Introduction Gastric astral vascular ectasia (GAVE) is a rare cause of gastrointestinal bleeding (GIB). Patients often require regular blood transfusions and many require oral or intravenous iron to manage symptomatic anaemia. Endoscopic therapy includes argon plasma coagulation (APC), YAG laser therapy and band ligation. In some patients, these measures are unsuccessful and they remain transfusion dependent with significant effects on health and quality of life. Radio-frequency ablation (RFA) may provide a solution.
Method Patients who had remained anaemic and/or transfusion or iron dependent were eligible. Other causes of GIB were excluded. Treatment with RFA was carried out with focal RFA at 12 J/cm2 with 3 applications to all visible areas of GAVE by a single endoscopist. Patients had up to 2 RFA treatments 6 weeks apart before entering the follow up phase. Data were collected before and after treatment.
The primary outcome was change in haemoglobin (Hb) 3 months after treatment. Secondary outcomes were a reduction in frequency of blood transfusions and/or iron (PO or IV) in the 3 months before and after RFA treatment. Endoscopic surface area (SA) regression of areas of GAVE were analysed by asking 3 expert endoscopists to estimate the overall% change in SA following treatment by examining endoscopic images of the stomach before and after treatment.
Results 16 patients have undergone RFA for refractory GAVE. The median age was 70 years (IQR 56–84). 13/16 (82%) were female. 12 were previously treated with APC, 3 with LASER and 2 of those treated with APC had also had band ligation. In the 3 months prior to treatment, 11/16 were on oral iron, 2/16 were on IV iron. 11/16 required blood transfusions. 8/16 required both iron and blood transfusions. The median number of RFA treatments was 1 (IQR 1–2). No patients had recorded complications of RFA.
The mean pre-treatment Hb was 99.5 g/L (95% CI 90.3–108.6 g/L). The mean post treatment Hb at 3 months after treatment was 121 g/L (95% CI 112–129 g/L). The mean change in Hb was +20.5 g/L (95% CI 8.8–30.0 g/L) (p=0.0026).
Post treatment, only 2/11 (18%) of the patients who required transfusion prior to the procedure had ongoing transfusions. Only 2/13 (15%) patients who were previously on iron had ongoing iron requirements.
The mean surface area regression when scored by 3 upper GI endoscopists was 51.6%. [95% CI 38.7–64.4]
Conclusion RFA for patients with symptomatic anaemia due to GAVE is a novel treatment therapy for a difficult cause of GI bleeding. This single centre experience suggests it can reduce transfusion dependence and iron supplementation in some patients with an improved Hb after treatment. The required number of treatments is small and it appears safe.
Disclosure of Interest None Declared
- Radio-frequency ablation