PMID: 36873644
PMCID: PMC9975712
DOI: 10.3389/fped.2023.1043067

Abstract

Background: Complications of Crohn’s disease (CD) often impair patients‘ quality of life. It is necessary to predict and prevent these complications (surgery, stricturing [B2]/penetrating [B3] disease behavior, perianal disease, growth retardation and hospitalization). Our study investigated previously suggested and additional predictors by analyzing data of the CEDATA-GPGE registry. 

Methods: Pediatric patients (< 18 years) diagnosed with CD with follow up data in the registry were included in the study. Potential risk factors for the selected complications were evaluated by performing Kaplan-Meier survival curves and cox regression models. 

Results: For the complication surgery, the potential risk factors older age, B3 disease, severe perianal disease and initial therapy with corticosteroids at the time of diagnosis were identified. Older age, initial therapy with corticosteroids, low weight-for-age, anemia and emesis predict B2 disease. Low weight-for-age and severe perianal disease were risk factors for B3 disease. Low weight-for-age, growth retardation, older age, nutritional therapy, and extraintestinal manifestations (EIM) of the skin were identified as risk factors for growth retardation during the disease course. High disease activity and treatment with biologicals were predictors for hospitalization. As risk factors for perianal disease, the factors male sex, corticosteroids, B3 disease, a positive family history and EIM of liver and skin were identified. 

Conclusion: We confirmed previously suggested predictors of CD course and identified new ones in one of the largest registries of pediatric CD patients. This may help to better stratify patients‘ according to their individual risk profile and choose appropriate treatment strategies. 

Keywords: complication; disease behavior; growth; hospitalization; outcome; perianal disease; prediction; surgery. 

Der Beitrag Predicting complications in pediatric Crohn’s disease patients followed in CEDATA-GPGE registry  erschien zuerst auf CEDATA GPGE – Patientenregister.

DATA FROM THE CEDATA GPGE REGISTRY

To date, failure of the first-line medical management of Crohn’s disease (CD) in children and adolescents has been followed by escalation treatment, with administration of immunomodulators, e.g., azathioprine, or biologics, such as infliximab (1). The most recent revision of the international guideline proposes stratification of the therapy, with selection of either accelerated step-up or primary treatment with biologics on the basis of predictors of poor outcome (POPO) (2). The objective of this study was to evaluate this concept of early immunosuppression with the aid of data from the largest registry of children and adolescents with chronic inflammatory bowel disease (IBD) in Europe.

Der Beitrag Early Immunosuppression in Children and Adolescents with Crohn’s Disease erschien zuerst auf CEDATA GPGE – Patientenregister.

Front Med (Lausanne). 2021; 8: 666190. 

Published online 2021 May 24. doi: 10.3389/fmed.2021.666190

PMCID: PMC8180568

PMID: 34109197

Abstract

Introduction: The rising incidence of pediatric inflammatory bowel diseases (PIBD) facilitates the need for new methods of improving diagnosis latency, quality of care and documentation. Machine learning models have shown to be applicable to classifying PIBD when using histological data or extensive serology. This study aims to evaluate the performance of algorithms based on promptly available data more suited to clinical applications. 

Methods: Data of inflammatory locations of the bowels from initial and follow-up visitations is extracted from the CEDATA-GPGE registry and two follow-up sets are split off containing only input from 2017 and 2018. Pre-processing excludes patients in remission and encodes the categorical data numerically. For classification of PIBD diagnosis, a support vector machine (SVM), a random forest algorithm (RF), extreme gradient boosting (XGBoost), a dense neural network (DNN) and a convolutional neural network (CNN) are employed. As best performer, a convolutional neural network is further improved using grid optimization. 

Results: The achieved accuracy of the optimized neural network reaches up to 90.57% on data inserted into the registry in 2018. Less performant methods reach 88.78% for the DNN down to 83.94% for the XGBoost. The accuracy of prediction for the 2018 follow-up dataset is higher than those for older datasets. Neural networks yield a higher standard deviation with 3.45 for the CNN compared to 0.83-0.86 of the support vector machine and ensemble methods. 

Discussion: The displayed accuracy of the convolutional neural network proofs the viability of machine learning classification in PIBD diagnostics using only timely available data.

Keywords: CEDATA-GPGE registry; convolutional neural network; diagnostic assistance; machine learning; pediatric inflammatory bowel disease.

Der Beitrag Machine Learning Classification of Inflammatory Bowel Disease in Children Based on a Large Real-World Pediatric Cohort CEDATA-GPGE Registry erschien zuerst auf CEDATA GPGE – Patientenregister.

• PMID: 28604511
• DOI: 10.1097/MPG.0000000000001649

Abstract

Objectives: Perianal disease (PD) with fistula and/or abscess formation is a severe complication in Crohn disease (CD). We examined prevalence, incidence, and risk factors for PD development in a pediatric CD cohort.

Methods: Patients with CD from the prospective, multicenter registry for inflammatory bowel disease from Germany and Austria (CEDATA-GPGE) were included if diagnosed at the age of 18 years or younger, registered within 3 months after diagnosis, and having at least 2 follow-up visits within the first year of registration. We examined potential risk factors for PD with Kaplan-Meier analysis and a final Cox model considering sex, family history of inflammatory bowel disease, extraintestinal manifestations, disease location, and induction therapy (corticosteroids or nutritional therapy).

Results: Of 2406 patients with CD, 742 fulfilled inclusion criteria (59% boys, mean age at diagnosis 12.4 ± 3.4 years). PD was present at diagnosis in 41 patients (5.5%; 80.9% boys), whereas 32 patients (4.3%, 81.3% male) developed PD during follow-up (mean 2.0 ± 1.6 years). The cumulative incidence of PD at 12 and 36 months after diagnosis was 3.5% and 7.5%, respectively. Potential risk factors for PD development during follow-up were male sex (hazard ratio = 3.2, [95%; confidence interval 1.2-7.8]) and induction therapy with corticosteroids (hazard ratio = 2.5 [1.1-5.5]). Diagnostic evaluation at PD diagnosis was incomplete in 40% of affected subjects. PD resolved within 1 year in 50% of cases.

Conclusions: Approximately 10% of CD patients in our cohort suffered from PD within the first 3 years of their disease. Male sex and initial corticosteroid therapy were associated with an increased risk to develop PD after diagnosis.

Der Beitrag Incidence and Risk Factors for Perianal Disease in Pediatric Crohn Disease Patients Followed in CEDATA-GPGE Registry erschien zuerst auf CEDATA GPGE – Patientenregister.

Abstract

Aim: To determine the prevalence of anemia and its association with disease severity in children and adolescents with IBD.

Methods: CEDATA-GPGE is a registry for pediatric patients with IBD in Germany and Austria from 90 specialized centers. As markers of disease severity, analysis included patient self-assessment on a Likert scale (1-5; 1 = very good) and physicians‘ general assessment (0 = no activity to 4 = severe disease) and the disease indices. Anemia was defined as hemoglobin concentration below the 3rd percentile.

Results: Prevalence of anemia was 65.2% in CD and 60.2% in UC. Anemic CD and UC patients showed significantly worse self-assessment than patients without anemia (average ± standard deviation; CD: 3.0 ± 0.9 versus 2.5 ± 0.9, p < 0.0001; UC: 2.9 ± 0.9 versus 2.3 ± 0.9, p < 0.0001). Accordingly, physicians‘ general assessment (PGA) was significantly worse in anemic than in nonanemic patients in CD (p < 0.0001) and UC (p < 0.0001). PCDAI in anemic CD, p < 0.0001, and PUCAI in anemic UC patients, p < 0.0001, were significantly higher than in nonanemic patients. 40.0% of anemic CD and 47.8% of anemic UC patients received iron during follow-up.

Conclusion: Almost 2/3 of pediatric IBD patients are anemic. Patients‘ self-assessment and disease severity as determined by PGA and activity indices are worse in anemic patients. Contrastingly, only a minority received iron therapy.

Der Beitrag Prevalence of Anemia in Pediatric IBD Patients and Impact on Disease Severity: Results of the Pediatric IBD-Registry CEDATA-GPGE erschien zuerst auf CEDATA GPGE – Patientenregister.