible for our results. Our study has some limitations. First, it is possible that needle manipulations to search for specific distal motor response
Vascular puncture Horner syndrome Dyspnea
Group P, (n = 48) 0 (0) 0 (0) 0 (0)
Group M, (n = 46) 3 (6.5) 1 (2.2) 1 (2.2)
P 0.113 0.489 0.489
Data are presented as the number of patients (%). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord.
might influence the spread of local anesthetics. Second, the assessment time of 50 minutes may account for the favorable results in both groups, at least in part. This long period of assessment may not be acceptable in clinical practice. Finally, we did not evaluate the influence of specific cord stimulation on postoperative outcomes such as block duration, pain score, time to first analgesia, and opioid consumption. In conclusion, this study demonstrated that needle manipulation to elicit either medial or posterior cord response provided similar block-related characteristics during neurostimulation-guided vertical ICB. E DISCLOSURES
Name: Chun Woo Yang, MD. Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript. Attestation: Chun Woo Yang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Sung Mee Jung, MD. Contribution: This author helped design the study, analyze the data, and write the manuscript. Attestation: Sung Mee Jung has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Figure 3. Percentage of patients with complete sensory block for each nerve over time (with 95% confidence interval). Group P = stimulation of the posterior cord; Group M = stimulation of the medial cord; MAC = medial antebrachial cutaneous nerve.
April 2014 • Volume 118 • Number 4
www.anesthesia-analgesia.org
877
E Brief Report Name: Hee Uk Kwon, MD, PhD. Contribution: This author helped analyze the data and write the manuscript. Attestation: Hee Uk Kwon has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Po Soon Kang, MD, PhD. Contribution: This author helped analyze the data and write the manuscript. Attestation: Po Soon Kang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Choon Kyu Cho, MD. Contribution: This author helped analyze the data and write the manuscript. Attestation: Choon Kyu Cho has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Jin Young Oh, MD. Contribution: This author helped collect the data. Attestation: Younsuk Lee has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Younsuk Lee, MD, PhD. Contribution: This author helped analyze the data. Attestation: Younsuk Lee has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Junghee Choi, MD. Contribution: This author helped analyze the data. Attestation: Junghee Choi has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. This manuscript was handled by: Terese T. Horlocker, MD.
878 www.anesthesia-analgesia.org
REFERENCES 1. Bloc S, Garnier T, Komly B, Leclerc P, Mercadal L, Morel B, Dhonneur G. Single-stimulation, low-volume infraclavicular plexus block: influence of the evoked distal motor response on success rate. Reg Anesth Pain Med 2006;31:433–7 2. Lecamwasam H, Mayfield J, Rosow L, Chang Y, Carter C, Rosow C. Stimulation of the posterior cord predicts successful infraclavicular block. Anesth Analg 2006;102:1564–8 3. Moayeri N, Renes S, van Geffen GJ, Groen GJ. Vertical infraclavicular brachial plexus block: needle redirection after elicitation of elbow flexion. Reg Anesth Pain Med 2009;34:236–41 4. Sauter AR, Smith HJ, Stubhaug A, Dodgson MS, Klaastad Ø. Use of magnetic resonance imaging to define the anatomical location closest to all three cords of the infraclavicular brachial plexus. Anesth Analg 2006;103:1574–6 5. Yang CW, Kang PS, Kwon HU, Lee KC, Lee MJ, Kim HY, Choi EK, Lim HK, Kim CW. Effects of increasing the dose of ropivacaine on vertical infraclavicular block using neurostimulation. Korean J Anesthesiol 2012;63:36–42 6. Weller RS, Gerancher JC. Brachial plexus block: “best” approach and “best” evoked response–where are we? Reg Anesth Pain Med 2004;29:520–3 7. Liisanantti O, Luukkonen J, Rosenberg PH. High-dose bupivacaine, levobupivacaine and ropivacaine in axillary brachial plexus block. Acta Anaesthesiol Scand 2004;48:601–6 8. Rettig HC, Gielen MJ, Boersma E, Klein J. A comparison of the vertical infraclavicular and axillary approaches for brachial plexus anaesthesia. Acta Anaesthesiol Scand 2005;49: 1501–8 9. Blackwelder WC. “Proving the null hypothesis” in clinical trials. Control Clin Trials 1982;3:345–53 10. Newcombe RG. Interval estimation for the difference between independent proportions: comparison of eleven methods. Stat Med 1998;17:873–90 11. Altman D, Machin D, Bryant T, Gardner M. Statistics with Confidence. Confidence Intervals and Statistical Guidelines, 2nd ed. London: BMJ Books, 2000
anesthesia & analgesia
