Background: Hemodynamically significant patent ductus arteriosus (hsPDA) is the most common functional cardiovascular disease in preterm infants. The necessity to treat hsPDA can neither be derived solely from clinical nor from echocardiographic criteria. Objective: The aim of this study was to establish non-invasive parameters which can differentiate hsPDA from non-hsPDA. Methods: Urinary protein levels of NT-proBNP, NGAL, and H-FABP were measured and correlated with the necessity of therapy for PDA. In 37 neonates (<1,500 g), urinary protein concentrations were tested on days 0, 2, and 7 by ELISA methodology. Of 37 infants, 12 required therapeutic interventions according to current treatment standards. Results: Infants receiving an intervention for PDA showed significantly higher levels of pro-BNP, NGAL, and H-FABP at all time points except for NT-proBNP on day 0. Infants requiring a second or third course of ibuprofen had significantly higher levels of H-FABP and NGAL. In all samples, the concentration of the three proteins correlated positively with each other. Conclusions: The present study shows that measurement of urinary proteins is a powerful and non-invasive method to quantify the effect of PDA on systemic perfusion in preterm infants. Furthermore, NGAL and H-FABP may be used to indicate the necessity of pharmacological or surgical treatment of PDA.