S0559 Dislodgement and Perforation in Endoscopic Percutaneous Enteral Tube Placement: Jejunal Extension Risk and Chemotherapy Administration Outcomes

Abstract

INTRODUCTION: Gastrointestinal perforation is a significant risk in almost all endoscopic procedures. Tube dislodgement is unique to percutaneous endoscopic gastrostomy (PEG) and percutaneous endoscopic gastrostomy with jejunal extension (PEG-J) placement. We assessed associated risks of PEG and PEG-J tube PD in all patients (pts) as well as those with malignancy who received chemotherapy. METHODS: 509 pts identified by diagnostic codes for initial attempt at PEG or PEG-J tube placement from 2012-2017 were retrospectively reviewed. 29 pts were excluded from analysis due to aborted procedures and lack of outcome data. Figure 1 PD related complications were defined as radiographical, endoscopic, or surgically confirmed perforation or frank tube migration requiring medical or procedural intervention within 6 months from procedure date. As only 3 perforations occurred with 2 having concurrent tube dislodgements, PD was considered as a composite outcome in analysis. Initial chemo, demographic characteristics, maneuver type, tube size, antiplatelet or anticoagulant use (AAU), malignancy status, baseline proton pump inhibitor (PPI) and histamine receptor 2 antagonist (H2RA) use, body mass index (BMI), diabetes status, and albumin levels were collected Table 1. Univariate and multivariate models were used to assess association of risk factors with PD outcome of interest. RESULTS: Of the 480 pts in the study, 22 (4.6%) developed tube dislodgement with 3 (0.6%) suffering perforation. Median time to PD was 22 days (IQR 7–52). 73 (15%) specifically underwent PEG-J placement. Over half the study population (56%) had malignancy and 167 (33%) received chemotherapy. 1 of 3 perforations occurred at time of procedure and 1 of 3 perforations involved PEG-J placement. Age ≥ 70, tube size (≤20 Fr vs > 20 Fr), baseline malignancy, PPI/H2RA use, AAU, albumin < 3, and BMI < 18 or >30 did not increase risk of PD on univariate analysis Table 2. Chemotherapy within 30 days prior to 15 days after the procedure did not increase odds of PD on univariate (0.642, CI 0.14–2.81, P < 0.55) or multivariate analysis (OR 0.789, CI 0.17–3.56, P < 0.75) while PEG-J procedures increased risk of PD in both (OR 3.94, CI 1.64–9.49 P < 0.0022 and OR 3.81, CI 1.55–9.33 P < 0.0034, respectively) Table 2. CONCLUSION: PD incidence in percutaneous endoscopic enteral tube placement is not influenced by chemotherapy administration even when given in the periprocedural period. PD remains higher in PEG-J procedures and is primarily related to jejunal extension malfunction.Figure 1.: Flowchart of study exclusion.Table 1.: Characteristics of patients undergoing initial percutaneous endoscopic enteral tube placement (2012–2017)Table 2.: Univariate and multivariate associations with perforation or dislodgement following endoscopic percutaneous enteral tube placement