Endoscopic ablation of Barrett’s esophagus using high power setting argon plasma coagulation: A prospective study

Abstract

AIM: This prospective study evaluated the effectiveness of 90 W argon plasma coagulation (APC) for the ablation of Barrett’s esophagus (BE) that is considered to be the main risk factor for the development of esophageal adenocarcinoma.

METHODS: The results from 25 patients, observed at the First Department of General Surgery, University of Verona, Italy, from October 2000 to October 2003, who underwent APC for histologically proven BE were prospectively analyzed.

RESULTS: The ablation treatment was completed in all the patients but one (96%). The mean number of APC sessions needed to complete ablation was 1.6 (total number: 40). The eradication was obtained in the majority of cases by one session only (60%), two sessions were required in 24% of the cases and three or more in 16%. About 43% of the sessions were complicated. Retrosternal pain (22.5%) and fever (17.5%) were the most frequent symptoms. Only one major complication occurred, it was an hemorrhage due to ulcer formation on the treated esophagus that required urgent endoscopic sclerosis and admission. The follow-up was accomplished in all the patients with a mean period of 26.3 mo and 20 patients (84%) with a follow-up period longer than 24 mo. Only one patient showed a relapse of metaplastic mucosa 12 mo after the completion of ablation. The patient was hence re-treated and now is free from recurrence 33 mo later.

CONCLUSION: High power setting (90 W) APC showed to be safe and effective. The effects persist at a mean follow-up period of two years with a comparable cost in term of complications with respect to standard power settings. Further studies with greater number of patients are required to confirm these results and to assess if ablation reduces the incidence of malignant progression.

Key Words: Barrett’s esophagus, Argon plasma coagulation, Endoscopic treatment

INTRODUCTION

While the incidence of the adenocarcinoma of the distal esophagus and gastric cardia is rising rapidly[1], long-term prognosis of surgically treated patients remains poor[2,3]. The principal risk factor for the development of esophageal adenocarcinoma is Barrett’s esophagus (BE). Although the exact estimation of this hazard has not been assessed yet, it is estimated that BE leads to an increased risk of 30-125 times compared to the general population[4,5]. Many questions remain regarding the best management of BE, since routine endoscopic follow-up is not definitely accepted[6,7] and medical therapies as well as anti-reflux surgery have not proven to induce the regression of BE[8,9].

Argon plasma coagulation (APC) is one of the several endoscopic approaches that have been proposed in order to reverse BE and to induce squamous re-epithelialization. This technique allows to ablate large areas of specialized intestinal metaplasia with a limited depth of injury that minimizes the risk of stricture and perforation. Conversely, the chance of having isolated microfoci of metaplastic mucosa beneath the new squamous epithelium points out the importance of eliminating the totality of BE[10]. Although APC has been reported in many series with different power settings (30-60 W), few studies reported high level of energy (70-90 W)[11,12].

The aim of this prospective study was to evaluate the effectiveness of 90 W APC for the ablation of BE.

MATERIALS AND METHODS

Study population

The study includes 25 patients observed at the First Department of General Surgery, University of Verona, Italy, between October 2000 and October 2003 who underwent APC for histologically proven BE.

Definition and inclusion criteria

BE was diagnosed if specialized intestinal metaplasia was observed in biopsy specimens obtained from the tubular esophagus above the gastro-esophageal junction. Long-segment BE was defined as the occurrence of intestinal metaplasia extending more than 3 cm in length from the gastro-esophageal junction. Short-segment BE was considered when the columnar lining was less than 3 cm[13].

The most important inclusion criteria were: histologically proven BE, age ranging from 18 to 75 years old, possibility of follow-up with a life expectancy longer than 5 years, informed consent, absence of previous malignancies or intercurrent diseases precluding the prognosis under study. Patients with low grade dysplasia were included while no patient underwent APC for high grade dysplasia.

Endoscopy and biopsy protocol

A baseline endoscopic examination was performed in all the cases with a Pentax EG endoscope (series 2901-2940) to allow precise measurement of the BE segment length and to classify it according to the presence of tongues (type I), islands (type II) or circumferential (type III) metaplastic mucosa.

Gastro-esophageal junction was defined by the end of the tubular esophagus or, in the presence of hiatal hernia, by the proximal extent of the gastric rugal folds. The length of BE was routinely evaluated considering the distance from the incisors on the basis of markings on the endoscope shaft while the endoscope was being withdrawn. Four 7-mm biopsy specimens were taken every 2 cm of BE segment. Further four biopsy specimens were taken just over the proximal margin of BE and four just below the GEJ. Additionally, specific biopsies were taken on any mucosal abnormality[14].

Ablation technique

APC was carried out as a day-case procedure under intravenous sedation with midazolam in all the cases. Heart rate, blood pressure and pulse oximetry were routinely monitored. Ablation was performed using the ERBE ‘Argon Beamer 2’ device (ERBE Electromedizin, Tübingen, Germany) with a power setting of 90 W and a gas flow of 2 L/min. After esophageal intubation the APC probe was moved forward for 1 cm from the endoscope and positioned at 1-2 mm from the mucosa. A longitudinal strip was ablated starting from the gastro-esophageal junction withdrawing the endoscope till the end of BE. The procedure was then repeated on the adjacent mucosa. When the length of BE was longer than 4 cm, its ablation was carried out in two or more sessions. Any remaining islets of metaplastic mucosa were treated in a final procedure.

No post-operative blood or instrumental examinations were routinely performed. A tepid semiliquid diet was advised on the day of intervention, while no particular dietetic restrictions (except usual GERD rules) were prescribed for the following days. All patients were established on high dose PPI regimens (pantoprazole or esomeprazole 40 mg×2) starting one week before the planned treatment and continuing till the confirmation of the complete ablation of the BE (usually one month). Thereafter prolonged medical therapy with standard PPI doses (pantoprazole or esomeprazole 20 mg) was recommended in any case except one who had undergone previous anti-reflux surgery.

Follow-up

After the completion of APC treatment and the complete BE eradication, confirmed histologically at one month, the patients were followed up at 6 mo, at 1 and 2 years and thereafter at two-year intervals. The follow-up protocol requires endoscopic examination as previously described and multiple 7-mm biopsy specimens taken according to the four quadrant technique considering the previously assessed length of metaplastic mucosa.

RESULTS

Patients’ clinical characteristics are shown in Table 1. Twenty-five patients underwent APC, 11 had a long-segment BE while in 14 the length of metaplastic mucosa was inferior to 3 cm. The mean BE length was 34 mm (median 25 mm, range 10-130 mm). Among the 25 patients only 3 cases had a previous histological diagnosis of low grade dysplasia.

Table 1 Clinical characteristics of the 25 patients¹.

	APC treatment
Age (yr)2	61.7 (34–74)
Male:Female	18:07
Presence of GERD symptoms	15 (60)
Presence of hiatal hernia	24 (96)
Length of BE (mm)2	34 (10–130)
Short-segment BE	14 (56)
Long-segment BE	11 (44)
Type of BE
Type I	13 (52)
Type II	4 (16)
Type III	8 (32)
Presence of low grade dysplasia	3 (12)

¹Number represented in parenthesis are percentages.

²Mean (range).

Data on APC treatment are available in Table 2. The ablation treatment was completed in all the patients but one (96%). This was the longest BE, measuring 130 mm, and its treatment was stopped after the fourth APC session without achieving the complete eradication because of voluntary withdrawal of the patient. The total number of treatment sessions was 40 with a mean number of 1.6. If the case just described were excluded, none of the patients required more than three APC sessions. Three sessions were necessary in 3 patients, two sessions in 6, while complete eradication of BE was obtained by a single session in 15 cases (60%).

Table 2 Results of ablation treatment in the 25 patients¹.

	APC treatment
Treatment completed	24 (96)
No. of APC sessions2	1.6 (40)
Single	15 (60)
Two	6 (24)
Three or more	4 (16)
Treatment-related morbidity3	17 (42.5)
Retrosternal pain	9 (22.5)
Fever	7 (17.5)
Dysphagia	2 (5)
Ulcer formation4	2 (5)
Eradication completed	23 (92)

¹Number represented in parenthesis are percentages.

²Mean (total).

³Morbidity was calculated on the total number of treatment sessions.

⁴In one case, ulcer formation was a cause of severe hemorrhage that required admission and endoscopic sclerosis.

The post-operative course was uneventful in 11 out of the 25 patients (44%) and in 23 out of 40 treatment sessions (57.5%). The most frequent side effects were retrosternal pain (22.5%) and fever (17.5%). Retrosternal pain was well controlled by medical therapy in all the cases, while fever never exceeded 38 °C. Two cases suffered from temporary dysphagia that resolved without any adjunctive treatment. Ulcer formation was observed in two cases, one of which needed urgent endoscopic sclerosis and re-admission for severe hemorrhage 7 d after the second APC session. The subsequent recovery was uneventful and the patient was discharged after three days without any blood transfusion.

None of the patients was lost to follow-up with an overall mean period of 26.3 mo (median 23 mo, range 9-45 mo). Twenty-one patients (84%) had a follow-up period longer than 12 mo. The patient who underwent incomplete ablation of his 130-mm BE was excluded from the subsequent analysis. The total number of endoscopic follow-up examinations for the remaining 24 patients was 82 with a mean number of 3.4 per patient (median 3 per patient, range 2-6 per patient).

Only one case out of the 24 patients (4.2%) with a complete regression of metaplastic mucosa showed a relapse of BE. This was the first treated patient who recurred one year after the completion of the treatment of a long-segment BE (35 mm). A further APC session was hence performed and, at the moment, there is no evidence of intestinal metaplasia after 33 mo.

DISCUSSION

Epidemiological studies clearly indicate that there has been a rising prevalence of adenocarcinoma of the distal esophagus and gastro-esophageal junction not only in United States and Western Europe[1,15] but also in some Oriental regions[16]. BE is considered the strongest risk factor for esophageal adenocarcinoma, but an association with cardia cancer is also reported[17]. Long as well as short-segment BE seem to be related to this risk and are hence managed similarly[17]. The exact hazard assessment has not been calculated yet and the reported increase of 30-125 times[4,5] is perhaps overestimated[18].

Many questions remain regarding the best management of BE, since the efficacy of routine endoscopic follow-up has not been clearly demonstrated[6,7] and medical therapies as well as anti-reflux surgery have not proven to induce regression of BE[8,9].

In the attempt to reduce the risk for esophageal adenocarcinoma several endoscopic ablative treatments such as Nd-YAG laser, multipolar electrocoagulation, photodynamic therapy and APC[19] have been proposed in order to induce squamous re-epithelialization of BE. One of the most popular treatments is APC, a technique that involves the application of a high frequency current carried to the tissue through a flow of ionized argon gas that allows to ablate metaplastic mucosa with a limited depth of injury[20]. It has been estimated that the depth of injury necessary to completely eradicate the metaplastic mucosa is 0.6 mm[21] and this is obtained by APC. Many studies analyzed APC with power settings of 30-60 W, but few studies reported higher level of energy (70-90 W)[12,13]. The need of an increase in the power setting of APC is due to the concern of residual nests of metaplastic mucosa beneath the neo-squamous epithelium[10] that maintains the possibility to progress to cancer[22].

Vargo[23] reviewed 10 case series totaling 304 cases with a power setting of 30-90 W and reported a complete macroscopic eradication of BE in 82.6% of the cases. But the described rate of microscopic foci of BE underlying the new squamous epithelium overwhelmed 50%. In our experience the treatment was completed in 24 out of the 25 patients (96%). The treatment of the longest segment BE (130 mm) was dropped out for voluntary withdrawal of the patient after four sessions. In this case, less than half of the metaplastic mucosa was ablated. Among the remaining 24 patients, APC led to a complete regression of metaplasia in all the cases. At a mean follow-up period of 26.3 mo with 84% of patients with more than one year of follow-up, only one patient showed a histologically proven relapse of BE (4.2%) one year after the completion of the ablation therapy. A further APC session was hence performed and, at the moment (33 mo later), there is no evidence of intestinal metaplasia. Our data compare favorably with literature results, accomplished with usual power settings[23] and are consistent with the 98.6% reported by Pereira-Lima who used 65-70 W[11] and with the 96.9% reported by Schulz who used 90 W[12].

Theoretically, the use of a powered wattage leads to an increased risk of complications such as stenosis and perforations[11,12,19]. In our experience, 42.5% of the treatment sessions were complicated (66% of the patients). These percentages are similar to literature results with a lower number of major complications[23]. As already reported the most frequent complications were retrosternal pain (22.5%) and fever (17.5%). Transient dysphagia as well as ulcer formation was observed in a minority of patients (5%). Notably, one of the two patients who developed an ulcer on the segment of ablated esophagus suffered from hemorrhage that required urgent endoscopic sclerosis and re-admission one week after the treatment. No blood transfusion was required and the patient was discharged 3 d later. No perforation or stenosis was observed in our group of patients. Perforation is rare with an incidence of 0.7%. Esophageal strictures are more frequent and they are described in the range of 0-9.5%[23] with a rate of 4.3% and 9.1% reported after high power setting APC by Schulz et al[12], and Pereira-Lima et al[11].

In conclusion, APC with high power setting (90 W) showed to be safe and effective. The effects persist at a mean follow-up period of two years with a comparable cost in term of complications with respect to standard power settings. Long-term follow-up with a greater number of patients is required to assess if ablation of BE can reduce the incidence of malignant progression and if high power setting has a role in it.