Neonatal brachial plexus palsy: An unpredictable injury

OBJECTIVE: Brachial plexus paralysis is a serious form of neonatal morbidity. We determined its incidence and persistence of disability and evaluated whether its occurrence could be predicted by maternal characteristics or partographic analysis in a large cohort of recent cases. STUDY DESIGN: During the years 1994 through 1998, all infants with neonatal evidence of obstetric brachial plexus injury were identified and followed greater than or equal to1 year. Obstetric details in these maternal-infant pairs were compared with 108 control pairs, matched for maternal and gestational age, parity, and birth weight. Partographs of cases and control subjects were reviewed, in a blinded manner, by 3 obstetricians who were asked to identify likely cases of nerve injury. A risk score comprising eight recognized associated clinical features was assigned. RESULTS: Fifty-four of 35,796 infants (1.5/1000) had evidence of brachial plexus injury; in 10 cases (19%), neurologic deficit persisted to 1 year. Although the risk factor profile was relatively higher in cases compared with control subjects, the highest score was 5 of 8 in six cases (2 cases, 4 control subjects). Obstetricians' partographic assessment identified "likely brachial plexus injury" in 13 of 54 cases (24%) and 16 of 108 control subjects (15%), and in only 3 cases (6%) did the assessors concur (positive predictive value, 7%-17%; negative predictive value, 5%-12%; sensitivity, 24%-50%; specificity, 66%-68%). Risk scores were similar among persistently and transiently injured cases. CONCLUSION: Our results indicate that brachial plexus injury is not predictable before delivery, either by risk factor scoring or by partographic analysis.