Background: Over the past decade, many studies have reported the association of brain natriuretic peptide (BNP) and N-terminal pro-brain natriuretic peptide (NT-proBNP) with clinical outcome of intracerebral hemorrhage (ICH). However, a broad consensus has not been reached.Objective: To evaluate the role of BNP/NT-proBNP levels in prognosis and disease severity assessment in patients with ICH.Methods: A systematic literature search was conducted utilizing PubMed, Embase, Web of Science and the Cochrane Library databases up to July 23, 2021. Studies that explored the association between BNP/NT-proBNP level and clinical outcome or disease severity in ICH patients were eligible. Outcome measures were all-cause mortality, poor functional outcome, adverse cardiac events and markers of disease severity.Results: Ten studies, involving 1,373 patients with ICH, met the inclusion criteria. Nine studies focused on clinical outcomes (five all-cause mortality, five functional outcomes, and one adverse cardiac event) and seven on disease severity. In terms of prognosis, all five studies showed an association between elevated BNP/NT-proBNP level and increased risk of all-cause mortality in ICH patients. Four of the five studies reported poor functional outcomes in patients with higher BNP/NT-proBNP levels and one study associated higher BNP/NT-proBNP levels with increased risk of adverse cardiac events. Moreover, two studies identified an additional predictive ability of BNP/NT-proBNP level beyond that of pre-existing prognostic variables. In terms of disease severity, five studies (71%) reported that BNP/NT-proBNP level correlated positively with hematoma volume in addition to ICH and GCS scores.Conclusion: Elevated BNP/NT-proBNP level is associated with increased risk of all-cause mortality, poor functional outcome, adverse cardiac events and disease severity in patients with ICH. Thus, BNP/NT-proBNP level is a promising prognostic indicator for ICH and also an effective marker of disease severity. Current evidence remains limited by the small number and high heterogeneity of included studies. Further appropriately designed, large-scale studies are required to confirm the current findings.