BACKGROUND While there is a growing body of research on borderline personality disorder (BPD) in children and adolescents, controversy remains regarding the validity and diagnosis of the disorder prior to adulthood. METHODS MEDLINE, EMBASE, Psych INFO and PubMed databases were systematically searched for articles pertaining to the clinical and psychosocial outcomes (i.e. predictive validity) of BPD first diagnosed in childhood or adolescence (i.e. prior to 19 years of age). All primary empirical studies were included in the review. A narrative synthesis of the data was completed. RESULTS A total of 8200 abstracts were screened. Out of 214 full-text articles, 18 satisfied the predetermined inclusion criteria. Quality assessment indicated that most studies had high risk of bias in at least one study domain. Consistent with the adult literature, the diagnostic stability of BPD prior to the age of 19 years was low to moderate, and mean-level and rank-order stability, moderate to high. Individuals with BPD symptoms in childhood or adolescence had significant social, educational, work and financial impairment in later life. CONCLUSIONS Studies indicate that borderline pathology prior to the age of 19 years is predictive of long-term deficits in functioning, and that a considerable proportion of individuals continue to manifest borderline symptoms up to 20 years later. These findings provide some support for the clinical utility of the BPD phenotype in younger populations, and suggest that an early intervention approach may be warranted. Further prospective studies are needed to delineate risk (and protective) factors pertinent to the chronicity of BPD across the lifespan.