Primary Treatment of Neonatal Brachial Plexus Palsy

Primary Treatment of Neonatal Brachial Plexus Palsy

Bryce R. Bell

Ji-Geng Yan

Hani Matloub


  • Normally, the brachial plexus has contributions from C5 to T1.

  • Can have prefixed or postfixed cords with contributions from C4 (22%) and T2 (1%), respectively

  • Roots combine to form upper (C5/C6), middle (C7), and lower trunks (C8/T1).

  • Near the level of the clavicle, these trunks give off anterior and posterior divisions to form lateral, medial, and posterior cords, so named for their relation to the axillary artery. The lateral cord is formed from the confluence of the anterior divisions of the upper and middle trunks. The posterior cord is formed from the posterior divisions of all three trunks. Finally, the medial cord consists of only the anterior division of the lower trunk.

  • Cords then divide into terminal branches (musculocutaneous, axillary, radial, median, and ulnar).

  • Smaller branches originate from various levels in the brachial plexus and are important in localizing the level of injury by clinical examination (FIG 1).

  • The sympathetic chain runs along the ventral aspect of the exiting roots and is often disrupted in root avulsions from the spinal cord. This accounts for the findings seen in Horner syndrome.


  • Mechanism of injury

    • Most commonly a downward force on the shoulder causes deviation of the neck away from the involved extremity. This typically involves at least the C5-C6 roots but can result in complete plexus injury.

    • More rarely, injury can occur due to extreme cranial traction on the extremity with neck extension during breech delivery. These injuries often involve the lower roots.

    • Traditionally, injury was thought to be caused by traction during delivery by the obstetrician, by either manual or instrumented delivery.

    • However, given that more than half of children with brachial plexus injuries do not have an identifiable risk factor, some debate exists on the presence of an intrauterine mechanism of injury that is more difficult to identify. One study estimated uterine contractions exert up to 9 times greater force on the shoulder than that applied by the obstetrician.4

    • In addition, Walsh et al. showed no significant decrease in incidence of NBPP despite structured training on management of shoulder dystocia and a rising cesarean section rate.5

    • These injuries result in varying degrees of weakness in the ipsilateral extremity, ranging from subtle weakness to flaccid paralysis.

  • Classifications of Seddon and Sunderland

    • Seddon divided nerve injuries into three categories: neurapraxia, axonotmesis, and neurotmesis.6

      FIG 1 • The most common anatomy of the brachial plexus. The branching of the brachial plexus can be highly variable, and a thorough understanding of the most common variants is crucial prior to proceeding with brachial plexus exploration.

    • Neurapraxic injuries have a transient loss of conduction, but do not undergo wallerian degeneration. In axonotmesis, the nerve is still externally in continuity but undergoes wallerian degeneration due to axonal disruption. Neurotmesis indicates a complete division of a nerve.

    • Sunderland classified neurapraxia as type I and neurotmesis as type V. The axonotmesis category is divided into three separate categories, types II to IV, based on degree of internal injury short of complete division of the nerve.7 MacKinnon and Dellon later added a type VI injury, indicating multiple degrees of injury in a single cross-sectional area of a nerve.8 This has been termed a neuroma-in-continuity.

    • Wallerian degeneration occurs in axonotmetic and neurotmetic lesions due to axonal disruption. While there is some chance of recovery in axonotmetic lesions depending on the degree of internal injury, neurotmetic lesions will not recover without surgical repair or grafting.

  • Contractures

    • Denervation leads to a progressive joint contracture due to muscle paralysis. This joint contracture was generally believed to be caused by muscle imbalance and fibrosis of denervated muscle groups. However, animal studies have shown that muscle contracture precedes fibrosis, indicating a separate, yet to be determined mechanism.9

    • Two types of joint contractures are seen in NBPP patients. The first is due to overpowering of affected muscles by their antagonists, as is seen with the internal rotation contracture at the shoulder in Erb palsy. The second is due to muscle contracture and shortening, as seen in the glenohumeral abduction contracture caused by the paralyzed deltoid or the elbow flexion contracture caused by the brachialis in Erb palsy.

    • Global shoulder dysfunction and glenohumeral deformity develop as early as 3 months due to unbalanced forces on the shoulder and may require open or arthroscopic reduction and tendon transfers for treatment.

    • It has also been shown that each segment of the upper extremity exhibits decreased length and girth compared to the unaffected extremity.10


  • In the past, full recovery had been reported to be as high as 95.7%.11 However, more recent reports claim less optimistic spontaneous recovery rates in the 60% to 70% range.12,13

  • Upon closer analysis, it is clear that location and severity of injury, as well as the degree of spontaneous recovery by 3 to 6 months, can all dramatically change the prognosis for a given patient. Assessing these factors is essential to recommending a plan of care to the patient’s parents.

  • In patients with only mild weakness and rapid recovery within weeks, complete neurologic recovery is highly likely without surgical intervention.

  • However, in patients with a flaccid upper extremity and Horner syndrome (indicating a complete brachial plexus injury with avulsion), acceptable recovery is extremely unlikely, and early surgical intervention at about 3 months is recommended. Even in the case where there is some degree of proximal muscle improvement, if the hand is still paralyzed at 3 months, one should proceed with surgical intervention to reinnervate the hand, as these patients are unlikely to recover meaningful hand function on their own.

  • The remainder of neonatal brachial plexus patients fall between these two extremes, and the decision on the appropriate timing for intervention, if any, is still controversial.


  • History should include a detailed maternal gestational history as well as patient birth history. The physician should seek specific information on the course of delivery, shoulder dystocia, instrumented delivery, respiratory distress or signs of hypoxia, and Apgar scores. This information can be helpful in ruling out other causes of apparent limb weakness or paralysis (ie, cerebral palsy, infection, fracture of the involved limb).

  • Careful, systematic examination is crucial for making treatment decisions and comparisons over time. Given the importance of examination over time and the negative effects of delayed treatment, early referral to a multidisciplinary brachial plexus clinic is recommended. This will not only facilitate early surgical intervention when necessary but also ensure that patients receive prompt physical and occupational therapy to mitigate joint contracture formation while awaiting recovery.

  • Physical examination should be performed with the child naked from the waist up to assess the entire upper extremity as well as the neck, chest, back, scapula, and abdomen. In the infant, physical examination relies heavily on observance of limb posture, spontaneous motion, neonatal reflexes, and passive motion.

  • Certain postures have traditionally been associated with different levels of injury. The “waiter’s tip” posture describes a patient with the shoulder internally rotated, elbow extended, and wrist flexed. This posture would be suggestive of an upper plexus injury. Inversely, a patient with a lower plexus injury may have near normal shoulder and elbow function with a flaccid hand. A totally flaccid extremity indicates a complete brachial plexus injury.

  • Horner syndrome describes a patient with ipsilateral ptosis, meiosis, and anhidrosis. This is indicative of a preganglionic avulsion from the spinal cord.

  • There are many assessment tools and classification systems that can help synthesize information from the physical examination and assist in treatment decision-making. The most commonly used scores include the modified Mallet scale, the Active Movement Scale, and the Toronto Test Score. All three of these scores have been assessed for interand intrarater reliability.

  • The modified Mallet scale is a widely used measure of global shoulder function and does not grade specific muscles. It requires that the patient is mature enough to follow directions.

  • The Active Movement Scale (AMS) is a detailed scoring of 15 different movements on a scale of 1 to 7. It was developed to quantify motor function in infants with NBPP and evaluate recovery over time. It can be used as early as 2 to 3 weeks of age by eliciting neonatal reflexes to assess specific joint movements.15

  • The Toronto Test Score was developed to help predict recovery in children with NBPP. It can be used in young children who are unable to follow directions and tests five specific motions: elbow flexion, elbow extension, wrist extension, thumb extension, and finger extension.14


  • Although the decision to proceed with primary nerve surgery for NBPP is a clinical one, further diagnostic testing can provide useful information in development of a surgical plan.

  • Electrodiagnostic testing can be useful in determining preganglionic vs postganglionic lesions because avulsions will show normal sensory nerve action potential with no motor nerve conduction at the same level. However, these studies tend to overestimate the degree of recovery in NBPP and thus are not routinely used in surgical decision-making.

  • Computed tomography (CT) myelography can also be useful in determining the presence of root avulsions preoperatively. One study showed 98% specificity, but only 37% sensitivity for CT myelography in diagnosing root avulsions if no rootlets are seen traversing the pseudomeningocele.16

  • Another study compared CT myelography to MR myelography and found similar results for CT myelography to prior studies. However, MR myelography on a 3T magnet showed equivalent sensitivity and specificity without the need for lumbar puncture, intrathecal contrast injection, or ionizing radiation. Therefore, with improvements in technology, many now favor MR over CT myelography.17

  • Other imaging modalities that some find useful are ultrasound to assess diaphragmatic motion and chest radiography to look for an elevated hemidiaphragm. Abnormal findings would indicate phrenic nerve injury. This is particularly important if one is considering using the phrenic or intercostal nerves for plexus reconstruction.


  • Patients should be referred early to an occupational therapist with experience in neonatal brachial plexus injuries. Ideally, this therapist will see the patient at the same time as the surgeon to facilitate communication.

  • Therapy should focus stretching and range of motion of the entire upper extremity to prevent contractures and maladaptive motions. It is particularly important to stabilize the scapula during shoulder range of motion and stretching to make sure motion is truly glenohumeral and not scapulothoracic.


Feb 27, 2020 | Posted by in Pediatric plastic surgery | Comments Off on Primary Treatment of Neonatal Brachial Plexus Palsy

Full access? Get Clinical Tree

Get Clinical Tree app for offline access