Outcomes of endoscopic treatment of gastroduodenal Dieulafoy's lesion with rubber band ligation and thermal/injection therapy.

BACKGROUND

Dieulafoy's lesion is a rare but important cause of upper gastrointestinal bleeding. Current endoscopic methods used to treat Dieulafoy's lesion include injection, with or without thermal methods, and mechanical methods. The latter include variceal ligation and hemoclips. There are no studies comparing the outcomes of rubber band ligation and injection with or without thermal therapy.

AIM

To report the outcomes of Dieulafoy's lesion treated endoscopically with rubber band ligation and injection with or without thermal therapy at a single institution.

METHODS

Patients with the diagnosis of Dieulafoy's lesion treated endoscopically at the Carl T. Hayden VA Medical Center in Phoenix, between August 1994 and August 2002 were analyzed. Demographic data, mode of presentation, risk factors for gastrointestinal bleeding, hemodynamic parameters, blood transfusion requirements, endoscopic findings, details of endoscopic therapy, length of stay in ICU/hospital, complications, recurrence of bleeding, and mortality rates were collected and compared between those receiving endoscopic band ligation (EBL group) and those receiving injection with or without thermal therapy (non-EBL group).

RESULTS

Twenty-three patients with Dieulafoy's lesion (14 in the EBL group and nine in the non-EBL group) were studied. All patients were men. The mean age, hemoglobin levels on admission, and the transfusion requirements before therapy were similar in both groups. Fourteen patients (eight in the EBL- and six in the non-EBL groups) presented with hematemesis and the remaining with melena. The majority of Dieulafoy's lesions (91.3%) were located in the stomach and two in the duodenum. Active bleeding at the time of endoscopy was seen in 61% of cases, and immediate hemostasis was achieved with either method in 100% of patients. Early rebleeding (within 72 hours of endoscopic therapy) occurred in only one patient treated with epinephrine plus heater probe therapy. The length of stay in ICU was longer in the non-EBL group (6.7 days) compared with the EBL group (1.8 days) (P = 0.2). There were six deaths (three in the non-EBL group and three in the EBL group) within 30 days of the index hospitalization. The causes of death included infection/sepsis (n = 3), complications of acute myocardial infarction (n = 2), and end-stage liver disease (n = 1).

CONCLUSIONS

Endoscopic rubber band ligation is as effective as injection with or without thermal therapy in the treatment of Dieulafoy's lesion.