Endoscopy: Gl Tract. Ablating Barrett's mucosa

By: S.C. Sydney Chung

Corresponding address: S.C. Sydney Chung, M.D.

Prince of Wales Hospital, Department of Surgery Shatin, New Territories, Hong Kong, China Fax: +852-26 35 00 75

There has been a dramatic increase in the incidence of adenocarcinoma of the distal esophagus in the last two decades. The culprit is Barrett's esophagus—intestinal metaplasia at the lower end of the esophagus as a result of acid reflux. Barrett's esophagus is not uncommon, being seen in approximately one in 100 endo­scopic examinations and in 10-15% of patients with reflux esophagitis. Periodic endoscopic surveillance is suggested for this premalignant condition. Life­long surveillance is fraught with difficulties, and is of ques­tionable benefit in influencing survival. Surgical resection, although curative, carries appre­ciable morbidity and mortality.

Recently, interest has focused on endoscopic destruction of Barrett's mucosa with Nd: YAG or argon laser, photodynamic therapy, or multipolar electro­coagulation. If gastric acid is suppressed, the mucosal surface is often restored by squamous epithelium, with regression of the Barrett's mucosa.

Van Laethem and his col­leagues at Erasmus Hospital in Brussels used endoscopic argon beam coagulation to obliterate Barrett's mucosa. The argon beam coagulator, which uses ionized argon gas to effect monopolar electrocoagulation in a non-contact manner, is suitable for mucosal obliteration, since the thermal damage is confined to a depth of 2-3 mm.

Thirty-one patients with Barrett's esophagus of at least 3 cm in length were treated. Argon beam coagulation was performed at monthly intervals, and patients required a mean of 2.4 sessions. On endoscopic examination, complete eradi­cation was achieved in 25 of 31 patients (80.7%), although residual glands of intestinal met­aplasia were found in six of the 25 on histological examination,

giving a success rate of only 61%. Complete eradication was more easily achieved in short Barrett segments and in non-circumferential lesions. Three complications occurred: one case of severe esophagitis and two esophageal stenoses. During the course of treatment, the patients received 40 mg of omeprazole daily and received either 40 mg or 10 mg of omep­razole on follow-up. Seventeen patients completed a one-year follow-up, and only nine were clear of metaplastic glands. Five of these patients took 10 mg omeprazole and four took 40 mg omeprazole daily.

Overholt and colleagues in Nashville, Tennessee used photodynamic therapy (PDT). Forty-eight hours after injection of the photosensitizer porfimer sodium (Photofrin), red light at 630 nm from an argon-pumped dye laser was shone into the area using a diffuser inserted through a special transparent centering balloon. All patients were treat­ed with omeprazole 20 mg twice a day for the first three months, and then once a day. Patients were followed up at three-monthly intervals with endosco­py, using Lugol's iodine staining. Residual areas of Barrett's esophagus was destroyed using Nd:YAG laser treatment.

One hundred patients, including 13 with superficial cancer, were treated. The surface area of the Barrett's esophagus was reduced in all cases. An estimated 75-80% of Barrett's mucosa converted to squamous epithelium, with complete elim­ination of the Barrett's mucosa

in 43 patients. Dysplasia was eliminated in 78 patients. Ten of the 13 malignancies were ablated. Adverse effects included chest pain, dysphagia, atrial fib­rillation, and pleural effusions. Esophageal strictures developed in 34 patients. During follow-up, two patients were found to have high-grade dysplasia under the cover of the new squamous epithelium 18 and 22 months after PDT, and another patient developed a subsquamous aden­ocarcinoma six months after PDT.

If the specialized columnar epithelium of Barrett's esopha­gus is obliterated and the area is reepithelialized with squamous mucosa, the risk of adenocarcin­oma is theoretically removed. Endoscopic methods of local destruction are opening up exciting new opportunities for the therapeutic endoscopist. An important question that remains unanswered is the depth of injury necessary to obliterate all viable columnar cells, as persistent intestinal metaplasia underneath the neosquamous epithelium may progress to cancer. How best to nurture the new squamous epithelium is another problem. Follow-up data are preliminary, but suggest that maintenance treatment with proton-pump inhibitors may not be sufficient. Perhaps laparos­copic Nissen fundoplication after squamous reepithelialization is a better strategy.