Intervention Review

You have free access to this content

Infraclavicular brachial plexus block for regional anaesthesia of the lower arm

  1. Ki Jinn Chin1,*,
  2. Husni Alakkad1,
  3. Sanjib D Adhikary2,
  4. Mandeep Singh1

Editorial Group: Cochrane Anaesthesia Group

Published Online: 28 AUG 2013

Assessed as up-to-date: 7 JUN 2013

DOI: 10.1002/14651858.CD005487.pub3


How to Cite

Chin KJ, Alakkad H, Adhikary SD, Singh M. Infraclavicular brachial plexus block for regional anaesthesia of the lower arm. Cochrane Database of Systematic Reviews 2013, Issue 8. Art. No.: CD005487. DOI: 10.1002/14651858.CD005487.pub3.

Author Information

  1. 1

    University of Toronto, Department of Anesthesia, Toronto, Ontario, Canada

  2. 2

    Penn State College of Medicine, Milton S Hershey Medical Centre, Department of Anesthesiology, Hershey, Pennsylvania, USA

*Ki Jinn Chin, Department of Anesthesia, University of Toronto, Toronto Western Hospital, 399 Bathurst Street, Toronto, Ontario, M5T 2S8, Canada. gasgenie@gmail.com.

Publication History

  1. Publication Status: New search for studies and content updated (conclusions changed)
  2. Published Online: 28 AUG 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Arcand 2005

MethodsRandomized controlled trial


ParticipantsN = 80; adult; BMI <35; 56 male, 24 female; ASA 1-3; surgery of the distal arm/forearm/hand; Canadian study


Interventions
  1. Infraclavicular block (De Andres 2002). Localization method: ultrasound-guided. Endpoint for injection: ultrasound + motor response to neurostimulation at <0.6mA. Single injection through needle.
  2. Supraclavicular block (De Andres 2002). Localization method: ultrasound-guided. Endpoint for injection: ultrasound + motor response to neurostimulation at <0.6mA. Single injection through needle.


Injectate in both blocks: bupivacaine 0.5% and lidocaine 2% in 1:3 ratio with 1:200,000 epinephrine in volume of 0.5 ml/kg to a maximum of 40 ml

Sedation for block: IV midazolam 0.5-2 mg and fentanyl 25-100 μg as needed

Intraoperative sedation: propofol infusion when needed


Outcomes
  1. Surgical anaesthesia, defined as surgery without patient discomfort or need for supplementation
  2. Sensory block to cold, scored as no/partial/complete block, at 5 min intervals over 30 min, in individual nerve territories (RN/UN/MN/MCN)
  3. Motor block, scored as no/partial/complete block, at 5 min intervals over 30 min
  4. Block performance time, defined as time between needle insertion and removal. If time >20 min, this was taken as block failure
  5. Block-associated pain, scored as visual analogue score (VAS) 0-10
  6. Duration of postoperative analgesia, defined as time between block completion and 1st postoperative analgesic medication
  7. Complications - only neuropathy and pneumothorax specifically mentioned. Chest x-ray performed only if respiratory distress occurred. Telephone interview at one week with specific questions regarding prolonged neurological deficit or symptoms, and respiratory difficulty


NotesBlocks performed by single physician, a resident with previous experience of 11 blocks in each technique.

Sensory block in individual territories at specific time points is defined for this review as no sensation (rather than blunt or no sensation), so as to be consistent with studies using a 2-point (all-or-none) scale of block intensity.

No dichotomous data on block-associated pain was available.

There were 2 failures to perform block (unable to visualize plexus) in the supraclavicular group; 1 failure to perform block (unable to obtain stimulation) in the infraclavicular group. These were included only in the analysis of block performance time. Other outcomes analysed on an available-case basis (infraclavicular n=39, supraclavicular n = 38).

Block performance time was only reported as mean values and subdivided by early or late stage of study. There was evidence of a learning effect on the block performance time, which became shorter as the study progressed.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo mention of randomization method in text

Allocation concealment (selection bias)Unclear riskNo mention of randomization method in text

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNo mention of blinding of patients or outcome assessors in text

Incomplete outcome data (attrition bias)
All outcomes
Low riskFailure to perform block in two patients (infraclavicular) and one patient (supraclavicular). Only block performance time, and no other outcomes, were reported for these patients

Selective reporting (reporting bias)Low riskAll pre-specified and relevant outcomes reported

Other biasUnclear riskA weight-based formula was used to calculate local anaesthetic volume: 0.5ml/kg up to a maximum of 40ml. The use of lower volumes (<40ml) may have reduced infraclavicular block success

Caruselli 2005

MethodsRandomized controlled trial


ParticipantsN = 36; paediatric; ASA 1; average age 5 yrs; 23 male, 13 female; emergency surgery for trauma to arm / elbow; Italian study


Interventions
  1. Infraclavicular block (Raj 1973). Localization method:surface landmarks - intersection of clavicle and line between Chassignac's tubercle and the axillary arterial pulsation. Endpoint for injection: motor response to neurostimulation at 0.3-0.5mA. Single injection through needle.
  2. Modified parascalene block (Dalens 1987). Localization method: surface landmarks - junction of lower third and upper two thirds of a line between Chassignaac's tubercle and the midpoint of the clavicle. End point for injection: motor response to neurostimulation at 0.3-0.5mA. Single injection through needle.


Injectate in both blocks: ropivacaine 2.7 mg/kg in volume of 0.5 ml/kg

Sedation for block: oral midazolam 0.3mg/kg 30 minutes prior to block, and IV midazolam 0.1mg/kg, ketamine 1 mg/kg, propofol 1 mg/kg just before block

Intraoperative sedation: none


Outcomes
  1. Block quality, defined as A = no sign of discomfort, complete motor block; B = required systemic IV analgesics; C = required general anaesthesia, and assessed 15 min after block completion
  2. Complications (unspecified)


NotesBlock quality A was taken as equivalent to the definition of surgical anaesthesia.

Block success was assessed at 15 min, earlier than in other studies (30 or 60 min). This may have reduced success rates.

Given that complications were unspecified, but that 1 case of Horner's syndrome was reported, data was only entered for the Horner's syndrome.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo mention of the method of random sequence generation

Allocation concealment (selection bias)Low riskQuote: " progressively numbered closed envelopes"

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNo mention of blinding of patients or outcome assessors in text

Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete follow-up for all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasUnclear riskTiming for measurement of outcomes may have been inappropriate - 15 minutes is too short a time for assessment of block efficacy in the reviewers' opinion

De Jose Maria 2008

MethodsRandomized controlled trial


ParticipantsN = 80; paediatric; ASA 1-2; average age 8-9 yrs; 56 male, 24 female; elective upper limb surgery, Spanish study


Interventions
  1. Infraclavicular block. Localization method: ultrasound using an out-of-plane approach with probe placed parallel to clavicle. Endpoint for injection: local anaesthetic spread around the plexus. Single-injection through needle.
  2. Supraclavicular block. localization method: ultrasound using an in-plane approach with probe in a coronal-oblique plane. Endpoint for injection: local anaesthetic spread around the plexus. Single-injection through needle.


Injectate in both blocks: ropivacaine 0.5% in a volume up to 0.5ml/kg. Total volume was that needed to achieve adequate local anaesthetic spread around the plexus. This was 6±2 ml in the supraclavicular group

Sedation for block and intraoperatively: all patients received a general anaesthetic before the block: 1 MAC sevoflurane in 50% oxygen/air. Fentanyl was given if heart rate or blood pressure increased by 10% or more


Outcomes
  1. Block success, defined as no need for intraoperative fentanyl
  2. Block performance time, defined as time from first needle insertion to removal at end of the block. Reported as mean (range)
  3. Duration of sensory block, defined as time from brachial plexus puncture to first dose of rescue analgesia after the first 4 postoperative hours
  4. Duration of motor block, defined as time from brachial plexus puncture to first movement of fingers or wrist
  5. Complications: pneumothorax, Horner's syndrome, neurological deficits. The last was assessed at a 1 week follow-up visit


NotesSurgery started within 15 min of block. This may have reduced success rates.

Two patients in the infraclavicular group had block procedure abandoned after arterial puncture. All outcomes except vascular puncture were analysed on an available-case basis (N=38).

Block performance times reported as mean, range and 95% CI for difference of means. Standard deviation calculated from this data, and also requested from author.

Duration of sensory block was only reported for the supraclavicular group. Additional data requested from author.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: “80 children…were prospectively randomized into two groups…”

No further details provided

Allocation concealment (selection bias)Unclear riskNot stated in text

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNo mention of blinding of patients or outcome assessors in text

Incomplete outcome data (attrition bias)
All outcomes
Low riskTwo patients in the infraclavicular group had block procedure abandoned after arterial puncture. All outcomes except vascular puncture were analysed on an available-case basis (N=38)

Selective reporting (reporting bias)Unclear riskNo data reported on duration of sensory and motor block, or on volume of local anaesthetic injected in the infraclavicular group. These are outcomes of lesser importance

Other biasHigh riskTiming for measurement of outcomes may have been inappropriate - 15 min is too short a time for assessment of block efficacy in the reviewers' opinion

The ultrasound-guided infraclavicular technique was an unusual one – out-of-plane – which may have contributed to the incidence of vascular puncture, which in turn led to abandonment and classification of the block as “failed” in these patients

Deleuze 2003

MethodsRandomized controlled trial


ParticipantsN = 100; adult; 56 male, 44 female; ASA 1-2; orthopaedic surgery of forearm/wrist/hand; French study


Interventions
  1. Infraclavicular block (Jandard 2002). Localization method: surface landmarks - slightly above and medial to coracoid process. Endpoint for injection: motor response in hand or wrist to neurostimulation at <0.6mA. SIngle injection through needle.
  2. Axillary block, perivascular. Localization method: above and below arterial pulsation in the axilla. Endpoint for injection: motor response to neurostimulation at <0.6mA. Triple injection through needle (median or ulnar + radial + musculocutaneous nerve), equally divided.


Injectate in both blocks: ropivacaine 0.75% in a volume of 40 ml. For axillary block, 10 ml per stimulation and 10 ml subcutaneously to block medial antebrachial (MABCN) and brachial cutaneous nerves (MBCN)

Sedation for block: IV midazolam 1mg

Intraoperative sedation: none


Outcomes
  1. Block success, defined as loss of cold and pinprick sensation in all of 5 nerve distributions (MCN, MN, UN, RN, MABCN) within 30 min
  2. Sensory loss (cold and pinprick), scored as complete or none
  3. Motor block, scored as complete or none
  4. Onset time of sensory and motor block of individual nerves (MCN, MN, UN, RN)
  5. Complications: pneumothorax, vascular puncture, intravascular injection, Horner's syndrome. All assessed clinically at 24 h


NotesPatients without a complete sensory block in all terminal nerve distributions at 30 min post-block had supplementation prior to surgery, and general anaesthesia if there was pain during surgery.

No mention of whether patients received any postoperative follow up or assessment in the methods but the results section states that "no side effects of complications....were noticed after 24 hours of clinical assessment".

No numerical data on incidence of sensory and motor block in individual nerve distributions at 30 min, nor the incidence of complete sensory block at intervals 5-20 min, although information is presented as a graph. Authors were contacted but no reply was received.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo mention of randomization method in text

Allocation concealment (selection bias)Unclear riskNo mention of randomization method in text

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNo mention of blinding of patients or outcome assessors in text

Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete follow up for all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasLow riskNone identified

Ertug 2005

MethodsRandomized controlled trial


ParticipantsN = 30; adult, no sex distribution data; ASA 1-2; orthopaedic surgery of forearm/hand; Turkish study


Interventions
  1. Infraclavicular block. Localization method: surface landmarks - 1cm below and 2cm medial to coracoid process. Endpoint for injection: motor response in hand or wrist to neurostimulation at ≤0.5mA. Single-injection through an indwelling catheter.
  2. Axillary block, perivascular. Localization method: surface landmarks - above arterial pulsation in the axilla. Endpoint for injection: motor response in hand or wrist to neurostimulation at ≤0.5mA. Single-injection through an indwelling catheter.


Injectate in both blocks: bupivacaine 0.375% in volume of 40ml through non-stimulating catheter

Sedation for block: IV midazolam 2mg

Intraoperative sedation: none


Outcomes
  1. Block success, defined as loss of cold and pinprick in all 4 nerve territories (MCN, MN, RN, UN) within 30 min
  2. Onset time of sensory and motor block in individual nerve territories (MCN, MN, RN, UN)
  3. Duration of sensory and motor block (MCN, MN, RN, UN), defined as complete regression
  4. Time required to place catheter
  5. Tolerance of turnstile (tourniquet)
  6. Complications at catheter insertion site (infection, haematoma, other) after 48h. Catheters were removed at 24h. Neurological complications assessed at 2 weeks


NotesBlock performance time taken as time required to place catheter.

Duration of block was taken as time to complete regression, not time to first postoperative analgesia.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo mention of randomization method in text

Allocation concealment (selection bias)Unclear riskNo mention of randomization method in text

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNo mention of blinding of patients or outcome assessors in text

Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete follow up for all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasLow riskNone identified

Fleischmann 2003

MethodsRandomized controlled trial


ParticipantsN = 40; paediatric; ASA 1; average age 6 yrs; 22 male, 18 female; emergency forearm/hand surgery following trauma, Austrian study


Interventions
  1. Infraclavicular block (Kapral 1999). Localization method: surface landmarks - 0.5cm below coracoid process. Endpoint for injection: motor response in hand or wrist to neurostimulation at 0.3mA. Single injection through needle.
  2. Axillary block, perivascular. Localization method: surface landmarks - between coracobrachialis muscle and axillary artery. Endpoint for injection: motor response in hand or wrist to neurostimulation at 0.3mA. Single injection through needle.


Injectate in both blocks: ropivacaine 0.5% in volume of 0.5ml/kg

Sedation for block: rectal midazolam 1mg/kg as premedication for children aged 1-6 yrs, and IV midazolam 0.05-0.1 mg/kg ± propofol 0.1-0.5 mg/kg just before the block

Intraoperative sedation: none


Outcomes
  1. Block success, defined as at least 2 out of 4 nerve territories (UN, MN, RN, MCN) blocked (Vester-Andersen's criteria (Vester-Andersen 1984)) and no pain at start of surgery
  2. Sensory block to pinprick, scored as pain or no pain, in 6 individual nerve territories (AxN, MCN, MBCN, MACN, RN, MN, UN) at 30 min
  3. Sensory block onset, defined as time from puncture until pain visual analogue scores (VAS) = 1/5 (evaluated q2min)
  4. Sensory block duration, defined as time between puncture and first postoperative analgesic (administered when VAS >3/5)
  5. Sensory block quality, defined as VAS at baseline versus 30 min after puncture versus intraoperative
  6. Motor block at 30 min, scored as normal/reduced/no contraction
  7. Complications (pneumothorax, vascular puncture)
  8. Pain VAS before, during and 30 min after needle puncture


NotesVAS was only assessed in children aged 3 or older. This was only possible in 15/20 infraclavicular and 16/20 axillary blocks. Onset and duration data are therefore only available for these patients.

Numerical data for VAS was not available as it was presented as a graph. Authors were contacted but without reply.

Detection of complications was by auscultation for pneumothorax, with CXR if inequality detected; and examination of puncture site for haematoma/swelling.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "The randomization protocol was prepared outside of the study centre and delivered in sealed, opaque and sequentially numbered envelopes"

Allocation concealment (selection bias)Low riskQuote: "The randomization protocol was prepared outside of the study centre and delivered in sealed, opaque and sequentially numbered envelopes"

Blinding (performance bias and detection bias)
All outcomes
Low riskQuote: "Sensory and motor blockade was assessed by a staff anaesthesiologist not otherwise involved in the study." In the reviewers' opinion, this constitutes sufficient blinding of the outcome assessor

Incomplete outcome data (attrition bias)
All outcomes
Low riskInability to measure VAS in children aged less than 3 meant that outcomes of onset and duration of sensory block were not assessed in all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasLow riskNone identified

Frederiksen 2010

MethodsRandomized controlled trial


ParticipantsN = 80; adult (19-80 yrs); 36 male, 44 female; elective or emergency elbow/forearm/hand surgery, Danish study


Interventions
  1. Infraclavicular block (Dingemans 2007). Localization method: ultrasound-guided. Endpoint for injection: U-shaped distribution of local anaesthetic posterior to axillary artery.
  2. Axillary block (Chan 2007). Localization method: ultrasound-guided. Endpoint for injection: Local anaesthetic distribution around visible nerves, or if nerves not visible, around the axillary artery. Injectate divided equally between each of 4 nerves: median, ulnar, radial, musculocutaneous.


Injectate in both blocks: mixture of ropivacaine 0.75% and mepivacaine 2% in a 1:1 ratio, with 1:200,000 epinephrine, in volume of 0.5ml/kg (maximum of 50 ml)

Sedation for block: none

Intraoperative sedation: intravenous midazolam if requested by patient


Outcomes
  1. Adequate surgical anaesthesia, defined as a sensory score of 1 or 2 in nerves distal to elbow at 30 minutes
  2. Complete motor block, defined as no movement in hand/wrist/elbow
  3. Number of needle passes
  4. Block performance time
  5. Block onset time
  6. Time to readiness for surgery, defined as sum of block performance and onset times
  7. Block-associated pain, scored on a 0-100 VAS
  8. Preference for a similar block in the future
  9. Complications: pneumothorax, dysaesthesiae, vascular puncture


NotesStudy authors state that the infraclavicular block is the preferred approach in their institution and that limited experience with the axillary block may have affected outcomes.

A weight-based formula was used to calculate local anaesthetic volume: 0.5ml/kg up to a maximum of 50ml. The use of lower volumes (<40ml) may have reduced infraclavicular block success.

Mean and standard deviation data for some outcomes was obtained from the corresponding author by email correspondence.

Abbreviations: VAS = visual analogue score


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: “computer-generated random number”

Allocation concealment (selection bias)Low riskQuote: “sealed envelope method”

Blinding (performance bias and detection bias)
All outcomes
Low riskQuote: “a blinded observer” performed all assessments, and performed all supplementary blocks

Only the outcome assessor was blinded. Further blinding not feasible. The review authors do not believe this will introduce significant bias

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing data

Selective reporting (reporting bias)Low riskThe data for surgical and sensory block success were not explicitly reported in the published study, but this information was obtained from the study authors

Other biasHigh riskQuote: “The infraclavicular approach is our standard technique for hand/forearm surgery, although the AX block is more often used in arthritic, obese or muscular patients. We teach our residents the IC block as the primary approach. Most blocks in this study were performed by the residents, although the number of AX blocks performed by the residents in this study is lower than the number of IC blocks. Their limited experience with the axillary approach might have affected the performance times and numbers of needle passes. All incomplete AX blocks occurred among the first half of the patients included, which supports this speculation”

Fredrickson 2009

MethodsRandomized controlled trial


ParticipantsN = 60; adult; 14 male, 46 female; elective wrist/hand surgery, New Zealand study


Interventions
  1. Infraclavicular block (Dingemans 2007). Localization method: ultrasound-guided. Endpoint for injection: visualization of local anaesthetic spread. Triple-injection technique, 50% of local anaesthetic at posterior aspect of axillary artery, 25% at lateral cord, 25% between artery and vein. Patient's arm adducted, curvilinear probe used, 16-18G Tuohy needle.
  2. Supraclavicular block. Localization method: ultrasound-guided. Endpoint for injection: visualization of local anaesthetic spread. Dual-injection technique, 50% of local anaesthetic at 'corner pocket' between subclavian artery and first rib, 50% at a point superior to the artery. Linear or curvilinear probe used, 16-18G Tuohy needle.


Injectate in both blocks: 30 ml of lidocaine 2% with epinephrine 1:200,000 (25 ml if patient's weight <60kg)

Sedation for block: intravenous midazolam 2mg and alfentanil 0.5mg

Intraoperative sedation: intravenous midazolam or propofol infusion if requested by patient


Outcomes
  1. Adequate surgical anaesthesia, defined as completion of surgery without requirement for supplementation or general anaesthesia
  2. Complete sensory block in terminal nerves
  3. Block performance time, defined as time from ultrasound probe placement to needle withdrawal from the skin
  4. Block onset time, defined as time to complete sensory block success
  5. Block-associated pain, scored on a 0-10 numerical rating scale
  6. Complications: pneumothorax (clinical diagnosis), vascular puncture, neurological symptoms at 10 days, local anaesthetic systemic toxicity


NotesAt skin closure, 5-10 ml of 0.5% bupivacaine was injected into the wound or around the median/radial or ulnar nerve; if >2 nerves needed to be blocked, an infraclavicular injection of 10-20 ml of ropivacaine 0.5% was injected (this means that block duration or analgesic duration cannot be assessed in this study).

The scanning time and needling time were reported separately, and as medians with interquartile ranges. We obtained total block performance time data, including mean and standard deviations, from the lead author by email correspondence.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "a computer-generated random number"

Allocation concealment (selection bias)Unclear riskInsufficient information provided to make a judgement

Blinding (performance bias and detection bias)
All outcomes
Low riskA blinded observer assessed sensory and motor block and block-associated pain and follow-up at 10 days

Blinding of patient and operator not feasible and unlikely to influence study outcomes

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing data

Selective reporting (reporting bias)Low riskAll pre-specified and relevant outcomes reported

Other biasUnclear riskThis was a single-operator study, and it is possible that there may have been preference for one approach over another. There is insufficient information to make a judgement

Heid 2005

MethodsRandomized controlled trial


ParticipantsN = 60; adult; 36 male, 24 female; ASA 1-3; upper limb surgery distal to elbow, German study


Interventions
  1. Infraclavicular block (Kilka 1995). Localization method: surface landmarks - midpoint of line between the acromion and the jugular notch. Endpoint for injection: motor response in hand to neurostimulation at <0.5mA. Single injection through needle.
  2. Axillary block, perivascular. Localization method: surface landmarks - between coracobrachialis muscle and the axillary artery. Endpoint for injection: motor response in hand to neurostimulation at <0.5mA. Single injection through a proximally-threaded catheter.


Injectate in both blocks: ropivacaine 0.75% in volume of 40 ml

Sedation for block: none

Intraoperative sedation: IV midazolam 1-2mg, propofol infusion 1-2 mg/kg/h if requested by patient


Outcomes
  1. Incidence of successful sensory block, defined as no sensation to pinprick, in 7 individual nerve distributions (AxN, MBCN, MABCN, MCN, MN, RN, UN) at 5 min intervals to 60 min
  2. Degree of sensory block, scored as 0 (none), 1 (partial), 2 (complete) in each nerve distribution. Sensory score (Koscielniak-N 2000) out of 14 computed at each time point
  3. Block completeness time, defined as time to complete sensory block in 50% of patients
  4. Complications (vascular puncture, seizure, dysrhythmias, oxygen saturation <90%, Horner's syndrome)


NotesSensory block in individual territories at specific time points is defined for this review as no sensation (rather than blunt or no sensation), so as to be consistent with studies using a 2-point (all-or-none) scale of block intensity.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "allocated by a computed randomization list"

Allocation concealment (selection bias)Unclear riskNo mention of allocation concealment

Blinding (performance bias and detection bias)
All outcomes
Low riskQuote: "a blinded observer evaluated sensory block". Measures were taken to ensure blinding including a sham catheter

Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete follow-up for all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasLow riskNone identified

Kapral 1999

MethodsRandomized controlled trial


ParticipantsN = 40; adult; 22 male, 18 female; ASA 1-2; surgery of forearm and hand following trauma; Austrian study


Interventions
  1. Infraclavicular block (Kapral 1999). Localization method: below coracoid process. Endpoint for injection: lateral cord motor response to neurostimulation at 0.5mA. Single-injection through needle.
  2. Axillary block, perivascular. Localization method: surface landmarks - next to arterial pulsation in the axilla. Endpoint for injection: motor response in hand or wrist to neurostimulation at ≤0.5mA. Single-injection through needle.


Injectate in both blocks: mepivacaine 1% in volume of 40ml

Sedation for block: none

Sedation for block: none


Outcomes
  1. Block success, defined as at least 2 out of 4 nerve territories (UN, MN, RN, MCN) blocked (Vester-Andersen's criteria (Vester-Andersen 1984))
  2. Sensory block to pinprick, scored on a 0-100% scale, in individual nerve territories (RN/UN/MN/MCN)
  3. Motor block of individual nerves (RN/UN/MN/MCN)
  4. Duration of block, defined as time between administration of local anaesthetic and the recurrence of pain
  5. Complications (pneumothorax and vascular puncture mentioned in the text)


NotesSensory and motor block assessed at 0,5,10,30,60,120,180, 360 min but incidences of block at each time point were not published. Authors were contacted but without reply.

A chest x-ray was done to look for pneumothorax in each patient.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo mention of randomization method in text

Allocation concealment (selection bias)Unclear riskNo mention of randomization method in text

Blinding (performance bias and detection bias)
All outcomes
Low riskQuote: "Evaluation of all blocks of the study was done by the same anaesthesiologist...who was not involved in the performance of the block"

Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete follow-up for all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasLow riskNone identified

Koscielniak-N 2000

MethodsRandomized controlled trial


ParticipantsN = 60; adult; 40 male, 19 female; ASA 1-2; surgery of forearm, wrist or hand; Danish study


Interventions
  1. Infraclavicular block. Localization method: surface landmarks - 2-3cm below medial border of the coracoid process. Endpoint for injection: motor response in 2 out of 4 nerve distributions (MCN/MN/UN/RN) to neurostimulation at 0.3-0.5mA. Dual-injection through needle, equally divided.
  2. Axillary block, perivascular. Localization method: surface landmarks - above arterial pulsation in the axilla. Endpoint for injection: motor response in all 4 nerve distributions (MCN/MN/UN/RN) to neurostimulation at 0.3-0.5mA. Quadruple-injection through needle, equally divided.


Injectate in both blocks: ropivacaine 0.75% by volume according to weight (ml/kg): 20ml /<50kg, 30ml/50-100kg, 40ml/>100kg

Sedation for block: oral diazepam 0.1-0.15 mg/kg on the morning of surgery

Intraoperative sedation: IV midazolam 1-2.5 mg increments if requested. Alfentanil 0.5-1mg if there was tourniquet pain


Outcomes
  1. Sensory block (to pinch with a plastic clamp), scored as 0 (none), 1 (analgesia), 2 (anaesthesia) in each nerve distribution (RN/UN/MN/MCN/AN/MBCN/MABCN). Assessed every 5 min up to 30 min post-block
  2. Incidence of sensory block, defined as analgesia or anaesthesia, in each nerve distribution (RN/UN/MN/MCN/AN/MBCN/MABCN) at 30 min post-block
  3. Incidence of surgical block, defined as analgesia or anaesthesia in 5 terminal nerve distributions (MCN/MACN/MN/UN/RN)
  4. Motor block intensity in hand, scored as good (limp hand), satisfactory (minor movement of digits), poor (no relaxation)
  5. Block performance time, defined as end of skin preparation to completion of the last injection
  6. Block onset (latency) time, defined as end of block performance to appearance of surgical block (see above)
  7. Block duration, defined as end of block performance to onset of pain or touch perception in fingers
  8. Pain during block performance, assessed by visual analogue score (VAS)
  9. Surgical pain, assessed by VAS
  10. Tourniquet pain, assessed by VAS
  11. Complications (vascular puncture, systemic toxicity) during and after block performance, including neurological sequelae (defined as persistent pain or dysaesthesiae) at surgical follow-up visits (5-10 days, 3-4 weeks)


NotesOne patient in the axillary group was excluded after randomization was excluded because of inability to lie still during the block.

Sensory block in individual territories at specific time points is defined for this review as no sensation (rather than blunt or no sensation), so as to be consistent with studies using a 2-point (all-or-none) scale of block intensity.

Additional data was requested and received from the author: mean and standard deviation of block-associated pain VAS, mean and standard deviation of block performance time, block onset time, and block duration.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "computer generated random numbers"

Allocation concealment (selection bias)Low riskQuote: "a closed envelope method"

Blinding (performance bias and detection bias)
All outcomes
Low riskQuote: "assessed by an anaesthetist who was unaware of the applied block"

Incomplete outcome data (attrition bias)
All outcomes
Low riskOne patient was excluded after randomization due to an inability to lie still for the block. No outcome data were available for this patient

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasUnclear riskA weight-based formula was used to calculate local anaesthetic volume: <50kg=20ml, 50-100kg=30ml, and <100kg=40ml. The use of lower volumes (<40ml) may have reduced infraclavicular block success

Koscielniak-N 2005

MethodsRandomized controlled trial


ParticipantsN = 80; adult; 48 male, 32 female; ASA 1-2; fast-track surgery of forearm, wrist or hand; Danish study


Interventions
  1. Infraclavicular block (Klaastad 2004). Localization method: surface landmarks - medial to the coracoid process. Endpoint for injection: motor response at hand or wrist to neurostimulation at 0.3-0.5mA. Single injection through needle.
  2. Axillary block, perivascular. Localization method: surface landmarks - above and below arterial pulsation in the axilla. Endpoint for injection: motor response in all 4 nerve distributions (MCN/MN/UN/RN) to neurostimulation at 0.3-0.5mA. Quadruple injection through needle.


Injectate in both blocks: ropivacaine 0.75% + mepivacaine 2% + epinephrine 1:200,000 in volumes of 0.5ml/kg (30-50ml)

Sedation for block: IV fentanyl 1 μg/kg

Intraoperative sedation: none


Outcomes
  1. Pain associated with block, and further subdivided according to components (needle passes, LA injection, neurostimulation), and assessed by visual analogue score (VAS)
  2. Incidence of sensory block, defined as analgesia or anaesthesia , in each nerve distribution (RN/UN/MN/MCN/AN/MBCN/MABCN) at 30 min post-block
  3. Incidence of surgical block, defined as analgesia or anaesthesia in 5 terminal nerve distributions (MCN/MACN/MN/UN/RN)
  4. Block performance time
  5. Block onset time
  6. Complications of block (pain, vascular puncture, pneumothorax, systemic toxicity)
  7. Number of patients requesting a similar block in the future


NotesSensory block in individual territories at specific time points is defined for this review as no sensation (rather than blunt or no sensation), so as to be consistent with studies using a 2-point (all-or-none) scale of block intensity.

Definitions were not explicit in this study, but presumed to be similar to that of their previous study in 2000.

Pain during block for this review was taken to include the occurrence of painful paraesthesia.

Block-associated pain VAS was expressed as median and ranges, not mean and SD. Authors were contacted for the additional information but without reply.

Assessment of pneumothorax was done by asking patients to contact hospital if they experienced unilateral chest pain or dyspnoea.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "randomized using computer-generated random numbers"

Allocation concealment (selection bias)Low riskNo mention of allocation concealment, but based on other studies by the same author, this is likely to have been done

Blinding (performance bias and detection bias)
All outcomes
Low riskQuote: "assessments by an anaesthesiologist unaware of the primary blocking technique"

Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete follow-up for all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasUnclear riskA weight-based formula was used to calculate local anaesthetic volume: 0.5 ml/kg, range 30-50 ml. The use of lower volumes (<40ml) may have reduced infraclavicular block success

Koscielniak-N 2009

MethodsRandomized controlled trial


ParticipantsN = 120; adult (age 45-51 years); 79 male, 41 female; ASA 1-2; elective or emergency surgery of elbow/forearm/hand; Danish study


Interventions
  1. Infraclavicular block (Dingemans 2007). Localization method: ultrasound-guided. Endpoint for injection: local anaesthetic spread. 50% of local anaesthetic was injected posterior to the axillary artery, and the rest was injected with needle repositioning to achieve U-shaped spread posterior to the artery.
  2. Supraclavicular block. Localization method: ultrasound-guided. Endpoint for injection: 50% of local anaesthetic was injected superficial to plexus, and the rest was injected with needle repositioning to obtain circumferential spread around the plexus/nerves.


Injectate in both blocks: mixture of ropivacaine 0.75% and mepivacaine 2% in a 1:1 ratio in a volume of 0.5 ml/kg (range 30-50 ml)

Sedation for block: intravenous fentanyl 25-50 mcg and midazolam 1-2 mg

Intraoperative sedation: none used


Outcomes
  1. Adequate surgical anaesthesia by 30 minutes, defined as anaesthesia or analgesia of the five nerves distal to the elbow.
  2. Complete sensory block of individual nerves, defined as anaesthesia or analgesia of axillary, medial brachial cutaneous, medial antebrachial cutaneous, radial, ulnar, median, musculocutaneous nerves.
  3. Block performance time, defined as from insertion of needle to removal. Does not include ultrasound scanning time.
  4. Block onset time.
  5. Complications: phrenic nerve palsy (clinical diagnosis), pneumothorax, vascular puncture, Horner's syndrome, neurological deficits, systemic toxicity.
  6. Preference for a similar anaesthetic in the future.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: “computer-generated random numbers”

Allocation concealment (selection bias)Low riskQuote: “closed envelope method”

Blinding (performance bias and detection bias)
All outcomes
Low riskQuote: “single-blinded” study

Patient, operator and block observer were unblinded. However it is not explicitly stated as to whether assessor for other outcomes was blinded. Based on previous work by primary author, this is likely to have been done

Incomplete outcome data (attrition bias)
All outcomes
Low riskQuote: “Three patients (two in the I group and one in the S group) were lost to follow-up.” This only impacts outcomes of transient neurological deficit and patient preference for a similar anaesthetic in the future. The review authors do not believe this will introduce bias

Selective reporting (reporting bias)Low riskAll major and important outcomes reported

Other biasHigh riskQuote: “We speculate that the poorer efficacy of the supraclavicular blocks in our patients was caused by lower experience with this approach and a higher number of colleagues performing the block. Our standard blocking technique for hand and/or forearm surgery is infraclavicular, although obese patients mostly receive supraclavicular or axillary blocks. Although staff anaesthesiologists listed as authors were skilled in both approaches, the infraclavicular approach is better known by other colleagues and the residents who carried out the blocks under supervision.”

Minville 2005

MethodsRandomized controlled trial


ParticipantsN = 120, adult; 68 male, 52 female; ASA 1-3; surgery of the inferior third of humerus to hand; French study


Interventions
  1. Infraclavicular block (Minville 2005). Localization method: surface landmarks - 1cm medial to the coracoid process and 1cm below the clavicle. Endpoint for injection: motor response in the distribution of MCN and 1 of 3 other nerve distributions (MN/UN/RN) to neurostimulation at 0.3-0.5mA. Dual injection through needle (10ml at MCN, 30ml at other site).2.
  2. Humeral block. Localization method: surface landmarks - axillary arterial pulsation at the junction of proximal and middle third of arm. Endpoint for injection: motor response in all 4 nerve distributions (MCN/MN/UN/RN) to neurostimulation at 0.3-0.5mA. Quadruple injection through needle, equally divided.


Injectate in both blocks: lidocaine 1.5% + epinephrine 1:200,000 in volume of 40ml

Sedation for block: IV sufentanil 0.1 μg/kg

Intraoperative sedation: none


Outcomes
  1. Block success, defined as absent sensation of cold and pinprick in 4 nerve distributions (RN, MN, UN, MCN) at 30 min
  2. Sensory block to cold and pinprick, scored as 0 (none), 1 (analgesia), 2 (anaesthesia) in each nerve distribution (RN/UN/MN/MCN/AN/MBCN/MABCN) Assessed every 5 min up to 30 min post-block
  3. Block performance time, defined as needle insertion to withdrawal
  4. Onset time of sensory block, defined as end of procedure to analgesia in 4 nerve distributions (RN, MN, UN, MCN)
  5. Motor block, scored on scale of 1-5, in 5 nerve distributions (RN, MN, UN, MCN, AN)
  6. Patient satisfaction post-surgery, scored on scale of 0-5
  7. Preference for similar block in future
  8. Complications (vascular puncture, systemic local anaesthetic toxicity, recurrent laryngeal nerve palsy, phrenic nerve palsy, paraesthesia, Horner's syndrome, pneumothorax)


NotesPatient satisfaction was converted to a dichotomous outcome as follows: satisfied = score of 4 or 5, dissatisfied = score of 3 or less.

It is not clear how they surveyed for complications. The list mentioned above was stated in the Results section and not in the Methods.

Additional data was obtained from the author on duration of sensory block (time to first request for postoperative analgesia) and the standard deviations for block onset and performance time.

Authors did not report the incidences of sensory block in the individual nerve distributions at the various time intervals. This data was not available on request.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo mention of randomization method in text

Allocation concealment (selection bias)Unclear riskNo mention of randomization method in text

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNo mention of blinding of the patient or outcome assessor in text

Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete follow-up for all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasLow riskNone identified

Minville 2006

MethodsRandomized controlled trial


ParticipantsN = 104; adult; 58 male, 46 female; ASA 1-3; trauma surgery on inferior third of humerus to hand; French study


Interventions
  1. Infraclavicular block (Minville 2005). Localization method: surface landmarks - 1cm medial to the coracoid process and 1cm below the clavicle. Endpoint for injection: motor response in the distribution of MCN and 1 of 3 other nerve distributions (MN/UN/RN) to neurostimulation at 0.3-0.5mA. Dual-injection through needle (10ml at MCN, 30ml at other site).
  2. Humeral block. Localization method: surface landmarks - axillary arterial pulsation at the junction of proximal and middle third of arm. Endpoint for injection: motor response in all 4 nerve distributions (MCN/MN/UN/RN) to neurostimulation at 0.3-0.5mA. Quadruple-injection through needle, equally divided.


Injectate in both blocks: lidocaine 1.5% + epinephrine 1:200,000 in volume of 40mls

Sedation for block: IV sufentanil 0.1 μg/kg

Intraoperative sedation: none


Outcomes
  1. Block success, defined as absent sensation of cold and pinprick in 4 nerve distributions (RN, MN, UN, MCN) at 30 min
  2. Sensory block to cold and pinprick, scored as scored as 0 (none), 1 (analgesia), 2 (anaesthesia) in each nerve distribution (RN/UN/MN/MCN/AN/MBCN/MABCN). Assessed every 5 min up to 30 min post-block
  3. Block performance time, defined as needle insertion to withdrawal
  4. Onset time of sensory block, defined as end of procedure to analgesia in 4 nerve distributions (RN, MN, UN, MCN)
  5. Duration of block (not defined)
  6. Pain associated with block, and further subdivided according to components (skin transfixion, needle redirection, local anaesthetic injection causing paraesthesia or dysaesthesia, electrolocation causing movement-associated pain), and assessed by visual analogue score (VAS)
  7. Patient satisfaction post-surgery, scored on scale of 0-5
  8. Preference for similar block in future
  9. Complications (vascular puncture, systemic local anaesthetic toxicity, recurrent laryngeal nerve palsy, phrenic nerve palsy, paraesthesia, Horner's syndrome, pneumothorax)


NotesPatient satisfaction was converted to a dichotomous outcome as follows: satisfied = score of 4 or 5, dissatisfied = score of 3 or less.

It is not clear how they surveyed for complications. The list mentioned above was stated in the Results section and not in the Methods.

Authors did not report the incidences of sensory block in the individual nerve distributions at the various time intervals. This data was not available on request.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo mention of randomization method in text

Allocation concealment (selection bias)Unclear riskNo mention of randomization method in text

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNo mention of blinding of the patient or outcome assessor in text

Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete follow-up for all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasLow riskNone identified

Niemi 2007

MethodsRandomized controlled trial


ParticipantsN = 60; adult; 40 male, 19 female; uraemic patients; AVF creation in forearm; Finnish study


Interventions
  1. Infraclavicular block (Wilson 1998). Localization method: surface landmarks - 2cm below and medial to the coracoid process. Endpoint for injection: motor response in any 1 of 4 nerve distributions (MCN/MN/UN/RN) to neurostimulation at ≤0.5mA. Single injection through needle.
  2. Axillary block, perivascular. Localization method: surface landmarks - above the arterial pulsation in the axilla. Endpoint for injection: motor response in any 1 of 3 nerve distributions (MN/UN/RN) to neurostimulation at ≤0.5mA. Single injection through needle.


Injectate in both blocks: mepivacaine 1% with epinephrine by volume according to weight (ml/kg): 35ml/40-49kg, 40ml/50- 59kg, 45ml/60-69kg, 50ml/ ≥70kg

Sedation for block: IV diazepam (unspecified dose)

Intraoperative sedation: IV midazolam (maximum total dose 5 mg), propofol infusion (low-dose) if needed for anxiolysis


Outcomes
  1. Incidence of sensory block to pinprick, scored as sharp/blunt/no sensation, in 4 nerve distributions (MN, UN, RN, MCN), at intervals between 0-60 min
  2. Motor power, scored as normal/decreased/none, of grip strength and flexion/extension at elbow, at intervals between 0-60 min
  3. Incidence of surgical block, defined as blunt or no sensation to pinprick in 4 nerve territories (MN, UN, RN, MCN) at 60 min post-block (3-point scale), recovery of motor power (by 1 grade), recovery of 2 or more sensory territories
  4. Patient satisfaction with anaesthetic technique, scored on 4-point scale, and obtained in the immediate postoperative period


NotesSensory block in individual territories at specific time points is defined for this review as no sensation (rather than blunt or no sensation), so as to be consistent with studies using a 2-point (all-or-none) scale of block intensity.

Patient satisfaction was converted to a dichotomous outcome by combining very satisfied/satisfied, and dissatisfied/quite satisfied.

It is unclear from the text whether any safety outcomes were evaluated.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "investigator-generated code"

Allocation concealment (selection bias)Low riskQuote: "investigator-generated code that was sealed in sequentially numbered opaque envelopes". In the reviewers' opinion this constitutes adequate allocation concealment

Blinding (performance bias and detection bias)
All outcomes
Low riskQuote: "assessed by an anaesthetist who was blinded regarding the block approach"

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskThe authors could not locate the brachial plexus in 1 patient in the infraclavicular group and no outcome data was available for this patient.

The intraoperative and postoperative data of a further 3 patients in the infraclavicular group was excluded because of operating room delays (two cases) and because of an unplanned general anaesthetic to allow tourniquet application (one case)

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasUnclear riskA weight-based formula was used to calculate local anaesthetic volume: 40-49kg = 35ml, 50-59kg = 40ml, 60-69kg = 45ml and >69kg = 50ml. The use of lower volumes (<40ml) may have reduced infraclavicular block success

Rettig 2005

MethodsRandomized controlled trial


ParticipantsN = 60; adult; 30 male, 30 female; ASA 1-2; surgery of the elbow, forearm or hand; Dutch study


Interventions
  1. Infraclavicular block (Kilka 1995). Localization method: surface landmarks - midpoint of line between acromion and jugular notch. Endpoint for injection: motor response in median nerve distribution to neurostimulation at ≤0.5mA. Single injection through needle.
  2. Axillary block, perivascular. Localization method: surface landmarks - above arterial pulsation in the axilla. Endpoint for injection: motor response in median nerve distribution to neurostimulation at ≤0.5mA. Single injection through needle.


Injectate in both blocks: ropivacaine 0.75% 0.5ml/kg

Sedation for block: IV alfentanil 0.5mg increments or midazolam 1mg as needed

Intraoperative sedation: none


Outcomes
  1. Sensory block to pinprick, scored as 0 (none), 1 (partial), 2 (complete), in dermatomes (C5-T1), at various time intervals up to 60 min
  2. Incidence of complete sensory block, defined as no sensation to pinprick in all dermatomes (C5-T1) at 60 min
  3. Motor block, scored as 0 (none), 1 (partial), 2 (complete), in individual nerve territories (RN, MN, UN, MCN), at various time intervals up to 60 min
  4. Incidence of surgical block, defined as no requirement for conversion to general anaesthesia
  5. Duration of postoperative analgesia, defined as time from block completion to pain visual analogue score (VAS) of >30mm
  6. Duration of motor block, defined as time from block completion to total recovery of function
  7. Patient satisfaction with anaesthetic technique, scored on a 4-point scale (0-3). Assessed immediately after block and on first postoperative day
  8. Block performance time, defined as from start of patient positioning to completion of injection. This included landmark identification
  9. Complications (vascular puncture, Horner's syndrome, pneumothorax, neurological deficit)


NotesAssessment for late complications was performed at 2-4 weeks (all) and 6-8 weeks (if necessary). Transient paraesthesia and dysaesthesia was reported in both groups, and is indicated as neurological deficit in this review. Symptoms disappeared within 6-8 weeks after surgery in all patients.

Patient satisfaction was converted to a dichotomous outcome by combining very satisfied/satisfied, and unsatisfied/very unsatisfied.

Additional data was obtained from authors on block performance time, duration of postoperative analgesia and duration of motor block.

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo mention of allocation concealment in text

Allocation concealment (selection bias)Unclear riskNo mention of allocation concealment in text

Blinding (performance bias and detection bias)
All outcomes
Low riskAssessments done by a blinded nurse

Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete follow-up in all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasLow riskNone identified

Song 2011

MethodsRandomized controlled trial


ParticipantsN = 22; adult; 16 male, 6 female; ASA 1-2; surgery of forearm; Korean study


Interventions
  1. Infraclavicular block (Tran 2008). Localization method: ultrasound-guided. Endpoint for injection: "double bubble" sign of local anaesthetic spread adjacent to posterior aspect of artery. Single injection through needle.
  2. Axillary block, perivascular. Localization method: ultrasound-guided. Endpoint for injection: local anaesthetic spread around axillary artery. Quadruple injection targeting left, right and posterior to artery, and musculocutaneous nerve separately.


Injectate in both blocks: 20 ml of 1.5% lidocaine with 1:200,000 epinephrine and 0.1 mEq/ml of sodium bicarbonate

Sedation for block: none

Intraoperative sedation: none


Outcomes
  1. Adequate surgical anaesthesia at 30 minutes, defined as analgesia or anaesthesia in the radial, median, ulnar, and musculocutaneous nerve territories
  2. Sensory block in individual nerve territories, defined as analgesia or anaesthesia
  3. Block performance time, defined as time from application of skin preparation solution to removal of block needle
  4. Block onset time, defined as time from block completion to analgesia and partial motor block in the individual nerves
  5. Total anaesthetic time, defined as performance time plus onset time
  6. Complications: pneumothorax, vascular puncture


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: “randomized using a randomization program on the Internet...”

Allocation concealment (selection bias)Unclear riskAllocation concealment not stated

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNo mention of blinding. It is unclear if the outcomes would be significantly influenced by lack of blinding

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll outcome data were complete

Selective reporting (reporting bias)Low riskStated outcomes were reported

Other biasUnclear riskSingle operator study; it is unclear if there was a preference for one block technique over another

Tedore 2009

MethodsRandomized controlled trial


ParticipantsN = 220; adult; 110 male, 110 female; ASA 1-3; surgery at or distal to elbow; American study


Interventions
  1. Infraclavicular block. Localization method: ultrasound-guided. Endpoint for injection: local anaesthetic spread. Dual-injection technique: 75% of local anaesthetic injected around posterior cord, 25% injected around medial cord.
  2. Axillary block, transarterial. Localization method: surface landmarks. Endpoint for injection: dual-injection technique with 75% of local anaesthetic injected posterior to artery and 25% injected anterior to artery. Continuous pressure applied with arm adducted for >5 minutes.


Injectate in both blocks: mepivacaine 1.5% with 1:200,000 epinephrine and 0.1 mEq/ml of sodium bicarbonate, injected in a body weight-adjusted volume of 40-50 ml for weight <50kg and 50-60 ml for weight >50kg

Sedation for block: intravenous midazolam up to 5mg

Intraoperative sedation: none


Outcomes
  1. Adequate surgical anaesthesia, defined as lack of need for supplementation or general anaesthesia
  2. Block performance time
  3. Complications (assessed at two days and 10 days): pain, bruising, tenderness at block site, neurological symptoms
  4. Preference for same block in future


NotesPrimary aim of study was to assess neurological complications of the two techniques; not anaesthetic efficacy.

There was no specified time between block completion and start of surgery, but it was probably within 15-20 minutes.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: “sealed envelope, computer-generated random number method...”

Allocation concealment (selection bias)Low riskQuote: “sealed envelope, computer-generated random number method...”

Blinding (performance bias and detection bias)
All outcomes
Unclear riskPatient, operator, and assessor during block and the peri-operative period was unblinded. The assessor of outcomes (patient satisfaction and complications) at 2 and 10 days following the block was blinded

Incomplete outcome data (attrition bias)
All outcomes
Low riskThe exclusions of patients were explained clearly in the Results. The review authors believe the risk of bias is low

Selective reporting (reporting bias)Low riskAll stated outcomes were reported

Other biasUnclear riskA weight-based formula was used to calculate local anaesthetic volume, but the volumes were high rather than low. The patients were prepped for surgery 10-15 minutes after completion of the block, and surgery commenced shortly thereafter; this shortened interval to surgery may have reduced the incidence of surgical anaesthesia.

Tran 2008

MethodsRandomized controlled trial


ParticipantsN = 70; adult; 47 male, 23 female; ASA not specified; surgery distal to the middle third of the humerus; Canadian study


Interventions
  1. Infraclavicular block. Localization method: ultrasound in-plane approach. Endpoint for injection: spread of local anaesthetic posterior to axillary artery. Single injection through needle.
  2. Axillary block (Sia 2001). Localization method: surface landmarks. Endpoint for injection: motor response to neurostimulation (minimum current threshold not specified) in distribution of MN, RN, and MCN. Triple injection through needle.


Injectate in both blocks: bupivacaine 0.5% and lidocaine 2% in 1:1 mixture in a volume of 35 ml. For the triple-injection axillary block, 14 ml was injected at MN and RN, and 7 ml at MCN

Sedation during block: IV midazolam (up to 2 mg) and fentanyl (up to 50 μg) as required or requested by patient

All patients received a general anaesthetic thereafter (communication from author)


Outcomes
  1. Block success, defined as complete sensory anaesthesia in all four terminal nerve distributions (RN, MN, UN, MCN) at 40 minutes. Surgical anaesthesia was not a defined outcome
  2. Sensory block to touch and cold, scored as 0 (none), 1 (partial), 2 (complete), in individual nerve territories (RN, MN, UN, MCN), at various time intervals up to 40 min
  3. Motor block, scored as 0 (none), 1 (partial), 2 (complete), in individual nerve territories (RN, MN, UN, MCN), at various time intervals up to 40 min.
  4. Pain during block performance, assessed by visual analogue score (VAS)
  5. Requirement for sedation during block performance
  6. Block performance time, defined as start of scan (infraclavicular block) or raising of skin wheal (supraclavicular block) to end of local anaesthetic injection
  7. Complications - not specified in Methods


NotesNo mention of whether supplemental analgesia or general anaesthesia was required for completion of surgery in text. Personal communication from lead author indicated that all patients received a general anaesthetic as surgical anaesthesia was not an outcome measure. Block success data (31/35 in the infraclavicular group, and 32/35 in the axillary group) was therefore not included in the meta-analysis or surgical anaesthesia.

Requirement for sedation during block performance was taken to equate presence of pain during the block.

40 min data on sensory block in individual nerve distributions was pooled with 45 min data from other studies.

Numerical data on sensory block in individual nerve distributions was obtained from author (presented graphically in published paper).

Abbreviations: AN = axillary nerve, MABCN = medial antebrachial cutaneous nerve, MBCN = medial brachial cutaneous nerve, MCN = musculocutaneous nerve, MN = medial nerve, RN = radial nerve, UN = ulnar nerve


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated in text

Allocation concealment (selection bias)Unclear riskNot stated in text

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNo mention of blinding of patients or outcome assessors

Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete follow-up in all patients

Selective reporting (reporting bias)Low riskAll pre-specified outcomes reported

Other biasLow riskNone identified

Tran 2009

MethodsRandomized controlled trial


ParticipantsN = 120; adult; 71 male, 49 female; ASA 1-3; surgery of the elbow/forearm/wrist/hand; Canadian study


Interventions
  1. Infraclavicular block. Localization method: ultrasound-guided. Endpoint for injection: spread of local anaesthetic posterior to axillary artery. Single injection through needle.
  2. Supraclavicular block. Localization method: ultrasound-guided. Endpoint for injection: spread of local anaesthetic in "corner pocket" between subclavian artery and first rib. Single injection through needle.
  3. Axillary block. Localization method: ultrasound-guided. Endpoint for injection: local anaesthetic spread, 40% of volume posterior to axillary artery, 20% in each of the upper quadrants, and 20% around the musculocutaneous nerve.


Injectate in both blocks: 35 ml of lidocaine 1.5% with 1;200,000 epinephrine

Sedation during block: IV midazolam 0.03mg/kg and fentanyl 0.6 mcg/kg

Intraoperative sedation: no details given


Outcomes
  1. Adequate surgical anaesthesia within 30 minutes, defined as completion of surgery without need for supplementation or general anaesthesia
  2. Complete sensory block in individual nerve territories: radial, median, ulnar, musculocutaneous nerves
  3. Block performance time, defined as ultrasound imaging time plus needling time
  4. Block onset time, defined as interval between block completion and a sensory score of 7-8
  5. Total anaesthesia-related time, defined as sum of block performance time and block onset time
  6. Block-associated pain, scored on a 0-10 point visual analogue scale
  7. Complications: vascular puncture, Horner's syndrome, neurological symptoms, systemic toxicity


NotesAll blocks performed by "experts", defined as having experience of 60 or more blocks for that particular technique.

The lead author confirmed by email correspondence that all blocks that did not result in adequate surgical anaesthesia received a general anaesthetic.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: “computer-generated sequence of random numbers, and sealed envelope”

Allocation concealment (selection bias)Low riskQuote: “computer-generated sequence of random numbers, and sealed envelope”

Blinding (performance bias and detection bias)
All outcomes
Low riskQuote: “A blinded observer recorded the onset time, block-related pain scores, success rate (surgical anaesthesia) and the incidence of complications.”

Incomplete outcome data (attrition bias)
All outcomes
Low riskReporting of outcome data was complete

Selective reporting (reporting bias)Low riskAll stated outcomes were reported

Other biasLow riskNo potential sources of other bias identified

Yang 2010

MethodsRandomized controlled trial


ParticipantsN = 100; adult; 53 male, 47 female; ASA 1-3; elective surgery of the elbow/forearm/hand; Korean study


Interventions
  1. Infraclavicular block (Kilka 1995). Localization method: neurostimulation. Endpoint for injection: motor response in hand or wrist at a current threshold of 0.5mA or less.
  2. Supraclavicular block (Brown 1993). Localization method: neurostimulation. Endpoint for injection: motor response in hand or wrist at a current threshold of 0.5mA or less.


Injectate in both blocks: 30 ml 0.5% ropivacaine

Sedation during block: IV midazolam 0.03mg/kg and fentanyl 0.6 mcg/kg

Intraoperative sedation: no details given


Outcomes
  1. Adequate surgical anaesthesia within 50 minutes, defined as completion of surgery without need for supplementation or general anaesthesia
  2. Duration of sensory block
  3. Duration of motor block
  4. Number of patients satisfied with anaesthetic
  5. Complications: pneumothorax, vascular puncture, Horner's syndrome


NotesAn interval of 50 min was used between block completion and testing.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote:  “the patients were randomized..”. No details were given on how this was done

Allocation concealment (selection bias)Unclear riskNone stated

Blinding (performance bias and detection bias)
All outcomes
Low riskQuote: “An assessor blinded to the block technique evaluated...” sensory and motor blockade

Incomplete outcome data (attrition bias)
All outcomes
Low riskOutcome data missing only for outcomes of block duration (4/100) and patient satisfaction (3/100). Authors do not believe this will introduce bias

Selective reporting (reporting bias)Unclear riskQuote: “block performance time, onset time, and time of readiness for surgery were not assessed. These are important factors..”

Other biasHigh riskQuote: “there was more experience with the infraclavicular approach than with supraclavicular approach at that time…”

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Fredrickson 2011This was a letter to the editor summarising the evidence for infraclavicular block over other approaches to the brachial plexus

Mariano 2008This was a non-randomized retrospective comparison of blocks performed by a single operator using either an axillary or infraclavicular approach to brachial plexus blockade

Mariano 2011aThis randomized controlled trial was excluded for two reasons:

  1. it was terminated prematurely for reasons unrelated to the outcomes studied
  2. it focused on comparing the efficacy of postoperative analgesia, and not anaesthetic efficacy, of peripheral nerve block catheters inserted using either the axillary or infraclavicular approach

Mariano 2011bThis RCT was excluded because it focused on comparing the efficacy of postoperative analgesia of peripheral nerve catheters inserted using either the supraclavicular or infraclavicular technique. Data were not collected or presented on the anaesthetic efficacy of these approaches

Neuburger 1998This study was not a RCT but rather a retrospective comparison of prospective case series

Rodriguez 2003This was a non-randomized comparative trial that did not assess clinical outcomes of efficacy, only the local anaesthetic distribution

 
Characteristics of studies awaiting assessment [ordered by study ID]
Astore 2012

MethodsRandomized controlled trial

ParticipantsN = 130; Argentinian study

Interventions
  1. Infraclavicular block. Localization method: neurostimulation. Endpoint for injection: a distal motor response of radial or median nerve at a current threshold of 0.3-0.5 mA. Single injection through needle.
  2. Humeral block. Localization method: neurostimulation. Endpoint for injection: motor response of targeted nerve at current threshold of 0.3-0.5mA. Quadruple injection through needle: median, ulnar, radial, musculocutaneous nerve.


Injectate in both blocks: 40 ml of bupivacaine 0.5% and lidocaine 2% in 1:1 ratio with 1:400,000 epinephrine

Outcomes1. Block success, defined as "complete sensory blockade" after 30 minutes

2. Block performance time, not defined in abstract

3. Block onset time

4. Duration of analgesia for bone surgery

5. Duration of analgesia for soft tissue surgery

NotesAbstract in conference proceedings; limited methodological information available. Authors did not respond to request for further information.

Danelli 2008

MethodsRandomized controlled trial

Participants52 adult patients undergoing upper extremity orthopaedic surgery

Interventions1. Ultrasound-guided infraclavicular block

2. Ultrasound-guided axillary block

Outcomes1. Onset time of sensory block

2. Onset time of motor block

3. Patient satisfaction (3-point scale)

4. Block-associated pain (0-10 scale)

5. Rescue analgesia requirements during surgery

NotesListed in Clinical Trials Registry as completed. No data available.

Lopez Morales 2011

MethodsRandomized controlled trial

ParticipantsN=40, patients undergoing upper extremity surgery of the elbow, forearm, wrist and hand

Interventions1. Ultrasound-guided axillary block

2. Ultrasound-guided infraclavicular block

Outcomes1. Total anaesthesia-related time

2. Onset time

3. Block-related pain scores

4. Surgical anaesthesia success rate

5. Duration of analgesia

6. Incidence of complications

NotesAbstract in conference proceedings; limited methodological information available. We were unable to contact authors.

 
Characteristics of ongoing studies [ordered by study ID]
Boivin 2013

Trial name or titleComparison of ultrasound-guided infraclavicular block and ultrasound-guided axillary block

MethodsRandomized controlled trial

ParticipantsN = 224, adult patients undergoing surgery at the elbow, forearm, wrist or hand under regional anaesthesia

Interventions1. Ultrasound-guided infraclavicular block

2. Ultrasound-guided double-injection axillary block

Outcomes1. Number of patients with complete sensory block to cold after 30 minutes

2. Number of patients with complete motor block after 30 minutes

3. Onset time of sensory and motor block

4. Procedure-related pain

5. Adverse events related to nerve block

6. Surgical block success rate

7. Performance time of nerve block

Starting dateSeptember 2012

Contact informationAriane Boivin (ariane.boivin.2@ulaval.ca), Centre Hospitalier Universitaire de Quebec

NotesListed in ClinicalTrials.gov; status: currently recruiting participants.

Hillel Yaffe 2013

Trial name or titleComparison of quality of supraclavicular, infraclavicular, and axillary approach of ultrasound-guided brachial plexus block

MethodsRandomized controlled trial

ParticipantsN = 150, patients undergoing surgery of the distal arm, forearm or hand

Interventions1. Ultrasound-guided supraclavicular block

2. Ultrasound-guided infraclavicular block

3. Ultrasound-guided axillary block

Outcomes1. Quality of block, defined as whether or not there is need for additional analgesic drugs or general anaesthesia

Starting dateDecember 2011

Contact informationAnatoly Stav (stav@hy.health.gov.il), Hillel Yaffe Medical Center

NotesListed in ClinicalTrials.gov; status: not yet open for participant enrolment.

 
Comparison 1. Infraclavicular block versus all other blocks

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Adequate surgical anaesthesia211662Risk Ratio (M-H, Random, 95% CI)0.88 [0.60, 1.30]

    1.1 At 30 minutes post-block assessment interval
141051Risk Ratio (M-H, Random, 95% CI)0.88 [0.51, 1.52]

    1.2 At 15 minutes post-block assessment interval
3332Risk Ratio (M-H, Random, 95% CI)1.25 [0.35, 4.42]

    1.3 At 60 minutes post-block assessment interval
4279Risk Ratio (M-H, Random, 95% CI)0.72 [0.44, 1.16]

 2 Adequate surgical anaesthesia (subgrouped by LA volume and block type)21Risk Ratio (M-H, Random, 95% CI)Subtotals only

    2.1 Local anaesthetic volume weight-based or < 40 mls
9717Risk Ratio (M-H, Random, 95% CI)0.80 [0.38, 1.68]

    2.2 Local anaesthetic volume ≥ 40 mls
6612Risk Ratio (M-H, Random, 95% CI)0.84 [0.52, 1.37]

    2.3 Single-injection axillary block
3110Risk Ratio (M-H, Random, 95% CI)0.69 [0.19, 2.45]

    2.4 Multiple-injection axillary block
6617Risk Ratio (M-H, Random, 95% CI)0.98 [0.45, 2.15]

    2.5 Supraclavicular block
6514Risk Ratio (M-H, Random, 95% CI)0.68 [0.33, 1.40]

    2.6 Mid-humeral block
2224Risk Ratio (M-H, Random, 95% CI)1.67 [0.63, 4.43]

    2.7 Ultrasound-guided infraclavicular block
8774Risk Ratio (M-H, Random, 95% CI)0.55 [0.29, 1.06]

    2.8 Neurostimulation-guided infraclavicular block
13888Risk Ratio (M-H, Random, 95% CI)1.17 [0.77, 1.78]

 3 Supplementation required to achieve adequate surgical anaesthesia171412Risk Ratio (M-H, Random, 95% CI)0.95 [0.62, 1.46]

    3.1 At 30 minutes post-block assessment interval
11861Risk Ratio (M-H, Random, 95% CI)0.96 [0.49, 1.86]

    3.2 At 15 minutes post-block assessment interval
3332Risk Ratio (M-H, Random, 95% CI)1.11 [0.34, 3.66]

    3.3 At 60 minutes post-block assessment interval
3219Risk Ratio (M-H, Random, 95% CI)0.79 [0.48, 1.31]

 4 General anaesthesia required to achieve adequate surgical anaesthesia201584Risk Ratio (M-H, Fixed, 95% CI)0.66 [0.36, 1.21]

    4.1 Single-injection axillary block
6289Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.13, 0.88]

    4.2 Other block techniques
141295Risk Ratio (M-H, Fixed, 95% CI)1.17 [0.50, 2.73]

 5 Complete sensory block in individual nerve territories within 30 minutes11Risk Ratio (M-H, Random, 95% CI)Subtotals only

    5.1 Musculocutaneous nerve (all blocks)
11786Risk Ratio (M-H, Random, 95% CI)0.91 [0.51, 1.62]

    5.2 Musculocutaneous nerve (single-injection axillary block)
3159Risk Ratio (M-H, Random, 95% CI)0.46 [0.27, 0.78]

    5.3 Axillary nerve (all blocks)
5300Risk Ratio (M-H, Random, 95% CI)0.65 [0.25, 1.68]

    5.4 Radial nerve (all blocks)
11786Risk Ratio (M-H, Random, 95% CI)0.91 [0.47, 1.76]

    5.5 Median nerve (all blocks)
11786Risk Ratio (M-H, Random, 95% CI)0.99 [0.67, 1.45]

    5.6 Ulnar nerve (all blocks)
11786Risk Ratio (M-H, Random, 95% CI)0.85 [0.52, 1.41]

    5.7 Medial brachial cutaneous nerve (all blocks)
4278Risk Ratio (M-H, Random, 95% CI)1.22 [0.63, 2.37]

    5.8 Medial antebrachial cutaneous nerve (all blocks)
4278Risk Ratio (M-H, Random, 95% CI)0.87 [0.33, 2.27]

 6 Tourniquet pain8615Risk Ratio (M-H, Fixed, 95% CI)0.66 [0.47, 0.92]

 7 Onset time of adequate surgical anaesthesia (minutes)9726Mean Difference (IV, Random, 95% CI)1.93 [0.23, 3.64]

    7.1 Neurostimulation-guided infraclavicular block
4335Mean Difference (IV, Random, 95% CI)3.85 [3.20, 4.50]

    7.2 Ultrasound-guided infraclavicular block
5391Mean Difference (IV, Random, 95% CI)0.52 [-2.24, 3.27]

 8 Duration of postoperative analgesia (minutes)9635Mean Difference (IV, Random, 95% CI)3.98 [-6.29, 14.25]

 9 Block performance time (minutes)12Mean Difference (IV, Random, 95% CI)Subtotals only

    9.1 Single-injection axillary block
290Mean Difference (IV, Random, 95% CI)3.17 [1.84, 4.50]

    9.2 Multiple-injection axillary block
6391Mean Difference (IV, Random, 95% CI)-2.67 [-3.36, -1.98]

    9.3 Supraclavicular block
3260Mean Difference (IV, Random, 95% CI)-0.81 [-1.35, -0.27]

    9.4 Mid-humeral block
2224Mean Difference (IV, Random, 95% CI)-4.80 [-6.04, -3.57]

    9.5 Ultrasound-guided infraclavicular block
7501Mean Difference (IV, Random, 95% CI)-1.61 [-2.63, -0.58]

 10 Pain associated with block performance (scored 0-10)8650Mean Difference (IV, Random, 95% CI)-0.56 [-1.25, 0.14]

 11 Horner's syndrome11Odds Ratio (M-H, Fixed, 95% CI)Subtotals only

    11.1 Block below clavicle
6514Odds Ratio (M-H, Fixed, 95% CI)2.03 [0.50, 8.25]

    11.2 Block above clavicle
5455Odds Ratio (M-H, Fixed, 95% CI)0.09 [0.04, 0.21]

 
Summary of findings for the main comparison. infraclavicular block versus all other brachial plexus blocks for regional anaesthesia of the lower arm

infraclavicular block versus all other brachial plexus blocks for regional anaesthesia of the lower arm

Patient or population: patients with regional anaesthesia of the lower arm
Settings:
Intervention: infraclavicular block versus all other brachial plexus blocks

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

Controlinfraclavicular block versus all other brachial plexus blocks

Adequate surgical anaesthesia - At 30 minutes post-block assessment intervalStudy populationRR 0.88
(0.51 to 1.52)
1051
(14 studies)
⊕⊕⊕⊕
high

871 per 1000766 per 1000
(444 to 1000)

Medium risk population

868 per 1000764 per 1000
(443 to 1000)

Supplementation required to achieve adequate surgical anaesthesiaStudy populationRR 0.95
(0.62 to 1.46)
1412
(17 studies)
⊕⊕⊕⊕
high

135 per 1000128 per 1000
(84 to 197)

Medium risk population

120 per 1000114 per 1000
(74 to 175)

Tourniquet painStudy populationRR 0.66
(0.47 to 0.92)
615
(8 studies)
⊕⊕⊕⊕
high

180 per 1000119 per 1000
(85 to 166)

Medium risk population

157 per 1000104 per 1000
(74 to 144)

Onset time of adequate surgical anaesthesia (minutes)The mean onset time of adequate surgical anaesthesia (minutes) in the intervention groups was
1.93 higher
(0.23 to 3.64 higher)
726
(9 studies)
⊕⊕⊕⊝
moderate1

Block performance time (minutes) - multiple-injection axillary blockThe mean block performance time (minutes) - multiple-injection axillary block in the intervention groups was
2.67 lower
(3.36 to 1.98 lower)
391
(6 studies)
⊕⊕⊕⊕
high

Block performance time (minutes) - mid-humeral blockThe mean block performance time (minutes) - mid-humeral block in the intervention groups was
4.8 lower
(6.04 to 3.57 lower)
224
(2 studies)
⊕⊕⊕⊝
moderate2

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 Subgroup analysis by method of localization showed that there was a significant difference in onset time in the studies using neurostimulation-guided infraclavicular block but not in the studies using ultrasound-guided infraclavicular block.
2 Only two studies in this review compared infraclavicular block to mid-humeral block. Both were by the same investigators.
 
Table 1. Complications of block procedurea

ComplicationInfraclavicular blockAll other blocksOverall rateRR (95% CI)cP value

Pneumothorax0/558(0)2/557 (0.4)2/1115 (0.2)0.20 (0.01, 4.06)0.29

Vascular puncture36/653 (5.5)47/691 (6.8)83/1344 (6.2)0.79 (0.53, 1.18)0.25

Horner's syndrome10/465 (2.2)61/504 (12.1)71/969 (7.3)0.19 (0.10, 0.35)<0.0001

Transient neurological deficit12/470 (2.6)9/509 (1.8)21/979 (2.1)1.35 (0.56, 3.25)0.51

Systemic LAb toxicity1/412 (0.2)3/542 (0.7)4/954 (0.4)0.37 (0.04, 3.50)0.38

Phrenic nerve palsy0/60 (0)7/59 (11.9)7/119 (5.9)0.07 (0.00, 1.12)0.06

 a. All complication rates are reported as n/N (%).
b. LA: local anaesthetic.
c. RR: risk ratio with respect to infraclavicular block, CI: confidence interval.
 
Table 2. Success rate of surgical anaesthesia

StudyInfraclavicular block (%)All other blocks (%)

Arcand 200579.586.8

Caruselli 200561.188.9

De Jose Maria 200892.195.0

Deleuze 200390.088.0

Ertug 200580.086.7

Fleischmann 2003100.080.0

Frederiksen 20109577.5

Fredrickson 200993.363.3

Heid 200596.7100.0

Kapral 199990.085.0

Koscielniak-N 200053.382.8

Koscielniak-N 200585.092.5

Koscielniak-N 200993.378.0

Minville 200591.795.0

Minville 200690.494.2

Niemi 200762.146.7

Rettig 200596.783.3

Song 2011100100

Tedore 200991.884.3

Tran 200995.096.3

Yang 201088.086.0

Overall88.286.0