Clinical characteristics associated with gastrointestinal bleeding in neonates with laboratory-confirmed norovirus...

Norovirus (NoV) is a major cause of acute gastroenteritis in children, but pathogen-confirmed data in neonates remain limited, particularly regarding clinical differences associated with gastrointestinal bleeding.

We conducted a single-center retrospective observational study of neonates younger than 28 days who were hospitalized at the Children's Hospital of Soochow University between December 31, 2024 and December 31, 2025, had positive stool NoV antigen results, and underwent confirmatory nucleic acid testing. After excluding clinically identifiable alternative causes of gastrointestinal bleeding, patients were classified into hemorrhagic and non-hemorrhagic groups. Baseline characteristics, clinical manifestations, comorbidities, and laboratory findings obtained within 24 h after admission were compared between groups. Laboratory variables showing significant between-group differences were entered into a multivariable logistic regression model.

A total of 139 neonates were included, comprising 52 in the hemorrhagic group and 87 in the non-hemorrhagic group. Among the hemorrhagic cases, 49 had overt gastrointestinal bleeding and three had isolated fecal occult blood positivity. The hemorrhagic group had a higher frequency of diarrhea and a lower frequency of jaundice. Hemoglobin, total bilirubin, indirect bilirubin, aspartate aminotransferase, the aspartate aminotransferase/alanine aminotransferase ratio, pH-corrected venous ionized calcium, chloride, and lactate were lower in the hemorrhagic group, whereas blood base excess was higher. In multivariable logistic regression, each 0.1 mmol/L increase in ionized calcium was associated with lower odds of the hemorrhagic phenotype (OR = 0.624, 95% CI: 0.410-0.950; P = 0.028).

Neonates with laboratory-confirmed NoV gastroenteritis showed different clinical and laboratory characteristics according to the presence of gastrointestinal bleeding. The association with lower ionized calcium should be interpreted cautiously because of the retrospective design and potential residual confounding.