Radiofrequency ablation for patients with refractory symptomatic anaemia secondary to gastric antral vascular ectasia

Abstract

Background

Gastric antral vascular ectasia (GAVE) is a rare cause of gastrointestinal bleeding, often causing iron deficiency anaemia. Previous studies have looked at the management of this with argon plasma coagulation, laser therapy and endoscopic band ligation.

Methods

This was a single-centre prospective study to evaluate the efficacy and safety of radiofrequency ablation (RFA) in patients with GAVE with persistent anaemia refractory to at least one session of first-line endoscopic therapy. Patients were treated with a through-the-scope (TTS) radiofrequency catheter at two endoscopic sessions six weeks apart. The primary outcome was change in haemoglobin at six months posttreatment. The secondary outcomes were reduction in blood or iron requirements, endoscopic surface area regression and complications.

Results

Twenty patients were treated. The mean change in haemoglobin at six months was +12.6 g/l (95% confidence interval 11.7–24.3 g/l), paired t test p < 0.001. At six months, three of 14 individuals who had required blood transfusions had ongoing blood transfusions and five of 17 who had required iron had ongoing iron needs. Surface area regression was scored as 74% ± 25% but no correlation was seen between this and other outcomes. Three of 20 patients experienced pain which was managed with oral analgesia. Of the 14 patients who had reached 12-month follow-up, three required retreatment (21%).

Discussion

This small study suggests that RFA is a safe and effective treatment for GAVE. Our study uses the TTS catheter compared to other studies, and demonstrates prolonged improvement in haemoglobin and reduction in blood and iron requirements with a novel assessment of surface area regression.

Introduction

Gastric antral vascular ectasia (GAVE) was first described in 1953 by Rider et al. ¹ Later studies described its endoscopic appearances as ‘watermelon stomach’ with ‘longitudinal rugal folds traversing the antrum and converging on the pylorus, each containing a visible convoluted column of vessels, the aggregate resembling the stripes on a watermelon’. ² Although this is usually diagnosed by pathognomonic endoscopic appearances (Figure 1), confirmation by biopsies shows dilated, ectatic capillaries in mucosa and submucosa with microthrombi and minimal inflammation.^3,4

Figure 1.

Examples of the characteristic endoscopic changes of gastric antral vascular ectasia at endoscopy.

The majority of patients with GAVE have concomitant underlying systemic diseases including cirrhosis of the liver, ⁵ autoimmune diseases, ⁶ chronic kidney disease and hypothyroidism, amongst others. ⁷ The underlying pathophysiology of GAVE is unclear with roles suggested for prostaglandin E2 levels, ⁸ cross-linking of antibodies to proteins in the gastric mucosa, ⁹ trauma of the gastric mucosa due to dysmotility ¹⁰ or vasoactive substances such as gastrin or 5-hydroxytryptamine. ¹¹

Patients with GAVE often develop iron deficiency anaemia and require treatments with blood transfusions and iron supplementation. ¹² Along with the clinical problems of chronic anaemia, GAVE can affect patients’ quality of life, requiring regular visits to hospitals. In addition, despite transfusions many patients remain anaemic, which can impact on fatigue and exercise tolerance. ¹³

The ideal management algorithm for GAVE is not clear. Treatment of the underlying disease is recommended. Medical therapies, including corticosteroids, tranexamic acid, thalidomide and octreotide, have been trialled and described in short case series or case reports or small clinical trials.^14–17

Surgical treatment (antrectomy) can be definitive for GAVE by removing the affected mucosa.^18,19 Particularly in those with diseases such as liver cirrhosis, however, the mortality and morbidity associated with surgery can be high and therefore outweigh any potential benefit. ²⁰

Endoscopic therapy (ET) is therefore a mainstay of treatment in patients with GAVE and symptomatic anaemia. ET mostly uses thermal modalities, aiming to achieve a widespread eradication of mucosal capillary ectasia network in the antrum, with subsequent re-epithelialisation, without causing damage deeper than the superficial submucosa. Argon plasma coagulation (APC) is a thermoablative method that uses a high-frequency current to pass through argon gas to cause thermocoagulation. It is the preferred endoscopic treatment of GAVE because it is easy to use, relatively safe and has an acceptable cost. However, it can often require multiple sessions, and discontinuation of transfusion dependence is currently achieved in only one-third of patients. ²¹ Endoscopic band ligation (EBL) has also been used to treat GAVE although, again, this can result in multiple sessions and can be painful for the patient.^22,23 Furthermore, APC is not without complications including sepsis, hyperplastic polyps (which can also bleed) and gastric outlet obstruction.^24–26 ET with a neodymium-doped yttrium aluminium garnet (Nd:YAG) laser has also been used but is less available, has deeper tissue injury which can result in perforation, and has a higher cost. ²⁷

Radiofrequency ablation (RFA) is the mainstay of endoscopic ablation treatment in diseases such as Barrett’s oesophagus, using the Barrx HALO system (Medtronic, Dublin, Ireland). ²⁸ This technique transmits high-power energy for a short period of time to ablate superficial lesions. This leads to a uniform depth of ablation with low risk of ulceration, no need for gas insufflation (e.g. as is needed with APC treatment), and there are a range of catheters which can be used to achieve appropriate tissue contact. RFA has also been used to treat ectatic rectal vessels in the context of radiation proctitis. ²⁹ RFA has been used in small numbers of patients for the treatment of GAVE, showing improvements in haemoglobin (Hb) and reductions in transfusion requirements in patients refractory to APC without complications.^30,31 Mucosal ablation can therefore likely provide a successful, safe method of treatment of GAVE associated with iron deficiency, particularly in those with disease refractory to first-line treatments. We report the results of our prospective study of these patients.

Aim

A single-centre prospective feasibility study was proposed to evaluate the efficacy and safety of RFA in patients with macroscopic GAVE with persistent anaemia refractory to first-line ETs.

Methods

Patient selection

Patients with confirmed GAVE were recruited to this study prospectively from referrals to our specialist centre. Formal written informed consent was taken from patients to join the study, and the study was approved by local and national ethics bodies. The study was approved by the London Central Research Ethics Committee. IRAS ID: 183819 REC Ref: 16/LO/0149, 3 February 2016, ISRCTN Registry number 32186. The study protocol conforms to the 1975 Declaration of Helsinki.

Inclusion criteria

Patients with the following criteria were screened for our study:

Existing diagnosis of GAVE

Persistent iron deficiency anaemia despite ongoing therapy with blood transfusions, intravenous (IV) or oral iron supplementation within the last six months

At least 18 years of age

Previous treatment for GAVE with either APC, YAG laser or EBL

Visible GAVE at time of treatment endoscopy commensurate with anaemia

Able to comply with study requirements and understand and sign the informed consent form

Exclusion criteria

Patients with the following criteria were excluded from our study:

Active malignancy (malignancy diagnosed in past five years or ongoing treatment)

Alternative cause for anaemia, e.g. coeliac disease, colonic pathology or haematological disease (all patients underwent bidirectional endoscopy prior to inclusion)

Pregnancy

Anatomical abnormalities precluding treatment with RFA

Any other reason that the investigator feels makes patient unsuitable for trial

RFA Treatment and investigation

Once informed written consent had been taken, details of transfusion/iron requirements and pretreatment Hb were recorded.

Treatment with RFA was carried out with focal RFA with a through-the-scope (TTS) catheter at 12 J/cm² to all visible areas of GAVE under conscious sedation or propofol sedation by experienced endoscopists. Prior to treatment, 2% N-acetylcysteine as a mucolytic and 1:10,000 adrenaline were applied through a dye-spray catheter. The latter acted as a topical vasoconstrictor to reduce contact bleeding with the catheter and improve visibility.

The device was apposed to the tissue to achieve good contact and three sequential applications at 12 J/cm² were used (see Figure 2). Enhanced imaging was also used in some cases to highlight areas of visible GAVE (see Figure 3). All patients received omeprazole (40 mg twice daily), ranitidine (300 mg once per day) and sucralfate liquid (1 g three times per day) post-endoscopy for six weeks as per our standard protocol for RFA in Barrett’s oesophagus.

Figure 2.

(a) Application of N-acetylcysteine and adrenaline 1:10,000. (b) Insertion of through-the-scope (TTS) catheter. (c) TTS catheter in endoscope. (d) Application of radiofrequency ablation to visible gastric antral vascular ectasia. (e) Image posttreatment. (f) Follow-up endoscopy.

Figure 3.

Gastric antral vascular ectasia (GAVE) treatment with radiofrequency ablation with enhanced imaging to identify areas of GAVE.

Following the initial RFA procedure, all patients returned for a repeat endoscopy after six weeks. In those for whom there was macroscopic GAVE present at this endoscopy, this was treated with RFA as at the initial therapy. Eight weeks following this, patients returned for a follow-up endoscopy during which images were captured of the gastric mucosa, and entered the follow-up phase during which Hb and ongoing transfusion/iron requirements were recorded at six months following the first procedure (see Figure 4).

Figure 4.

Flowchart demonstrating trial algorithm.

Outcomes

The primary outcome was a sustained increase in Hb six months after at least two sequential RFA treatments.

Secondary outcomes were a reduction in frequency of blood transfusions and/or iron (orally or IV) in the six months before and after RFA treatments. Endoscopic surface area (SA) regression of areas of GAVE at the follow-up endoscopy were analysed by asking two expert endoscopists to estimate the overall percentage change in GAVE SA from index treatment by examining single endoscopic images of the stomach before and after the treatment sessions. The endoscopists were asked to score 0%, 20%, 40%, 60%, 80% or 100% regression in a manner similar to that previously published in an assessment of regression for Barrett’s oesophagus. ³² When there was more than 30% disagreement, a consensus meeting was held to score with review of the images. Follow-up Hb and treatment success were correlated to the percentage of regression at the follow-up endoscopy.

Statistical methods

The data were entered into an anonymised database and analysis was performed using Statistical Package of the Social Sciences, 10.0.1 for Mac. The change in Hb score values before and after treatment was compared using paired t test for parametric variables. An α-error of less than or equal to 0.05 was considered to be statistically significant.

Results

Patient demographics and characteristics

Twenty-two patients were recruited into this feasibility study in our centre. Two patients were excluded at first endoscopy because of absence of macroscopic GAVE.

The patient characteristics are shown in Table 1.

Table 1.

Patient characteristics.

Age (years) median	69 (IQR 62–81)
Gender male	3 (15%)
Gender female	17 (85%)
Previously treated with APC	13 (65%)
Previously treated with YAG laser	4 (20%)
Previously treated with endoscopic band ligation	3 (15%)
Systemic sclerosis	5 (25%)
Liver cirrhosis	4 (20%)
Hypothyroidism	3 (15%)
Type 2 diabetes	2 (10%)
Chronic kidney disease	1 (5%)
No associated diseases	5 (25%)
Oral iron supplementation pretreatment	11 (55%)
Intravenous iron supplementation pretreatment	6 (30%)
Blood transfusions pre-endoscopy	14 (70%)
Iron supplementation and blood transfusions	11 (55%)

APC: argon plasma coagulation; IQR: interquartile range; YAG: yttrium aluminium garnet.

Outcomes

The primary outcome in our study was changes in Hb following treatment with RFA. The mean pretreatment Hb was 95.1 g/l (95% confidence interval (CI) 87.4–102.6 g/l). The mean posttreatment Hb at six months post-RFA treatment protocol was 112 g/l (95% CI 103.9–120.1 g/l). The mean change in Hb was +12.6 g/l ((95% CI 11.7–24.3 g/l) (p < 0.001) (paired t test)). See Figure 5.

Figure 5.

Change in haemoglobin at six months post-radiofrequency ablation (RFA) treatment.

At six months post-RFA, only three of 14 (21%) patients who had required blood transfusions prior to the procedure needed ongoing transfusions (see Figure 6). Five of 17 (29%) patients who were previously on iron had ongoing iron needs.

Figure 6.

Change in blood requirements at six months posttreatment.

The mean SA regression of GAVE when scored by two experienced upper gastrointestinal endoscopists on still endoscopic images taken before and after treatment was 74% ± 25%. No correlation was seen between higher SA regression and Hb change, and blood or iron requirements. Pearson log rank was p = not significant. The median number of treatments needed in these patients was two (interquartile range (IQR): 1–2). The mean number of ablations needed per treatment was 106 ± 50. Figure 7 shows examples of the images used for regression analysis.

Figure 7.

Images before and after treatment of gastric antral vascular ectasia with radiofrequency ablation.

Durability of anaemia reversal

At 12 months post-RFA, three of 14 (21%) patients had recurrence of anaemia with requirements for blood or IV iron. All three patients were then treated with RFA (outside of this study protocol). Patients with a lower Hb pretreatment appeared to be less likely to suffer a recurrence (p < 0.05 Mann-Whitney U test). Two patients were treated at eight months and one at 10 months post-initial RFA treatment.

Adverse events

Three of 20 patients in our study experienced pain which was managed with oral pain medications (paracetamol or codeine). There were no complications of perforation, acute bleeding or stricture formation.

Discussion

This prospective feasibility study presents further evidence for the use of RFA in GAVE with symptomatic anaemia. In our group of 20 patients, all of whom had failed other ET for GAVE, remaining anaemic with blood and iron requirements, RFA over only two sessions was shown to produce a highly significant change in Hb. There was a surprising negative correlation between pretreatment Hb and improvement in Hb which may tentatively suggest that RFA for GAVE is most effective in patients with more severe anaemia, although the study was not formally powered to address such questions. In addition, the number of patients requiring ongoing oral and IV iron supplementation was greatly reduced, as was the number of patients requiring blood transfusions. The number of treatments needed for this significant response was low compared to those reported in studies of other therapies^25–27 and our own clinical experience, meaning fewer visits to the endoscopy department for patients who often, because of other comorbidities, have multiple frequent hospital appointments. The treatment was well tolerated, with a small number of patients reporting pain which was controllable with oral analgesia. Many patients reported improved quality of life although this was not formally assessed. Improvement was seen across a broad range of clinical conditions associated with GAVE, with our group of patients involving more individuals with systemic sclerosis than previous studies. Of those patients who had reached 12-month follow-up, only 21% had recurrence of anaemia which required blood transfusion or recommencement of iron supplementation and further treatment, demonstrating durability of the therapy.

RFA has previously been studied in GAVE. An initial study by Gross et al. found similar results in a small case series of six patients. ³³ A further study by McGorisk et al. ³⁰ showed improvements in Hb and reductions in transfusions in patients who had previously failed APC treatment using the Halo 90 catheter, while Jana et al. ³⁴ demonstrated reductions in transfusion requirements and improvements in Hb in seven patients who had failed APC treatment. All these studies reported acceptable safety profiles.

Our study includes patients with multiple different treatment modalities prior to intervention with RFA, including laser and EBL in addition to APC. In addition, there were strict inclusion and exclusion criteria, especially the exclusion of other causes of anaemia through bidirectional endoscopy. In addition, we believe this to be the first to investigate changes in IV and oral iron requirements, which we feel is important given the potential impact on quality of life from continuous oral iron supplementation or visits to the hospital for IV iron infusions. SA regression has been previously used in the assessment of Barrett’s oesophagus, ³² but has not been used previously to assess therapeutic response in GAVE. Within the strict criteria of our trial, patients were permitted two treatment sessions before regression analysis. Further studies might assess whether SA regression predicts clinical outcomes. This was not the case in our study, although this may be due to insufficient numbers. Furthermore, our study used the TTS catheter while many of the previous studies used the Halo 90 catheter.

The TTS device is ideal for this treatment protocol as there is only one intubation of the oesophagus with a standard gastroscope (rather than the second intubation with a gastroscope with a Halo Ultra catheter attached externally, which expands the diameter of the device and in our experience from the treatment of Barrett’s oesophagus can be uncomfortable for the patient). In addition, the improved flexibility and field of movement of the catheter allows treatment across the non-flat parts of the stomach. Another advantage of the TTS device is that there is no need to reposition the catheter on the endoscope from the 6 o’clock to the 12 or 3 o’clock positions, again reducing the need for further intubations of the oesophagus.

The 12 J/cm²× 3 protocol allows treatment of superficial disease with a uniform depth of ablation across a large area rapidly, which is ideal for the treatment of superficial diseases such as GAVE without deeper treatment which may lead to more risk of ulceration and pain. This was chosen on the basis of dosimetry used in the treatment of Barrett’s oesophagus, and cleaning was not performed as we felt this might reduce the risk of bleeding. We plan, however, to investigate the use of higher-energy protocols, particularly in those patients in whom the disease recurs.

The most frequently used alternative to RFA is APC. Trials have demonstrated the efficacy of this but, as with RFA, most of these are small, single-centre studies or retrospective analyses. In addition, many patients required frequent treatment sessions ²³ and showed a significant level of recurrence of more than 50% at one year. ²¹ Although RFA is more expensive at the time of endoscopy, we believe that a formal cost analysis of these patients may show savings due to reduced blood and iron infusions, reduced hospital stays and reduced number of treatments. This is the first study to show the feasibility of this treatment within the United Kingdom National Health Service.

The limitations of our study include the small number of patients as can be expected with a rare disease such as GAVE. In addition, this is a single-centre study. Further work, which we are commencing, is needed to assess the cost effectiveness of RFA for GAVE, which in turn could help design further studies including randomised studies comparing it with APC or EBL across different centres, and the relative benefits of different RFA catheters and different energy protocols.