Back to Hand Surgery Index
Back to Hand Surgery IndexObstetrical brachial palsy (also known in its various forms as Erb’s palsy, Klumpke’s paralysis, Erb-Duchenne palsy) complicates a very small proportion of births. Furthermore it seems likely that many cases recover with little in the way of remaining deficit but it is equally certain that some cases will not recover. There is increasingly good evidence that microsurgical help at an early stage can improve the prognosis in some cases. Scepticism about the role of surgery abounds however and our Children’s Hand Clinic sees many cases referred too late for primary nerve surgery. We believe that specialist surgical advice, specialist physiotherapy and where appropriate, either early nerve surgery or secondary reconstructive procedures at a later date can improve the condition of many of these children. In this document we outline the background and the rationale for our policy as well as the policy itself as it stands at present.
The condition was probably first described by Smellie in 1779 (1) who cited a case of bilateral arm paralysis at birth which rapidly resolved in a matter of days. Danyau (1851) (2) described the post-mortem findings in an infant born with birth palsy who had blood within the plexus which was not however ruptured. Duchenne in 1872 (3) attributed the injury to traction on the arm. Although Flaubert had presented a case of total flaccid paralysis of the upper limb in an adult in 1827, Erb in 1874 (4) was the first to describe the typical upper root palsy and to localise the lesion in an adult to the junction of the C5 and C6 root with the upper trunk of the brachial plexus. In a postscript to this description he acknowledges Duchennes prior description and cites one of his own obstetrical cases which he attributes to pressure upon the plexus during "the Prague manoeuvre" of version and extraction. If an eponym should attach to the condition it should probably be Duchenne-Erbs palsy.
Theories on aetiology have abounded. The main theories have been that compression (either direct or indirect from instruments, fingers or between the bony structures) or traction are the cause (5). Some authors proposed that infective or ischemic causes might be found and even as recently as 1992 (6) a proposal has been made that there are two populations of OBP, those caused at birth and those caused during intrauterine life. This theory has not received wide currency at present however (7). Some elegant cadaver experiments showed (8) that lateral flexion with traction always ruptured the suprascapular nerve first, and that this was facilitated by fracture of the clavicle. Later theories addressed the possible role of disruption of the shoulder capsule or epiphysis as an attempt to explain the late deformity of the shoulder.
One of the most famous examples of Obstetrical Palsy was that of Kaiser Wilhelm II, grandson of Queen Victoria, whose breech presentation was complicated by nuchal arms, and whose left arm was used to turn the body.
The reported incidence varies. Adler and Patterson (9) reported a decreasing incidence of the condition in New York between the years 1938 (1.56 per 1000 live births) to 1962 (0.38 per 1000). They attributed this to improved obstetrical care. Bennett and Harrold (10) found an incidence at St Mary's Hospital of 0.61 per 1000 live births "including all but the most transient palsies". Of their 24 cases all bar 5 had complicated deliveries with overweight babies (4.2 Kg mean) predominating. All these children including the breech cases (2 babies) had upper plexus lesions in either Group I or Group II. Sjoberg et al (11) reported 48 cases in 25,736 live births over a ten year period in Malmo, for an incidence of 0.19%. They remarked that 25% had persistent palsy (1:2000 live births) and that high birth weight, shoulder dystocia and multiparity were associated.
Specht (12) performed a retrospective search of ten years births at the University of California Hospitals. He found 11 cases of brachial plexus lesions giving the incidence of 0.57 per 1000 live births (which because of the nature of the study almost certainly underestimates the true incidence.) More recently a study undertaken by Al-Rajeh and colleagues (13) looked at the incidence in the Eastern province of Saudi Arabia. They found 57 cases giving a frequency of 1.19 per 1000 live births and a prevalence rate of 0.32 per 1000 live births. They found evidence of either difficult deliveries (56%), large babies (35%) and assisted delivery with forceps or vacuum extraction (35%). Most presentations were vertex although there were some breech and some shoulder presentations. They also found associated injuries including fracture of the clavicle, the humerus, the skull and Horner’s syndrome. Only one patient had the hand muscles alone effected (Klumpke’s paralysis) but two cases had bilateral paralysis.
Camus et al (14) saw 33 cases in 30207 live births giving an incidence of 1.09 per 1000 live births. They noted the same risk factors already described and felt that careful obstetric monitoring and increased vigilance and training might prove preventative in many cases.
In Great Britain the incidence of this condition is not known (15). A reporting mechanism exists for congenital abnormalities of the limbs but this was established in 1962 in order to cope with scenarios of the Thalidomide type and reporting is undertaken within the first 10 days. There is no linkage in the system so that later updating is not possible and the list of recommended conditions for notifying does not include specifically obstetric brachial palsy.
The diagnosis of brachial plexus injury is easily made in a new born infant when the one upper extremity is not moved actively and the passive range of motion is equal on both sides. If the active and passive motion is equally restricted injury to the proximal humeral epiphyses should be suspected (9). Radiographs will help to exclude this lesion as well as a fractured clavicle. Accurate documentation of level of function in each joint is valuable for later comparison. Horner’s syndrome should be assiduously sought.
Other limbs should be examined as well, in order to rule out neonatal tetraplegia (which has a poor prognosis) and hemiparesis. In addition (16), radiographs should be taken to check for associated arm fractures or for paralysis of the hemidiaphragm (which indicates injury to the phrenic nerve). Eng et al (17) found a high instance of associated fractures (12 cases out of 135) including clavicle, humerus and metacarpal bones. In addition 3 cases had facial palsy and 2 had ipsilateral-diaphragmatic paralysis, whilst one child had subluxation of the cervical spine with cervical cord trauma.
Gilbert et al (18) routinely used EMG (including transoesophageal EMG of the diaphragm) prior to surgery but recognise that these often lead to false hopes of recovery and have been unable to correlate the EMG findings, with prognosis with the sole exception that a completely negative EMG showing no voluntary activity may indicate root avulsion and correspondingly bad prognosis. Eng et al (17) found electromyography useful for documenting the progress of recovery and also for selecting muscles for transfer. Laurent and Lee (19) have found electromyogram and SSEP ( Somatosensory evoked potentials) useful in refining the clinical evaluation but have not found it of prognostic or predictive value as others have reported (20).
Myelography is routine in our adult brachial plexus patients, but is difficult in children. It has been performed routinely by Gilbert et al pre-operatively. They usually perform this under general anaesthesia and report a significant false positive rate (i.e. signs of root avulsion that prove to be unfounded) with a small false negative rate (root avulsion present but not seen on myelography). They do however point out that there are some complications to this procedure including sterile meningitis and convulsions, albeit rare. Myelography may be combined with computerised tomography and may improve the diagnostic accuracy which still correlates relatively poorly with intraoperative findings (19,21). At present magnetic resonance imaging does not offer an improved diagnostic rate despite the ability of the fast spin-echo technique to demonstrate pseudomeningoceles (22,23).
Forty-eight hours post-partum, a more detailed neurological examination should be carried out. This will reveal further information about the nature and extent of the lesion. Out of 24 Cases, Bennett and Harrold (10) found 2 cases of radial nerve palsy and one case of septic arthritis of the shoulder causing diagnostic error.
It is widely accepted that there are generally three main types of lesion:
* C5-6 the arm is adducted and internally rotated at the shoulder, the elbow is extended, the forearm pronated, and the wrist (and sometimes fingers) flexed. This is the classical 'waiters tip' posture.
* C5-7 as above, although the elbow may be slightly flexed.
* C5-T1 the arm is totally flail with a claw hand. The arm has a marbled appearance due to vasomotor disturbance. It may or may not be accompanied by Horner's syndrome.
Narakas (24) classified brachial plexus lesions into four groups:
Group I: C5-6; paralysis of shoulder and biceps
Group II: C5-7; paralysis of shoulder, biceps and forearm extensors
Group III: C5-T1; complete paralysis of limb
Group IV: C5-T1; as above with Horner's syndrome
All major series highlight important risk factors. High birth weight (greater than 3600 grams), shoulder presentation, forceps or vacuum assisted delivery and multiparity have all been cited as risk factors. In addition breech presentation may also be a risk factor (9). According to Gilbert and Razaboni (18) large babies in the vertex presentation and small babies in the breech presentation are particularly at risk. In the latter category avulsion of the upper roots occurs in approximately 80% and this injury carries a substantially worse prognosis than the rupture of these roots which occurs in the vertex presentation (25). Ubachs et al (26) found that breech delivery was more likely to cause the more localised C5 C6 lesion, and these lesions were mainly avulsion. They postulated an explanation of this mechanism in the differential strong attachment of the C5 and C6 spinal nerves to the transverse process allowing a central mechanism (27) to be responsible for the avulsions whilst a peripheral mechanism was responsible for the ruptures in the cephalic presentation.
Undoubtedly the commonest origin of obstetric brachial palsy is the large baby with vertex delivery who suffers shoulder dystocia (19,28-30). Thus the common risk factors include multiparity, prolonged labour (31), high birth weight (31,32), assisted delivery (19,26,32,33), breech delivery (26,33), previous child with obstetric brachial palsy (34) and shoulder dystocia. There has been some suggestion that the incidence of brachial plexus palsy may be rising in association with a rise in birth weight although this has not been established in the United Kingdom where such data are not available. The prenatal diagnosis of large birth weight and prediction of the risk of shoulder dystocia is fraught with difficulties and it seems unlikely that we are any nearer a preventative strategy for this rare complication (35,36). In one study where prenatal ultrasound was used to estimate foetal weight, it was found that both medically and economically prophylactic cesarean section in non-diabetic women for this indication was unsound (37). Obstetric palsy can occur after cesarean section although it is exceptionally rare. In these circumstances a careful evaluation for other causes of limb paralysis should be made (38).
The reported instance of complete recovery varies enormously. Wickstrom (39) reported complete recovery in 12.9% and Arthuis (40) reported recovery in as many as 80% of cases. Gjorup (41) found that a third of patients obtained a usable arm (and all of these consider themselves handicapped in some way) and approximately a third had a useless arm with the other third lying somewhere between these two extremes. This extraordinary variation in reported incidence of useful recovery is probably attributable to variation in the original diagnosis (if one includes all cases diagnosed at birth many of them will improve by the time of a two week check to the point where the condition is no longer recognised) and to varying definitions of what constitutes a full or useful recovery. One large study reported good functional return in 95% of infants with more than 90% of those recovering showing improvement by four months of age (28).
Bennett and Harrold (10) found full recovery in 18 of 24 cases. Twelve of 14 Group I cases recovered fully, compared with 3 of 4 Group II cases. The other 3 cases in these 2 Groups made partial recoveries. They noted that in cases of eventual full recovery the outcome was apparent by 5 months, (and these children were only followed up for another 9 months for a final check), and all of these children showed some recovery in the first 2 weeks. In the 6 children with incomplete recovery, some improvement was seen but it halted at 1 year. These authors noted that recovery appeared to be centripetal, which they attributed to slow recovery in the universally paralysed suprascapular nerve, postulated to be at particular risk because of its anatomy.
Boome and Kaye (42) studied a group of cases selected for conservative care because of early return of activity in muscles innervated by C5 C6. In 21 cases with Group I lesions, the level of final function was related to age at which C5 C6 activity first was seen: in those with contraction by 3 months, then a recovery Grade 4/5 (combines MRC Grades 4 and 5) was seen. In 6 cases with C5 C6 contractions delayed until 4 months, then 5 had residual weakness at final review. Nine of the 21 children had weak or absent external rotation requiring muscle transfer to correct it. Interestingly, they report that of those with involvement of the lower roots also, all bar one recovered full function.
There are several proposed indicators of prognosis, and opinion varies about their usefulness. Some examples are:
1. Nature of injury e.g. avulsion vs rupture.
2. Extent of injury e.g. upper, middle or lower plexus lesion (25)
3. Associated ipsilateral Horner's syndrome (43)
4. Associated bone fractures e.g. ribs, clavicle, humerus. Al-Qattan et al (44) have shown in a study of 183 consecutive obstetric brachial plexus birth injuries that concurrent clavicular fracture has no prognostic value.
Brown (16) stated that poor prognostic signs included:
* Lower paralysis, which according to Ehrenfest almost never recovered.
* Total paralysis at birth, which never recovered completely, again according to Ehrenfest (45).
* Persistence of pupillary signs, as proposed by Sever (5)
Narakas (24) noted that recovery in group IV was especially poor, with 50% having little or no function. This would tend to suggest that the presence of Horner's is a bad prognostic sign, an opinion that Gilbert and Razaboni (18) agree with, and in such cases they scheduled surgery as soon as possible.
Jackson and Hoffer (43) analysed a group of 19 patients with obstetric brachial plexus palsy. They found that patients with a lesion of the upper brachial plexus had the best prognosis. Out of 13 limbs with a C5-6 palsy, all fully recovered, whereas only 4 out of 8 limbs with a lower plexus palsy had complete recovery. In addition to this, one of their patients had a clavicular fracture, and another had a clavicular fracture with Horner's. Neither achieved a full recovery. Unfortunately, with numbers as small as this it is impossible to draw any definite conclusions about the effect of clavicular fracture on prognosis. The work of Al-Qattan (44) shows no correlation between clavicular fracture or prognosis however (31).
Laurent, Lee and Shenaq (46) stated that ruptures were more common in upper roots and avulsions more common in lower. This finding was reinforced by experimental work of Metaizeau et al (47). This could be the explanation for why upper plexus lesions appear to carry a greater chance of recovery than lower ones.
Finally Michelow and Clarke (48) quoted phrenic nerve injury leading to hemidiaphragm paralysis and ipsilateral Horner's syndrome as indications of a severe lesion.
Affected infants should be followed up at clinic over a period of a couple of months, at regular intervals, in order to assess the extent of recovery, if any. Physiotherapy should be instigated on the affected limb to help prevent joint contractures. Some upper lesions may recover spontaneously within a few days or weeks, Narakas (24) has observed a 90% recovery rate in this group. However, in many cases it is not as straightforward as this, and the lesions may never recover fully, leaving the child with residual impairment. It is at this point that surgery may have a role to play.
Splinting and bracing in the "Statue of Liberty" position was at one stage popular but after treatment of 1100 patients, Sever (8) recommended against bracing because of supposed delay in recovery, whilst Milgram (49) reported abduction contracture of the shoulder was frequent. Adler and Patterson (9) agree that this latter serious complication is common after splinting is in that position.
In deciding the care of patients with OBP it is generally accepted that direct nerve surgery is less successful the later it is performed. Although the upper age limits and nerve repair has never been satisfactorily evaluated and some authors have anecdotally reported good results with repair as late as a year of even 24 months (19). However it is also accepted that the majority of cases recover spontaneously to a greater or lesser degree (19,31,42). It is for this reason that great attention has been devoted to prognostication at an early stage and studies of the natural history have assumed great importance. Adler et al (9) and others have documented well the late consequences of obstetrical palsy in untreated and treated patients including internal rotation contractures, hypoplasia of the arm, altered sensibility, flexion contractures of the elbow, dislocations of the radial head and very importantly, psychological and social consequences. Brown (50) has detailed these factors in terms of shoulder, elbow and forearm, and wrist and hand deformity very thoroughly. The correct management of these complications both in terms of prevention by physiotherapy, and correction by tendon transfers or osteotomy is beyond the scope of this brief review but is well documented in other places (16,50-56).
Unfortunately, the treatment of obstetrical brachial plexus injuries and the timing of surgical intervention has proved to be a very controversial field over the last century. Surgeons, like Kennedy (57), were operating on the brachial plexus as early as 1903 and reporting some success. However, during the 1920's and 30's surgery fell out of favour, and a more conservative approach was deemed the most suitable. This often involved splinting the shoulder in an abducted and externally rotated position, to prevent the internally rotated, adducted shoulder associated with obstetric brachial plexus palsy.
Adler and Patterson (9) stated that it was impossible to predict a patient's chances of recovery. They were, however, aware of the risks of development of contractures and deformities whilst awaiting any spontaneous recovery. Nor did they agree with the so-called Statue-of-Liberty splints mentioned above, recognising the danger of an abduction contracture of the shoulder with this procedure. With this in mind they recommended putting the affected arm through a series of passive movements to help prevent the development of these joint contractures. The parent was taught these, and the authors felt that complete recovery may take up to eighteen months. If recovery did not take place, the child could have secondary reconstructive surgery at about four years of age. Although most of their patients went on to lead normal, active lives, many had residual impairment or disability, especially involving the elbow.
Bennet and Harrold (10) advocated careful repeated examination. Postural changes were noted, and only muscles that never produced joint movement were classified as paralysed. Once again, mothers were taught simple physiotherapy to prevent joint contractures whilst awaiting recovery. They claimed a 75% rate of full recovery. Those who didn't recover fully improved up to one year, leaving them with internal rotation contractures and loss of elbow extension. They underwent tendon transfers at a later date.
It is clear from this that although some patients do recover spontaneously and regain a normal arm, others do not and are left with, in some cases, severe residual deformities. Thus, with the advent of new microsurgical techniques enabling primary plexus repair, and perhaps promising a better rate of recovery, surgical treatment returned to the fore. That is not to say that conservative treatment was considered obsolete, in fact quite the opposite. The challenge now lay in deciding which children would recover spontaneously, and which would need surgery to aid their progress.
Terzis and Liberson (58) used a combination of video tapes and EMG to assess recovery. The videos were used to examine the infants' behaviour, while the EMG was used to:
* discover the severity of the lesion
* test the extent of muscle reinnervation following the initial injury
Therefore the EMG was used primarily to detect the amount of clinical improvement, but only up to a period of nine months. If improvement did not meet their requirements (which they do not state) then surgery was scheduled. However, the authors themselves admit the difficulty involved in EMG testing on babies, due to reaction to stimulation and introduction of needles. Also, the EMG readings are not necessarily indicative of useful function obtained from the limb. In other words they are not quantitative.
Tassin in his thesis (59) reported on 44 infants followed from birth to the age of five, and classified shoulder function according to Mallet's scale (60). He stated that:
* complete recovery was only seen in patients with contraction of biceps and deltoid in the first month and normal contraction of these muscles by the third month.
* a good shoulder (Mallet class IV) wasn't obtained unless biceps and deltoid contraction began by three months and was normal by five months.
* average (class III) or poor (class II) shoulder resulted if biceps or deltoid began recovery after three months
Gilbert and Razaboni (18) modified Tassin's criteria due to difficulties in testing the deltoid. They proposed three indications for surgery:
The complete palsy with flail arm and Horners syndrome.The complete C5 C6 palsy without muscle contraction by 3 months and with a negative EMG (often, they say, corresponding to a complete root avulsion).C5 C6 palsy with no recovery in biceps at 3 months: biceps alone is chosen because examination of deltoid to the exclusion of pectoralis major is difficult at this age.
Myelography and EMG are undertaken prior to surgery. The authors (18) then describe their surgical results in a large number of cases before concluding that they are superior to spontaneous recovery in comparable cases. This paper with these criteria is so often quoted that it is worth observing that for surgery the claim is made that "in C5-C6 at 5 years we found 80% of good or near-normal results as compared with 0% in spontaneous recovery. Those figures are 45% for versus 0% for C5-C6-C7". The controls are presumed to be those of Tassin, who, they say, found poor (Grade II and III ) shoulder recovery if biceps recovery was delayed more than 3 months. In other words, if intervention wasn't undertaken, recovery would occur spontaneously, but it would be incomplete or poor.
Boome and Kaye (42) in their study of 70 children with 72 palsies at a secondary referral centre reported a group of 6 patients with C5-C6 lesions in whom first signs of muscle recovery were delayed until 4 months, but who were not selected for surgery. Only one made a full recovery. (The possibility exists that this one case was incorrectly assessed at the three month examination). From their study they stated that if recovery in the muscles innervated by C5-6 was delayed beyond three months then root disruption was likely, although only this small group of 6 patients with delay of recovery beyond 3 months was reported, since the rest of their "non-recoverers" went on to surgery. They therefore agreed with Gilbert and Razaboni's criteria for surgery.
A different proposal came from Zancolli and Zancolli (54). They suggest that for each level of the plexus there was a different key muscle so that groups one, two and three would have different key indicators for direct nerve surgery to the plexus. They were as follows:
* Upper Plexus (C5-6)
Biceps and deltoid were the key muscles. They claimed that 75% of biceps recovery began after five months, therefore surgery wasn't advisable before this, as recovery could still be continuing. Instead, if elbow flexion paralysis was definitive, this was used as a marker for surgery.
* Middle plexus (C5-7)
Key muscle was the triceps. · Complete paralysis (C5-T1) Key muscles were finger flexors and thumb extensors. If the hand recovered before three months, final function was generally good. If hand recovery had not appeared at 6 months, final recovery was very poor or non-existent.
In general they operated at between six to eight months if there were no clinical or electrophysiological signs of recovery, or if recovery had stopped for some time at a value of M2 or less on the British Muscle Movement Scale.
Laurent, Lee and Shenaq (46) also devised their own system using the three muscle groups of the upper brachial plexus; biceps, triceps and deltoid. Surgery was indicated if there wasn't improvement of at least one grade on the BMMS in two of the three muscle groups. The improvement had to be by four to six months of age, and continued improvement had to occur and be maintained at a level greater than M3 for another four months for surgery to be avoided. In a group of 44 patients who were judged not to need surgery, recovery had started by four months of age and progressed in all, which tends to cast doubt on Gilbert's limit of three months.
The most recent proposal is that of Michelow and Clarke (48) from Toronto. They worked in a unit which had a policy of scheduling children for primary plexus surgery based on recovery quality between 9 months and a year of age. They were therefore able to analyse the records of 39 patients over a 3 year period. At 3 monthly intervals from birth the child was examined and a selection of movements scored with a simple grading and scoring system. T
his system was used to measure shoulder adduction and abduction, elbow, wrist and finger extension and flexion. There were thus ten different movements, each with a score. This enabled each child at each stage to acquire a score for the limb. The outcomes of this group were simply divided into poor and good depending on the quality of shoulder abduction and elbow flexion (after Narakas (56)). Multivariate analysis was used to find those factors at three months of age most likely to predict good and poor recovery at one year of age. From this they discovered that using elbow flexion combined with elbow, wrist , thumb and finger extension gave the most accurate prediction of recovery. Those children with a score of 0 for elbow flexion (i.e. no recovery seen) at 3 months were eventually evenly distributed between the good and poor recoverers. Put another way, in this study absent elbow flexion at 3 months wrongly predicted recovery quality in 12.8% of cases, [four with M0 at 3 months went on to fit the good recovery group, whilst one with M1 at three months went on to the poor recovery group]. However when five different functions were scored, and those scores added up to less than 3.5 at 3 months, then the error rate was reduced to 5%, and no child with a lower score went on to a good recovery [although 2 children with higher scores recovered poorly]. The exact scoring system has since been slightly modified (61).
When compared with Gilbert’s criteria, this has important philosophical implications in that using this system the error rate would be smaller, unnecessary surgery is likely to be avoided, but there remains a small possibility that some patients who would benefit from surgery do not receive it, at the cost of protecting those who do not need surgery from unnecessary operations. For some surgeons this will be a more intrinsically conservative and therefore attractive policy. Differences between the conclusions of this group and of Gilbert may not be as real as they seem. The Toronto group split their outcome measures into two simple groups that may have differed slightly from those of Gilbert, and this is likely to be the main source of disagreement. Both groups had small total populations (Tassin 44 cases, Michelow 39) and conclusions must be guarded. In a commentary on the Toronto paper Laurent (46) from Houston agreed that all centres are now moving toward multiple muscle assessment as an indicator for surgery, and we should remember that Gilbert originally wished to examine deltoid as well as biceps, but dropped this because of the difficulty of such examination at 3 months (at which age the effects of pectoralis major are very difficult to separate from those of anterior fibres of deltoid.) The Houston group (19,46), having seen satisfactory recovery start at 4 months, do not consider children for surgery before that time.
Kennedy in 1903 published an account of surgery in three cases with one showing recovery of some function. One patient died (57). Taylor reported the results of exploration in 70 cases with improvement in many, but reported 3 deaths! Since then surgery suffered a dark period before the advent of microsurgery prompted a renaissance in the 1980s. Boome and Kaye (42) reported surgery in 22 children at Groote Schuur Hospital. The decision to operate was made if there was no recovery at 3 months in the muscles innervated by C5 and C6 (i.e. deltoid and Biceps Brachialis). The mean age by the time of surgery was 5 months, and still no recovery was seen. Two patients were in Group I, 13 in Group II and 7 in Group III + IV. After surgery and recovery they graded power into 3 grades, combining MRC Grades 1 and 2 to Grade 1/2, leaving Grade 3, and combining Grades 4 and 5 to Grade 4/5. External rotation was restored to power 4/5 in 25%, (the rest requiring muscle transfers for satisfactory function). 80% recovered good function in deltoid (Grade 4/5), whilst biceps was more disappointing with recovery to Grade 4/5 seen in 55%, and 20% were Grade 1/2 or less. They commented that in these four cases with no useful biceps recovery the proximal root stump had also been used for either wrist or deltoid reinnervation, probably diluting the benefit to biceps: a remediable factor therefore. Ten of 20 cases followed up had full or near full recovery in both biceps and deltoid. Surgery to the lower roots was infrequent and disappointing.
Gilbert and Razaboni (18) reported on 178 surgical cases between the years 1977 and 1986. Of these 120 had a follow up greater than 2 years. They used a slight modification of Mallets shoulder classification (detailed in their paper) and observed that for the Group I (C5-C6) lesions in those (49) cases with rupture 80% achieved a good or near-normal shoulder. For cases of avulsion, only 40% achieved a Grade IV (good) shoulder, by nerve transfer. They also reported that the biceps "recovered" in all bar one of the Group I (C5-C6) cases. For Group II cases 65% achieved a good or near-normal shoulder (Grades IV or IV+), but triceps recovered in all cases, and thumb extension was restored in 70%.
Gilbert and Whitaker (62) then published results using Mallets original unmodified scoring system showing, for 290 reviewed cases (out of 350 operated cases) results for the Group I (C5-C6) 81% with class III, IV or V shoulders, compared to only 16% with these classes in the Group II (C5C6C7) cases.
Laurent et al have reported the results for conservative treatment, neurolysis, direct repair and sural nerve grafting. The most dramatic improvements were achieved with sural nerve grafting and again, in all groups by twelve months post-operatively biceps had recovered to at least muscle score Grade of 3 (active against gravity). The most dramatic results were evident with nerve grafting and those patients undergoing neurolysis or decompression of the scarred interscaline interval achieved less dramatic results but had less severe conditions pre-operatively (46). Other authors have found improvement following neurolysis in the C5, C6 lesion but no benefit in the complete plexus palsy (63).
Since the Renaissance in microsurgical treatment of obstetric brachial palsy in the 1980s advancement in the treatment of the adult lesion and the infantile lesion have developed in parallel. The discovery of the efficacy of nerve transfers in the upper limb has prompted their use in obstetrical brachial palsy and particularly beneficial results may be seen in children from nerve transfers such as the eleventh nerve to suprascapular nerve, the intercostal nerves to the musculocutaneous nerve and intraplexal transfers. At a recent conference (64) several surgeons described results of such transfer which have particular value in the case of avulsed roots but may also be used in preference to direct nerve repair (for instance in the eleventh cranial nerve to suprascapular nerve transfer). It is possible to perceive an important change in philosophy in the extensive lesion where every effort may be made by the use of nerve grafting, intraplexal nerve transfers and extraplexal nerve transfers, to direct as much regenerating neuronal material to the forearm and hand as is possible.
Future developments in this field seem likely to progress apace. Diagnostic accuracy and therefore prognostication and a selection of patients for surgery is likely to come from a combination of improvement in imaging, refinement of electrophysiological studies (65) and developments in the understanding of the natural history. Despite pleas to the contrary, it is unlikely given the present findings that a randomised control study of surgery versus conservative treatment will be possible because of difficulties in case selection and admission criteria. Nonetheless rigorous documentation of pre-operative condition and operative results is necessary and will require the universal adoption of scoring methods. These have been begun by Narakas with the use of the modified Mallet classification and continued by other authors with scoring systems appropriate for grading recovery in arm and hand.
The remarkable discoveries of regeneration from spinal chord and ventral horn to peripheral nerve shown so elegantly in animals and in adult human cases (66-68) may yet find an even more important place in the treatment of the obstetrical lesion where one might be optimistic about the regenerative abilities of the child’s neuronal tissue.
Although obstetric brachial palsies, and indeed adult brachial palsies, represent a small proportion of the reconstructive nerve work of the upper limb in any population, there is no doubt that the anatomical complexity of this area in combination with the intellectual complexity of our knowledge of the subject at present, and the rapid and exciting pace of change make this an area of reconstructive work suitable for very specialised units treating more than a small number of cases per annum. Just in other areas of reconstructive work (such as cleft lip and palate or craniofacial surgery, or in cancer surgery (ref)) recommendations for the minimum establishment and minimum number of cases treated per annum have been made to ensure some constituency in the expertise and management of such complex cases, so there is a case for calling upon the Department of Health and Royal Colleges to impose similar guidelines on the profession in the management of obstetric and adult brachial plexus palsy.
In our Children’s Hand Clinic we like to see obstetrical brachial palsy patients as early as possible. In keeping with other children’s hand anomalies, we believe that early discussion with a clinician who is very familiar with the condition and its implications has a positive and reassuring benefit on the parents. All children in our Children’s Hand Clinic receive the option of support from a specialist physiotherapist and from a clinical psychologist experienced in the counselling of parents and patients suffering from these conditions. We also encourage patients to join the self-help groups (Erb’s Palsy Support Group) which provide invaluable emotional support for the children and the families. We do not use static splinting in our cases but instruct parents at a very early stage in passive stretching and maintenance, particularly in external rotation of the shoulder.
Later clinic appointments are scheduled for three months of age when we employ the criteria of Michelow and Clarke. In order for accurate assessment to be made, all these assessments are made by the same physiotherapist during an open ended clinic interview prior to the main clinic. These assessments are then checked and discussed in the main clinic and based on this the rationale of surgery is explained to the parents and if appropriate an offer of primary nerve surgery is made.
Primary nerve surgery is then scheduled if appropriate within a month of that appointment and the child is reassessed on admission. In the rare cases in which there has been a dramatic improvement in that intervening month the implications of that improvement are discussed and a judgement as to whether to continue with surgery is made. We are not didactic in this respect, and would prefer to spend some time explaining to the parents the paucity of knowledge available in this area before formulating a care plan for the child.
For those children who do not proceed to surgery, regular follow-up and monitoring is continued. Physiotherapy continues to be undertaken by the parents with close contact and supervision by our therapist.
For later palliative surgery, if internal rotation of supination contractures have been avoided then muscle transfers can be delayed until the age of approximately three years. For children who develop internal rotation contractures resistant to physiotherapy, or who are referred late with contractures, surgical release of those contractures may be undertaken very early in order to prevent malformation of the effected joints. This leaves the way free for further physiotherapy and later tendon transfers as required.
We frequently see children presenting to the clinic very late with severe lesions and little recovery. We have the facility in our clinic to offer a combination of microvascular muscle transplantation with proximal nerve transfers for a small selected group of these patients and in appropriate cases have seen useful restoration of function that might not otherwise be achieved. We are therefore glad to see these patients even at a late date in order to attempt to palliate their condition. In rare occasions rotation osteotomies may prove valuable.
Unfortunately, not only may families that present late have received no early discussion with a specialist surgeon, but they may also not have received early support from a physiotherapist and clinical psychologist experienced in these conditions. Even in families that do not receive early medial input, the Erb’s Palsy Support Group have noted that lack of counselling and emotional support is a particular problem in most parts of the country. They would like to see this more widely available. In our own Children’s Hand Clinic such input is available from the first clinic appointment. However, we would also welcome earlier referral so that parents can be visited in the perinatal period when they are acutely distressed, confused and vulnerable and perhaps most in need of an opportunity to ventilate their feelings, ask questions and feel supported. In addition to contacting any member of the team at this point, this would also be an appropriate time for families to receive information about the Erb’s Palsy Support group. Families are then aware that such help is available, which can be reassuring in itself, and can make contact at whatever point feels appropriate for them.
In the Children’s Hand Clinic we recognise that there is no right answer to many of the questions posed by obstetrical brachial palsy. We believe that the establishment of centralised care for this rare condition and the development of statistically significant series of carefully monitored data is important.
1. Smellie W. A collection of cases and observation in midwifery to illustrate
his former treatise or first volume on that subject. New ed. London: W Strahan,
T Cadell, G Nicol; 1779;p. 446-8.
2. Danyau M. Paralysie du membre superieur, chez le nouveau-ne. Bull Soc Chir
1851;2:148.
3. Duchenne GBA. 3rd ed. Paris: J.B. Balliere et fils; 1872;De l'electrisation
localisee et de son Application a la pathologie et la therapeutique. p. 357-62.
4. Erb W. Ueber eine eigenthümliche localisation von Lähmun gen
implexus brachialis. Verhandlungen des naturhisorischen vereins von Heidelberg
1874;2:130-7.
5. Sever JW. Obstetric paralysis: its etiology, pathology, clinical aspects
and treatment, with report of four hundred and seventy cases. American Journal
of Diseases in Children 1916;12:541-78.
6. Jennett RJ, Tarby TJ, Kreinick CJ. Brachial plexus palsy: an old problem
revisited [see comments]. American Journal of Obstetrics & Gynecology
1992;166(6:Pt 1):Pt 1):1673-6.
7. Slooff AC, Ubachs JM. Brachial plexus impairment - a birth trauma? [letter;comment].
American Journal of Obstetrics and Gynecology 1993;169(1):230
8. Sever JW. Obstetric paralysis: report of eleven hundred cases. Journal
of the American Medical Association 1925;1862-5.
9. Adler JB, Patterson RL, Jr. Erb's palsy. Long-term results of treatment
in eighty-eight cases. Journal of Bone & Joint Surgery - American Volume
1967;49(6):1052-64.
10. Bennet GC, Harrold AJ. Prognosis and early management of birth injuries
to the brachial plexus. British Medical Journal 1976;1(6024):1520-1.
11. Sjoberg I, Erichs K, Bjerre I. Cause and effect of obstetric (neonatal)
brachial plexus palsy. Acta Paediatrica Scandinavica 1988;77(3):357-64.
12. Specht EE. Brachial plexus palsy in the newborn. Incidence and prognosis.
Clinical Orthopaedics & Related Research 1975;(110):32-4.
13. al-Rajeh S, Corea JR, al-Sibai MH, al-Umran K, Sankarankutty M. Congenital
brachial palsy in the eastern province of Saudi Arabia. Journal of Child Neurology
1990;5(1):35-8.
14. Camus M, Vauthier D, Lefebvre G, Veron P, Darbois Y. [Retrospective study
of 33 cases of obstetric paralysis of the brachial plexus]. [French]. Journal
de Gynecologie, Obstetrique et Biologie de la Reproduction 1988;17(2):220-8.
15. Birch R. Surgery for Brachial Plexus. Journal of Bone & Joint Surgery
- British Volume 1993;75B(3):346-8.
16. Brown KL. Review of Obstetrical Palsies: Non-operative Treatment. Clinics
in Plastic Surgery 1984;3(1):181-5.
17. Eng GD, Koch B, Smokvina MD. Brachial plexus palsy in neonates and children.
Archives of Physical Medicine & Rehabilitation 1978;59(10):458-64.
18. Gilbert A, Razaboni R, Amar-Khodja S. Indications and results of brachial
plexus surgery in obstetrical palsy. Orthopedic Clinics of North America 1988;19(1):91-105.
19. Laurent JP, Lee RT. Birth-related upper brachial plexus injuries in infants:
operative and nonoperative approaches [see comments]. Journal of Child Neurology
1994;9(2):111-7.
20. Terzis JK, Liberson WT, Levine R. Terzis JK, editors.Microreconstruction
of nerve injuries. Philadelphia: W B Saunders Co; 1987; 38, Our experience
in obstetrical bachial plexus palsy. p. 513-28.
21. Hashimoto T, Mitomo M, Hirabuki N, Miura T, Kawai R, Nakamura H, Kawai
H, Ono K, Kozuka T. Nerve root avulsion of birth palsy: Comparison of myelography
with CT myelography and somatosensory evoked potential. Radiology 1991;178:841-5.
22. Francel PC, Koby M, Park TS, Lee BC, Noetzel MJ, Mackinnon SE, Hene MM,
Kaufman BA. Fast spin-echo magnetic resonance imaging for radiological assessment
of neonatal brachial plexus injury. Journal of Neurosurgery 1995;83(3):461-6.
23. Miller SF, Glasier CM, Griebel ML, Boop FA. Brachial plexopathy in infants
after traumatic delivery: Evaluation with MR imaging. Radiology 1993;189:481-4.
24. Narakas AO. Lamb DW, editors.The paralysed hand. Edinburgh: Churchill
Livingstone; 1987; 9, Obstetrical brachial plexus injuries. p. 116-35.
25. Geutjens G, Gilbert A, Helsen K. Obstetric brachial plexus palsy associated
with breech delivery. Journal of Bone and Joint Surgery 1996;78B(2):303-6.
26. Ubachs JM, Slooff AC, Peeters LL. Obstetric antecedents of surgically
treated obstetric brachial plexus injuries. British Journal of Obstetrics
and Gynaecology 1995;102(10):813-7.
27. Sunderland S. Nerve Injuries and their repair. Edinburgh: Churchill Livingstone;
1991;Traction Nerve Injury. p. 147-58.
28. Gordon M, Rich H, Deutschberger J. The immediate and long term outcome
of obstetric birth trauma. American Journal of Obstetrics and Gynecology 1973;117:51-66.
29. Sharpe W. The operative treatment of brachial plexus paralysis. JAMA 1916;66:876-84.
30. Zancolli EA. Classification and management of the shoulder in birth palsy.
Orthopedic Clinics of North America 1981;12:433-57.
31. Walle T, Hartikainen-Sorri AL. Obstetric shoulder injury. Associated risk
factors, prediction and prognosis. Acta Obstetricia et Gynecologica Scandinavica
1993;72(6):450-4.
32. Iffy L, Varadi V, Jakobovits A. Common intrapartum denominators of shoulder
dystocia related birth injuries. Zentralblatt fur Gynakologie 1994;116(1):33-7.
33. Benjamin B, Khan MR. Pattern of external birth trauma in Southwestern
Saudi Arabia. Journal of Trauma 1993;35(5):737-41.
34. Al-Qattan MM, al-Kharfy TM. Obstetric brachial plexus injury in subsequent
deliveries. Annals of Plastic Surgery 1996;37(5):545-8.
35. Gonen R, Spiegel D, Abend M. Is macrosomia predictable, and are shoulder
dystocia and birth trauma preventable? Obstetrics and Gynecology 1996;88((4
Pt 1)):526-9.
36. Baskett TF, Allen AC. Perinatal implications of shoulder dystocia. Obstetrics
and Gynecology 1995;86(1):14-7.
37. Rouse DJ, Owen J, Goldenberg RL, Cliver SP. The effectiveness and costs
of elective cesarean delivery for fetal macrosomia diagnosed by ultrasound.
JAMA 1996;276(18):1480-6.
38. Al-Qattan MM, el-Sayed AA, al-Kharfy TM, al-Jurayyan NA. Obstetrical brachial
plexus injury in newborn babies delivered by caesarean section. Journal of
Hand Surgery 1996;21B(2):263-5.
39. Wickstrom J, Haslam ET, Hutchinson RH. The surgical management of residual
deformities of the shoulder following birth injuries of the brachial plexus.
The Journal of Bone and Joint Surgery 1955;37A(1):27-36.
40. Arthuis M. Paralysis obstetricale du plexus brachial. Etude diagnostique.
SOFCOT XLVI Reunion Annuelle. Rev Chir Orthop 1972;59(Suppl):124-8.
41. Gjorup L. Obstetrical lesion of the brachial plexus. Acta Neurologica
Scandinavica 1966;42:Suppl:18:1-80.
42. Boome RS, Kaye JC. Obstetric traction injuries of the brachial plexus.
Natural history, indications for surgical repair and results. Journal of Bone
& Joint Surgery - British Volume 1988;70(4):571-6.
43. Jackson ST, Hoffer MN, Parrish N. Brachial-plexus palsy in the newborn.
Journal of Bone and Joint Surgery 1988;70:1217-20.
44. Al-Qattan MM, Clarke HM, Curtis CG. The prognostic value of concurrent
clavicular fractures in newborns with obstetric brachial plexus palsy. Journal
of Hand Surgery 1994;19(6):729-30.
45. Ehrenfest H. Birth injuries in the Child. New York: D Appleton and Co;
1922;Brachial plexus palsy. p. 165
46. Laurent JP, Lee R, Shenaq S, Parke JT, Solis IS, Kowalik L. Neurosurgical
correction of upper brachial plexus birth injuries. J Neurosurg 1993;79:197-202.
47. Metaizeau JP, Gayet C, Plenat F. [Brachial plexus birth injuries. An experimental
study (author's transl)]. [French]. Chirurgie Pediatrique 1979;20(3):159-63.
48. Michelow BJ, Clarke HM, Curtis CG, Zuker RM, Seifu Y, Andrews DF. The
natural history of obstetrical brachial plexus palsy. Plastic & Reconstructive
Surgery 1994;93(4):675-80.
49. Milgram JE. Discussion of L'Episcopo. New York J Med 1939;39 50.
50. Brown LB. Terzis JK, editors.Microreconstruction of nerve injuries. Philadelphia:
W B Saunders; 1987; Review of Obstetrical Palsies: Nonoperative treatment.
p. 499-511.
51. Liberson WT, Terzis JK. Terzis JK, editors.Microreconstruction of nerve
injuries. Philadelphia: W B Saunders; 1987; Contribution of Clinical Neurophysiology
and Rehabilitation Medicine to the Management of Brachial Plexus Palsy. p.
555-67.
52. Strecker WB, McAllister JW, Manske PR, Schoenecker PL, Dailey LA. Sever-L'Episcopo
transfers in obstetrical palsy: a retrospective review of twenty cases. Journal
of Pediatric Orthopedics 1990;10(4):442-4.
53. Buschmann WR, Sager G. Orthopaedic considerations in obstetric brachial
plexus palsy. Orthopaedic Review 1987;16(5):290-2.
54. Zancolli EA, Zancolli ER, Jr. Palliative surgical procedures in sequelae
of obstetric palsy. Hand Clinics 1988;4(4):643-69.
55. Pollock AN, Reed MH. Shoulder deformities from obstetrical brachial plexus
paralysis. Skeletal radiology 1989;18(4):295-7.
56. Narakas AO. The treatment of brachial plexus injuries. International Orthopaedics
1985;9(1):29-36.
57. Kennedy M. Suture of the brachial plexus in birth paralysis of the upper
extremity. BMJ 1903;1:298-301.
58. Terzis JK, Liberson WT, Levine R. Obstetric brachial plexus palsy. Hand
Clinics 1986;2(4):773-86.
59. Tassin, J.L. Paralysies Obstetricales du Plexus Brachial. Evolution Spontanee,
Resultats des Interventions Reparatrices Precoces. 1983; University Paris
VII;
60. Mallet J. Paralysie obstetricale du plexus brachial. Symposium sous la
direction de J Mallet. Revue de Chirurgie Orthopedique et Reparatrice de l
Appareil Moteur 1972;58(Suppl):115-204.
61. Clarke HM, Curtis CG. An approach to obstetrical brachial plexus injuries.
Hand Clinics 1995;11(4):563-81.
62. Gilbert A, Whitaker I. Obstetrical brachial plexus lesions. Journal of
Hand Surgery - British Volume 1991;16(5):489-91.
63. Clarke HM, Al-Qattan MM, Curtis CG, Zuker RM. Obstetric brachial plexus
palsy: results following neurolysis of conducting neuromas-in-continuity.
Plastic and Reconstructive Surgery 1996;97(5):974-84.
64. Brachial Plexus Symposium. Vienna: March 8-11th 1997.
65. Smith SJM. The role of neurophysiological investigations in traumatic
brachial plexus lesions in adults and children. Journal of Hand Surgery 1996;21B(2):145-7.
66. Carlstedt T. Nerve fibre regeneration across the peripheral-central transitional
zone. Journal of Anatomy 1997;190(1):51-6.
67. Carlstedt T, Grane P, Hallin RG, Noren G. Return of function after spinal
cord implantation of avulsed spinal nerve roots. Lancet 1995;346(8986):1323-5.
68. Carlstedt T. Spinal nerve root injuries in brachial plexus lesions: basic
science and clinical application of new surgical strategies. A review. Microsurgery
1995;16(1):13-6.
Professor Simon Kay, Consultant Plastic Surgeon and the Children’s Hand Clinic team would like to thank Alexander Ford and Emma Dawdry for their work on this project.
TOP
 |
 |
|
 |
 |
 |
 |
|
Indications for Surgery
Results of Surgery
Future Directions in Nerve Surgery
Policy
Conclusion
Bibliography
Acknowledgements
Introduction
Aetiology and Incidence
Diagnosis
Classification
Risk Factors
Prognostic Factors and Natural History
Early Management|
|
Cosmetic,
Plastic and Reconstructive Surgery including: |
