Background: The role of hemostasis in the closure of patent ductus arteriosus (PDA) in preterm infants is controversial. Objective: To assess thromboelastography (TEG) at birth in very-low-birth-weight (VLBW) infants affected by PDA. Methods: This was an ancillary study of a prospective observational study aimed at defining the TEG profile in healthy VLBW infants in the first month of life. In this analysis, we included neonates of <33 weeks’ gestational age (GA) with PDA and compared TEG traces based on (1) spontaneous closure versus the need for pharmacological treatment and (2) treatment response. We collected blood samples in the 1st day of life to perform recalcified native-blood TEG (reaction time, maximum amplitude, and lysis at 30 min [Ly30)]), standard coagulation tests, and a full blood count. Results: We enrolled 151 infants with a PDA at the first echocardiogram; 111 experienced spontaneous PDA closure while 40 required treatment. Mean GA was 29.7 ± 1.7 and 27.6 ± 2.1 weeks, and birth weight was 1,158 ± 256 and 933 ± 263 g in the 2 groups, respectively (p < 0.001). The hemostatic profile was similar between groups. Median hematocrit (44.6 and 48.7%; p = 0.01) and platelet count (187 and 216 × 10<sup>3</sup>/μL; p = 0.04) were lower in the treated group, although differences lost significance after controlling for GA and illness severity in the multivariate analysis. Responders to PDA treatment (n = 20) had a significantly lower median Ly30 than nonresponders (0 and 0.7%; p = 0.02). Conclusion: TEG at birth does not predict spontaneous PDA closure in preterm newborns. Fibrinolysis is enhanced in nonresponders to PDA treatment; this observation warrants further investigation.