Brachial Plexus Palsy: Erb Palsy, Klumpke Palsy, Obstetric Palsy, and Birth-Related Brachial Plexus Palsy

GENESIS

The frequency of brachial plexus palsy has been decreasing with improved obstetric management from 1 per 1000 births before 2000, to 0.5 per 1000 births after 2000. Supraclavicular traction or stretching of the brachial plexus during delivery can injure nerve fibers; hence this injury is sometimes termed obstetric palsy . Historically risk factors have included shoulder dystocia, fetal macrosomia, labor abnormalities, forceps or vacuum delivery, and prior neonatal brachial plexus palsy, but these risk factors have not been shown to be reliable predictors. Other factors include familial congenital brachial plexus palsy, maternal uterine malformation, congenital varicella syndrome, osteomyelitis involving the proximal head of the humerus or cervical vertebral bodies, exostosis of the first rib, tumors and hemangioma in the region of the brachial plexus, and intrauterine maladaptation, but Kaiser Wilhelm syndrome (neonatal brachial plexus palsy due to placental insufficiency) is no longer considered to be a cause of brachial plexus palsy. Most cases occur in infants weighing less than 4500 g, whose mothers are not diabetic with no other identifiable risk factors, and cesarean section reduces but does not completely eliminate the risk for neonatal brachial plexus palsy. In a summary of 63 publications covering 17 million births, the likelihood of not having concomitant shoulder dystocia was 76% overall. In studies from the United States, the rate of permanent neonatal brachial plexus palsy is 1.1–2.2 per 10,000 births and 2.9–3.7 per 10,000 births in other countries.

The brachial plexus is formed by joining the anterior branches of the roots of C5, C6, C7, C8, and T1, and it emerges between the anterior and middle scalene muscles. In some cases, it receives a contribution from C4 (termed prefixed), while the contribution from T2 is termed postfixed. The brachial plexus is usually injured by excessive traction during labor, which can occur in up to 15% of newborns with shoulder dystocia. In most cases the lesion resolves within 6–12 months, but severe cases may require surgery, with risk of permanent damage in up to 10% of occurrences. The fibers that originate from the 5th and 6th cervical segments, which innervate the supraspinatus and infraspinatus muscles, are usually the most commonly (90% of cases) and severely affected. Occasionally fibers from C7, C8, and T1 also can be affected. Lesions that affect the upper segments (C5–C7) result in Erb palsy, whereas lesions that affect the lower spinal segments (C7–T1) result in Klumpke palsy. There is also total (mixed) palsy, which results in both motor and sensory palsy of the entire affected limb due to injury to all of the brachial plexus. There may be associated injuries suggesting a difficult delivery, such as fracture of the clavicle or humerus (9–21% of cases), diaphragmatic paralysis (5–9%), or facial palsy (5–14%). The position at delivery is related to the risk of brachial plexus injury, and infants delivered vaginally from an occipitoposterior position have a higher incidence of Erb palsy and facial palsy than those delivered from the occipitoanterior position.

Traction to the plexus, especially the upper plexus, occurs during delivery when the angle between the neck and shoulder is suddenly and forcibly increased, with the arms in an adducted position. This can occur during vertex deliveries when traction is placed on the head to deliver the after-coming shoulder, particularly when the shoulders are caught against the pelvic brim in shoulder dystocia, as forceful contractions push the head and trunk forward. Brachial plexus palsy also can occur during breech deliveries when the adducted arm is pulled forcefully downward to free the after-coming head (accounting for 24% of brachial plexus palsies) or during other malpresentations when the head is rotated to achieve an occipitoanterior presentation. The lower plexus is most susceptible to injury when traction is exerted on an abducted arm, such as occurs in vertex deliveries when traction is applied to an abducted prolapsed arm, or during breech deliveries when traction is applied to the trunk or legs while the after-coming arm is fixed in abduction. Spinal nerves are attached to the vertebral transverse process distal to the intervertebral foramen, encased in funnel-shaped dural sleeves, and enmeshed in a network of rami, cords, and trunks to form the brachial plexus; these factors serve to protect the plexus from traction injury. When traction is excessively rapid and forceful, then diffuse multifocal injury occurs, including avulsion of the roots from the cord in the most severe injuries. Previous risk factors (before 2000) have included technically difficult (57%) or breech (9%) deliveries, fetal macrosomia (weight greater than 4 kg) (55%), shoulder dystocia, multiparous mothers, prolonged labor, or fetal hypotonia leading to loss of the normal cushioning effect of intact muscle tone. Among 751 cases of neonatal brachial plexus palsy, 248 patients were born following routine deliveries and 503 patients were born following difficult deliveries. The routine delivery infants were more likely to have upper Erb palsy, whereas the difficult delivery group of infants were significantly more likely to develop total palsy. Poor functional recovery was more common in the difficult delivery group regarding shoulder, wrist, and hand function, suggesting that higher peak forces applied by the clinician in difficult deliveries affect the extent of neonatal brachial plexus palsy. There are also reports of prenatal-onset brachial plexus injuries in which denervation was demonstrated by electromyography (EMG) shortly after birth. One child demonstrated left brachial plexus injury, left Horner syndrome, left phrenic nerve injury, and hypoplasia of the left hand in addition to distortion of the first four ribs because of pressure on the left side of the neck and shoulder from the septum of a bicornuate uterus. In a retrospective study of the predictive value of prenatal ultrasound in 152 deliveries (2004–2012) complicated by shoulder dystocia, birthweight (odds ratio 12.1 for greater than or equal to 4000 g; 95% CI, 4.18–35.0) and vacuum extraction (odds ratio 3.98; 95% CI, 1.25–12.7) were the most significant clinical risk factors. Among cases with shoulder dystocia, the incidence of brachial plexus palsies was high (40%) and the impact of diabetes as a risk factor for shoulder dystocia decreased, reflecting improved screening and treatment.

The advent of microsurgical techniques and neuroelectrodiagnostic techniques has fostered the development of new neurosurgical techniques to repair brachial plexus injuries, although most infants recover spontaneously by 4 months of age. Severe lateral flexion of the infant’s neck at delivery makes avulsion of the lower brachial plexus nerve roots more likely than avulsion of the upper plexus. Among 91 infants observed through 2 years of age who sustained a brachial plexus birth injury and were treated with only physical and occupational therapy, 63 children with an upper or middle plexus injury recovered good to excellent shoulder and hand function. Of the remaining 28 infants, 12 sustained global injury, resulting in a useless arm, and 16 infants showed inadequate recovery of deltoid and biceps function by 6 months of age. These authors concluded that children with global injury would clearly benefit from early nerve reconstruction, and careful examination by age 6 months, in the sitting position, could confirm the potential for almost full recovery in most infants. Recovery of motor and sensory nerve function is attributed to axonal regeneration with re-innervations of original target muscle tissue, and functional improvement, which may continue for 5 years or longer. This longer period of recovery mirrors adaptational mechanisms at the spinal and supraspinal levels, which overcome initial motor neuron loss. According to the literature, hand functions are conserved in upper-root brachial plexus injury, and there is no need to evaluate them, but there can be activity restrictions related to hand functions involving forearm rotation. In children with total plexus injury, grasp was absent and thumb function was deficient. After perinatal upper brachial plexus injury to spinal roots C5 and C6, spinal root C7 contributes to biceps and deltoid innervations, but this does not occur in the adult. The absence of biceps clinical recovery in the 6th month of life and the presence of nerve root rupture signs are indications for surgery. Magnetic resonance imaging is the most sensitive and specific technique for identifying preganglionic nerve injuries such as rootlet avulsion, which has the poorest prognosis and may dictate the need for surgical intervention. Brachial plexus palsy can progress to significant sequelae, such as muscle contractures and glenohumeral dysplasia. Timely characterization of these entities based on different imaging modalities is a high priority for optimal patient outcomes.