The presence of leak after any kind of gastric bypass can be classified based on three parameters as suggested by Csendes et al. [6–8].

Risk factors for leaks can be subdivided as surgical and non-surgical factors. The common surgical factors include anastomotic tension and ischaemia [10]. Anastomotic tension may result in stress that exceeds the disruptive forces of a stapled or sutured anastomosis contributing to a leak [3]. The position of the alimentary limb (antecolic vs retrocolic) has also been debated but no conclusive evidence exist [11–14]. Non surgical risk factors include advanced age, super-obesity, male sex, presence of multiple co-morbidities and previous bariatric operations [4, 5, 10, 15–18].

A large number of patients do not present with the typical features of peritonitis and routine post-operative oral contrast studies fail to identify a significant proportion of leaks, which can delay diagnosis and treatment [13, 18–20]. Hence a high index of suspicion is important based on the clinical parameters. Mickevicius A et al reported that a pulse rate of >90 on day 1 had a sensitivity of 100 % and specificity of 87 % [21]. The same has been reported by others as well, that unexplained tachycardia as an early indicator of leak [22, 23]. Significant differences in temperature on day 2 and higher pain scores on day 3 are additional factors. Serum C-reactive protein (CRP) concentrations were also significantly high on day 2 and 3 in patients with leaks [21] In fact Jacobsen et al suggested that in patients with tachycardia exceeding 120/min, used more pain medication than expected and/or unable to be mobilized within 2 h after surgery were considered to have a bleeding or leak and was an indication for surgical exploration in the first 24 h [9].

Some surgeons prefer to perform a routine upper GI series (UGS) in the postoperative period as a routine when early leaks can be identified [3]. But it should be noted that the sensitivity of this routine UGI series has a low sensitivity and hence not routinely followed in may centers [3, 13]. If routinely followed, it is suggested that small localized leaks can be better diagnosed with barium sulfate, and not with liquid contrast medium like Gastrograffin or Hypaque [7].

Considering the morbidity associated with a missed leak being quite significant, CT scanning can be performed. Findings suggestive of an anastomotic leak include contrast extravasation from the gastrojejunostomy or the jejunojejunostomy, collections adjacent to the gastric pouch, diffuse abdominal fluid and the presence of free intraperitoneal gas. However, the sensitivity and specificity of both UGS and CT are directly dependent on the radiologist experience with post-operative anatomical changes after roux en Y gastric bypass (RYGB) [3].

With the diagnosis of leak after RYGB, the choice is between conservative approach which is usually a combination of endoscopic/radiologic interventions or surgery(open/laparoscopy). Hamilton et al performed open re-exploration in all cases of leaks, at which time the abdomen was irrigated, leak repaired with placement of gastric feeding tubes and closed suction drains [13]. But it is interesting to note that in Ballesta’s series that the hospital stay was prolonged in patients managed operatively with a mortality of 8.5 % [3]. This could be because of the more severe nature of the leak on whom surgery was performed with the hemodynamically stable patients being managed conservatively. Gonzalez et al reported that 12 % of patients had unsuccessful non-operative treatment and required subsequent operation because of systemic toxicity or poor clinical outcome [18].