Epidemiology

Early postoperative bleeding occurs in 1% to 5% of cases after RYGB. The variable reported rates for early bleeding depend on the definition used for “early” hemorrhage, type of surgery performed (laparoscopic vs open), type of bleeding described (eg, intraluminal or extraluminal), perioperative use of antithrombotic agents, and threshold used to differentiate bleeding from normal postoperative hemodilution.

In a meta-analysis involving 3464 RYGB patients, the reported rate for acute postoperative GI bleeding was 1.9%. In another study, early postoperative hemorrhage occurred in 33 of 1025 patients (3.2%) who underwent RYGB; bleeding was extraluminal in almost 50% of cases. Similar to other study findings, the postoperative bleeding rate was higher in the laparoscopic group than in the open group (5.1% vs 2.4%). Rebleeding episodes may also occur more commonly in this patient population. In a retrospective study involving 933 LRYGB patients, 30 (3.2%) developed acute postoperative bleeding. Of these, a single bleeding episode occurred in 14 (47%) patients; two bleeding episodes occurred in 13 (43%) patients; and three bleeding episodes occurred in 3 (10%) patients. In another study of 518 patients who underwent LRYGB, the rate of early postoperative bleeding was 3.9%; 80% of the bleeding was extraluminal and 20% was intraluminal.

Postoperative bleeding is a rare complication after vertical banded gastroplasty, laparoscopic adjustable gastric banding, and laparoscopic sleeve gastrectomy. Death after bariatric surgery as a consequence of acute bleeding is uncommon. In a retrospective study of 13,871 patients who underwent bariatric surgery, 34 deaths were reported. Of these, only one was due to a bleeding gastric ulcer after biliopancreatic diversion.

Predisposing factors

Obesity itself is an independent risk factor for venous thromboembolism. The risk for a thromboembolic event, including deep vein thrombosis and pulmonary embolism, is increased further in patients undergoing bariatric surgery. Because the risk may be as high as 3.1%, several measures are implemented to reduce the rate of postoperative venous thromboembolism, including early ambulation, compression stockings, and administration of prophylactic heparin. Drugs, such as heparin and clopidogrel, have been used to prevent thromboembolic events but the risk of postoperative bleeding using these agents must be balanced against the risk of postoperative thromboembolic events.

In a retrospective study, no correlation was seen between serum levels of anti-Xa and use of low molecular weight heparin (LMWH) and the risk of postoperative bleeding. In contrast, another retrospective study demonstrated an increased incidence of early hemorrhage after RYGB in patients who received preoperative LMWH. In patients receiving prophylactic antithrombotic agents after LRYGB, unfractionated heparin was associated with a lower incidence of acute postoperative bleeding than enoxaparin. In a recent prospective study, an extended 3-week LMWH protocol was used for prophylaxis of venous thromboembolism in 735 patients who underwent laparoscopic bariatric surgery. The incidence of postoperative venous thromboembolism was 0% and only 3 adverse bleeding events occurred. In the presence of clinically overt GI bleeding, the administration of heparin products should be discontinued.

Limited data are available regarding the risk for early postoperative GI bleeding associated with the use of antiplatelet agents, such as clopidogrel. One study reported patients undergoing gastric bypass having a high bleeding rate while taking clopidogrel, but all bleeding cases were late events occurring at 25 to 234 days from surgery.

Ketorolac, an injectable nonsteroidal anti-inflammatory drug with strong antiplatelet activity, is often used as an alternative to opioid analgesia for relief of postsurgical pain. After a single dose of ketorolac, platelet aggregation is disturbed for at least 24 hours. In one study, however, no correlation was seen between the use of ketorolac and perioperative bleeding in postbariatric surgery patients despite the concomitant use of heparin.

Predisposing factors

Obesity itself is an independent risk factor for venous thromboembolism. The risk for a thromboembolic event, including deep vein thrombosis and pulmonary embolism, is increased further in patients undergoing bariatric surgery. Because the risk may be as high as 3.1%, several measures are implemented to reduce the rate of postoperative venous thromboembolism, including early ambulation, compression stockings, and administration of prophylactic heparin. Drugs, such as heparin and clopidogrel, have been used to prevent thromboembolic events but the risk of postoperative bleeding using these agents must be balanced against the risk of postoperative thromboembolic events.

In a retrospective study, no correlation was seen between serum levels of anti-Xa and use of low molecular weight heparin (LMWH) and the risk of postoperative bleeding. In contrast, another retrospective study demonstrated an increased incidence of early hemorrhage after RYGB in patients who received preoperative LMWH. In patients receiving prophylactic antithrombotic agents after LRYGB, unfractionated heparin was associated with a lower incidence of acute postoperative bleeding than enoxaparin. In a recent prospective study, an extended 3-week LMWH protocol was used for prophylaxis of venous thromboembolism in 735 patients who underwent laparoscopic bariatric surgery. The incidence of postoperative venous thromboembolism was 0% and only 3 adverse bleeding events occurred. In the presence of clinically overt GI bleeding, the administration of heparin products should be discontinued.

Limited data are available regarding the risk for early postoperative GI bleeding associated with the use of antiplatelet agents, such as clopidogrel. One study reported patients undergoing gastric bypass having a high bleeding rate while taking clopidogrel, but all bleeding cases were late events occurring at 25 to 234 days from surgery.

Ketorolac, an injectable nonsteroidal anti-inflammatory drug with strong antiplatelet activity, is often used as an alternative to opioid analgesia for relief of postsurgical pain. After a single dose of ketorolac, platelet aggregation is disturbed for at least 24 hours. In one study, however, no correlation was seen between the use of ketorolac and perioperative bleeding in postbariatric surgery patients despite the concomitant use of heparin.

Diagnosis and initial management

The majority of GI bleeding occurs in the intraoperative or immediate postoperative period. In one study, 70% of bleeding cases manifested themselves within 4 hours after surgery. Early GI bleeding may be intraluminal or extraluminal. When extraluminal, an increase in abdominal drain output of blood becomes an essential indicator of postoperative bleeding and allows bleeding to be recognized early so that prophylactic anticoagulants can be discontinued. In addition, based on the drainage output, it is possible to determine the rate of extraluminal bleeding and the need for transfusion and re-exploration. When blood clots occlude the drain, however, when kinking of the drain occurs, or when incorrectly positioned, extraluminal bleeding may not be recognized. Alternatively, the diagnosis of intraluminal hemorrhage is readily apparent in the presence of hematemesis, melena, or hematochezia. Associated symptoms and signs of bleeding may include tachycardia, hypotension, drop in hemoglobin, and even small bowel obstruction from blood clots. Extraluminal (intra-abdominal) hemorrhage occurs in as many as half of early postoperative bleeding cases and requires a high index of suspicion because cardinal signs of GI bleeding (eg, melena) are absent. Patients complaining of more abdominal pain than expected and those with diffuse abdominal tenderness with hypoactive bowel sounds and a pattern of cyclic tachycardia (100 to 120 beats per minute) should be considered to have postsurgical bleeding even in the absence of an increase in abdominal drain output.

Obtaining an abdominal imaging study (eg, CT scan) should be considered, particularly when alarm symptoms, such as tachycardia, oliguria, abdominal distention, and a precipitous fall in hematocrit, are present.

In most patients, early postoperative GI bleeding is mild and can be managed conservatively, including the withholding of antithrombotic agents. Some investigators suggest that diagnostic studies are unnecessary, because the source of bleeding is more likely at the staple line. In a review that included 89 GI bleeding cases after LRYGB, clinical observation alone was sufficient in 18 (20%); 49 patients (55%) required only fluid replacement and blood transfusion. Surgery was necessary in 18 (20%) patients who were hemodynamically unstable or remained unstable with ongoing bleeding despite fluid and blood transfusion. In another study, 7/20 patients (35%) required surgical intervention to manage bleeding.