Risk Factors for Poor Prognosis in Neonatal Necrotizing Enterocolitis

• Study Rationale Necrotizing Enterocolitis (NEC) is a life-threatening gastrointestinal disorder primarily affecting very low birth weight (VLBW) neonates (incidence: 5-12%). Around 20-40% of cases progress to severe disease requiring surgery, and survivors often face long-term complications (e.g., short-bowel syndrome, neurodevelopmental impairments). Despite advances in neonatal care, early identification of high-risk patients remains challenging due to nonspecific clinical/radiographic signs and insufficiently sensitive/specific conventional biomarkers.

NEC pathophysiology involves intestinal ischemia-reperfusion injury, excessive immune activation, and microbial dysbiosis. Serum lactate (a marker of hypoperfusion/anaerobic metabolism) correlates with NEC severity but lacks specificity; peripheral blood cell (CBC) indices are accessible but understudied in combination with lactate. The Neonatal Sequential Organ Failure Assessment (nSOFA) score predicts overall NEC mortality but fails to stratify risk for perforated NEC (PNEC), a high-risk subtype. This study aims to address gaps by integrating metabolic (lactate), hematologic (CBC), and organ dysfunction (nSOFA) metrics for NEC prognosis.

• Study Design and Conduct This is a retrospective cohort study conducted at the Guangzhou Women and Children's Medical Center, approved by the institutional ethics committee and compliant with the Helsinki Declaration. The study period is January 2017 and December 2022; eligible participants are identified via retrospective review of electronic medical records (EMRs), with data extracted in a structured manner.
• Data Collection

Structured EMR extraction captures key variables:

Neonatal history: Sex, gestational age, birth weight, birth asphyxia (Apgar scores), small for gestational age (SGA) status, preterm complications (e.g., patent ductus arteriosus [PDA], respiratory distress syndrome [RDS]), feeding pattern, mechanical ventilation use.

Maternal history: Delivery mode, chorioamnionitis, premature rupture of membranes.

NEC-related parameters: Bell's staging, clinical manifestations (e.g., abdominal distension, pneumoperitoneum), complications (sepsis, shock), therapeutic interventions (surgery), in-hospital mortality.

Laboratory/organ dysfunction indicators: Serum lactate (at NEC onset), CBC parameters (WBC, neutrophils, lymphocytes, platelets) and derived inflammatory indices, nSOFA score (at NEC onset).

• Statistical Analysis

Analyses use SPSS 26.0 and GraphPad Prism 9.0:

Data expression: Normally distributed variables (mean ± SD), skewed variables (median [IQR]), categorical variables (frequency %).

Intergroup comparisons: Chi-square/Fisher's exact test (categorical), t-test/Mann-Whitney U test (continuous).

Multivariate logistic regression: Includes variables with p<0.05 from univariate analysis to identify independent predictors.

Mediation analysis (PROCESS macro 4.1, 1000 bootstraps): Decomposes total effects into direct/indirect paths.

ROC curve analysis: Evaluates predictive performance of single/combined biomarkers (AUC as metric). All tests are two-tailed; p<0.05 is significant.

• Study Objectives Primary: Identify indicators predicting NEC progression and in-hospital mortality via analysis of clinical, metabolic, and laboratory data.

Secondary: Explore integrated biomarkers for NEC severity stratification (e.g., Bell's staging, surgical need) to optimize clinical decision-making.

• Participant Protection All data are de-identified before analysis and stored in a password-protected database (accessible only to the study team). Informed consent is obtained from participants' legal representatives, who are informed of the right to withdraw at any time without penalty.

Inclusion criteria

(1) Diagnosis of NEC;

Exclusion criteria

• Congenital gastrointestinal malformations (e.g., intestinal atresia, Hirschsprung's disease) or spontaneous intestinal perforation;
• Hereditary metabolic disorders;
• refusal of participation.