Purpose <p>To evaluate the diagnostic accuracy, safety, and organisational aspects of point-of-care ultrasound (POCUS) in neonates admitted to neonatal intensive care units (NICUs), focusing on pulmonary, cardiac, neurological, and abdominal conditions.</p> Methods <p>A systematic review was conducted according to the PRISMA statement. MEDLINE, EMBASE, Cochrane Library, Web of Science, and CINAHL were searched from inception to May 2024. Grey literature and trial registries were also explored. Studies assessing the diagnostic accuracy of POCUS in NICU neonates were included. Risk of bias was assessed using QUADAS-2. Due to substantial heterogeneity, a narrative synthesis was performed.</p> Results <p>Thirteen diagnostic accuracy studies were included, evaluating pulmonary (<i>n</i> = 8), cardiac (<i>n</i> = 2), neurological (<i>n</i> = 2), and abdominal (<i>n</i> = 1) conditions. Most studies were observational and single-centre. In pulmonary conditions, sensitivity ranged from 80.0 to 100% and specificity from 63.6 to 100%. For patent ductus arteriosus, sensitivity ranged from 68.7 to 100% and specificity from 87.5 to 93.7%. In neurological pathology, POCUS showed high accuracy for severe intraventricular haemorrhage, but lower sensitivity for smaller or localised lesions. In abdominal pathology, sensitivity ranged from 45.0 to 90.9% and specificity from 85.7 to 89.2%. No adverse events attributable to POCUS were reported, although safety was not systematically assessed. Evidence on organisational aspects was limited and inconsistently reported.</p> Conclusion <p>POCUS may provide valuable bedside diagnostic information in NICUs, particularly in pulmonary conditions. However, the evidence remains limited by heterogeneity, small sample sizes, and operator dependence. Further prospective multicentre studies with standardised protocols are needed.</p>