Impact of Body Mass Index as Predictor for Surgical Interventions in Crohn’s Disease Strictures

Background: Half of patients with Crohn’s disease (CD) develop strictures within 20 years from disease diagnosis. Many of these patients require either endoscopic and/or surgical intervention. This study was designed to determine risk factors associated with the need for surgical intervention in CD-related strictures. Methods: We utilized i2b2 to obtain a registry of patients with CD strictures. Initially, 209 patients were found utilizing this search. Out of the 209 patients, 30 met the inclusion criteria with complete medical records. Patients’ demographics and study outcomes including hospitalizations, endoscopic and surgical interventions were collected. Patients’ cohort was classified based on their need for surgical intervention for CD strictures into surgical (Sx) and non-surgical (non-Sx) groups. Cox regression analysis was performed to examine the association between various factors (age, race, BMI, smoking status, severity of stricture, number of obstructive episodes and hospitalizations, presence of pre-stenotic dilatation or inflammation on imaging, serum inflammatory markers, medications and number of endoscopic dilations) and surgical intervention. Time to surgical intervention was generated by Kaplan-Meier method. Results: Of the 30 patients with Crohn’s related strictures, 40% (n=12) required surgical intervention for their strictures. There was no statistically significant difference between patients’ demographics in both groups. Although patients in the Sx group had higher percentages of severe strictures (diameter from 1-5 mm) on colonoscopy and inflamed strictures with pre-stenotic dilatation on imaging compared to the non-Sx groups, there was no statistically significant difference. Hospitalization due to small bowel obstruction was significantly higher in the Sx group 92% vs. 44% in the non-Sx group, P = 0.009. Being overweight or obese was associated with a higher likelihood of surgical intervention [HR 6.85, 95% CI (1.12-41.88), P = 0.037] and [HR 24.90, 95% CI (1.20-518.23), P = 0.038]. Patients with high BMI have a shorter duration to surgery compared to patients with normal BMI on Kaplan-Meier curve (2.26 months vs. 1.1 months, P = 0.019). Other factors such as smoking, the severity of stricture, prior endoscopic dilation, and IBD medications at the time of stricture did not show statistically significant associations with surgical intervention. Conclusions: Patients with CD strictures and high BMI have a higher risk of surgical intervention and shorter duration to surgery compared to patients with normal BMI. These findings need to be explored in a larger sample size.