FIRST RESULTS OF THE EUROPEAN NEC REGISTRY AND PLANS FOR AI-BASED INNOVATIONS – AN EXPLORATORY STUDY
Zeitschrift für Geburtshilfe und Neonatologie · 01.06.2026 · 230 (03) · S109
DOI: 10.1055/s-0046-1823540 ↗Kurz gefasst
Fachartikel in der Zeitschrift Zeitschrift für Geburtshilfe und Neonatologie, erschienen am 01.06.2026. Verfasst von H P Trapp und 5 weiteren Autor:innen.
Abstract (Originalsprache)
Aim
Despite decades of experience, necrotizing enterocolitis (NEC) remains one of the most severe gastrointestinal complications of preterm infants, with high mortality rates. As with most rare neonatal diseases, scientific evidence is largely based on single-center studies with small patient numbers. Our study’s aim is a first comprehensive characterization of European NEC patients in a large cohort of 1,310 neonates with severe intestinal disease based on data from the International Registry for Necrotizing Enterocolitis (IRNEC).
Methods
Data were collected from 14 European neonatal centers for preterm infants born between 2000 and 2025 treated for NEC and other gastrointestinal diseases. Variables included patient characteristics, perinatal data, maternal factors, outcomes, and treatment modalities. Results were compared between German and other European neonatal units to evaluate inter-site variability. Our first descriptive analysis is based on statistical methods, but further machine learning approaches regarding disease classification are currently being explored.
Results
A total of 1,310 neonatal cases were extracted from the IRNEC. Of those 580 patients had confirmed NEC (median gestational age 27 weeks (IQR 5); median birth weight 935 g (IQR 600)). Postnatal care and patient characteristics were mainly similar across German and other European units. Among other characteristics, 36% of mothers had chorioamnionitis, 22% had hypertension and 87% received antenatal corticosteroids. Congenital anomalies were observed in 20% of patients, and a patent ductus arteriosus (PDA) in nearly half (49%). During hospitalization, 71% of infants developed clinical signs of sepsis. Regarding treatment, 57% of confirmed NEC-patients were managed conservatively, and 43% underwent surgical intervention. German centers opted for surgery significantly more often than other units (64% vs. 45%, p=0.034), without adjustment for cohort disparities. Despite these therapeutic differences, survival was comparable across centers, with an overall in-hospital mortality of 34%.
Conclusion
This study provides new insights into the characteristics of NEC patients based on a representative large cohort. Besides highlighting a rather high level of standardization of care in the participating units, it serves as a foundation for thorough analysis of the patients included. To gain further insight especially in disease classification and risk factors, we are currently developing an AI model using Random Forest algorithms, trained on 500 datasets with a variety of variables. We expect the use of AI to advance statistical analysis by enabling faster and more accurate data assessment. Additional projects include generative AI–based data extraction, which could simplify data entry and improve the precision of registry data which is crucial for AI applications. Publication History Article published online: 10 June 2026 © 2026. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany
Zitieren
APA
Trapp, H. P., Pernthaler, M., Anzinger, F., Zwack, S. I., Muensterer, O. J., & Flemmer, A. W. (2026). FIRST RESULTS OF THE EUROPEAN NEC REGISTRY AND PLANS FOR AI-BASED INNOVATIONS – AN EXPLORATORY STUDY. Zeitschrift für Geburtshilfe und Neonatologie, 230(03), S109. https://doi.org/10.1055/s-0046-1823540
Vancouver
Trapp HP, Pernthaler M, Anzinger F, Zwack SI, Muensterer OJ, Flemmer AW. FIRST RESULTS OF THE EUROPEAN NEC REGISTRY AND PLANS FOR AI-BASED INNOVATIONS – AN EXPLORATORY STUDY. Zeitschrift für Geburtshilfe und Neonatologie. 2026;230(03):S109. doi:10.1055/s-0046-1823540
BibTeX
@article{Trapp2026,
author = {Trapp, H P and Pernthaler, M and Anzinger, F-X and Zwack, S I and Muensterer, O J and Flemmer, A W},
title = {FIRST RESULTS OF THE EUROPEAN NEC REGISTRY AND PLANS FOR AI-BASED INNOVATIONS – AN EXPLORATORY STUDY},
journal = {Zeitschrift für Geburtshilfe und Neonatologie},
year = {2026},
volume = {230},
number = {03},
pages = {S109},
doi = {10.1055/s-0046-1823540}
}
EUVIMED