The results of this study showed that 68% of patients who underwent EBD for treatment of CD related strictures did not require surgery within a five-year follow-up period. These findings align with existing literature, which demonstrated a 42.9% surgical rate following EBD, underscoring the utility of EBD in managing CD-related strictures with a low rate of surgical intervention post-EBD (11).
Treatment history at the time of diagnosis varied across patients, with some receiving anti-TNF, anti-integrin, IL-23 inhibitor, immunomodulators, or a combination thereof. Steroiduse at the time of dilation was also noted to provide insight into the pre-EBD treatment landscape as discussed below.
Consistent with previous studies, steroid therapy emerged as a predictor of poorer outcomes (11). The analysis revealed a significant association between the use of prednisone at the time of dilation and the subsequent need for surgical intervention within five years. Specifically, a higher proportion of patients who required surgery post-dilation (24.4%) were on prednisone compared to those who did not require surgery (11.6%). However, prednisone dosing (< 20 mg vs ≥ 20 mg) at the time of dilation was not significantly associated with surgery. Also, budesonide use at time of dilation was not associated with surgical intervention in the five-year timeframe. These findings suggest that prednisone use may be a predictor of poorer outcomes post-EBD, potentially due to its indication in more active, refractory disease, or perhaps due to its systemic effects that might influence stricture behavior or healing.
The type of stricture significantly impacted patient outcomes post-EBD. De novo strictures had a higher likelihood of leading to surgery compared to anastomotic strictures, a finding that has also been demonstrated in other studies (12). This finding emphasizes the importance of stricture etiology in guiding management and prognostication for CD patients undergoing EBD. Anastomotic strictures, perhaps due to their specific inflammatory and fibrotic pathways or their anatomical characteristics, may respond more favorably to EBD compared to de novo strictures (13). In addition, anastomotic strictures are more commonly a result of post-surgical complication compared to de novo strictures which result from active Crohn’s disease. This distinction between anastomotic and de novo strictures further highlights the importance of stricture etiology in guiding management decisions.
Our analysis included a significant number of total dilations, with strictures dilated to a maximal diameter based on the endoscopists’ judgment. Stricture characteristics such as location and severity were carefully evaluated, with the largest proportion being anastomotic and predominantly ileocecal. The need for repeat dilation was observed in 42.9% of the cases. This underscores the potential recurrent nature of CD-related strictures and highlights the necessity for ongoing surveillance and possibly iterative endoscopic interventions in managing this chronic condition. In addition, stepwise graded dilations are a safer approach for strictures with starting small diameters. Typically, we increase the diameter by 3 mm per session to prevent complications. The end goal is 18 mm for most of our endoscopists.
These findings offer valuable insights into the predictors of EBD success and the need for subsequent surgical intervention. Understanding these predictors can aid clinicians in selecting appropriate candidates for EBD and tailoring follow-up care to optimize patient outcomes. Further research may be needed to explore the mechanisms underlying these associations and to develop strategies to mitigate the risk of poor outcomes.
Data was mainly collected retrospectively via extraction from patient charts in the EMR which could contribute to potential information bias. This bias was mitigated through the uniform collection of data from consecutive patient files, so this potential bias was non-differential. In a small proportion of cases, there was no documentation of stricture length. Furthermore, this was a single center study, therefore limiting external validity. These potential confounders were controlled by multivariable adjustment. In addition, there were multiple endoscopists with no standard protocol for management of the strictures.
This is one of the largest single-center studies looking at EBD outcomes for treatment of CD-related strictures in individuals with IBD. This study also includes a long term follow up period. Future prospective studies should aim to validate these predictors and explore the impact of novel therapeutic strategies on EBD outcomes.
In summary, EBD represents an effective intervention for CD-related strictures, with specific patient and stricture characteristics predictive of outcomes. Identifying these predictors can guide clinicians in optimizing management strategies for individuals with CD, potentially reducing the need for surgical intervention.