Microsurgical reconstruction is indicated for infants with brachial plexus birth palsy (BPBP) that demonstrate limited spontaneous neurological recovery. This investigation defines the demographic, perinatal, and physical examination characteristics leading to microsurgical reconstruction. Infants enrolled in a prospective multicenter investigation of BPBP were evaluated. Microsurgery was performed at the discretion of the treating provider/center. Inclusion required enrollment prior to six months of age and follow-up evaluation beyond twelve months of age. Demographic, perinatal, and examination characteristics were investigated as possible predictors of microsurgical reconstruction. Toronto Test scores and Hospital for Sick Children Active Movement Scale (AMS) scores were used if obtained prior to three months of age. Univariate and multivariate logistic regression analyses were performed. 365 patients from six regional medical centers met the inclusion criteria. 127 of 365 (35%) underwent microsurgery at a median age of 5.4 months, with microsurgery rates and timing varying significantly by site. Univariate analysis demonstrated that several factors were associated with microsurgery including race, gestational diabetes, neonatal asphyxia, neonatal intensive care unit admission, Horner's syndrome, Toronto Test score, and AMS scores for finger/thumb/wrist flexion, finger/thumb extension, wrist extension, elbow flexion, and elbow extension. In multivariate analysis, four factors independently predicted microsurgical intervention including Horner's syndrome, mean AMS score for finger/thumb/ wrist flexion <4.5, AMS score for wrist extension <4.5, and AMS score for elbow flexion <4.5. In this cohort, microsurgical rates increased as the number of these four factors present increased from zero to four: 0/4 factors = 0%, 1/4 factors = 22%, 2/4 factors = 43%, 3/4 factors = 76%, and 4/4 factors = 93%. In patients with BPBP, early physical examination findings independently predict microsurgical intervention. These factors can be used to provide counseling in early infancy for families regarding injury severity and plan for potential microsurgical intervention.Level of Evidence: Prognostic Level I.