Machine learning-based analysis for prediction of surgical necrotizing enterocolitis in very low birth weight infants using perinatal factors: a nationwide cohort study.

Early prediction of surgical necrotizing enterocolitis (sNEC) in preterm infants is important. However, owing to the complexity of the disease, identifying infants with NEC at a high risk for surgical intervention is difficult. We developed a machine learning (ML) algorithm to predict sNEC using perinatal factors obtained from the national cohort registry of very low birth weight (VLBW) infants. Data were collected from the medical records of 16,385 VLBW infants registered in the Korean Neonatal Network (KNN). Infants who underwent surgical intervention were identified with sNEC, and infants who received medical treatment, with medical NEC (mNEC). We used 38 variables, including maternal, prenatal, and postnatal factors that were obtained within 1 week of birth, for training. A total of 1085 patients had NEC (654 with sNEC and 431 with mNEC). VLBW infants showed a higher incidence of sNEC at a lower gestational age (GA) (p < 0.001). Our proposed ensemble model showed an area under the receiver operating characteristic curve of 0.721 for sNEC prediction.    Conclusion: Proposed ensemble model may help predict which infants with NEC are likely to develop sNEC. Through early prediction and prompt intervention, prognosis of sNEC may be improved. What is Known: • Machine learning (ML)-based techniques have been employed in NEC research for prediction, diagnosis, and prognosis, with promising outcomes. • While most studies have utilized abdominal radiographs and clinical manifestations of NEC as data sources, and have demonstrated their usefulness, they may prove weak in terms of early prediction. What is New: • We analyzed the perinatal factors of VLBW infants acquired within 7 days of birth and used ML-based analysis to identify which infants with NEC are vulnerable to clinical deterioration and at high risk for surgical intervention using nationwide cohort data.

Features of colonic diverticulitis in children and adolescents: A multicenter study.

BACKGROUND: Colonic diverticulitis (CD), typically seen in the elderly of Western countries, is increasingly prevalent worldwide, yet data on CD in children and adolescents are scarce. This study explores the characteristics of CD in this younger demographic. METHODS: In a multicenter, retrospective review, 104 patients under 20 years diagnosed with CD at four Korean tertiary hospitals from June 2003 to December 2020 were analyzed. Abdominal CT scans were used for diagnosis, with the modified Hinchey classification assessing the severity of CD. RESULTS: CD was found in the cecum or ascending colon in 103 (99%) of cases. The mean patient age was 17.24 ±â€¯2.4 years, with males constituting 59.6% of cases. Solitary lesions were noted in 93 (89.4%) of patients. Severity was classified as modified Hinchey stage 0 in 58.7%, stage Ia in 29.8%, and stage Ib in 11.5%, with no cases of stage II or higher. Misdiagnosis as acute appendicitis occurred in six instances. IV antibiotics were administered to 68.3%, and oral antibiotics were sufficient for 24%. Surgical treatment was necessary for two patients. A 7.8% recurrence rate was noted among first-time CD patients, yet all cases were amenable to conservative management. CONCLUSION: While uncommon, CD in children and adolescents is a growing concern, with most cases presenting as solitary lesions in the cecum or ascending colon. The severity is typically less than that in adults, and conservative treatment is generally effective. These findings underscore the need for specific management guidelines for pediatric CD, advocating for non-surgical initial approaches.

Learning-Based Longitudinal Prediction Models for Mortality Risk in Very-Low-Birth-Weight Infants: A Nationwide Cohort Study.

INTRODUCTION: Prediction models assessing the mortality of very-low-birth-weight (VLBW) infants were confined to models using only pre- and perinatal variables. We aimed to construct a prediction model comprising multifactorial clinical events with data obtainable at various time points. METHODS: We included 15,790 (including 2,045 in-hospital deaths) VLBW infants born between 2013 and 2020 who were enrolled in the Korean Neonatal Network, a nationwide registry. In total, 53 prenatal and postnatal variables were sequentially added into the three discrete models stratified by hospital days: (1) within 24 h (TL-1d), (2) from day 2 to day 7 after birth (TL-7d), (3) from day 8 after birth to discharge from the neonatal intensive care unit (TL-dc). Each model predicted the mortality of VLBW infants within the affected period. Multilayer perception (MLP)-based network analysis was used for modeling, and ensemble analysis with traditional machine learning (ML) analysis was additionally applied. The performance of models was compared using the area under the receiver operating characteristic curve (AUROC) values. The Shapley method was applied to reveal the contribution of each variable. RESULTS: Overall, the in-hospital mortality was 13.0% (1.2% in TL-1d, 4.1% in TL-7d, and 7.7% in TL-dc). Our MLP-based mortality prediction model combined with ML ensemble analysis had AUROC values of 0.932 (TL-1d), 0.973 (TL-7d), and 0.950 (TL-dc), respectively, outperforming traditional ML analysis in each timeline. Birth weight and gestational age were constant and significant risk factors, whereas the impact of the other variables varied. CONCLUSION: The findings of the study suggest that our MLP-based models could be applied in predicting in-hospital mortality for high-risk VLBW infants. We highlight that mortality prediction should be customized according to the timing of occurrence.

Two-stage learning-based prediction of bronchopulmonary dysplasia in very low birth weight infants: a nationwide cohort study.

Introduction: The aim of this study is to develop an enhanced machine learning-based prediction models for bronchopulmonary dysplasia (BPD) and its severity through a two-stage approach integrated with the duration of respiratory support (RSd) using prenatal and early postnatal variables from a nationwide very low birth weight (VLBW) infant cohort. Methods: We included 16,384 VLBW infants admitted to the neonatal intensive care unit (NICU) of the Korean Neonatal Network (KNN), a nationwide VLBW infant registry (2013-2020). Overall, 45 prenatal and early perinatal clinical variables were selected. A multilayer perceptron (MLP)-based network analysis, which was recently introduced to predict diseases in preterm infants, was used for modeling and a stepwise approach. Additionally, we applied a complementary MLP network and established new BPD prediction models (PMbpd). The performances of the models were compared using the area under the receiver operating characteristic curve (AUROC) values. The Shapley method was used to determine the contribution of each variable. Results: We included 11,177 VLBW infants (3,724 without BPD (BPD 0), 3,383 with mild BPD (BPD 1), 1,375 with moderate BPD (BPD 2), and 2,695 with severe BPD (BPD 3) cases). Compared to conventional machine learning (ML) models, our PMbpd and two-stage PMbpd with RSd (TS-PMbpd) model outperformed both binary (0 vs. 1,2,3; 0,1 vs. 2,3; 0,1,2 vs. 3) and each severity (0 vs. 1 vs. 2 vs. 3) prediction (AUROC = 0.895 and 0.897, 0.824 and 0.825, 0.828 and 0.823, 0.783, and 0.786, respectively). GA, birth weight, and patent ductus arteriosus (PDA) treatment were significant variables for the occurrence of BPD. Birth weight, low blood pressure, and intraventricular hemorrhage were significant for BPD ≥2, birth weight, low blood pressure, and PDA ligation for BPD ≥3. GA, birth weight, and pulmonary hypertension were the principal variables that predicted BPD severity in VLBW infants. Conclusions: We developed a new two-stage ML model reflecting crucial BPD indicators (RSd) and found significant clinical variables for the early prediction of BPD and its severity with high predictive accuracy. Our model can be used as an adjunctive predictive model in the practical NICU field.

Development of artificial neural networks for early prediction of intestinal perforation in preterm infants.

Intestinal perforation (IP) in preterm infants is a life-threatening condition that may result in serious complications and increased mortality. Early Prediction of IP in infants is important, but challenging due to its multifactorial and complex nature of the disease. Thus, there are no reliable tools to predict IP in infants. In this study, we developed new machine learning (ML) models for predicting IP in very low birth weight (VLBW) infants and compared their performance to that of classic ML methods. We developed artificial neural networks (ANNs) using VLBW infant data from a nationwide cohort and prospective web-based registry. The new ANN models, which outperformed all other classic ML methods, showed an area under the receiver operating characteristic curve (AUROC) of 0.8832 for predicting IP associated with necrotizing enterocolitis (NEC-IP) and 0.8797 for spontaneous IP (SIP). We tested these algorithms using patient data from our institution, which were not included in the training dataset, and obtained an AUROC of 1.0000 for NEC-IP and 0.9364 for SIP. NEC-IP and SIP in VLBW infants can be predicted at an excellent performance level with these newly developed ML models. https://github.com/kdhRick2222/Early-Prediction-of-Intestinal-Perforation-in-Preterm-Infants .

Prediction of Recurrence of Completely Resected Pancreatic Solid Pseudopapillary Neoplasms in Pediatric Patients: A Single Center Analysis.

BACKGROUND: Many previous studies have investigated the risk factors for the recurrence of pancreatic solid pseudopapillary neoplasms (SPNs), although a consensus has not yet been reached, despite this effort. We aimed to identify the predictive factors for recurrence in patients with SPNs who underwent complete surgical resection of the tumor. METHODS: We retrospectively analyzed the records of pediatric patients with SPNs who underwent surgical resection at a single center between 2001 and 2018. RESULTS: During the study period, 47 patients with SPNs underwent radical resection of the tumor. The median age of the patients was 14 (8-18) years. R0 resection was confirmed in every case and none of the patients presented with systemic metastasis at the time of diagnosis. The median follow-up period was 53.1 (30.8-150.8) months. Of the 47 patients, only two (4.2%) experienced recurrence. Using comparative analysis, we found that some factors such as a large tumor size, peripancreatic tissue invasion, and capsule invasion did not increase the risk of recurrence of SPNs. Lymph node metastasis was the only significant factor for recurrence in our study (p = 0.043). CONCLUSION: During our single center analysis, we found that only lymph node metastasis was a predictive factor for recurrence of SPNs among patients who underwent complete tumor resection. Long-term follow-up is required to determine whether SPNs will recur if lymph node metastasis is observed after surgery. Furthermore, therapeutic benefits of routine lymphadenectomy or sentinel lymph node biopsy should be investigated in future studies to reduce the risk of recurrence in patients with SPNs.

Abnormal Orientation of the Superior Mesenteric Vessels Detected in Asymptomatic Infants: What Is Its Destiny?

Background: Ultrasonography (USG) has been described as an alternative diagnostic tool for malrotation that evaluates the orientation of the superior mesenteric vessels. However, literature concerning the management of patients who do not have abdominal symptoms is limited. We aimed to review the clinical course of infants showing abnormal orientation of the superior mesenteric vessels on USG who were asymptomatic at the time of diagnosis. Methods: Seventy asymptomatic infants with abnormal orientation of the superior mesenteric vessels in a single center between 2014 and 2018 were retrospectively analyzed. Results: The 70 patients, 21 underwent upper gastrointestinal series (UGIS) and 11 underwent abdominal surgery for other surgical conditions. Among the 32 (45.7%) patients who underwent UGIS or abdominal surgery, 11 were proven to have malrotation. Of the 38 (54.3%) patients who did not undergo UGIS or abdominal surgery, six patients were too unstable to undergo UGIS, five died due to cardiac complications, and the remaining patient developed midgut volvulus and died 3 days after emergency surgery. The remaining 32 patients who did not undergo UGIS or abdominal surgery were discharged without additional tests, and all were asymptomatic until their last follow-up. In the multivariate analysis, history of heart surgery and the presence of more than three anomalies were significantly associated with malrotation. Conclusion: A significant number of malrotation were diagnosed in asymptomatic infants with abnormal orientation of the superior mesenteric vessels on USG. Infants with major cardiac or multiple anomalies need special attention and should undergo UGIS in a promptly manner to confirm malrotation.

Thoracoscopic repair of esophageal atresia with distal tracheoesophageal fistula: is it a safe procedure in infants weighing less than 2000 g?

BACKGROUND: Since Rothenberg first performed thoracoscopic repair for esophageal atresia with distal tracheoesophageal fistula (EA/TEF) successfully in 2000, thoracoscopic repair has achieved status as a routine procedure worldwide. Previously, an international multicenter study reported that this procedure was not inferior to conventional open surgery. However, thoracoscopic surgery is a highly difficult operation for surgeons and anesthesiologists; as a result, the safety and efficacy of the surgery is still under debate. Considering these circumstances, the purpose of this study was to analyze the results of single-center thoracoscopic surgery and to compare the outcomes relative to the patient's weight at the time of surgery. METHODS: We retrospectively analyzed patients with EA/TEF who underwent thoracoscopic surgery in a single center between October 2008 and February 2017. RESULTS: In total, 41 cases of thoracoscopic repair of EA/TEF were performed. Upon subgrouping by over and under 2000 g of body weight at the time of operation, 34 were found to be over 2000 g and seven were under 2000 g. Intraoperative factors and events were not significantly different between the two groups. Additionally, most of the postoperative outcomes, including the rate of postoperative leakage and strictures, showed no difference. On the other hand, the under 2000 g group had more gastroesophageal reflux requiring fundoplication than did the heavier group (P = 0.04). CONCLUSIONS: The results of this center's thoracoscopic repair of EA/TEF were not inferior to other centers' outcomes. Additionally, the intraoperative and postoperative outcomes were similar despite differences in weight at operation. Therefore, thoracoscopic repair might be a feasible surgical option for infants weighing less than 2000 g when performed by a surgeon and anesthesiologist team who are experienced in pediatric thoracoscopic surgery.

The learning curve for thoracoscopic repair of esophageal atresia with distal tracheoesophageal fistula: A cumulative sum analysis.

BACKGROUND: Thoracoscopic repair of esophageal atresia with tracheoesophageal fistula (EA/TEF) remains technically challenging due to the rarity of these procedures. The aim of this study is to report our experience with thoracoscopic repair of type C EA/TEF and to evaluate the learning curve based upon the surgeon's skill level. METHODS: We retrospectively reviewed data of thoracoscopic EA/TEF repair performed in our center between October 2008 and May 2019. The learning curve was evaluated using the cumulative sum (CUSUM) method based on operative time. RESULTS: Of the 50 consecutive cases evaluated, the mean birth weight was 2634 ±â€¯608 g and the median age at operation was 3 days (range, 1-29 days). The mean operation time was 144 ±â€¯65 min. Anastomosis leakage occurred in 3 cases (6%) and strictures requiring balloon dilatations occurred in 16 cases (32%). The CUSUM analysis evaluated a learning curve of approximately 10 cases of thoracoscopic type C EA/TEF repair. A lower gestational age was associated with longer operation time. CONCLUSIONS: Thoracoscopic repair of type C EA/TEF is a feasible and safe procedure. The number of procedures required to achieve a stable learning curve was 10. The learning phase may be shortened by adequate set-up under the supervision of an expert endoscopic surgeon. TYPE OF STUDY: Retrospective Comparative Treatment Study. LEVEL OF EVIDENCE: III.

Minimal dissection of posterior wall of rectum reduces rectal prolapse in laparoscopic assisted anorectal pull-through.

PURPOSE: To determine if minimal dissection of the posterior wall of rectum can reduce rectal prolapse after laparoscopic assisted anorectal pull-through (LAARP) in male anorectal malformation (ARM) with rectourethral fistula. METHODS: Eighty-six male patients with ARM who underwent LAARP in our center between 2007 and 2018 were retrospectively analyzed. There were 45 cases of prostatic urethral fistula, 24 bulbar urethral fistulas, and 15 bladder neck fistulas. Two patients had no fistula. To prevent rectal prolapses, we markedly shortened the length of posterior rectal dissection from mid-2016. Dissection of posterior wall of rectum was performed minimally around the level of the fistula and the dissected portion of the posterior rectum was significantly shorter than the previous cases. For comparative analysis, patients were divided into two groups (before and after application of minimal dissection of posterior wall of rectum): Group A, from 2007 to mid-2016 and Group B, from mid-2016 to 2018. RESULTS: There were 60 patients in Group A and 26 patients in Group B. Demographic characteristics were not significantly different between the two groups. The median follow-up duration was 52.4 months for Group A and 26.9 months for Group B. Group B had lower incidence of rectal prolapse (11.5%) than Group A (68.3%) (p < 0.001). Upon our subgroup analysis based on types of fistula, patients with recto-prostatic urethral fistula and recto-bulbar urethral fistula showed significant reduction in the incidence of rectal prolapse (both p < 0.001). However, patients with recto-bladder neck fistula showed no statistical significance (p = 0.264). CONCLUSION: Minimal dissection of the posterior wall of rectum can reduce rectal prolapse in LAARP. LEVEL OF EVIDENCE: III. Retrospective Comparative Treatment Study.