Evaluation of gestational age, serum amyloid A, and hemoglobin decline (ΔHb) as diagnostic markers for NEC secondary to late-onset sepsis in preterm infants.

Objective: To evaluate gestational age, serum amyloid A (SAA), and hemoglobin decline (ΔHb) as diagnostic markers for necrotizing enterocolitis (NEC) secondary to late-onset sepsis (LOS) in preterm infants. Methods: A retrospective study was conducted on 77 preterm infants with LOS admitted to Anhui Provincial Children's Hospital from January 1, 2019, to October 31, 2024. The infants were divided into an NEC group (24 cases) and a non-NEC group (53 cases). Perinatal factors, initial blood counts, C-reactive protein, and SAA levels during early LOS were recorded. ΔHb was calculated as the difference between pre-LOS Hb concentration and initial Hb concentration at LOS onset. Differences were analyzed using the Mann-Whitney U test, χ 2 test, or Fisher's exact test. ROC curves were used to evaluate the predictive value. Results: The NEC group had significantly lower birth weight, gestational age, white blood cell count, neutrophil count, and lymphocyte count compared to the non-NEC group. In contrast, SAA, ΔHb, asphyxia incidence, and ventilator use rate were significantly higher in the NEC group (P < 0.05). Logistic regression analysis indicated that gestational age, SAA, and ΔHb were independent risk factors for NEC secondary to LOS. ROC curve analysis showed that the optimal cut-off values were 236 days for gestational age, 78.3 mg/L for SAA, and 15 g/L for ΔHb. The combined model had an area under the curve of 0.888 (95% CI: 0.810-0.953, P < 0.001), and the Youden's index was 0.637. Conclusion: Gestational age, SAA, and ΔHb appear to be useful indicators for predicting NEC secondary to LOS in preterm infants. The ease of obtaining these indicators and their low cost make them suitable for clinical application. However, this study was retrospective with certain uncontrollable factors and a relatively small sample size. Larger prospective studies are recommended for further validation.