Utilization and morbidity associated with placement of a feeding jejunostomy at the time of gastroesophageal resection.

BACKGROUND: The purpose of the study was to evaluate the utilization and morbidity associated with feeding jejunostomy tubes (JT) placed at the time of gastroesophageal resection (GER).
METHODS: Under institutional review board approval, a prospective database of patients undergoing GER from January 2004 to September 2010 was reviewed. Data analyzed included patient demographics, postoperative complications, JT use, and JT specific complications. Fisher's exact tests explored associations with utilization of a JT following resection.
RESULTS: Seventy-three patients (51 men, 22 women, median age of 59) underwent placement of a JT at the time of GER (total gastrectomy = 28, Ivor-Lewis = 28, subtotal gastrectomy = 8, proximal gastrectomy  = 6, and transhiatal esophagectomy  = 3) of both malignant (97%) and benign (3%) disease processes. Twenty-one JT specific complications (11 minor and 10 major) were identified. Reoperation was required in the management of two complications (small bowel obstructions), while all other complications were easily managed by an interventional radiologist (n = 8), bedside procedure (n = 5), or did not require intervention (n = 6). Eighty-six percent of patients were discharged tolerating a postgastrectomy diet, 10% nothing per orem, and 4% a liquid diet. Inpatient enteral nutrition (EN) was initiated in 68%, but continued on discharge in only 54% secondary to failure to thrive (54%), dysphagia (21%), anastomic leak (15%), chyle leak (3%), esophagostomy (3%), and duodenal stump leak (3%). The mean time to discontinuance of EN and removal of the JT was 44 days (range, 4-203) and 71 days (range, 15-337) respectively. Although only 13% (n = 5) of patients requiring adjuvant therapy were utilizing their JT at the commencement of therapy, 75% (n = 21) required EN during its course. The median time to adjuvant therapy was found to be slightly longer in those who required outpatient EN versus those who did not (61 vs. 90 days, p = 0.08). However, the median time to adjuvant therapy did not differ between those who were and were not receiving EN at the time of adjuvant therapy commencement (80 vs. 92 days, p = 0.2). Age (p = 0.4), number of co-morbidities (p = 0.2), preoperative percent body weight loss (p = 0.9), and clinical stage (p = 0.8) were not significantly associated with outpatient JT use. Patients who suffered a postoperative complication were most likely to require EN (p = 0.002), an association that strengthened as the number of complications increased (p = 0.0008). Although not statistically significant, a trend towards increased outpatient EN was noted in patients who underwent transhiatal esophagectomy and total gastrectomy (p = 0.06).
CONCLUSIONS: JT placement carries a considerable morbidity in patients undergoing GER. However, because it is difficult to preoperatively ascertain who will need prolonged EN, the routine placement of a JT is recommended, particularly in those who will likely require adjuvant therapy or are at high risk for postoperative complications. Despite patient desires for early removal of an unused JT, caution should be taken if adjuvant therapy is being considered.