Necrotizing enterocolitis (NEC), with the main manifestations of bloody stool, abdominal distension, and vomiting, is one of the leading causes of death in neonates, and early identification and diagnosis are crucial for the prognosis of NEC. The emergence and development of machine learning has provided the potential for early, rapid, and accurate identification of this disease. This article summarizes the algorithms of machine learning recently used in NEC, analyzes the high-risk predictive factors revealed by these algorithms, evaluates the ability and characteristics of machine learning in the etiology, definition, and diagnosis of NEC, and discusses the challenges and prospects for the future application of machine learning in NEC.
Infant, Newborn ; Humans ; Enterocolitis, Necrotizing/therapy* ; Infant, Newborn, Diseases ; Prognosis ; Gastrointestinal Hemorrhage/diagnosis* ; Machine Learning

Objective: To retrospectively analyze the clinical data of different types of selective intrauterine growth restriction (sIUGR) pregnant women under expectant management, including the natural evolution, typing conversion and perinatal outcomes. Methods: The clinical data of 153 pregnant women with sIUGR under expected treatment in Women's Hospital, Zhejiang University School of Medicine from January 2014 to December 2018 were collected. Maternal characteristics including maternal age, gravidity, parity, method of conception, pregnancy complication, gestational age at delivery, indication for delivery, birth weight, the rate of intrauterine and neonatal death and neonatal outcomes were recorded. Pregnant women with sIUGR were divided into three types according to end-diastolic umbilical artery flow Doppler ultrasonography, and the differences of typing conversion and perinatal outcomes of sIUGR pregnant women based on the first diagnosis were compared. Results: (1) Clinical characteristics and pregnancy outcomes: among 153 pregnant women with sIUGR, 100 cases (65.3%) were diagnosed with type Ⅰ, 35 cases (22.9%) with type Ⅱ, and 18 cases (11.8%) with type Ⅲ. There were no significant differences in age, conception mode, pregnancy complications, first diagnosis gestational age, characteristics of umbilical cord insertion, delivery indications, fetal intrauterine mortality and neonatal mortality among three types of sIUGR pregnant women (all P>0.05). The average gestational age at delivery of type Ⅰ sIUGR was (33.5±1.9) weeks, which was significantly later than those of type Ⅱ and Ⅲ [(31.3±1.8), (31.2±1.1) weeks, P<0.001]. The percentage disordance in estimated fetal weight (EFW) of type Ⅰ sIUGR was significantly lower than those of type Ⅱ and type Ⅲ (P<0.001). The incidence rate of neonatal intensive care unit (NICU) admission, cerebral leukomalacia and respiratory complications of both fetus and necrotizing enterocolitis of large fetus in type Ⅰ were significantly lower than those in type Ⅱ and type Ⅲ (all P<0.05). (2) Typing conversion: in 100 cases of type Ⅰ sIUGR, 18 cases progressed to type Ⅱ and 10 cases progressed to type Ⅲ. Compared with 72 stable type Ⅰ sIUGR, those with progressed type Ⅰ sIUGR had higher incidence of NICU admission and lung disease in both fetuses, and cerebral leukomalacia and necrotizing enterocolitis in large fetus (all P<0.05). The proportion of inconsistent cord insertion was significantly higher in those type Ⅰ progressed to type Ⅲ (6/10) than in those with stable type Ⅰ (19.4%, 14/72) and type Ⅰ progressed to type Ⅱ sIUGR [0 (0/18), P=0.001]. Four cases of type Ⅱ sIUGR reversed to type Ⅰ and 6 cases reversed to type Ⅲ. Compared with type Ⅱ reversed to type Ⅰ sIUGR, those stable type Ⅱ and type Ⅱ reversed to type Ⅲ sIUGR had a higher incidence of NICU admission in large fetus (P<0.05). Two cases of type Ⅲ sIUGR reversed to type Ⅰ and 6 cases progressed to type Ⅱ. There were no significant differences in fetal serious complications in type Ⅲ sIUGR with or without doppler changes (all P>0.05). Conclusions: The different types of sIUGR could convert to each other. The frequency of ultrasound examinations should be increased for patients with the type Ⅰ sIUGR, especially when the percentage discordance in EFW is substantial or with discordant cord insersion.
Pregnancy ; Female ; Infant, Newborn ; Humans ; Fetal Growth Retardation/epidemiology* ; Pregnancy Outcome ; Retrospective Studies ; Enterocolitis, Necrotizing ; Twins, Monozygotic ; Umbilical Arteries/diagnostic imaging* ; Gestational Age ; Ultrasonography, Prenatal/methods* ; Pregnancy, Twin

OBJECTIVES: To investigate the clinical characteristics and risk factors for early-onset necrotizing enterocolitis (NEC) in preterm infants with very/extremely low birth weight (VLBW/ELBW). METHODS: A retrospective analysis was performed on the medical data of 194 VLBW/ELBW preterm infants with NEC who were admitted to Children's Hospital Affiliated to Zhengzhou University from January 2014 to December 2021. These infants were divided into early-onset group (onset in the first two weeks of life; n=62) and late-onset group (onset two weeks after birth; n=132) based on their onset time. The two groups were compared in terms of perinatal conditions, clinical characteristics, laboratory examination results, and clinical outcomes. Sixty-two non-NEC infants with similar gestational age and birth weight who were hospitalized at the same period as these NEC preterm infants were selected as the control group. The risk factors for the development of early-onset NEC were identified using multivariate logistic regression analysis. RESULTS: Compared with the late-onset group, the early-onset group had significantly higher proportions of infants with 1-minute Apgar score ≤3, stage III NEC, surgical intervention, grade ≥3 intraventricular hemorrhage, apnea, and fever or hypothermia (P<0.05). The multivariate logistic regression analysis showed that feeding intolerance, blood culture-positive early-onset sepsis, severe anemia, and hemodynamically significant patent ductus arteriosus were independent risk factors for the development of early-onset NEC in VLBW/ELBW preterm infants (P<0.05). CONCLUSIONS VLBW/ELBW preterm infants with early-onset NEC have more severe conditions compared with those with late-onset NEC. Neonates with feeding intolerance, blood culture-positive early-onset sepsis, severe anemia, or hemodynamically significant patent ductus arteriosus have a higher risk of early-onset NEC.
Child ; Infant ; Female ; Pregnancy ; Infant, Newborn ; Humans ; Infant, Premature ; Infant, Extremely Low Birth Weight ; Ductus Arteriosus, Patent ; Enterocolitis, Necrotizing/etiology* ; Retrospective Studies ; Infant, Newborn, Diseases ; Infant, Premature, Diseases/etiology* ; Risk Factors

OBJECTIVES: To study the effect of gut microbiota on hematopoiesis in a neonatal rat model of necrotizing enterocolitis (NEC). METHODS: Neonatal Sprague-Dawley rats were randomly divided into a control group and a model group (NEC group), with 6 rats in each group. Formula milk combined with hypoxia and cold stimulation was used to establish a neonatal rat model of NEC. Hematoxylin and eosin staining was used to observe the pathological changes of intestinal tissue and hematopoiesis-related organs. Routine blood tests were conducted for each group. Immunohistochemistry was used to observe the changes in specific cells in hematopoiesis-related organs. Flow cytometry was used to measure the changes in specific cells in bone marrow. 16S rDNA sequencing was used to observe the composition and abundance of gut microbiota. RESULTS: Compared with the control group, the NEC group had intestinal congestion and necrosis, damage, atrophy, and shedding of intestinal villi, and a significant increase in NEC histological score. Compared with the control group, the NEC group had significantly lower numbers of peripheral blood leukocytes and lymphocytes (P<0.05), nucleated cells in the spleen, thymus, and bone marrow, and small cell aggregates with basophilic nuclei in the liver (P<0.05). The NEC group had significant reductions in CD71⁺ erythroid progenitor cells in the liver, CD45⁺ lymphocytes in the spleen and bone marrow, CD3⁺ T lymphocytes in thymus, and the proportion of CD45⁺CD3^-CD43⁺SSC^hi neutrophils in bone marrow (P<0.05). There was a significant difference in the composition of gut microbiota between the NEC and control groups, and the NEC group had a significant reduction in the abundance of Ligilactobacillus and a significant increase in the abundance of Escherichia-Shigella (P<0.05), which replaced Ligilactobacillus and became the dominant flora. CONCLUSIONS Multi-lineage hematopoietic disorder may be observed in a neonatal rat model of NEC, which may be associated with gut microbiota dysbiosis and abnormal multiplication of the pathogenic bacteria Escherichia-Shigella.
Rats ; Animals ; Enterocolitis, Necrotizing/etiology* ; Gastrointestinal Microbiome ; Rats, Sprague-Dawley ; Animals, Newborn ; Infant, Newborn, Diseases

OBJECTIVES: To systematically evaluate the risk factors for necrotizing enterocolitis (NEC) in preterm infants. METHODS: PubMed, Embase, Cochrane Library, China National Knowledge Infrastructure, and Wanfang Data were searched for case-control studies and cohort studies on the risk factors for NEC in preterm infants published up to December 2021. RevMan 5.3 software was used to perform the Meta analysis. RESULTS: A total of 38 studies were included (28 case-control studies and 10 cohort studies). The Meta analysis showed that maternal gestational diabetes (OR=2.96, P<0.001), intrahepatic cholestasis during pregnancy (OR=2.53, P<0.001), preeclampsia (OR=1.73, P=0.020), history of neonatal asphyxia (OR=2.13, P<0.001), low gestational age (OR=1.23, P=0.010), sepsis (OR=5.32, P<0.001), patent ductus arteriosus (OR=1.57, P=0.001), congenital heart disease (OR=3.78, P<0.001), mechanical ventilation (OR=2.23, P=0.020), history of antibiotic use (OR=1.07, P<0.001), use of vasopressors (OR=2.34, P=0.040), and fasting (OR=1.08, P<0.001) were risk factors for NEC in preterm infants, while cesarean section (OR=0.73, P=0.004), use of pulmonary surfactant (OR=0.43, P=0.008), and breastfeeding (OR=0.24, P=0.020) were protective factors against NEC. CONCLUSIONS Maternal gestational diabetes, intrahepatic cholestasis during pregnancy, preeclampsia, low gestational age, fasting, sepsis, patent ductus arteriosus, congenital heart disease, and histories of asphyxia, mechanical ventilation, antibiotic use, and use of vasopressors may increase the risk of NEC in preterm infants, while cesarean section, use of pulmonary surfactant, and breastfeeding may decrease the risk of NEC in preterm infants.
Anti-Bacterial Agents ; Asphyxia ; Cesarean Section ; Cholestasis, Intrahepatic ; Diabetes, Gestational ; Ductus Arteriosus, Patent ; Enterocolitis, Necrotizing ; Female ; Humans ; Infant ; Infant, Newborn ; Infant, Newborn, Diseases ; Infant, Premature ; Pre-Eclampsia ; Pregnancy ; Pulmonary Surfactants ; Risk Factors ; Sepsis

OBJECTIVES: To study the clinical value of intestinal regional oxygen saturation (rSO₂) and C-reactive protein (CRP) in the diagnosis of necrotizing enterocolitis (NEC) in preterm infants. METHODS: A prospective observational study was conducted among the preterm infants who were hospitalized in Children's Hospital Affiliated to Anhui Medical University, from October 2020 to December 2021, with 22 infants in the NEC group and 35 infants in the non-NEC group. Intestinal rSO₂ was monitored 24 hours after a confirmed diagnosis of NEC in the NEC group, and serum CRP levels were measured before anti-infection therapy. In the non-NEC group, intestinal rSO₂ monitoring and serum CRP level measurement were performed at the corrospording time points. The two groups were compared in terms of intestinal rSO₂ and serum CRP level. The receiver operating characteristic (ROC) curve was used to analyze the value of intestinal rSO₂ alone, serum CRP alone, and intestinal rSO₂ combined with CRP in the diagnosis of NEC in preterm infants. RESULTS: Compared with the non-NEC group, the NEC group had a significantly lower level of intestinal rSO₂ (P<0.05) and a higher serum CRP level (P<0.05). The ROC curve analysis showed that intestinal rSO₂ had an optimal cut-off value of 50.75% in the diagnosis of NEC in preterm infants, with a sensitivity of 81.8%, a specificity of 85.7%, and an area under the ROC curve (AUC) of 89.4%; CRP had an optimal cut-off value of 12.05 mg/L in the diagnosis of NEC in preterm infant, with a sensitivity of 72.7%, a specificity of 74.3%, and an AUC of 74.8%; intestinal rSO₂ combined with CRP had a sensitivity of 90.9%, a specificity of 77.1%, and an AUC of 91.9% in the diagnosis of NEC. The AUC of intestinal rSO₂ alone in the diagnosis NEC was higher than that of CRP (P<0.05). There was no significant difference in the AUC between intestinal rSO₂ alone and intestinal rSO₂ combined with CRP (P>0.05). CONCLUSIONS The value of intestinal rSO₂ in the diagnosis NEC is higher than that of CRP, and is equivalent to that of the combination of intestinal rSO₂ and CRP in preterm infants.
Infant ; Child ; Infant, Newborn ; Humans ; Enterocolitis, Necrotizing/diagnosis* ; Infant, Premature ; C-Reactive Protein/analysis* ; Oxygen Saturation ; Infant, Newborn, Diseases

OBJECTIVES: To investigate whether evidence-based standardized nutrition protocol can facilitate the establishment of full enteral nutrition and its effect on short-term clinical outcomes in very preterm/very low birth weight infants. METHODS: A retrospective analysis was performed on the medical data of 312 preterm infants with a gestational age of ≤32 weeks or a birth weight of <1 500 g. The standardized nutrition protocol for preterm infants was implemented in May 2020; 160 infants who were treated from May 1, 2019 to April 30, 2020 were enrolled as the control group, and 152 infants who were treated from June 1, 2020 to May 31, 2021 were enrolled as the test group. The two groups were compared in terms of the time to full enteral feeding, the time to the start of enteral feeding, duration of parenteral nutrition, the time to recovery to birth weight, the duration of central venous catheterization, and the incidence rates of common complications in preterm infants. RESULTS: Compared with the control group, the test group had significantly shorter time to full enteral feeding, time to the start of enteral feeding, duration of parenteral nutrition, and duration of central venous catheterization and a significantly lower incidence rate of catheter-related bloodstream infection (P<0.05). There were no significant differences between the two groups in the mortality rate and the incidence rate of common complications in preterm infants including grade II-III necrotizing enterocolitis (P>0.05). CONCLUSIONS Implementation of the standardized nutrition protocol can facilitate the establishment of full enteral feeding, shorten the duration of parenteral nutrition, and reduce catheter-related bloodstream infection in very preterm/very low birth weight infants, without increasing the risk of necrotizing enterocolitis.
Birth Weight ; Enteral Nutrition/methods* ; Enterocolitis, Necrotizing/prevention & control* ; Humans ; Infant ; Infant, Extremely Premature ; Infant, Newborn ; Infant, Very Low Birth Weight ; Retrospective Studies ; Sepsis/epidemiology*

OBJECTIVES: To investigate the risk factors for necrotizing enterocolitis (NEC) in very preterm infants and establish a nomogram model for predicting the risk of NEC. METHODS: A total of 752 very preterm infants who were hospitalized from January 2015 to December 2021 were enrolled as subjects, among whom 654 were born in 2015-2020 (development set) and 98 were born in 2021 (validation set). According to the presence or absence of NEC, the development set was divided into two groups: NEC (n=77) and non-NEC (n=577). A multivariate logistic regression analysis was used to investigate the independent risk factors for NEC in very preterm infants. R software was used to plot the nomogram model. The nomogram model was then validated by the data of the validation set. The receiver operating characteristic (ROC) curve, the Hosmer-Lemeshow goodness-of-fit test, and the calibration curve were used to evaluate the performance of the nomogram model, and the clinical decision curve was used to assess the clinical practicability of the model. RESULTS: The multivariate logistic regression analysis showed that neonatal asphyxia, sepsis, shock, hypoalbuminemia, severe anemia, and formula feeding were independent risk factors for NEC in very preterm infants (P<0.05). The ROC curve of the development set had an area under the curve (AUC) of 0.833 (95%CI: 0.715-0.952), and the ROC curve of the validation set had an AUC of 0.826 (95%CI: 0.797-0.862), suggesting that the nomogram model had a good discriminatory ability. The calibration curve analysis and the Hosmer-Lemeshow goodness-of-fit test showed good accuracy and consistency between the predicted value of the model and the actual value. CONCLUSIONS Neonatal asphyxia, sepsis, shock, hypoalbuminemia, severe anemia, and formula feeding are independent risk factors for NEC in very preterm infant. The nomogram model based on the multivariate logistic regression analysis provides a quantitative, simple, and intuitive tool for early assessment of the development of NEC in very preterm infants in clinical practice.
Asphyxia/complications* ; Child ; Enterocolitis, Necrotizing/etiology* ; Female ; Fetal Growth Retardation ; Humans ; Hypoalbuminemia ; Infant ; Infant, Newborn ; Infant, Newborn, Diseases ; Infant, Premature ; Infant, Premature, Diseases/etiology* ; Nomograms ; Sepsis/complications*

Objective: To explore the composition of intestinal microflora prior to onset of necrotizing enterocolitis (NEC) in very low birth weight preterm infants. Methods: This was a multicenter prospective nested case-control study. A total of 46 very low birth weight preterm infants (birth weight <1 500 g and gestional age <35 weeks) within 24 h of life admitted into Neonatal Intensive Care Unit of Children's Hospital of Soochow University and Suzhou Municipal Hospital from April 20 to November 20, 2018 were enrolled. Baseline clinical data and fecal samples of these infants were collected. The subsequent sampling time points were 1st, 4th and 7th day in the first week of life then once per week consecutively. The endpoint of sampling was NEC occurrence, patient discharge or the 8th week post-discharge, whichever came first. Fecal samples were analyzed by 16 S rDNA high-throughput nucleotide sequencing. The control cases were infants without NEC who were matched to the NEC cases with a ratio of 1∶1. The operational taxonomic units (OTU), sequence number and shannon diversity index of the fecal samples were analyzed. Continuous variables were compared with t-test or non-parametric test, and χ² test or Fisher's exact test was used for categorical variables. Results: There were 23 patients in each group. The gestational age was (29.4±1.8) weeks in NEC group and (29.9±1.6) weeks in control group, including 13 males (57%) and 11 males (48%) in each group, respectively. Species abundance showed that the Firmicutes in both groups decreased temporarily at 7 days of age and then increased with age in control group, but not in NEC group, the Proteobacteria in both groups increased at 7 days of age and then decreased in control group, but kept increasing in NEC group. Regarding the other levels of taxonomy, compared with that of the control group, the NEC group had lower abundance of Proteobacteria, γ-proteobacteria and Enterobacteriaceae at 7 days of age, while higer abundance of Faecalibacterium at 14 days of age, meanwhile, lower Clostridium and Streptococcus at 21 days of age, lower Firmicutes, Clostridia and Clostridium perfringens and higher Proteobacteria and γ-proteobacteria at 28 days of age, these differences were all statistically significant (U=43.00, 43.00, 45.00, 80.00, 74.00, 76.00, 19.00, 8.00, 36.00, 25.00, 25.00,all P<0.05). The shannon index of NEC group was both lower than that of the controls at 21 days of age (2.4 (1.4, 3.0) vs. 3.1 (2.6, 4.0), U=67.00, P=0.027) and 28 days of age (2.4 (1.4, 2.8) vs. 3.9 (3.3, 4.2), U=12.00, P=0.001). Conclusions: The intestinal microflora profile of very low birth weight preterm infants has already changed prior to NEC development. The emergence of differential flora and the reduction of microflora diversity may facilitate early identification and prevention of NEC.
Aftercare ; Case-Control Studies ; Child ; Enterocolitis, Necrotizing/epidemiology* ; Gastrointestinal Microbiome ; Humans ; Infant ; Infant, Newborn ; Infant, Premature ; Infant, Very Low Birth Weight ; Male ; Patient Discharge ; Prospective Studies

Dyskeratosis Congenita/genetics* ; Enterocolitis/genetics* ; Fetal Growth Retardation ; Humans ; Intellectual Disability/genetics* ; Microcephaly/genetics*