|
|
|
|
Back to NESS Scientific Program
Routine feeding jejunostomy tube placement may not be beneficial in pancreaticoduodenectomy.
James T McPhee1, Deep Adhikari1, Theodore P McDade1, Joshua S. Hill1, Giles F Whalen1, Dana K Andersen2, Hiroomi Tada3, Mary E Sullivan1, Jennifer F Tseng1
1University of Massachusetts Medical School, Worcester, MA;2Johns Hopkins Bayview Medical Center, Baltimore, MD;3Temple University Hospital, Philadelphia, PA
Objective: To determine if routine placement of feeding jejunostomy during pancreaticoduodenectomy (PD) is beneficial.
Design: Single-institution retrospective study.
Setting: University teaching hospital, 2002-2006.
Patients: 53 sequential patients undergoing pancreaticoduodenectomy.
Interventions: Standard or pylorus-preserving pancreaticoduodenectomy.
Main Outcome Measures: Effect of feeding tube (FT) placement on post-operative outcomes.
Results: Mean patient age was 62.9 years. 53% were women. 64% underwent standard PD while 36% underwent a pylorus-preserving procedure. Twenty-nine patients (55%) had FT placed during surgery. The FT group was older (mean age, 66 vs. 59, P=.04), and more underwent standard PD (79% vs. 46%, P=.01). The two groups had similar performance status (P=.26), and prevalence of co-morbidities (cardiac disease (P=.16), pulmonary disease (P=1), hypertension (P=.43), and diabetes (P=.84)). Mean operative time (FT 550 minutes vs. non-FT 512, P=.28) and estimated blood loss (FT 996 mL vs. non-FT 838, P=.37) did not differ significantly. The prevalence of major perioperative complications including leak (P=1), reoperation (P=0.29) and cardiopulmonary morbidity (P=0.44) were similar for the two groups. Nutritionally, the FT group had a mean postoperative serum albumin level significantly lower than the non-FT group (2.7 vs. 3.3, P=0.02), and 57% of the FT group had an albumin level that had decreased from the preoperative value compared to 24% of the non-FT group (P=0.05). 6 patients in the FT group (21%) experienced a tube-related complication including dislodgement (n=3), rupture/leak (n=2) and occlusion (n=1).
Conclusions: Routine feeding tube placement does not significantly increase operative time or blood loss. However, it does not confer an identifiable nutritional advantage and is associated with a relatively high (21%) tube-related complication rate. Our data do not support the routine placement of feeding tubes during pancreaticoduodenectomy.
Back to NESS Scientific Program
|
