S1320 Does a Response to Botulinum Toxin Injection Predict a Future Response to Per Oral Pyloromyotomy (POP) in Patients With Gastroparesis?

Abstract

INTRODUCTION: Gastroparesis is a chronic gastrointestinal condition characterized by varying degrees of symptoms. It is often difficult to treat with limited viable options. Recently, a new technique called peroral endoscopic pyloromyotomy (POP) was developed which involves cutting the pyloric muscle by creating a submucosal tunnel. It has been thought that responsiveness to pyloric botulinum toxin (BT) injection may predict a response to POP. Our hypothesis is that a clinical response to BT injection of the pyloric sphincter is not a predictor of a clinical response to POP. METHODS: A retrospective study was performed at Cleveland Clinic from January 2016 to September 2019. Data from all patients who underwent POP was collected. We included patients who underwent BT injection prior to undergoing POP. We excluded all patients who did not undergo BT injection. Data such as age, sex, use of opioids and prokinetics at the time the POP is performed, etiology and prior interventions to treat gastroparesis including presence of gastric pacemaker and BT response was collected. All data analyses were conducted using SAS version 9.4. RESULTS: A total of 471 patients underwent POP at our institutions over the study period out of which 125 subjects were included in the analysis. Patients had a median age of 42.0 years, and 83.2% were females. The predominant etiology was idiopathic (59.2%) followed by diabetes (20.8%). Overall, 78 patients had improvement with BT, 82 had improvement with POP and 56 with both. On regression analysis, a response to BT did not predict improvement with POP (P = 0.18) and there was no difference seen with age, gender, etiology, prokinetic use, opioid use or presence of a pacemaker. CONCLUSION: POP may be a viable procedure for medically refractory gastroparesis in eligible patients. However, in this cohort of patients, a positive response to pyloric BT injection did not correspond to a positive response to POP. A positive response to POP was not predicted by any factors collected.Figure 1.: Patient Inclusion.Table 1.: Results of the logistic regression for main analysis demonstrating the odds ratio (OR) and 95% Confidence Intervals (CI) for covariates. The OR represents likelihood of improvement following POP