Selective approach to use of upper gastroesophageal imaging study after laparoscopic Roux-en-Y gastric bypass

Many institutions routinely perform upper gastroesophageal imaging (UGI) studies on their laparoscopic Roux-en-Y gastric bypass (LRYGB) patients after surgery. We had routinely studied our patients with UGI on postoperative day 1 to rule out an anastomotic leak or obstruction, until recently when we abandoned this practice. As previously reported, we found that routine UGI did not contribute significantly to patient care. The purpose of this study was to determine whether patient outcomes were affected by this change in protocol.

From March 2004 to September 2005, 508 LRYGB cases were performed at our institution. Linear cutting staplers were used to create both the gastrojejunostomy and the jejunojejunostomy. In each case, the Roux limb was brought up in an antecolic, antegastric configuration. Before changing our protocol, 194 LRYGB cases were performed, and each patient underwent a routine UGI study (group 1). After abandoning this practice, 314 LRYGB cases were performed (group 2), and an UGI study was obtained only if clinical indicators (e.g., tachypnea, tachycardia, nausea, vomiting, low urine output, and/or abdominal pain) were present. The patient demographics, including gender, age, body mass index, length of hospital stay, and complications were recorded in our bariatric database and reviewed retrospectively.

A postoperative UGI study was obtained in 204 patients--in 194 patients routinely (group 1) and in 10 patients because of clinical indications (group 2). No obstructions or leaks were found in any of these 204 patients. In group 2, 304 patients were discharged without an UGI series and did well without any leak or obstruction, except for 1 patient who returned 3 months postoperatively with a stricture at his jejunojejunostomy. No statistically significant differences were found between the 2 groups.

The results of our study have shown that routine UGI studies after LRYGB do not contribute significantly to postoperative patient care at our institution. We now perform them selectively according to clinical indications, without this change adversely affecting our clinical outcomes.