ORIGINAL PAPER

Microsurgical outcome in posttraumatic brachial plexusinjuries in children

Kanwaljeet Garg & Sumit Sinha & Ashok Kumar Mahapatra &

Bhawani Shankar Sharma

Received: 19 September 2013 /Accepted: 11 November 2013# Springer-Verlag Berlin Heidelberg 2013

AbstractPurpose The purpose of the study was to analyze the surgicaloutcomes in children (≤18 years) with brachial plexus injuryoperated between April 2008 and March 2012 at our center.Methods All children <18 years of age admitted to our centerand surgically treated with a diagnosis of posttraumatic bra-chial plexus injury were included in the study. The demo-graphic details of these patients were retrieved from the com-puterized database of our hospital. The results were analyzedin terms of the mode of injury, type of injury, surgical proce-dure performed, and motor recovery after the surgery (MRCGrading). Motor recovery with MRC >3/5 was termed asgood outcome.Observations A total of 33 patients were surgically treated.The mean age at presentation was 15.1 (range 4–18) years.Boys constituted 79 % (n =26) of our patient population.High-velocity injury was the commonest mode of injury.Panbrachial injury was the commonest seen in 82 % (n =27)of patients. Mean duration between injury and surgical inter-vention was 6 (range 2–13, SD±2.6) months. Majority of

patients underwent neurotization procedure. Mean follow-upwas 32 (range 6–51) months.Conclusions High-velocity trauma is the most commonmodeon injury. Global palsy involving all the plexal elements waspresent in 82 % of the children. Neurotization was the mostcommonly performed surgical procedure. Good motor out-come (MRC grade ≥3/5) was seen in 62 % of patients.

Keywords Brachial plexus . Pediatric .Management .

Surgery

Introduction

Brachial plexus injuries have remained a complex and devas-tating entity with innumerable cases reported in literaturesince the days of Hippocrates’ scripts. In ancient militaristicSparta, babies with paralyzed limbs were thrown alive into thegorge to maintain the purity of their society [8].

The concept of treating brachial plexus injuries has beenrecorded for thousands of years, evidence of which can beseen in Homer’s iliad. The first surgical descriptions weremade by Smellie in 1768 and Flaubert in 1827 [13]. The firstpublished article discussing attempts at reconstruction waspublished by Thorburn in 1900 [15]. Thereafter, interest inthe surgical treatment temporarily waned off. This wascompounded in 1925 by Sever’s paper showcasing a highlynegative outlook in a large series of 1,100 cases [12]. How-ever, the landmark article by Seddon in 1947 helped reestab-lish surgical repair with nerve grafts as a viable option forbrachial plexus injuries [10]. More recently, with new micro-surgical techniques, there has been a resurgence in brachialplexus reconstruction [14].

Brachial plexus injuries in children are mostly obstetricalinjuries. Posttraumatic palsies in childhood are very rare andneither their characteristics nor their surgical indications have

A portion of this work was presented as proceedings at the 40thInternational Society of Pediatric Neurosurgery Conference 2012,Sydney, Australia, on September 12, 2012.

K. Garg : S. Sinha :B. S. SharmaDepartment of Neurosurgery, All India Institute of Medical Sciences,New Delhi, India

A. K. MahapatraAll India Institute of Medical Sciences, Bhuvneshwar, India

S. Sinha (*)JPNATrauma Center, AIIMS, R No-307, IIIrd Floor,New Delhi, India 110029e-mail: sumitaiims@yahoo.com

S. Sinhae-mail: sumitneuro@gmail.com

Childs Nerv SystDOI 10.1007/s00381-013-2325-z

been well described in literature. Pediatric brachial plexusinjuries are becoming more common because of increase inroad traffic accidents, blast, and fire arm injuries, especially indeveloping countries like India. Most of the injuries are closedtraction type. Other causes include fall, postsurgical and post-injection, postirradiation, and knife cuts, etc.

In this study, we present a series of 33 cases of posttrau-matic brachial plexus injuries in children treated by nervegrafting and/or nerve transfers at our center. The aim of thestudy was to evaluate the microsurgical outcome in pediatricbrachial plexus injuries and its correlation with time sinceinjury. To the best of our knowledge, this is the largest serieson posttraumatic pediatric brachial plexus injuries.

Materials and methods

Thirty-three children <18 years of age admitted to our centerwith a diagnosis of posttraumatic brachial plexus injury wereincluded in study. The period of study ranged fromApril 2008to March 2012. The clinical details of all these patients wereretrieved and analyzed retrospectively from the computerizeddatabase of our hospital. All patients with birth injuries wereexcluded. The ethical clearance was taken from the InstituteEthical Committee prior to the start of the study. The ageranged from 4 to 18 years (mean ± SD 15.1±4.3 years)(Fig. 1). There were 79 % (n =26) males and 21 % (n =7)females.

A detailed history and neurological examination, i.e., mo-tor, sensory, and autonomic deficit were recorded in a pre-designed proforma for each patient. Muscle power gradingwas done using British Medical Research Council grading(grades 1 to 5). Radiography like X-ray on the chest, neck,and shoulder were done for bony fractures or dislocationsassociated with nerve injuries. The electrophysiological stud-ies included nerve conduction velocities to specify the site andextent of nerve injury and electromyography (EMG) to con-firm the neurogenic type of EMG and rule out the muscledisease as a cause of weakness.

The indications for surgery were either the presence ofunequivocal clinical and radiological signs of root avulsion

or the absence of clinical/EMG recoverywithin 12weeks aftertrauma. Postoperatively, the paralyzed limb was keptimmobilized in optimal functional position for a period of4 weeks, after which the physiotherapy was started under awell-structured physiotherapy protocol of the institute.

The primary endpoint of the study was the improvement inmotor power of the affected muscle groups. The patients werefollowed up on outpatient basis every 6 months. The outcomewas classified as good (motor power of ≥M3) or poor (motorpower <M3) at the time of last follow-up.

All analyses were performed using the statistical programSPSS version 17. Percentages were compared using the chi-square or Fisher exact test. For intergroup comparison, theStudent’s t test was used for parametric data. Probabilityvalues <0.05 in these analyses were considered statisticallysignificant. Statistical analysis was done to find the relationbetween early or late surgery and improvement.

Results

The most common mode of injury was road traffic accidents(82 %) with most of the patients being pedestrians hit by a car.Other common modes of injury included fall from height,crush injury, and gunshot injury (Fig. 2). Associated injuriespresent were bony injuries (33 %), head injury (12 %), andvascular injury (6 %) (Fig. 3).

Panbrachial involvement was seen in 82 % of patients(Fig. 4). Mean duration from injury to surgery was 6 (range2–13, SD±2.6) months. Early surgical intervention (within6 months of injury) was performed in 64 % of patients and therest were operated 6 months after injury. Neurolysis was donein 21 % patients. Neurotization without nerve graft was donein 70 and 9 % patients underwent neurotization with nervegraft (Fig. 5). Table 1 enlists various donor and recipientnerves used in the neurotization procedures. There were nocomplications related to surgery in the present series.

The mean follow-up was 32 (range 6–51) months. Twenty-one patients could be followed up. Overall, 62 % (13 patients)had good motor outcome with elbow flexion and shoulder

Fig. 1 Pie chart showing age distribution of patients Fig. 2 Graph showing distribution of different mechanisms of injury

Childs Nerv Syst

abduction >3/5. Fifty-six percent patients with pan plexalinjury had good outcomes, while 80 % of patients with partialplexal injuries had good outcome. Seventy percent of thepatients showed good outcome if operated within 6 monthsafter injury as compared to only 50 % operated after 6 monthsof injury. However, this difference was not statistically signif-icant (p <0.22) (Table 2). Patients with panbrachial and partialinjuries were analyzed separately to find the relation betweenmotor outcome and timing of intervention. There was nostatistical significant difference in the surgical outcomes inpanbrachial group, and it was significant in the partial injurygroup, but since the number of patients was small, it can bemisleading (Table 2).

Discussion

Motor vehicle accidents are most common mode of traumaticbrachial plexus injuries in children [2, 3]. The kinetic energyinvolved in these accidents is more severe than that involvedin birth palsy, and therefore, results in a spectrum of moresevere injuries [6]. The occurrence of associated skeletalinjuries is more common in children. The brachial plexuslesion often occurs in the setting of multiple traumas whereassociation with a head injury, long bone fractures, and some-times abdominal trauma is common. The diagnosis is some-times delayed owing to severity of other lesions; patients aresometimes even in coma. Hence, high index of suspicion is

required to detect a brachial plexus injury in time. The inci-dence of root avulsion is high in pediatric patients as com-pared to adults [6]. Dumontier and Gilbert reported rootavulsion in 63 % of their cases [5]. The same finding has beensubstantiated in the present study where 85 % of children hadpanbrachial injury, which is more than panbrachial involve-ment seen in adults treated surgically at our center.Panbrachial involvement was seen in 52 % of patients in theseries by Dumontier and Gilbert and in 20 % of patients in theseries by El-Gammal et al. [5, 6]. In the series by El-Gammalet al. [6], eight patients (80%) had two or more root avulsions,mainly of the upper roots, and two of them had additionalsevere infraclavicular injuries. Like obstetric palsies, whichare mostly supraclavicular, traumatic plexus injuries in chil-dren in our series were also mostly supraclavicular. Althoughroot avulsion is frequently associated with pain in adults, theincidence of this finding was grossly less in our series seen inonly 9 % (n =3) of our cases. Dumontier and Gilbert reportedtwo teenage cases with moderate pain [5].

In our series, we found no statistical significant differencein the outcomes of patients operated within 6 months of injuryor those after 6 months. In patients with partial plexal

Fig. 3 Graph showing distribution of associated injuries

Fig. 4 Graph showing distribution of different injuries in the study group

Fig. 5 Graph showing distribution of different surgical procedures donein study group

Table 1 Various donor and recipient nerves used in the neurotizationprocedures

Neurotization procedure Number ofpatients

Medial pectoral nerve to axillary nerve 1

Phrenic nerve to anterior division of upper trunk 1

Phrenic nerve to lateral cord 1

Phrenic nerve to musculocutaneous nerve 2

Phrenic nerve to suprascapular nerve 8

Spinal accessory nerve to C5 1

Spinal accessory nerve to suprascapular nerve 9

Spinal accessory nerve to posterior cord 1

Ulnar and median nerve with sural nerve cable grafts 1

Ulnar nerve fascicle to musculocutaneous nerve 5

Intercostal (second, third, and fourth) to musculocutaneousnerve

7

Intercostal (second and third) to nerve to biceps 1

Medial pectoral nerve to musculocutaneous nerve 2

Childs Nerv Syst

involvement, early surgery was associated with a significantlybetter outcome. However, this could be because of very smallnumber of patients in this group. These findings are alsoconcordant with several previous reports in literature. Boomeindicated that the delay between injury and direct surgery onthe plexus is not as critical as in adults [2]. He mentioned thatsome children had nerve grafts 30 months after injury and yethad fair recovery. The denervation time in the series by Tareket al. [6] did not exceed 8 months. There was no difference inoutcome between children operated upon at 13 months andthose operated upon at 68 months after their accidents in theirseries [6].

The foremost aim of surgery is to restore the elbow flexionand shoulder abduction, followed by hand sensation andfinger flexion. In the present study, neurotization was donein 79 % of patients given the frequency of root avulsion. Theresults of the different neurotization procedures for elbow andshoulder reconstruction were slightly better than those report-ed in adults.

In this study, there were no untoward short- and long-termsequelae of harvesting the phrenic nerve, spinal accessorynerve, and intercostal nerves in children older than 3 years. Inour previous experience, we did not find any long-term com-promise in respiratory function with simultaneous or isolatedphrenic and intercostal nerve harvesting in children. Severalreports in literature have also supported this viewpoint [16].However, there have been several other reports which haveshown that harvesting the phrenic nerve in infants has beenshown to be associated with severe respiratory distress [3].

There has been a very sparse literature on pediatric post-traumatic brachial plexus injuries. Some children have hadsurgical treatment and then reported in adult series, but theirresults are not stated separately. Boome reported an incidenceof 1.1 % of all brachial plexus injuries [2]. He reported on 16children seen over 14 years, of whom nine underwent brachialplexus reconstruction. Dumontier and Gilbert reported on 25

children seen over 15 years, of whom 16 had surgical repair ofthe plexus [5]. d’Aubigne and Deburge [4] in 1967 reportedon 13 cases of traumatic pediatric brachial plexus injury over a20-year period. In this series, only one patient was under10 years old. Rigault [9] in 1969 presented seven cases thatwere treated and followed up. Some large adult series includea few children, but their treatment and results are not reportedseparately [1, 7, 11].

The limitations of study are that this is a study comprisingsmall number of patients. In the future, a larger patient cohortwith a longer follow-up is desirable for the conclusions to bemore statistically evident.

Conclusion

This is the largest series of pediatric posttraumatic brachialplexus injuries reported in English literature until date. Thereare differences in pediatric plexal injuries as compared to theadult counterparts. Traumatic brachial plexus palsy in childrenis rare, but typically severe. High-velocity trauma is the mostcommon mode on injury. Associated lesions are frequent dueto the violent nature of the trauma. Majority of children havepanplexal injuries (85 %). Good outcome could be seen in ahigh number of cases (62 %) despite a large number ofpanplexal injuries. The surgical outcome is not very muchdifferent in children operated within or after 6 months ofinjury. Therefore, management priorities differ in children ascompared to adults where the results of delayed plexal recon-struction are not as good as compared to that in children.

Conflict of interest The authors have no personal, financial, or institu-tional interest in any of the drugs, materials, or devices described in thisarticle. The authors report no conflict of interest concerning the materialsor methods used in this study or the findings specified in this paper.

References

1. Alnot JY (1977) Traumatic paralysis of the brachial plexus in theadult. I. Lesions and their mechanism. Rev Chir Orthop ReparatriceAppar Mot 63:39–43

2. Boome R (2000) Traumatic brachial plexus injuries in children. In:Gupta A, Kay SPJ, Scheker LR (eds) The growing hand: diagnosismanagement of the upper extremity in children. Mosby, London, pp653–657

3. Chuang DC, Lee GW, Hashem F, Wei FC (1995) Restoration ofshoulder abduction by nerve transfer in avulsed brachial plexusinjury: evaluation of 99 patients with various nerve transfers. PlastReconstr Surg 96:122–128

4. D’Aubigne M, Deburge A (1967) Etiology, development and prog-nosis of traumatic paralysis of the brachial plexus. Rev Chir OrthopReparatrice Appar Mot 53:23–42

5. Dumontier C, Gilbert A (1990) Traumatic brachial plexus palsy inchildren. Ann Chir Main Memb Super 9:351–357

Table 2 Results of surgery in different injury groups

Study group Improved(n)

No improvement(n)

p value

Whole study group (62 % improvement)

Within 6 months(n =13)

9 (70 %) 4 (30 %) 0.22

6–12 months (n=8) 4 (50 %) 4 (50 %)

Panbrachial injuries (56 % improvement)

Within 6 months(n =10)

6 (60 %) 4 (40 %) 0.39

6–12 months (n=6) 3 (50 %) 3 (50 %)

Partial injuries (80 % improvement)

Within 6 months(n =4)

4 (100 %) 0 0.02

6–12 months (n=1) 0 1 (100 %)

Childs Nerv Syst

6. El-Gammal TA, El-Sayed A, Kotb MM (2003) Surgical treatment ofbrachial plexus traction injuries in children, excluding obstetric palsy.Microsurgery 23:14–17

7. Narakas A (1982) Neurotization or nerve transfer for brachial plexuslesions. Ann Chir Main 1:101–118

8. Plutarch (1914) Parallel Lives, I: Lycurgus. Loeb Classical library9. Rigault P (1969) Traumatic paralysis of the brachial plexus in children (a

study of 7 cases). Rev Chir Orthop Reparatrice Appar Mot 55:125–13010. Seddon HJ (1947) The use of autogenous grafts for the repair of large

gaps in peripheral nerves. Br J Surg 35:151–16711. Sedel L (1977) Palliative treatment of a series of 103 cases of

paralysis from elongation of the brachial plexus. Spontaneous chang-es and results. Rev Chir Orthop Reparatrice Appar Mot 63:651–665

12. Sever JW (1925) Obstetrical paralysis: report of eleven hundredcases. JAMA 85:1962–1970

13. Shenaq SM, Bullocks JM, Dhillon G, Lee RT, Laurent JP (2005)Management of infant brachial plexus injuries. Clin Plast Surg 32:79–98, ix

14. Terzis JK, Kostopoulos VK (2007) The surgical treatment of brachialplexus injuries in adults. Plast Reconstr Surg 119:73e–92e

15. Thorburn W (1900) A clinical lecture on secondary suture of thebrachial plexus. Br Med J 1:1073–1075

16. Zheng MX, Qiu YQ, Xu WD, Xu JG (2012) Long-term observationof respiratory function after unilateral phrenic nerve and multipleintercostal nerve transfer for avulsed brachial plexus injury.Neurosurgery 70:796–801, discussion 801

Childs Nerv Syst