We aimed to investigate the hemostatic status of diseased neonates using nonactivated rotational thromboelastometry (ROTEM) assay (NATEM) assay and, in addition, to evaluate the discriminative power of NATEM parameters in predicting the risk of bleeding in critically ill neonates and compare it to that of EXTEM (extrinsically activated ROTEM) parameters. This cohort study included 158 consecutive, critically ill neonates with presumed sepsis, perinatal hypoxia, or respiratory distress syndrome. The EXTEM and NATEM assays were performed on the first day of disease onset. The neonatal bleeding assessment tool was used to record and assess clinical bleeding events on the day of ROTEM analysis. Several EXTEM and NATEM ROTEM parameters differed between neonates with and without clinical bleeding events, indicating a hypo-coagulable state in neonates with clinical bleeding. NATEM parameters had comparable predictive performance for clinical bleeding events with EXTEM parameters for clotting time, clot formation time (CFT), A10 (clot amplitude at 10minutes), maximum clot firmness, lysis index at 60minutes, and maximum clot elasticity (p>0.05). However, NATEM A20, A30, and α angle demonstrated better predictive ability than EXTEM A20, A30, and α angle, respectively (p<0.05). A NATEM CFT value ≥147seconds presented 95.2% sensitivity (95% confidence interval [CI]: 76.1–99.8%) and 65.6% specificity (95% CI: 57.1–73.5%) to detect neonates with clinical bleeding, while a NATEM A10 value ≤42mm had 80.8% sensitivity (95% CI: 71.8–85.9%) and 76.0% specificity (95% CI: 52.8–91.7%) to detect neonates with clinical bleeding events. The NATEM assay has shown remarkable sensitivity in predicting bleeding in critically ill neonates, exceeding EXTEM performance in some selected parameters. The incorporation of NATEM test parameters in predictive models for neonatal hemorrhage seems promising.