Influence of Endoscopic Mucosal Resection On Esophagectomy Rates in a Specialized Barrett's Esophagus Unit

Abstract

Influence of Endoscopic Mucosal Resection On Esophagectomy Rates in a Specialized Barrett’s Esophagus Unit Ganapathy A. Prasad, Kenneth K. Wang, Navtej S. Buttar, Louis M. Wongkeesong, Lori S. Lutzke, Lynn S. Borkenhagen, Sarah M. Papenfuss Rationale: Endoscopic mucosal resection (EMR) has become widely utilized in the treatment and diagnosis of neoplasia in Barrett’s Esophagus (BE). Limited information is available on the influence of EMR on management of neoplasia in BE. Aim: To compare esophagectomy rates following the introduction of EMR in the management of BE related neoplasia in a specialized unit. Methods: Records of 384 patients who underwent 663 EMRs for BE in a specialized unit from 1995-2005 were reviewed. All EMRs were performed by a single experienced endoscopist (KKW) using standard techniques. 42% of patients underwent photodynamic therapy (PDT). Data extracted from a prospectively maintained database included patient demographics, clinical outcomes (frequency of esophagectomy and EMR related complications). Outcomes were assessed in three successive time intervals (TI) based on the number of procedures performed, TI-1: 4/95-2/03 with 221 EMRs, TI-2: 2/03-10/04 with 221 EMRs, and TI-3: 0/04-11/05 with 221 EMRs. Results: Mean age of patients was 67.9 years (SEM 11.4). 82% were males. Age, gender, length of BE and indications for referral (HGD 56-70%;Ca 18-31%)were similar across the 3 time intervals (p Z NS). EMR was performed using the Olympus EMR kit in 87% of cases, band & snare technique in 10% and Wilson Cook Duette kit in 3%. Lesions visible on EGD remained similar over the 10 year period (nodules in 60-70%). Size of EMR specimens and number of EMRs performed/session increased over the TIs. The overall stricture rate was 6.2% but decreased over the TIs (see table). Hemodynamically significant bleeds occurred in 8 (2%)patients. There were no perforations. Fewer patients went on to esophagectomy in TI-2 and 3 compared to TI-1 (see table). The proportion of cancers found at esophagectomy was comparable across the 3 groups (50-75%) (p Z 0.33). Discussion: This is the largest experience of EMRs in BE reported and demonstrates that the use of this technique was associated with a significant decrease in the proportion of patients undergoing esophagectomy for HGD/Cancer in BE over the last 10 years. This is most likely due to the greater accuracy of disease diagnosis and staging obtained with this technique. Increased EMR specimen size and number probably reflects increasing endoscopist experience with low complication rates.