Abstract
Introduction Constipation represents 3% of all office visits to pediatricians and 10% - 45% of consultations with pediatric gastroenterologists. It has been reliably established that the role of abdominal x-rays (AXR) in the diagnosis of constipation in pediatrics is limited; yet, significant overdiagnosis of constipation exists when plain abdominal x-rays are used in the acute setting for abdominal pain or to screen for other disorders. This results in loss of time, resources, exposure to unnecessary radiation, and potentially missing the primary diagnosis. The purpose of this study is to determine the sensitivity and specificity of AXR in diagnosing constipation in the acute setting. Objectives To determine 1) the sensitivity and specificity of plain AXR in the diagnosis of constipation and 2) the effect of age, race, gender, comorbid conditions, and practice setting on the diagnosis of constipation. Methods This study was a historical cohort study of children (two to 18 years of age) who were seen at Ascension St. John Children's Hospital between March 2015 - March 2018 and who had a plain AXR performed during an emergency department (ED) visit or inpatient stay. If AXR results contained keywords, such as "constipation," "stool load," "fecal retention," and "fecal load," the ambulatory medical record, Athena® (http://www.athenahealth.com), was searched to determine if the child had an ambulatory visit in the ensuing 45 days. Chart review was conducted to assess if the diagnosis of constipation was later confirmed by history and physical examination by a pediatrician or gastroenterologist at that visit. By comparing data from both encounters, the sensitivity and specificity of plain AXR in diagnosing constipation was assessed. All data were analyzed using the Statistical Package for Social Sciences (SPSS), v. 25.0 (IBM SPSS Statistics, Armonk, NY) and a p-value of 0.05 or less was considered to indicate statistical significance. Results Over the three-year study period, 1,383 AXRs were performed on 1,116 patients. The sensitivity of AXR in the diagnosis of constipation was 73.8%, specificity 26.8%, positive predictive value 46.4%, and negative predictive value of 54.3%. Pediatric gastroenterologists were more likely to diagnose constipation (63.2%) compared to pediatricians (41.4%) and pediatric surgeons (33.3%) (p = 0.04). Conclusions AXRs are not a reliable means of diagnosing constipation. Overall, we found similar sensitivity and specificity of AXR in diagnosing constipation compared to previous studies. Yet, our study gives new insight into the practices around diagnosing constipation in a single-center community hospital pediatric acute setting and the radiology department. This further emphasizes the need to review current practices and impart more education both in the acute care setting and radiology department.