S1090 Factors Associated With Severe Submucosal Fibrosis in the Treatment of Barrett’s Esophagus

Abstract

Introduction: Severe submucosal fibrosis (SSF) is established to increase complications with endoscopic submucosal dissection (ESD) throughout the GI tract. This is often attributed to prior treatment with endoscopic mucosal resection (EMR). However, there is little information regarding which forms of treatment increases the risk of SSFin Barrett’s esophagus therapy. We performed a retrospective cohort study to determine the risk factors for SSF. Methods: We identified consecutive patients from the Barrett's Esophagus unit who underwent ESD by a single experienced endoscopist (KKW) who graded the submucosal fibrosis. Submucosal fibrosis can only be reliably identified on ESD to adequately examine the collagen fibers. All resections were performed using a scissor type device. Submucosal fibrosis was evaluated as F0 (no fibrosis), F1 (moderate fibrosis with individual strands), and F2 (SSF) with solid sheets of fibrosis) by international standards. All patients had prior therapy identified and categorized as EMR, thermal treatments including RFA in combination with multipolar coagulation and APC, as well as cryotherapy either balloon or spray. Results: A total of 205 patients were identified with a mean age of 69±0.7 years with 156 males and a Barrett’s segment length of 4.4±0.3 cm. Diaphragmatic hernia length was 2±0.2 centimeters. 119 patients had either F0 (n=57, 28%) or F1 (n=55, 27%), while F2 fibrosis was found in 93 (45%) patients. The patients in each group were similar in terms of age, length of Barrett’s esophagus, and histology (Table). On univariate analysis, SSF was significantly associated with female gender (p< 0.03), thermal ablation (p< 0.04), and larger hiatal hernia (p< 0.005). Cryotherapy was not associated with SSF. EMR was performed in over half the patients with SSF but was also performed in 41% of those without (p< 0.08). On multivariate analysis, thermal ablation was significantly associated with SSF and hiatal hernia size remained significant but gender was not significant. Conclusion: SSF can dramatically decrease the ability for endoscopic resection. Thermal ablation and size of hiatal hernia are associated with SSF. Careful inspection and resection prior to beginning ablation should decrease the need to perform ESD after ablation. The association with hernia highlights the need for acid control. Table 1. - Assessment of Factors Producing Severe Submucosal Fibrosis Patients (n,%) None or Mild Fibrosis Severe Fibrosis < p 112 (54%) 93 (45%) Age (yrs) 69.5±0.9 68.4±1.0 0.46 Male gender (n,%) 92 (82%) 64 (69%) 0.03 Length of BE (cm) 4.5±0.4 4.2±0.4 0.57 Length of DH (cm) 1.6±0.2 2.5±0.2 0.005 LGD or ND (n,%) 50 (45%) 35 (38%) 0.31 HGD or ACA (n,%) 62 (55%) 58 (62%) Thermal Ablation (RFA) (n,%) 24 (21%) 32 (34%) 0.04 Cryotherapy 17 (15%) 21 (23%) 0.17 EMR 43 (38%) 47 (51%) 0.08