Background and Aims: Despite the potential benefits of thromboelastography (TEG) for bedside hemostatic assessment in critical care settings, its accuracy remains to be determined, especially in critically ill neonates. We determined the intra-assay reproducibility of TEG parameters: Reaction time (R), clot kinetics (K) and Maximum Amplitude (MA) in a cohort of very low birth weight (VLBW) infants., Study Design: Observational study., Subjects: One hundred VLBW newborns., Outcome Measures: We performed TEG duplicate measurements for blood samples from VLBW newborns. To assess for correlation, we calculated the coefficients of correlation by plotting the values of the first vs the second measurement. Paired samples were compared with t-test and the coefficient of variation (CV) on paired results was also calculated as a measure of variability. To evaluate the agreement between duplicates, Bland-Altman (BA) analysis was performed., Results: We evaluated 228 TEG pairs. Both the coefficient of correlation and the BA analysis showed an acceptable level of agreement between duplicates. TEG variability (CV, mean ± SD) was highest for K (10.4%, ±12.9), lowest for MA (3.6%, ±8.0) and moderate for R (7.9%, ±9.0). The results from ANOVA one-way analysis describe different variability trends: K-CV increased at higher values, while MA-CV and R-CV increased at lower values., Conclusions: In VLBW newborns, the agreement between TEG duplicate measurements for R and MA parameters is adequate for clinical purposes. TEG is a promising tool to quickly assess hemostasis ensuring a significant blood sparing in critically ill neonates., (Copyright © 2018 Elsevier B.V. All rights reserved.)