Early detection of complications after laparoscopic Roux-en-Y gastric bypass (LRYGB) can be difficult because of the subtle clinical findings in obese patients. Consequently, routine postoperative upper gastrointestinal contrast studies (UGI) have been advocated for detection of leak from the gastrojejunostomy. The medical records of 368 consecutive patients undergoing LRYGB were analyzed to determine the efficacy of selective use of radiological studies after LRYGB. Forty-one patients (11%) developed signs suggestive of complications. Of the 41 symptomatic patients, two were explored urgently, 39 (10%) had radiological studies, and 16 of them (41%) were diagnosed with postoperative complications. Overall morbidity of the series was 4.8 per cent. Four patients (1.1%) developed a leak from the gastrojejunostomy and were correctly diagnosed by ‘ computerized tomography (CT). The sensitivity and specificity of CT in determining leak was 100 per cent, with positive and negative predictive value of 100 per cent. The mortality of the series was 0 per cent. No radiologie studies were performed in asymptomatic patients, and no complications developed in these patients. Our results show that selective radiological evaluation in patients with suspected complications after LRYGB is safe. High sensitivity makes CT the test of choice in patients with suspected complication after LRYGB. Routine radiological studies are not warranted.
LAPAROSCOPIC ROUX-EN-Y GASTRIC bypass (LRYGB) has become the most often performed bariatric operation in the United States.1,2 Easier recovery, lower morbidity, shorter hospital stay, less pain, less disability, and superior cosmetic results have made minimally invasive surgery the preferred surgical approach for the treatment of morbid obesity. However, LRYGB is not risk-free, especially during the learning curve. Gastrointestinal leak and small bowel obstruction are the most common complications after open or LRYGB. Early detection of these complications allows prompt correction and reduces the associated morbidity and mortality.3
Following open Roux-en-Y gastric bypass at most bariatric programs, postoperative upper gastrointestinal contrast studies (UGI) are routinely performed to detect a leak or early obstruction because the subtle early clinical signs of these complications after open procedures are obscured by incisional pain and narcotic administration.4-8 Laparoscopic surgeons have adopted the practice of routine postoperative radiological contrast studies, perhaps because of this custom following open bariatric surgery.1,9 We undertook this study to evaluate the safety and efficacy of selective radiological studies following LRYGB.
Patients and Methods
The study group consisted of consecutive patients who underwent LRYGB in our institution from January 2001 to December 2003. The data were collected concurrently in a custom database designed specifically for bariatric surgery (FileMaker Pro 6, FileMaker Inc., Santa Clara, CA). A protocol of obtaining imaging studies selectively was in place. Outcome measures included type of imaging study, result of study, morbidity, and mortality. The radiographie reports of the patients who had CT or UGI performed in the early postoperative period were evaluated for effectiveness of detection of postoperative complications.
The majority of LRYGBs were performed in a standard technique that involves creation of a 15-cc gastric pouch, a stapled gastrojejunostomy using linear endoscopie cutter and hand-sewn anterior layer, a 75-cm antecolic antegastric Roux limb, and a stapled side-toside jejunojejunostomy. The first 68 operations were performed with retrocolic antegastric Roux-en-Y reconstruction.
CT or UGI studies were performed selectively in hemodynamically stable patients based on the clinical signs of new or persistent tachycardia, tachypnea, chest or abdominal pain, nausea, vomiting, fever, elevated white blood cell count, or drop in hemoglobin. Unstable patients were reexplored without additional diagnostic studies. The results were analyzed for specificity, sensitivity, and negative and positive predictive value.
Overall, 368 LRYGBs were performed. Three hundred twenty-seven patients (89%) had an uneventful postoperative period and required no imaging studies. Forty-one patients (11%) developed symptoms and signs in the early postoperative period suggestive of complications. In 23 (56% of symptomatic patients), no complications were detected, and the symptoms resolved spontaneously. Of the 41 symptomatic patients, 18 (44%) were diagnosed with postoperative complications (overall morbidity of 4.8%). The mortality of the series was O per cent. The complications are summarized in Table 1.
The data on clinical signs and symptoms of suspected complications were available for 33 patients. Most of the symptomatic patients had abdominal pain, tachycardia, nausea or vomiting, and elevated white cell count (Table 2).
Two patients had five clinical signs and symptoms. Both of these patients had postoperative complications. Of four patients with four clinical signs, two had complications. Of 10 patients with three clinical signs, four (40%) had complications. With one or two signs the chance of having postoperative complication was low (Table 3).
TABLE 1. Early Postoperative Morbidity
TABLE 2. Clinical Signs/Symptoms in Symptomatic Patients
TABLE 3. Clinical Signs/Symptoms and Complications for 33 Patients
Radiological contrast studies in the early postoperative period were performed in 39 of the 41 symptomatic patients. CT was performed in 34 (87%), UGI in addition to CT in 6 (15%), and UGI alone in 5 (13%). Table 4 correlates the study with the complications detected. Two patients with intraabdominal bleeding developed hemodynamic instability shortly after operation and underwent immediate laparoscopic exploration without imaging.
Four patients (1.1%) developed a leak from the gastrojejunostomy. CT detected this complication in three. All three patients were reoperated with uneventful recovery. A leak was diagnosed in the fourth patient clinically when gastric contents were identified in a drain placed intraoperatively. CT confirmed leak by extravasation of the contrast material. This patient was successfully managed nonoperatively. The cessation of leak in all four patients was confirmed by UGI. When CT was used to determine if a leak was present, there were no false-positive results. The sensitivity and specificity of CT in determining leak was 1OO per cent, with positive and negative predictive value of 100 per cent.
Early postoperative small bowel obstruction developed in three (0.8%) patients. Two of them had Roux limb obstruction, one diagnosed by CT and the other by UGI. The third patient had an obstruction of the biliopancreatic limb. UGI failed to detect it, but CT correctly diagnosed this complication. All patients with small bowel obstruction were successfully reoperated laparoscopically. All cases of small bowel obstruction occurred in patients with retrocolic Roux limb placement. No early or late small bowel obstruelions have developed in the 300 patients with antecolic antegastric Roux limb reconstruction.
TABLE 4. Imaging Modalities Used far Diagnosis of Complications
Early partial obstruction of the gastrojejunostomy was correctly diagnosed by UGI in two patients. Both patients were treated nonoperatively. Obstruction resolved spontaneously by the fifth postoperative day as confirmed with UGI.
Other complications occurred as well. Gastrogastric fistula was diagnosed in one patient by CT and confirmed by UGI. CT identified an abdominal wall hematoma in two patients, an intraabdominal collection in one, and atelectasis in three.
The majority (89%) of 368 patients had an uneventful postoperative period after LRYGB. Only 11 per cent of the whole group developed clinical symptoms and signs of complications. Of the 41 symptomatic patients, 18 (44%) had postoperative complications. All four leaks from the gastrojejunostomy were accurately diagnosed by CT. CT also diagnosed other complications such as gastrogastric fistula, small bowel obstruction, intraabdominal fluid collection, and abdominal wall hematoma. These results show that thorough clinical evaluation in combination with selective use of radiological contrast studies following LRYGB is a safe clinical practice. Selective use of CT and UGI has allowed accurate early detection and treatment of all complications, resulting in low morbidity and in no mortality. Most importantly no complications were seen in asymptomatic patients, nor did a selective imaging protocol result in a delay in diagnosis.
Selective use of postoperative radiology is not only safe but also cost and resource effective. The cost of an UGI study at our institution is $175. Selective use of radiological studies resulted in a cost saving of $57,575. At a time of increasing costs and decreasing health care dollars, it is very important to have clinical evidence to justify diagnostic procedures, especially in recommending a practice standard.
Reducing morbidity and mortality is important in all surgery but especially in bariatric surgery, as morbidly obese patients have little reserve. Gastroin\testinal leak can be a devastating complication of LRYGB. The incidence of leak after LRYGB ranges from 1.5 to 5.1 per cent and is the major reason for hospital mortality.3, 10 But gastrointestinal leak is difficult to recognize after gastric bypass because clinical findings in obese patients are often subtle. Persistent tachycardia and tachypnea are the most common early signs. Lowgrade fever and abdominal pain radiating to the back, chest, and left shoulder can also be early symptoms. Physical exam rarely shows abdominal tenderness unless the bypassed stomach leaks. The white cell count is often elevated but may be within normal limits. The occult nature of leaks has led many surgeons to routinely order postoperative contrast studies.5-7, 11- 13
We adopted a selective imaging approach with CT as our first choice in evaluating a symptomatic patient. Advantages of CT are high sensitivity for extraluminal gas and contrast material and the ability to detect other abnormalities not generally diagnosed by UGI.14 Other series report the value of CT as part of the protocol for the diagnosis and treatment of the gastrointestinal leak after gastric bypass.15 However, CT has the disadvantage of a weight limit for the table top and an aperture that limits the size of patients that can be imaged.
UGIs can be used when CT is technically impossible. However, the ability of UGI contrast studies to detect leaks suffers from low sensitivity. In a series of 201 LRYGBs, routine UGI studies detected leak in only 33.3 per cent of patients who actually had a leak.11 Another study showed that routine UGI after open gastric bypass in 100 patients diagnosed three of four leaks.6 Similarly, other authors have reported that routine UGI does not eliminate the risk of misdiagnosed gastrointestinal leak.5, 12, 14, 16, 17 Routine UGI has not been shown to be beneficial in early detection of post- LRYGB complications,13 although several studies have shown that obtaining UGI studies in symptomatic patients decreases the number of unnecessary x-ray studies without compromising patient care.4, 15, 18
We conclude that selective radiological evaluation in the patients with suspected complications after LRYGB is safe. Patients with clinical signs and symptoms of complications following LRYGB should be studied promptly with CT or, if technically impossible, with UGI. High sensitivity makes CT the test of choice in patients with suspected complication after LRYGB. Routine radiological studies are not warranted.
1. DeMaria EJ, Sugerman HJ, Kellum JM, et al. Results of 281 consecutive total laparoscopic Roux-en-Y gastric bypasses to treat morbid obesity. Ann Surg 2002;235:640-5.
2. Mason E. IBSR newsletter, 2003. htlp://www.surgery. uiowa.edu/ ibsr/wsummer03.htm
3. Marshall JS, Srivastava A, Gupta SK, et al. Roux-en-Y gastric bypass leak complications. Arch Surg 2003; 138:520-3.
4. Singh R, Fisher BL. Sensitivity and specificity of postoperative upper GI scries following gastric bypass. Obes Surg 2003; 13: 73-5.
5. Ganci-Cerrud G, Herrera MF. Role of radiologic contrast studies in the early postoperative period after bariatric surgery. Obes Surg 1999;9:532-4.
6. Serafini F, Anderson W, Ghassemi P, et al. The utility of contrast studies and drains in the management of patients after Roux- en-Y gastric bypass. Obes Surg 2002; 12:34-8.
7. Toppino M, Cesarani F, Comba A, et al. The role of early radiological studies after gastric bariatric surgery. Obes Surg 2001 11:447-54.
8. Buckwalter JA, Herbst CA Jr. Leaks occurring after gastric bariatric operations. Surgery 1988;103:156-60.
9. Witlgrove AC, Clark GW, Schubert KR. Laparoscopic gastrie bypass, Roux-en-Y: technique and results in 75 patients with 3-30 months follow-up. Obes Surg 1996;6:500-4.
10. Brolin RE. Gastric bypass. Surg Clin North Am 2001;81: 1077- 95.
11. Sims TL, Mullicun MA. Hamilton EC, et al. Routine upper gastrointestinal gastrografin swallow after laparoscopic Rouxen-Y gastric bypass. Obes Surg 2003; 13:66-72.
12. Hawthorne A, Kuhn J, McCarty TM. The role of routine upper GI series following Roux-en-Y gastric bypass. Obes Surg 2003; 13:222- 3.
13. Adusumilli P, Kurian M, Leitman M, et al. Docs early upper gastrointestinal series help in the management of gastric bypass patients. Obes Surg 2002; 12:207.
14. Blachar A, Federle MP, Pealer KM, et al. Gastrointestinal complications of laparoscopic Roux-cn-Y gastric bypass surgery: clinical and imaging findings. Radiology 2002;223:625-32.
15. Liu DC, Glantz G, Sawicki M, Livingston E. Management of post- operative leaks in Roux-en-Y gastric bypass patients: argument for selective radiologie studies. Obes Surg 2002; 12:215.
16. Schauer PR, Ikramuddin S, Gourash W, et al. Outcomes after laparoscopic Roux-en-Y gastric bypass Tor morbid obesity. Ann Surg 2000;232:515-29.
17. Suter M, Giusti V, Heraief E, et al. Laparoscopic Rouxen-Y gastric bypass: initial 2-year experience. Surg Endosc 2003; 17:603- 9.
18. McCormick JT, Papasavas PK, Pastor CG, et al. Analysis of radiographie studies used to evaluate complications after laparoscopic Roux-en-Y gastric bypass. Obes Surg 2003; 13:524.
SERGEY LYASS, M.D., THEODORE M. KHALILI, M.D., SCOTT CUNNEEN, M.D., FUMIHIKO FUJITA, M.D., KOJI OTSUKA, M.D., RITU CHOPRA, M.D., BRIAN LAHMANN, M.D., MATTHEW LUBLIN, M.D., GARY FURMAN, M.D., EDWARD H. PHILLIPS, M.D.
From the Center for Minimally Invasive Bariatric Surgery, Department of General Surgery, Cedars-Sinai Medical Center, Los Angeles, California
Presented at the Annual Meeting, Southern California Chapter of the American College of Surgeons, January 16-18, 2004, Santa Barbara, CA,
Address correspondence and reprint requests to Sergey Lyass, M.D., Department of Surgery, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Suite 8215, Los Angeles, CA 90048.
Copyright The Southeastern Surgical Congress Oct 2004