The Frequency, Severity, and Spectrum of Adverse Events Associated with Direct Percutaneous Endoscopic Jejunostomy (DPEJ)

Abstract

Background: DPEJ is increasingly utilized as a method for obtaining enteral access. Though it has been described for more than 15 years, scant data exists regarding the safety profile of DPEJ. Aims: To describe the frequency and type of complications associated with DPEJ. Methods: Institutional databases identified 316 attempted DPEJ placements at Mayo Clinic Rochester from 1996 to August 2004. All but 9 patients had consented to have their records used for research. These records were abstracted with attention to adverse events (AEs). Severe AEs were defined as those requiring urgent surgery or endoscopy or seriously threatening a patient's health. Moderate AEs were defined as those requiring non-urgent surgery or endoscopy, or any hospitalization (not meeting criteria for severe). All other AEs were categorized as mild. Mean follow-up (f/u) time was 251 days, median f/u time was 88 days, and 91% of patients had at least 48 hours of f/u. Results: 307 attempts at DPEJ were made on 286 patients. Of these, 209 succeeded (68%) and 98 failed (32%). Overall, 89 AEs were associated with DPEJ placement or removal in 73 (23.7%) of the cases, with a clear causal relationship in 65 cases. In 16 cases more than one AE occurred. There were 13 serious AEs (4.2% of all cases), 18 moderate AEs, and 58 mild AEs. Serious AEs included 7 bowel perforations, 3 cases of jejunal volvulus, and 3 major bleeding events. The only death in this group was due to profound jejunal mesenteric bleeding after an unsuccessful trocar pass. Moderate AEs included 8 chronic enterocutaneous fistulae that required surgical or endoscopic closure and 2 significant site infections requiring hospitalization for IV antibiotics and/or abscess drainage. Many of the mild AEs were site infections requiring oral antibiotics (22/58) or persistent site pain (14/58). Conclusions: DPEJ was associated with a moderate or severe complication in 29/307 (9.4%) of cases. While DPEJ is a useful technique to gain enteral access that often obviates the need for surgery and is more reliable than percutaneous endoscopic gastrostomy with jejunal extension, patients and physicians should be aware of the risks involved. Given incomplete follow-up in many patients, these complication rates may be an underestimate.