ROLE OF PELVIC ULTRASOUND IN THE TREATMENT MONITORING OF CHILDREN WITH CHRONIC Idiopathic Constipation

Constipation is a frequently encountered problem in childhood, with a prevalence ranging between 1 and 30%. From an etiological point of view, we distinguish the form functional/idiopathic (≈ 95%), defined according to the Rome IV Criteria, and shape secondary (≈ 5%). Fecal retention in chronic constipation often causes a dilatation of the rectum (megarectum), the extent of which affects the response to treatment. Several studies have proposed pelvic ultrasound, (simple, non-invasive and reproducible) both to define the presence of megarectum, and to follow the answer to treatment, but the real utility remains to be defined, especially in the follow-up. In case of persistence of megarectum and intractable constipation despite treatment conventional, anorectal manometry should be considered, aiming to identify conditions of colonic dysmotility. The latter could be excluded adopted unconventional biofeedback treatments, the effectiveness of which, however, is still little standardized.

Primary objective: To evaluate whether, in the conventional treatment of functional constipation of the child, the variations of the rectal diameter, measured through the use of the pelvic ultrasound, are a good marker of disease severity and efficacy of the treatment. Secondary objectives: 1) to define the prevalence of chronic idiopathic intractable constipation in our center; 2) Document whether the biofeedback treatment is effective in terms of rectal diameter reduction, ultrasound documented, and response clinical outcome, assessed by means of a validated questionnaire, in the subclass of patients with constipation chronic idiopathic resistant to conventional treatment. Inclusion criteria: pediatric patients (between 4 and 18 years of age) accessing services outpatient or pediatric first aid for constipation (according to Rome IV criteria) Exclusion criteria: age < 4 years; previous anorectal surgery; suspicion of anamnestic and/or clinical examination of an organic cause of constipation.

123 pediatric patients with chronic idiopathic constipation. The characteristics of the patients included in the study will be initially analysed (gender, age, anthropometric parameters), ultrasound data and the degree of severity of disease through a descriptive analysis. Mean and standard deviation or median e interquartile range will be used to summarize the quantitative type variables, upon verification of the assumption of normality with the Shapiro-Wilk test, while absolute frequency and percentage will be used for the qualitative variables. The data ultrasound and clinical scores will be analyzed at each time point and will be calculated the change from the previous time point. Statistically significant changes will be evaluated through the Student's t-test (quantitative variables normally distributed), the Mann-Whitney test (quantitative variables not normally distributed) or the Chi-square test (qualitative variables). Spearman's correlation index will come calculated to evaluate the concordance of the aforementioned variations. An effect model mixed for repeated measures will be used to evaluate the clinical improvement of patients over time (t1, t2, t3). Clinical improvement is indicated as a reduction in the diameter of the rectum identified by ultrasound analysis and a decrease of the severity score of the disease identified through the validated questionnaire. In the patients undergoing the biofeedback cycle will be evaluated for clinical improvement through the comparison between the ultrasound data and the scores collected before the start of the cycle with data collected at the end of the cycle using Student's t-test (quantitative variables normally distributed), the Mann-Whitney test (quantitative variables not normally distributed) or the Chi-square test (qualitative variables). Within the scope of the entire study the statistical significance is identified at a level of 5%. The prevalence estimates will be accompanied by the relative 95% confidence intervals.

Statistical analysis will be performed with R software version 4.1.1.