BACKGROUND The most suitable mechanical endoscopic hemostasis for a bleeding Dieulafoy's lesion (DL) is not yet well established.

OBJECTIVE To compare the hemostatic efficacy and clinical outcome of endoscopic hemoclip placement (EHP) and endoscopic band ligation (EBL).

DESIGN Retrospective, single-center study.

SETTING A tertiary-care referral university hospital.

PATIENTS Sixty-six patients who received mechanical endoscopic hemostasis for bleeding DLs.

INTERVENTIONS Endoscopic hemostasis.

MAIN OUTCOME MEASUREMENT Primary hemostasis and rebleeding rates.

RESULTS DLs accounted for 3.8% of cases of acute nonvariceal upper GI bleeding during the study period. Active bleeding from DLs was noted in 34 patients (51.5%). EHP and EBL were performed as a method of endoscopic hemostasis in 34 and 32 patients, respectively. There were no significant differences between the 2 groups with respect to baseline characteristics (except comorbidities) and endoscopic features of DLs. Primary hemostasis was achieved in all 66 patients (100%). There were 6 cases of recurrent bleeding: 5 (14.7%) and 1 (3.1%) in the EHP and EBL groups, respectively. Secondary hemostasis was achieved with endoscopic treatment and angiographic embolization in 5 patients and 1 patient, respectively, and no patients required surgery. The mean procedure time of endoscopic hemostasis was significantly longer in the EHP group (19.1 vs 11.5 minutes, P = .015). There was no bleeding-related mortality.

LIMITATIONS Retrospective analysis.

CONCLUSIONS Both EHP and EBL are suitable for the treatment of bleeding DLs. EBL can be used as an initial hemostatic method for bleeding DLs because of a favorable clinical outcome comparable to that with EHP and a shorter procedure time.