J Neurosurg (1 Suppl Pediatrics) 105:33–40, 2006

Recovery of hand function following nerve grafting and

transfer in obstetric brachial plexus lesions

WILLEM PONDAAG, M.D., AND MARTIJN J. A. MALESSY, M.D., PH.D.

Department of Neurosurgery, Leiden University Medical Center, Leiden, The Netherlands

Object. Infants with obstetric brachial plexus lesions (OBPLs) commonly undergo surgical repair. Outcome data
have been documented extensively for shoulder and biceps function, but information on hand function following nerve
repair is limited. Hand function is impaired in approximately 15% of patients. The authors present a surgical strategy
aimed primarily at restoration of hand function and analyze their methods and outcome to determine specific factors
affecting functional recovery.
Methods. Surgical strategy and outcome data were reviewed for 33 patients who underwent surgery for flail arm
during a 10-year period. Nerve repair was performed at a mean age of 4.4 months. In 16 patients, the period of follow
up (mean 50 months) was considered sufficiently long for final analysis.
Of these 16 patients, 13 (Group 1) had complete discontinuity of the C-7, C-8, and T-1 spinal nerves. In three patients
(Group 2), the C-8 and/or T-1 nerve was left in place because of shortage of nerve grafts or limited availability of prox-
imal donor stumps. Postoperatively, a Raimondi hand function grade of 3 or higher was attained by nine of the 13 patients
in Group 1 (69%) and one of the three patients in Group 2 (33%).
Conclusions. Useful hand function was restored in 69% of the patients in the presented series in whom reanimation
of the hand could be fully attributed to the surgical reconstruction. The authors conclude that restoration of hand func-
tion should be the first goal of nerve repair in infants with a flail arm caused by an OBPL, but that the optimal strate-
gy for different types of lesion remains to be determined.

KEY WORDS • brachial plexus avulsion • obstetric brachial plexus lesion •

brachial plexus repair • nerve surgery • hand function • outcome •
pediatric neurosurgery

BSTETRIC brachial plexus lesions are caused by trac-
O tion during delivery.7,30 The resulting nerve injury
may vary from neurapraxia or axonotmesis to neu-
rotmesis and avulsion of rootlets from the spinal cord. The
degree of spontaneous recovery correlates inversely with the
severity of the nerve lesion. Fortunately, most children show
good spontaneous recovery. In a systematic literature re-
view,36 we discussed the methodological flaws in the avail-
able natural history studies. Analysis of the most methodo-
logically sound studies led us to estimate the percentage of
children with residual deficits at 20 to 30%.36
The upper part of the brachial plexus is most commonly
affected, resulting in paresis of shoulder abduction, external
rotation, and elbow flexion. In more severe cases, the other
parts of the plexus are also involved. Hand function is im-
paired in approximately 15% of patients.2,9,16,37 In these in-
fants, the lesion usually involves the entire brachial plexus;
the clinical presentation is that of a flail arm, often accom-
panied by Horner syndrome. Children with a total lesion at
Abbreviations used in this paper: CT = computed tomography;
MRC = Medical Research Council; OBPL = obstetric brachial plex-
us lesion.
birth carry a greater risk of persistent palsy than those with
a lesion limited to the upper part of the plexus.9,37 Isolated
injury to the lower part of the brachial plexus (Dejerine–
Klumpke type) is extremely rare.1 It results in a paralyzed
hand and Horner syndrome with unimpaired proximal mus-
cles.
Currently, most authors advocate surgical repair for pa-
tients who do not recover spontaneously, although recom-
mendations differ with regard to the timing of surgery.8,13,24,41
Some authors advocate early surgery (at 3 or 4 months of
age),13,24 while others advise waiting until patients are older
to allow more time for spontaneous recovery. For patients
with upper brachial plexus palsy, it has been suggested that
surgery should be performed at 6 months41 or 9 months8 of
age. When a total lesion is present most surgeons tend to
perform surgery at an early stage (~ 3 months). Methods of
repair include nerve grafting after neuroma resection in pa-
tients with neurotmesis and nerve transfer in those with root
avulsion.13,18,19,34 Results achieved using these surgical ap-
proaches are claimed to be superior to outcomes in conser-
vatively treated patients with lesions of similar severity.14,41,42
This comparison does, however, rely on a historical control
series;21 no randomized study has yet been performed.5,20,29

J. Neurosurg: Pediatrics / Volume 105/ July, 2006 33

Brought to you by Rutgers University | Unauthenticated | Downloaded 05/05/20 05:10 AM UTC
 W. Pondaag and M. J. A. Malessy

Like other investigators,32 we had experienced disap- TABLE 1

pointing results following attempts to restore hand function Analysis of surgical reconstruction in 33 cases*
in adults with total brachial plexus lesions. Gilbert,13 how-
ever, has suggested that restoration of hand function should Root Trunk End-Nerve
be attempted in cases of OBPL, in view of the superior neu- Group Case AD AD LP MP
roregenerative capacities of infants as compared with adults No. No. Year C-7 C-8 T-1 MT IT IT MC MN MN
and the limited function of an upper limb with a paralytic
1 2002 x† x
hand. 2002 x† x
A satisfying outcome of nerve repair aiming at restora- 2003 x† x
tion of shoulder and biceps function has been reported in a 2003 x† x x
number of clinical studies.13,19,24,35 But there has been a pau- 2004 x x
city of publications concerning outcome of hand function 2004 x
following nerve repair in infants with OBPLs. Other than a 2004 x
2003 x‡ x‡
published meeting abstract,33 there is only one recent report 2003 x‡ x
describing the results of nerve surgery in total lesions.15 2003 x x
Useful recovery of hand function was obtained in 76% of 2003 x x
patients. The results of nerve repair in this paper are diffi- 2003 x x
cult to interpret, however, because details of the nerve le- 7 2000 x x
sions and the repair strategy are not provided. In addition, 2004 x
4 2002 x
the authors pooled results from primary (nerve repair) and 9 2002 x x
secondary (muscle transposition) surgery.15 A scientifically 1 2000 x
based optimal surgical strategy for recovery of hand func- 10 2000 x x
tion is, therefore, not yet available. 11 1998 x
The objective of surgical treatment of OBPLs is to estab- 2 1995 x
lish the ability to use the affected hand to assist in bimanual 8 2002 x
3 2000 x x
activity.23 Strong finger flexion, in combination with good 13 2000 x
elbow flexion, is mandatory for a supportive role in the bi- 2003 x x
manual execution of activities of daily living. Without rean- 12 1998 x x
imation of the hand, the maximal function that can be ob- 5 2002 x x x
tained for the affected limb is that of a hook. 6 1998 x x
According to our current treatment philosophy, restora- 2 2004 x‡ LIP
2004 x‡ LIP
tion of hand function is the primary aim of surgery in pa- 2004 x x‡ LIP
tients with a flail arm due to OBPL. Reanimation of elbow 15 1999 x LIP x
flexion and shoulder function are the second and third goals 16 1998 LIP LIP x
of surgery, respectively. 14 1995 LIP x
In the present paper, we describe the surgical strategy we * The “x” represents in each patient the target nerve for hand reanimation.
used in patients with a flail arm during a 10-year period. AD = anterior division; IT = inferior trunk; LIP = left in place; LP = lateral
Surgical neurotization of C-8, T-1, inferior trunk, or middle part; MC = medial cord; MN = median nerve; MP = medial part; MT = mid-
trunk was performed aiming at restoration of functions in- dle trunk.
† Direct coaptation of C-4 to C-7.
nervated by the median and/or ulnar nerves. We also ana- ‡ Direct coaptation to anterior (motor) filaments.
lyze our methods and the functional outcome attained in
this series of patients to determine specific factors affecting
functional recovery.
Characteristics of these 16 patients (11 male and five fe-
male infants) are provided in Table 2. Delivery presentation
Clinical Material and Methods was cephalic in 15 cases and breech in one (Case 10); de-
Retrospective analysis was performed for cases of OBPL livery was aided with either forceps or a vacuum cup in
in which infants had undergone surgery at the Leiden Uni- seven cases. Birth weight averaged 4614 g (range 3950–
versity Medical Center’s neurosurgery clinic, a multidis- 5850 g). In nine cases the injury was on the right side, and
ciplinary tertiary referral center for nerve injuries in The in 11 cases the child had Horner syndrome.
Netherlands. To date, over 300 children have been surgical- Preoperative examination for all patients included ul-
ly treated, the vast majority in the last 10 years. trasound examination of diaphragm movement and CT
Included in the current series were patients with a flail myelography after induction of general anesthesia.6,40 The
arm and absence of muscle contraction in response to direct combination of missing C-8 or T-1 root filaments and a
nerve stimulation of the inferior and middle trunk of the bra- pseudomeningocele was labeled as an avulsion. When a
chial plexus during surgery. Thirty-three patients matched pseudomeningocele was present with at least some visible
these inclusion criteria. We studied their surgical records to root filaments, or when no root filaments were visible yet
analyze the surgical reconstructions and their outcome (Ta- there was no pseudomeningocele, the result of the CT my-
bles 1 and 2). elography was said to be suggestive of avulsion (Table 2).
For the analysis of postoperative recovery of hand func- The mean age at surgery was 4.4 months (range 3–8
tion we selected all children who underwent surgery before months). During the operation, the severity of the nerve le-
July 1, 2002. This date was chosen to ensure that we would sion (that is, root avulsion or neurotmesis) was determined
have sufficient follow-up data for analysis. on the basis of a combination of characteristics. Root avul-

34 J. Neurosurg: Pediatrics / Volume 105 / July, 2006

Brought to you by Rutgers University | Unauthenticated | Downloaded 05/05/20 05:10 AM UTC
Hand function following nerve grafting and transfer in OBPLs

TABLE 2
Patient characteristics and results of surgery in 16 cases*
CTM Results
Avulsion Op Findings Nerve Reconstruction
MRC
Group Case Patient Horner Nerve Left Elbow FU Raimondi Grade
No. No. Age (mos) Syndrome C-8 T-1 C-5 C-6 C-7 C-8 T-1 in Place Hand Graft(s)† Flexion (mos) Grade (biceps)

1 1 5 yes yes yes N A A A A –– C5–IT 433 ICN–MCN 42 3 4

2 6 yes yes yes N A A A A –– C5–MC 635 ICN–MCN 105 4 4
3 6 no yes yes N A A A A –– C5–MP MN; 1 3 6; ICN–MCN 52 3 3
C5–LP MN 136
4 4 yes yes S N N N A N C5–IT 333 C6–AD ST 34 3 3
5 4 yes yes yes N N A A A C5–AD MT; 1 3 4; C5–AD ST 36 1 3
C6–8; C6–T1 2 3 3;
133
6 4 no yes no N N A A IN –– C6–LP MN; 1 3 2.5; C5–MCN 61 3 3
C6–8 3 3 2.5
7 3 yes no no N N N A A –– C6–8; C6–T1 2 3 2; C5–AD ST 49 3 3
232
8 4 yes yes yes N N N A A C6–MC 2 3 4, UG C5–LC 36 2 3
9 4 yes yes yes N N N A A C6–AD IT; 4 3 2.5; C5–AD ST 36 0 4
C7–AD MT 134
10 4 yes yes yes N N N A A –– C6–AD MT; 1 3 4.5; C6–AD ST 58 1 2
C7–AD MT; 1 3 4.5;
C7–AD IT 1 3 4.5
11 5 no no no N N N A A –– C7–AD IT 5 3 1.5 C6–AD ST 63 3 4
12 3 yes no yes N N N N A –– C7–AD MT; 1 3 2.5; C6–AD ST 50 3 3
C8–T1 4 3 0.5
13 3 no no no N N N N IN –– C8–MP MN 1 3 6, UG C6–AD ST 52 3 4
2 14 8 no yes yes N A A A A T-1 C5–MP MN 137 ICN–MCN 24 1 4
15 5 yes yes yes N N A A A T-1 C5–AD MT; 2 3 3.5; C5–AD ST 52 1 4
C6–8 4 3 2.5
16 3 yes S S IN N N A A C8–T1 C7–AD MT 3 3 3.5 C6–AD ST 53 3 3
* Group 1: reconstruction was performed, with complete interruption of C-8 and/or T-1; Group 2: reconstruction was performed, but C-8 and/or T-1 were
left in place. Age = age at surgery. CTM = CT myelography findings of C-8 and T-1. Abbreviations: A = avulsion; FU = follow up; ICN = intercostal nerves;
IN = intraforaminal neurotmesis; MCN = musculocutaneous nerve; MN = median nerve; N = neurotmesis; PD = posterior division; S = suggestive; SSN =
suprascapular nerve; ST = superior trunk; UG = ulnar graft; –– = not applicable.
† a 3 b = a grafts of b cm length.

sion was diagnosed when the spinal nerve at the juxtaforam-
inal and intraforaminal level exhibited root filaments and a
dorsal root ganglion, there was no neuroma, and there was
no muscle contraction in response to direct stimulation of
the spinal nerve. In the majority of the spinal nerves, these
findings corresponded with the absence of root filaments as
demonstrated on CT myelography (Table 2). Lesions were
said to be neurotmetic when the spinal nerve appeared nor-
mal at the intraforaminal level, but there was enlargement
of the nerve at the level of the trunk fusion due to neuroma
formation. In a frozen-section examination of the proximal
stumps, the presence of ganglion cells (indicative of total
avulsion) and myelin-content, which corresponds to the via-
bility of the proximal stump,28 were assessed.
There were four cases of neurotmesis of C-5 and avul-
sion of the C6–T1 nerve roots; two cases of neurotmesis of
C-5 and C-6 together with avulsion of the C7–T1 roots; and
one case of neurotmesis of C-5 and C-6 together with avul-
sion of C-7 and C-8 and intraforaminal neurotmesis of T-1.
In six cases, the C-8 and T-1 roots were avulsed and there
was neurotmesis of C-5 through C-7. Neurotmesis of C-5
through C-8 had occurred in two cases in combination with
avulsion or intraforaminal neurotmesis of T-1. Avulsion of
C-8 was found together with intraforaminal neurotmesis of
T-1 on one occasion.
Avulsed roots were cut as proximally as possible. When
the dorsal root ganglion could be morphologically identi-
fied, it was dissected from the ventral root and removed.
Following confirmation by frozen section of the presence of
ganglion cells in the resected specimen, it was certain that
the distal stump consisted only of the ventral root. In cases
of postganglionic rupture, neuroma tissue was resected in
a stepwise manner in order to minimize the gap between
proximal and distal stumps.
Distal target stumps were selected according to predeter-
mined goals. The primary goal was restoration of hand
function. Toward this end, we performed neurotization of C-
8, T-1, and the inferior or middle trunk, aiming at restoration
of innervation to the median and/or ulnar nerve. The second
priority was restoration of elbow flexion; toward this end,
we chose the anterior division of the superior trunk, the lat-
eral cord, or the musculocutaneous nerve as target nerves.
The third goal was to recover shoulder movements; neurot-
ization of the posterior division of the superior trunk and the
suprascapular and/or the axillary nerve was directed toward
this goal.
The preferred option was direct coaptation between an
available proximal nerve stump and the avulsed root (intra-
plexal nerve transfer). If this was not possible, nerve grafts
were extended from viable proximal nerve stumps to distal
target stumps. When the number of proximal stumps was
limited, intraplexal transfer or nerve grafting was used to

J. Neurosurg: Pediatrics / Volume 105 / July, 2006 35

Brought to you by Rutgers University | Unauthenticated | Downloaded 05/05/20 05:10 AM UTC

reinnervate the hand, and extra–intraplexal nerve transfers
were performed to restore shoulder and elbow flexion. Di-
rect coaptation without a nerve graft was performed in cas-
es in which the intercostal or spinal accessory nerves were
used as donor nerve.
In all patients, both sural nerves were harvested for use as
grafts; the procedure was routinely performed with the aid
of an endoscope. In addition, the cutaneous cervical plexus
or the cutaneous nerves of the arm or forearm were used in
some cases. Twice the ulnar nerve was employed as a free
graft because there was a shortage of graft material to cover
all the proximal stumps.
Postoperatively, each child’s upper body was placed in a
prefabricated body cast for 2 weeks to limit movements of
the head and affected arm.
Patients were examined at our outpatient clinic at 6-
month intervals. The active and passive range of joint move-
ments was noted in degrees and the MRC grade was deter-
mined. In addition, hand function was evaluated by means
of the Raimondi hand function scale (Table 3).
Results
The results are presented in Tables 1 and 2. The cases
were divided into two groups. Group 1 consisted of those
cases in which there was complete discontinuity of C-7, C-
8, and T-1 at the root level. Group 2 consisted of cases in
which C-8 and/or T-1 had been left in situ during surgery,
because of the insufficient availability of proximal stumps
or the shortage of graft material. The appearance of these
nerves was normal up to the level of the neural foramen, de-
spite CT myelographic findings showing avulsion or sug-
gestive of avulsion injury. In addition, no reaction could be
elicited upon electrical stimulation.
The different nerve reconstructive procedures performed
in our 33 patients are presented in Table 1. Outflow to the
hand was reconstructed at the root level in 16 patients. In
eight patients, the distal target was at the trunk level; in two
patients, it was at the cord level; and in three patients, the
end nerve was selected. In four patients, the distal target was
at multiple levels. As shown in Table 1, a proximal target
nerve was selected more often than a distal target. Table 1
also shows that the choice for nerve repair at the level of the
root was made more frequently in the later years of our ex-
perience. Intraplexal transfer by direct coaptation of the spi-
nal nerve to a viable proximal stump was made from 2002
onwards. In five patients, the anterior filaments of C-8 or T-
1 were directly coapted to C-5, C-6, or C-7.
In the 16 patients for whom duration of follow up was
long enough to allow evaluation of hand function, the meth-
ods for reanimation of the biceps muscle and shoulder
movement were also analyzed (Table 2). In 12 patients, suf-
ficient viable proximal stumps were available to enable
nerve reconstruction for elbow flexion with nerve grafting.
In the other four patients, extra–intraplexal nerve transfers
were performed using the intercostal nerves. Shoulder func-
tion was reanimated seven times using intraplexal grafting,
and seven times with extra–intraplexal nerve transfer. (Ac-
cessory nerve to suprascapular nerve transfer was used in
five of these seven cases.) In two cases, the hypoglossal
nerve was used as a donor stump to neurotize the supra-
scapular or axillary nerve, a method no longer used in our
clinic.27
36
Brought to you by Rutgers University | Unauthenticated | Downloaded 05/05/20 05:10 AM UTC
W. Pondaag and M. J. A. Malessy
TABLE 3
Raimondi hand function scale*
Grade Description
0 complete paralysis or slight finger flexion of no use, useless
thumb, no pinch, some or no sensation
1 limited active flexion of fingers, no extension of wrist or fingers,
lateral pinch of thumb present or absent
2 active extension of wrist w/ passive flexion of fingers (by means
of tenodesis), passive lateral pinch of thumb (by means of
pronation)
3 active complete flexion of wrist & fingers, mobile thumb w/
partial abduction–opposition, intrinsic balance, no active
supination, good potential for secondary surgery
4 active complete flexion of wrist & fingers, active wrist exten-
sion, weak or absent finger extension, good thumb opposition
w/ active ulnar intrinsics, partial pronation/supination
5 all function described in Grade 4 plus finger extension & almost
complete pronation/supination
* Adapted from Raimondi P: Evaluation of results in obstetric brachial
plexus palsy. The hand. Presented at the International Meeting on Obstetric
Brachial Plexus Palsy, Heerlen, The Netherlands, 1993.
The mean follow-up period was 50 months (range
24–105 months). In Group 1, nine of 13 patients (69%) at-
tained a Raimondi hand scale grade of at least 3. Treatment
failures occurred four times in Group 1 and twice in Group
2. Combining the surgical results of Groups 1 and 2, a Rai-
mondi grade of at least 3 was achieved in 10 of 16 patients
(63%). Elbow flexion of MRC Grade 3 or more was
achieved in 15 of 16 patients (94%).
Discussion
In this paper, we present our surgical strategy for nerve
reconstructive surgery to restore hand function in 33 infants
with OBPLs along with the results of nerve surgery in half
of these patients. Findings from the analysis of our surgical
results demonstrate that useful reanimation of the hand was
obtained in 69% of the patients in whom reanimation could
be fully attributed to surgical reconstruction. Our results are
not as good as those reported by Haerle and Gilbert,15 who
reported good recovery of hand function in 76% of patients.
One possible explanation for the difference in outcome is
that our series was limited to patients who underwent sur-
gery for nerve repair only, whereas the series Haerle and
Gilbert reported on included patients who had undergone
secondary surgery (that is, tendon transfers) at a later age in
addition to the primary nerve repair.
From the presentation of surgical reconstructions per-
formed, it is clear that we now tend to perform a more prox-
imal reconstruction and prefer to use intraplexal transfer of
the spinal nerve with direct coaptation to restore hand func-
tion. We modified our strategy after it became apparent that
such transfers are technically possible. The theoretical ad-
vantages, which will be outlined below, are based on basic
principles of peripheral nerve regeneration.38 The applica-
tion of these techniques has not––to our knowledge––been
previously discussed.
In 13 of the 16 cases for which outcome data are pre-
sented, either discontinuity of the outflow of the C-7, C-8,
and T-1 spinal nerves was present due to avulsion injury or
ruptured parts of the nerves were resected. (These cases are
J. Neurosurg: Pediatrics / Volume 105 / July, 2006
Hand function following nerve grafting and transfer in OBPLs

in Group 1.) The postoperative recovery of hand function such additional extra–intraplexal nerve transfers can only
in these cases can, therefore, only be attributed to the nerve be evaluated in adulthood.
reconstruction. As shown in Table 2, surgery resulted in a
functional hand (Raimondi grade $ 3) in nine of these 13 Optimal Nerve Reconstruction for Reanimation of the
cases (69%). Hand
In the three other cases for which outcome data are pre-
sented, the C-8 and T-1 nerve roots were continuous up to The optimal method of nerve reconstruction cannot be
their neural foramen. In Cases 14 and 15, only T-1 was left determined from our results because of the limited number
intact. In Case 16, both C-8 and T-1 were left intact. (These of patients in our study and the diversity of surgical proce-
three cases are in Group 2.) The lesions in these cases prob- dures performed. To date, no systematic research has been
ably consisted of Birch Type III or IV avulsions.4 We can- undertaken to define the optimal distal target nerves for re-
not exclude the possibility that some spontaneous recovery covery hand function. Choice of a distal target should there-
occurred in the untouched C-8 and T-1 outflow, and postop- fore be based on the consideration of several issues.
erative recovery of hand function cannot, therefore, be at- First, the function of the distal target must be taken into
tributed with certainty to the surgical procedure alone. We account (Fig. 1). The median nerve innervates the extrinsic
believe that hand function in OBPL patients with these flexor muscles of the fingers and the flexion and opposition
types of lesions should be analyzed separately from hand of the thumb;12,31 these movements are likely to be of pri-
function in those who had clear C-8 and T-1 axonal dis- mary importance when the nondominant hand is used to as-
continuity. sist the dominant hand in bimanual tasks. The ulnar nerve
Treatment failures, defined as a Raimondi grade of less provides reinnervation of intrinsic hand muscles,12,31 which
than 3, occurred four times in Group 1 and twice in Group
2 (Table 2). In general, failures can be explained by graft
shortage, excessive length of grafts, or technical failure
(such as displacement of the grafts postsurgery).
Weaknesses in our study were the limited number of pa-
tients and the retrospective study design. Although the anal-
ysis was performed retrospectively, however, the records of
all OBPL patients were recorded in a database, and the pa-
tients were evaluated in our outpatient clinic according to a
standard protocol.
Evaluation of Hand Function
The Raimondi scale was used to express the recovery of
hand function. According to this system, hand function is
graded from 0 to 5, depending on power of both intrinsic
hand muscles and forearm muscles. Although the Raimon-
di scale has not yet been validated, it was specifically de-
signed for evaluation of hand function related to severe
OBPLs and is internationally accepted.
More detailed evaluation systems of hand function26
require full cooperation during examination and are there-
fore only applicable in the adult population. One scoring
system presented for evaluation of an adult population di-
vides hand function into no function, minimal nondominant
arm function, supportive arm function (nondominant side),
minimal dominant arm function, useful arm function as
performed by the dominant arm, and normal arm function.10
A similar scoring system,23 which emphasizes functional
use of the limb, was suggested for the evaluation of chil-
dren with OBPLs and cerebral palsy but has not been wide-
ly used. Its authors maintain that for children with unilater-
al hand dysfunction, an optimal outcome has been achieved
when patients regain a high degree of ability to use the
affected hand to assist in bimanual activities. We agree with
this point of view that gain of an assisting hand can be con-
sidered a satisfactory result; therefore, we used a Raimondi
grade of at least 3 in our present evaluation.
The Raimondi scale includes only minimal evaluation of FIG. 1. Schematic drawing showing the theoretical distal target
sensory function. The potential influence of limited senso- nerves for reanimation of hand function. ADIT = anterior division
ry recovery on hand function after nerve surgery for OBPL of the inferior trunk; ADMT = anterior division of the middle trunk;
is not known. If only one proximal root was available, in LPMN = lateral part of the median nerve; MC = medial cord;
later years we chose to perform a direct transfer of C-7 to MN = median nerve; MPMN = medial part of the median nerve;
C-4 to augment sensory function of the hand. The value of UN = ulnar nerve.

J. Neurosurg: Pediatrics / Volume 105 / July, 2006 37

Brought to you by Rutgers University | Unauthenticated | Downloaded 05/05/20 05:10 AM UTC
 W. Pondaag and M. J. A. Malessy

are required for fine movement. In addition, due to the long

outgrowth trajectory, and thus the long interval before rein-
nervation is accomplished, severe atrophy is more likely to
occur in the intrinsic muscles (innervated by the ulnar
nerve) than in the more proximal extrinsic hand muscles
(innervated by the median nerve). Therefore, it is our opin-
ion that more emphasis should be placed on reconstructing
median nerve function rather than ulnar nerve function.
This philosophy is supported by the results achieved in
Case 13, in which the ulnar nerve was employed as a nerve
graft, and the patient regained good hand function.
A similar argument can be made for the preference of re-
pairing spinal nerve C-8 (which predominantly controls ex-
trinsic flexion), as opposed to T-1, which mainly governs
intrinsic muscles.12,31
Second, the location of the target nerve should be consid-
ered: should a more proximal target (for example, the nerve
rootlet) or a more distal target nerve (such as the end nerve)
be selected? Choosing the C-8 spinal nerve as a distal target
instead of the median nerve means the target will be closer
to the proximal stump, and a shorter graft can be used. This
can be expected to have a favorable effect on the result of
the grafting procedure, because the length of the nerve graft
has been shown to correlate inversely with the outcome.38
When C-8 is used as a distal target, some axons will grow
into the posterior part of the lower trunk and will not con-
tribute to grasp function. This can be referred to as “disper-
sion.” When a more distal target is chosen (for instance, the
median nerve), all outgrowing axons will contribute to hand
function. This implies, however, that the length of the grafts
will increase, potentially nullifying the benefit of selective FIG. 2. Schematic drawing demonstrating the preferred recon-
targeting. struction for cases in which only one proximal stump is available.
Obviously, the issue of whether to choose a proximal tar- Nerve grafting is performed for the C5–8 spinal nerves together
get nerve or a distal one depends on the length of the lesion. with nerve transfers as follows: the accessory nerve (XIN) to the
When the lesion length permits, direct coaptation of C-8 or suprascapular nerve (SSN) and the intercostal nerves 3, 4, and 5
(ICN 3, 4, 5) to the musculocutaneous nerve (MCN).
T-1 to C-6 or C-7 is, in our opinion, the best option. In the
present series such an intraplexal nerve transfer was per-
formed in five patients, but the results have not yet been
evaluated. The procedure is technically less demanding than ports of patient series comparing different methods of re-
nerve grafting, and it entails the advantage that sprouting construction, however, our treatment philosophy was nec-
axons have to bridge only one coaptation site. (Superior re-
sults have been documented for nerve grafting with one co- essarily developed on the basis of theoretical grounds only.
aptation site as opposed to two.38) Comparison of OBPLs and Traumatic Lesions in Adults
Third, the choice of the best option for nerve repair in
a given case depends on the number of viable proximal Our surgical approach in OBPLs, in which reconstruc-
stumps and the availability of nerve grafts. In cases of neu- tion of hand function favors reconstruction of biceps and
rotmesis of C-5 and avulsion of C-6 through T-1, only one shoulder function, differs from that used for lesions of trau-
proximal stump is available, and it should be used exclu- matic origin in adult patients. In adults, reported functional
sively for neurotization of the hand. Nerve transfers (for outcome after neurotization of the hand is disappointing.32
example, accessory to suprascapular nerve and intercostal Successful reinnervation after nerve grafting to restore fin-
to musculocutaneous nerve), should then be used for rein- ger flexion was reported in only four of 31 patients in one
nervation of shoulder and biceps function (Fig. 2). In cases series.3 In another report, functional recovery of finger flex-
of neurotmesis of C-5 and C-6 with C7–T1 avulsion, two ion was reported in as many as 35% of patients after nerve
proximal stumps are available. In this case, C-5 should be surgery and additional palliative surgery.39 One reason for
used as the donor stump for the superior trunk, with its an- diminished recovery in adults compared with infants may
terior division generally selected for recovery of elbow be found in the mechanism of injury. In adults, the brachial
flexion, and C-6 should be used as the donor stump for hand plexus lesion is usually caused by a high-velocity motor ve-
reanimation. Nerve transfer from the accessory nerve to the hicle accident, involving a large amount of kinetic energy,
suprascapular nerve should be performed for recovery of which may cause multilevel nerve damage and/or injury to
external rotation (Fig. 3). myoskeletal elements.
As can be concluded from Table 1, it has become our The greater ability of the young nervous system to adapt
preference to neurotize the hand by repairing the root of the and recover is likely to have a positive effect on the outcome
C-8 spinal nerve (20 of 35 patients). In the absence of re- of OBPL surgery. It is known that the plasticity of the young

38 J. Neurosurg: Pediatrics / Volume 105 / July, 2006

Brought to you by Rutgers University | Unauthenticated | Downloaded 05/05/20 05:10 AM UTC
Hand function following nerve grafting and transfer in OBPLs
FIG. 3. Schematic drawing illustrating the preferred reconstruc-
tion for cases in which two proximal stumps are available. Nerve
grafting from C-5 to the superior trunk, direct coaptation to C-6 to
C-8, and an additional nerve transfer from the accessory nerve to the
suprascapular nerve are depicted.
brain is superior to that of the adult brain,17,25 and a recent
paper documented greater plasticity at the level of the spinal
cord in young animals compared with mature animals.22 At
the level of peripheral nerve regeneration, newborn animals
also show better recovery compared with older animals af-
ter the repair of experimentally induced injuries of nerve
roots.11
Conclusions
Although the present series comprises only a limited num-
ber of patients, some important conclusions may be drawn.
Our results demonstrate that nerve reconstruction aiming
at recovery of hand function is effective in infants with
OBPLs. Using this strategy, functional recovery of the hand
may be achieved in the majority of cases. In our opinion,
this implies that reconstruction of hand function should be
the first goal of nerve surgery in these patients.
Intraplexal transfer on the root level, with direct coapta-
tion of the proximal donor nerve to the target nerve, is tech-
nically possible in some OBPL patients. This type of re-
construction is relatively straightforward and has potential
benefits in comparison with nerve grafting. However, the
optimal surgical repair strategy in specific cases remains to
be determined.
J. Neurosurg: Pediatrics / Volume 105 / July, 2006
Brought to you by Rutgers University | Unauthenticated | Downloaded 05/05/20 05:10 AM UTC
References
1. Al-Qattan MM, Clarke HM, Curtis CG: Klumpke’s birth palsy.
Does it really exist? J Hand Surg [Br] 20:19–23, 1995
2. Bager B: Perinatally acquired brachial plexus palsy—a persist-
ing challenge. Acta Paediatr 86:1214–1219, 1997
3. Bentolila V, Nizard R, Bizot P, Sedel L: Complete traumatic
brachial plexus palsy. Treatment and outcome after repair. J
Bone Joint Surg Am 81:20–28, 1999
4. Birch R, Bonney G, Parry CBW (eds): Traumatic lesions of the
brachial plexus, in Surgical Disorders of The Peripheral
Nerves. Edinburgh: Churchill Livingstone, 1998, pp 157–159
5. Bodensteiner JB, Rich KM, Landau WM: Early infantile surgery
for birth-related brachial plexus injuries: justification requires a
prospective controlled study. J Child Neurol 9:109–110, 1994
6. Chow BC, Blaser S, Clarke HM: Predictive value of computed
tomographic myelography in obstetrical brachial plexus palsy.
Plast Reconstr Surg 106:971–979, 2000
7. Clark LP, Taylor AS, Prout TP: A study on brachial birth palsy.
Am J Med Sci 130:670–674, 1904
8. Clarke HM, Curtis CG: An approach to obstetrical brachial
plexus injuries. Hand Clin 11:563–580, 1995
9. DiTaranto P, Campagna L, Price AE, Grossman JA: Outcome
following nonoperative treatment of brachial plexus birth in-
juries. J Child Neurol 19:87–90, 2004
10. Eggers IM, Mennen U: The evaluation of function of the flail
upper limb classification system: its application to unilateral
brachial plexus injuries. J Hand Surg [Am] 26:68–76, 2001
11. Fullarton AC, Myles LM, Lenihan DV, Hems TEJ, Glasby MA:
Obstetric brachial plexus palsy: a comparison of the degree of
recovery after repair of a C6 ventral root avulsion in newborn
and adult sheep. Br J Plast Surg 54:697–704, 2001
12. Gardner E, Bunge RP: Gross anatomy of the peripheral nervous
system, in Dyck PJ, Thomas PK, Lambert EH, et al (eds): Pe-
ripheral Neuropathy. Philadelphia: Saunders, 1984, pp 11–38
13. Gilbert A: Long-term evaluation of brachial plexus surgery in
obstetrical palsy. Hand Clin 11:583–595, 1995
14. Gilbert A, Tassin JL: [Surgery repair of the brachial plexus after
obstetric paralysis.] Chirurgie 110:70–75, 1984 (Fr)
15. Haerle M, Gilbert A: Management of complete obstetric brachi-
al plexus lesions. J Pediatr Orthop 24:194–200, 2004
16. Jacobsen S: [Occurrence of obstetrical injuries to the brachial
plexus on the islands of Lolland and Falster 1960–1970.] Nord
Med 86:1200–1201, 1971 (Danish)
17. Johnston MV: Clinical disorders of brain plasticity. Brain Dev
26:73–80, 2004
18. Kawabata H, Kawai H, Masatomi T, Yasui N: Accessory nerve
neurotization in infants with brachial plexus birth palsy. Micro-
surgery 15:768–772, 1994
19. Kawabata H, Shibata T, Matsui Y, Yasui N: Use of intercostal
nerves for neurotization of the musculocutaneous nerve in in-
fants with birth-related brachial plexus palsy. J Neurosurg 94:
386–391, 2001
20. Kay SPJ: Obstetrical brachial palsy. Br J Plast Surg 51:43–50,
1998
21. Kline DG: Different methods and results in the treatment of obstet-
rical brachial plexus palsy. J Reconstr Microsurg 16:420–422,
2000 (Letter)
22. Korak KJ, Tam SL, Gordon T, Frey M, Aszmann OC: Changes
in spinal cord architecture after brachial plexus injury in the
newborn. Brain 127:1488–1495, 2004
23. Krumlinde-Sundholm L, Eliasson A: Development of the assis-
ting hand assessment: a Rasch-built measure intended for chil-
dren with unilateral upper limb impairments. Scand J Occup
Ther 10:16–26, 2003
24. Laurent JP, Lee R, Shenaq S, Parke JT, Solis IS, Kowalik L:
Neurosurgical correction of upper brachial plexus birth injuries.
J Neurosurg 79:197–203, 1993
25. Lundborg G, Richard P: Bunge memorial lecture. Nerve injury
39

and repair—a challenge to the plastic brain. J Peripher Nerv
Syst 8:209–226, 2003
26. Macey AC, Burke FD, Abbott K, Barton NJ, Bradbury E, Brad-
ley A, et al: Outcomes of hand surgery. J Hand Surg [Br] 20:
841–855, 1995
27. Malessy MJ, Hoffmann CF, Thomeer RT: Initial report on the
limited value of hypoglossal nerve transfer to treat brachial
plexus root avulsions. J Neurosurg 91:601–604, 1999
28. Malessy MJ, van Duinen SG, Feirabend HK, Thomeer RT: Cor-
relation between histopathological findings in C-5 and C-6
nerve stumps and motor recovery following nerve grafting for
repair of brachial plexus injury. J Neurosurg 91:636–644, 1999
29. McNeely PD, Drake JM: A systematic review of brachial plex-
us surgery for birth-related brachial plexus injury. Pediatr Neu-
rosurg 38:57–62, 2003
30. Metaizeau JP, Gayet C, Plenat F: [Brachial plexus birth injuries.
An experimental study.] Chir Pediatr 20:159–163, 1979 (Fr)
31. Mumenthaler M, Schliack H: [Lesions of Peripheral Nerves.]
Stuttgart: Georg Thieme Verlag, 1977 (Ger)
32. Narakas A, Allieu Y, Alnot JY, Brunelli G, Merle M, Santos-Pa-
lazzi A, et al: [Complete supraclavicular paralysis. Surgical pos-
sibilities and results], in Alnot JY, Narakas A (eds): [Paralysis
of the Brachial Plexus.] Paris: Expansion Scientifique Fran-
çaise, 1989, pp 130–162 (Fr)
33. Palazzi S, Bonnard C, Raimondi P: Symposium on brachial plex-
us surgery. A Narakas Club. Barcelona (Spain) 13 and 14 March
1999. Chir Main 18:167–171, 1999
34. Piatt JH Jr: Neurosurgical management of birth injuries of the
brachial plexus. Neurosurg Clin N Am 2:175–185, 1991
35. Pondaag W, De Boer R, Van Wijlen-Hempel MS, Hofstede-
Buitenhuis SM, Malessy MJ: External rotation as a result of su-
40
Brought to you by Rutgers University | Unauthenticated | Downloaded 05/05/20 05:10 AM UTC
W. Pondaag and M. J. A. Malessy
prascapular nerve neurotization in obstetric brachial plexus le-
sions. Neurosurgery 57:530–537, 2005
36. Pondaag W, Malessy MJA, van Dijk JG, Thomeer RTWM: Nat-
ural history of obstetric brachial plexus palsy: a systematic re-
view. Dev Med Child Neurol 46:138–144, 2004
37. Sjoberg I, Erichs K, Bjerre I: Cause and effect of obstetric (neo-
natal) brachial plexus palsy. Acta Paediatr Scand 77:357–364,
1988
38. Sunderland S: Nerve injuries and Their Repair: A Critical
Appraisal. Edingburgh: Churchill Livingstone, 1991
39. Terzis JK, Vekris MD, Soucacos PN: Outcomes of brachial
plexus reconstruction in 204 patients with devastating paralysis.
Plast Reconstr Surg 104:1221–1240, 1999
40. Walker AT, Chaloupka JC, de Lotbiniere AC, Wolfe SW, Gold-
man R, Kier EL: Detection of nerve rootlet avulsion on CT mye-
lography in patients with birth palsy and brachial plexus injury
after trauma. AJR Am J Roentgenol 167:1283–1287, 1996
41. Waters PM: Comparison of the natural history, the outcome of
microsurgical repair, and the outcome of operative reconstruc-
tion in brachial plexus birth palsy. J Bone Joint Surg Am 81:
649–659, 1999
42. Xu J, Cheng X, Gu Y: Different methods and results in the treat-
ment of obstetrical brachial plexus palsy. J Reconstr Micro-
surg 16:417–420, 2000
Manuscript received August 16, 2005.
Accepted in final form March 27, 2006.
Address reprint requests to: Willem Pondaag, M.D., Department
of Neurosurgery, Leiden University Medical Center, P.O. Box 9600,
2300 RC Leiden, The Netherlands. email: w.pondaag@lumc.nl.
J. Neurosurg: Pediatrics / Volume 105 / July, 2006