Coagulation dysfunction in term neonatal sepsis: a prospective cohort study

To characterize clinical features, conventional coagulation profiles, and rotational thromboelastometry (ROTEM) parameters in term neonates with sepsis, and to evaluate their associations with all-cause mortality within 28 days after sepsis diagnosis, with particular emphasis on whether a fibrin-based ROTEM parameter provides incremental prognostic information beyond organ dysfunction severity and conventional coagulation testing.

This prospective hospital-based cohort study was conducted at the Neonatal Center of Vietnam National Children's Hospital between June 2024 and December 2025. Term neonates (gestational age 37-42 weeks) diagnosed with neonatal sepsis according to the European Medicines Agency 2010 criteria were enrolled. Clinical severity was assessed using the neonatal Sequential Organ Failure Assessment (nSOFA) score. Conventional coagulation tests included platelet count, prothrombin time/international normalized ratio (INR), aPTT ratio, fibrinogen, D-dimer, and selected natural anticoagulants. Viscoelastic coagulation was evaluated using ROTEM (INTEM, EXTEM, and FIBTEM assays). The primary outcome was all-cause mortality within 28 days after sepsis diagnosis. Associations between clinical, laboratory, and ROTEM parameters and mortality were examined using univariable analyses and prespecified multivariable logistic regression models. Model discrimination was assessed using the area under the receiver operating characteristic curve (AUC).

A total of 168 term neonates with sepsis were included, of whom 23 died within 28 days (mortality rate, 13.7%). Non-survivors had significantly higher nSOFA scores at diagnosis than survivors (median 12 vs. 6; P < 0.001) and universally presented with septic shock and required mechanical ventilation. Conventional coagulation testing demonstrated a hypocoagulable profile among non-survivors, characterized by lower platelet counts, prolonged PT/INR and aPTT ratio, lower fibrinogen levels, and reduced protein S and antithrombin III activities. ROTEM analyses revealed reduced clot strength (lower A5, A10, and maximum clot firmness) and impaired clot propagation (prolonged clot formation time and reduced alpha angle) across INTEM, EXTEM, and FIBTEM assays. In multivariable logistic regression using complete-case data (n = 161; deaths=20), the nSOFA score remained independently associated with mortality (OR per 1-point increase, 1.53; 95% CI, 1.25-1.87), whereas INR was not. Addition of FIBTEM maximum clot firmness (per 10-mm increase) showed a protective direction of association with mortality (OR, 0.43; 95% CI, 0.17-1.09) but resulted in only a modest, non-significant improvement in discrimination (AUC 0.862 vs. 0.878; DeLong P = 0.455).

Among term neonates with sepsis, early organ dysfunction severity assessed by nSOFA was the strongest determinant of short-term mortality. Non-survivors exhibited a consistent hypocoagulable phenotype characterized by prolonged conventional coagulation times, reduced clot strength, and impaired clot propagation on ROTEM. Although FIBTEM maximum clot firmness reflected disease severity and showed a clinically meaningful association with mortality, its incremental prognostic value beyond nSOFA and conventional coagulation measures was limited. ROTEM may therefore be most useful as a complementary tool for hemostatic phenotyping and assessment of coagulation dysfunction rather than as a stand-alone mortality prediction instrument. Larger multicenter studies incorporating serial viscoelastic assessments are warranted to further define its prognostic and clinical utility.