Clinical Classification Versus MRI Patterns of Injury in Brachial Plexus Birth Injury.

PURPOSE

The Narakas classification describes brachial plexus birth injury (BPBI) according to nerve root injury by the pattern of motor weakness on clinical examination. However, it is unknown whether the classification truly corresponds to the described nerve roots. The distribution of nerve root injuries on magnetic resonance imaging (MRI) in infants with BPBI was compared with the clinical classification.

METHODS

Infants with BPBI were prospectively enrolled at three children's hospitals, and the Narakas group was determined by physical examination. Infants underwent MRI prior to age 16 weeks. Neuroradiologists determined the injury severity (intact, rupture, avulsion) at each nerve root on MRI. The nerve root findings on MRI were compared with the expected nerve root injuries, based on the clinical Narakas classification.

RESULTS

Sixty-eight infants completed the MRI revealing 19 distinct patterns of nerve injury. The nerve root injury findings on MRI did not always correspond with the nerve roots involvement expected based on the Narakas classification. In Narakas 1 patients, 23% had injury to C5-C6 only, and 55% had additional injuries to C7, C8, and/or T1. In Narakas 2 patients, only 26% had an injury specifically to C5-C7 only. In the Narakas 3 and 4 groups, 43% had a C5-T1 global injury as expected by the Narakas classification. The mean number of nerve roots affected, and mean avulsions increased with higher Narakas grades. The most commonly injured and avulsed nerve roots were C6 (n = 60) and C8 (n = 15), respectively.

CONCLUSIONS

In 68 infants, 19 different patterns of injury were identified, suggesting that the pathoanatomy of BPBI is more nuanced than classically described. For Narakas 1 and 2 infants, the nerve root injury on MRI was often more extensive than expected based on clinical examination. Our results suggest the Narakas classification may not precisely correspond with the injury at the root level, as seen on MRI.

TYPE OF STUDY/LEVEL OF EVIDENCE: Diagnostic II.