Obstetric Brachial Plexus Injuries
D A Sherlock, T E Hems, Royal Hospital for Sick Children, Glasgow
SMJ 2004 49(4): 123-125
Despite improvements in obstetric care birth injuries of the brachial plexus (OBPP) seem to be on the increase.1 This is probably due to increases in birth weight since there is a close association between OBPP and high birth weight. The first case of OBBP was described in 1765 by Smellie, but it was not until 1872 when Duchenne described four cases of complete paralysis of the shoulder and elbow that it was recognized that this was a birth injury and not a congenital lesion. Erb in 1875 and 1876 defined his eponymous lesion involving the upper roots and trunks of the brachial plexus. Klumpke in 1885 described a similar lesion of the lower trunks. He also recognized the importance of Horner’s sign. Trombetta in 1880 reported that high birth weight and breech babies were particularly at risk. One of the first attempts at operative repair was performed by Kennedy in Glasgow in 1903, but it was not really until 1981 that the indications and outcomes of surgery for severe lesions were published by Narakas.2
The incidence of OBPP is unknown in Britain because it is not recorded as a separate condition. Worldwide the figures range from 0.1 to 4% of the live births. We estimate that in Scotland there are 50-70 cases per year. As noted previously breech births and high birth weight predict OBPP. Under 4kgs only 3% of babies develop OBPP compared to 4.7% of those weighing between 4 and 4.5kgs and 9.4% of those above 4.5kgs. Shoulder dystocia does not predict OBPP, but when present, it may increase the risk in high birth weight babies.
Some cases of OBPP occur antenatally. In these cases EMG studies at birth show evidence of nerve recovery, which is only possible if the lesion was pre natal. Some lesions are thought to occur as a result of a very precipitate 2nd stage causing rapid tilting of the neck as the baby traverses the birth canal. The majority occur as a result of force applied to deliver a child who is stuck and in distress. The force tends to separate the neck from the shoulder girdle, thereby stretching the brachial plexus. The upper roots are usually most at risk.
There are no clear strategies to prevent OBBP. Early delivery by induction or Caesarian section where a breech or high birth weight is suspected may reduce OBBP but at the risk of a non-physiological birth or the increased risks of surgery. If the baby ’s life is in danger, an OBBP is a small price to save a life.
The natural history of OBPP is variable. In general, there is substantial spontaneous recovery in the majority of cases. The degree of recovery depends primarily on the severity of the lesion. The most useful classification for severity of the lesion is by Narakas.3 This classification can only be applied after 2-4 weeks. Group 1 C5, 6. Biceps and deltoid paralysis. Ninety percent recover fully by two months. Group 2 C5, 6, 7. Only the long finger flexors work. Sixty-five percent recover fully by 3-6 months. The remainder have shoulder control defects. Group 3 Whole plexus involved with slight finger flexion only. Fewer than 50% recover fully and most have significant impairment of the shoulder, elbow, forearm, wrist and hand function. Group 4 Whole plexus involved plus Horner’s syndrome. None recover fully and all have permanent severe functional deficits. Spinal cord avulsions (2%) may affect ipsilateral leg function.
The age at which biceps function recovers is also a strong predictor of outcome. If biceps is working by the age of two months, full recovery is very likely. If biceps does not function until four months or later, then full recovery will not occur.(Waters4) When recovery is incomplete it is the usually the shoulder that fails to recover fully; in particular, the suprascapular nerve with consequent weakness of external rotation. Skeletal deformity may also develop at the shoulder, which further compromises function. This will be discussed below.
The diagnosis is made by the classical waiter’s tip position of the arm and paralysis of the various muscle groups. In 2% OBPP is bilateral. Some babies have phrenic nerve palsies. The differential diagnosis includes pseudo-paralysis due to infection or fractures of the clavicle or humerus. Cerebral palsy and arthrogryposis can also mimic OBPP.
Treatment is sub divided into initial management, early surgical treatment and late reconstructive surgery. Initial management Initial management comprises physiotherapy to prevent shoulder contractures, particularly internal rotation contractures. Splints are not helpful and may cause contractures. Because sensory nerves are also damaged input to the brain is abnormal and can result in inattention, even after a satisfactory recovery. Sensory input and movement encourage the development of central nervous connections. If grasp recovers by four weeks hand function will be excellent. Recovery of elbow flexion and shoulder abduction by three months is also a favourable sign. Lack of biceps recovery by three months warrants further investigation. Neurophysiological studies are difficult to interpret. However, work reported from the Royal National Orthopaedic Hospital (Bisinella et al5) indicates correct predictions of outcome for over 90% of C6 and C7 lesions. Predictions for C5 were less accurate, probably because outcome is influenced by skeletal deformity at the shoulder as well as nerve recovery. A MRI may show root avulsions and a cord haematoma.
There remains some disagreement about when early surgery to explore and repair the brachial plexus should be instituted. Waters4 found that surgical repair improved the outcome for patients who had not recovered biceps function by six months. Repair before 6 months did not improve the natural history, though failure to recover biceps function by three months was associated with some permanent deficit. There is good evidence that outcome is improved by nerve repair surgery in Narakas group 4 cases (Bisinella and Birch,6 Anand and Birch7). There is little evidence to support nerve surgery in less severe lesions. Overall less than 10% of cases of OBPP require nerve surgery If indicated, exploration of the brachial plexus should be carried out at about the age of six months. However, in cases of complete paralysis of the limb earlier operation may be considered. Nerve ruptures are repaired with nerve grafts. Root avulsions from the cord can only be repaired by grafts from other roots or nerve transfers (for example, the accessory or intercostal nerves). Nerve grafts are aimed particularly at reinnervating the hand, while nerve transfers are used for the shoulder and elbow. The results for nerve surgery are still poor though Gilbert8 found that after reinnervation of the lower plexus 75% of patients gained useful finger flexion and 50% had some intrinsic function. Waters4 assessed only shoulder function and showed that this was improved in patients with no biceps recovery at six months compared to the natural history. Anand and Birch6 found improved sensory recovery after surgery in group 4 cases. Recovery is very slow and can take up to four years. Sensory recovery is usually good but motor recovery is more variable and rarely complete.
As mentioned above, recovery of the suprascapular nerve, which innervates the shoulder external rotators, is frequently poor even in Narakas groups 1 to 3. It remains unclear whether early operation to transfer the accessory to suprascapular nerve might improve the outcome for shoulder function.
Late surgical reconstruction primarily concerns the shoulder and is required in about 25% of cases of OBPP. Muscle imbalance due to weak external rotators and strong internal rotators causes internal rotation contracture of the shoulder. Deformity of the glenoid and proximal humerus frequently occurs and in the worst cases is followed by subluxation and posterior dislocation of the shoulder. This seriously compromises the outcome and is preventable. Since recovery of suprascapular function can occur up to two years, surgery is rarely indicated before that time unless there is already an obvious shoulder dislocation. In some of these cases the dislocation may have been part of the traumatic birth process rather than due to muscle imbalance. If it is clear that external rotation, particularly active, is becoming restricted it is wise to assess the shoulder with x-rays. More information can be obtained with MRI which shows the cartilaginous parts of the glenoid and femoral head. However, this requires a general anaesthetic. Changes occur as early as four months. If the child is below the age of six, surgery usually takes the form of a release or lengthening of the subscapularis muscle anteriorly. This may be combined with a latissimus dorsi and teres major transfer so they become external rotators if there is no evidence of recovery in infraspinatus, the normal shoulder external rotator. Secondary deformities in the glenoid and humeral neck can make achieving reduction of a subluxed or dislocated shoulder difficult. A derotation osteotomy of the proximal humerus is often necessary in these circumstances but a reduction is not always possible. Most children gain significant benefit from this surgery because it enables them to place their arm into a functional position so that it can work in concert with the other hand. Shoulder elevation is not usually much changed. For a child over eight muscle transfers are less successful and a simple rotation osteotomy to move the arm into a more functional arc of movement may be the best option. Some loss of internal rotation is inevitable so it is usually necessary to have an excess of internal rotation to perform this procedure successfully.
Persistent winging of the scapula is common in after OBPP. This does not appear to result from damage to the long thoracic nerve. The mechanism is poorly understood. In addition to shoulder problems patients may have weak elbow flexion which can benefit from a transfer of pectoralis minor. Minor degrees of fixed flexion at the elbow are common but rarely cause any functional deficit and are best ignored. Dislocation of the radial head or a fixed supination or pronation deformity can be helped by a radius, or radius and ulna, rotation osteotomy. Weak wrist and finger extension can be helped by flexor to extensor tendon transfers.
Summary
It is clear that OBPP will continue to be a challenge for the foreseeable future. Because it is a rare condition the best results occur in specialised units for the fairly small number of patients who do not make a full recovery within two to three months of birth. This figure varies from 10 to 80%, according to the literature. Early referral to a specialist unit if there is not full, or almost full recovery, by two months will ensure that patients with OBPP obtain the optimal treatment available.
REFERENCES
1 Birch R et al. Birth Lesions of the Brachial Plexus in Surgical Disorders of the Peripheral Nerves. Churchill Livingstone: 1998, 209-233.
2 Narakas A. Brachial Plexus Surgery. Orthopaedic Clinics of North America 12: 1981, 303-322.
3 Narakas A. Obstetric Brachial Plexus Injuries in The Paralysed Hand. Churchill Livingstone: 1987, 116-135.
4 Waters P. J. Bone and Joint Surgery 81A: 1999, 649-659.
5 Binsella G, Birch R & Smith S. J Hand Surgery 28B: 2003, 148-152.
6 Bisinella GL & Birch R, J Hand Surgery 28B: 2003, 40-45.
7 Anand P & Birch R. Brain 125: 2002, 1-10.
8 Haerle M & Gilbert A. J. Paediatric Orthopaedics 24: 2004, 194-200.