Endoscopic Band Ligation Could Decrease Recurrent Bleeding in Mallory–Weiss Syndrome as Compared to Haemostasis by Hemoclips Plus Epinephrine

Endoscopic Band Ligation Versus Hemoclips Plus Epinephrine

S. Lecleire, M. Antonietti; I. Iwanicki-Caron; A. Duclos; S. Ramirez; E. Ben-Soussan; S. Hervé; P. Ducrotté


Aliment Pharmacol Ther. 2009;30(4):399-405. 

In This Article

Patients and Methods


Between January 2001 and April 2008, 218 patients with an endoscopically confirmed MWS, defined as a mucosal tear on the oesophago-gastric junction occurring after vomiting efforts followed by haematemesis, were hospitalized in our specialized Gastro-Intestinal Bleeding Unit. Among these 218 patients, 56 (26%) underwent an endoscopic haemostasis because of an active bleeding (spurting or oozing) or the presence of a visible vessel on the MW tear. The endoscopy was performed within 12 h of admission, after administration of basic life support. Clinical characteristics, including gender, age, hemodynamic shock defined by a systolic blood pressure of less than 90 mm Hg and a pulse rate greater than 100 beats/min, accompanied by pallor and/or cold sweating, coagulopathy or bleeding diathesis defined as a prothrombin time of greater than 14.5 s (normal 11–13 s) and/or a platelet count less than 60 000/mm3 (130 000–490 000),[4,6] the use of antiplatelets or anticoagulant drugs, a liver cirrhosis history and routine laboratory values were systematically registered in a specific medical observation of our Gastro-Intestinal Bleeding Unit. We retrospectively analysed these data.


The endoscopy was performed using an upper video-endoscope (GIF-Q180, Olympus Optical Co., Ltd., Tokyo, Japan), by five trained senior endoscopists (SL, MA, SR, EBS, SH). An endoscopic haemostasis was performed in patients with active bleeding (spurting or oozing) or with a visible vessel. Patients with adherent clots were only treated by endoscopy if an active bleeding was seen at the clot base. Endoscopists had the choice for the haemostatic procedure between a single band ligation (Banding group) and the association of hemoclip placement and epinephrine injection (H&E group).

In patients from Banding group, a single-band ligator was used with no overtube. After endoscopic identification of the MW tear, the endoscope was withdrawn and then reinserted after attachment of the band ligator. The hood of the ligating device was placed over the bleeding site or the visible vessel and then suction was applied to draw the bleeding mucosa into the banding cylinder. Clinicians could place as many bands as required to obtain the haemostasis.

In patients from H&E group, hemoclips (MDS50, Olympus) were applied to bleeding site with a clip application device (HX-3L, Olympus). Clinicians could place as many clips as required to obtain the haemostasis. An epinephrine injection was associated with the hemoclips placement in all patients from H&E group, before or after hemoclips placement, using a 1:10 000 solution of epinephrine injected into the surrounding area close to the bleeding point using a 23-gauge disposable needle (MTW, Wesel, Germany). The volume of injected epinephrine solution was decided by the endoscopist who performed the haemostasis procedure.


Primary haemostasis was defined as endoscopically verified cessation of bleeding for at least few minutes after band ligation or after hemoclip placement plus epinephrine injection during the first endoscopic session.

Recurrent bleeding was defined by one or more of the following signs of ongoing bleeding: haematemesis, haematochezia, instability of vital signs (systolic blood pressure of less than 90 mm Hg and a pulse rate greater than 100 beats/min), associated or not with a decrease in haemoglobin concentration of more than 2 g/dL within 24 to 72 h after successfully achieving primary haemostasis. Patients presenting with a first episode of recurrent bleeding underwent a second endoscopy. If active bleeding was found at the MW tear, recurrence of bleeding was confirmed. Haemostasis was then attempted using the other procedure than the one initially performed. Secondary haemostasis was defined as cessation of first episode of recurrent bleeding after a second cession of therapeutic endoscopy. Patients who did not respond to endoscopic re-treatment or presenting with a second episode of recurrent bleeding, defined by the same criteria than the first one, were evaluated for a salvage haemostasis procedure, encompassing transcatheter arterial embolization or emergency surgery. After endoscopy, all patients were observed in our Gastrointestinal Bleeding Unit. They were all given a single-dose of proton pump inhibitors for al least 2 weeks by oral route. Data on clinical outcome were routinely collected until patient's discharge, including the rate of primary haemostasis, the rate of recurrent bleeding, the need for a salvage haemostasis procedure during hospitalization, the blood transfusion requirements with the number of transfused red cells units, the hospital mortality rate and the length of hospital stay.

Statistical Analysis

Results are expressed as percentages or means ± standard deviation (s.d.) for continuous variables. Chi-square test, Fisher's exact test and Student t test were used to compare noncontinuous and continuous data, as appropriate. A P value less than 0.05 was considered significant. A univariate analysis was performed in all patients to identify the predictive factors of recurrent bleeding. Multivariate analysis of predictive factors of recurrent bleeding was then performed by logistic regression with calculation of odds ratio (OR) and a confidence interval (CI) of 95% for all tested data.