The Overuse of Empiric Antibiotic Treatment for Enterocolitis Diagnosed by Abdominal CT (OVERACCT)

Abstract

Introduction: The current ACG guidelines for the management of acute diarrhea discourage the use of empiric antibiotic therapy, unless justified by the clinical presentation, including duration of symptoms, recent traveling, fever or bloody diarrhea. Computed tomography (CT) is a widespread and easily accessible imaging tool for the investigation of abdominal symptoms. However, the incidental finding of nonspecific bowel wall thickening, often results in a radiological diagnosis of enterocolitis, irrespective of the symptoms, with subsequent inappropriate use of antibiotic therapy. Methods: A retrospective, observational, IRB approved study was conducted at an inner-city hospital from July 2015 to July 2017. A total of 184 subjects who experienced gastrointestinal symptoms such as abdominal pain, nausea, vomiting, or diarrhea were identified by the enterocolitis ICD diagnosis code. Patients with comorbidities such as AIDS, cancer, or immunocompromised were excluded from the analysis. Of the remaining subjects, 121 underwent abdominal CT to rule out acute abdominal conditions. Chart review included medical history, recent travel, inflammatory markers and stool studies. Clinical outcomes comprised symptoms resolution, length of hospital stay and final GI diagnosis. Chi-square test was used for data analysis. Results: Among 184 patients identified, 116 (63%) were women with a mean age of 46 ± 26 years. The majority of patients (66%) presented with acute diarrhea, lasting less than 72 hours, without a history of recent travel. A radiological diagnosis of enterocolitis was reached with abdominal CT in 80 patients, of which 71% received antibiotic treatment. There was a significant correlation between positive CT findings and antibiotic use (p=0.005). Surprisingly, 61% of patient with the radiological findings received antibiotic treatment despite the absence of diarrhea (p=0.006). Antibiotic treatment did not impact the clinical outcomes (p=ns). Almost 65% of the patients required only supportive treatment based on ACG guidelines. Conclusion: Patients with CT findings suggestive of enterocolitis were more likely to receive antibiotic therapy, irrespective of the presence or absence of diarrhea. This study highlights the widespread antibiotic misuse for the treatment of enterocolitis based on imaging modalities without adequate clinical correlation. Unwarranted radiological testing and antibiotic therapy were identified in 65% of this cohort, while their clinical outcomes remained unchanged.167_A Figure 1 No Caption available.167_B Figure 2. χ2 Calculations. Two way contingency table.167_C Figure 3. χ2 Calculations. Two way contingency table.