The two most common methods of performing EMR are cap-assisted EMR and ligation-assisted EMR. Cap-assisted EMR involves submucosal injection, suction of the lesion into a cap, and then snare electrocautery. The lesion is initially lifted with a submucosal injection. The submucosal injection can be performed with saline; however, other agents can also be utilized (use of hyaluronic acid; saline with the addition of epinephrine or dye such as methylene blue). After a submucosal injection with lifting has been performed, the lesion is suctioned into a clear plastic cap affixed to the end of the endoscope and then a snare is opened and positioned within the internal ridge of the cap (various snare shapes and sizes are available). The snare is then opened and the lesion is suctioned into the cap, allowing the snare to be closed around it. Electrocautery is then utilized to remove the lesion. Cap-assisted EMR mucosectomy devices with various different cap sizes (outer diameter ranging from 12.9 to 18 mm), shape (flat circular- or oval-shaped tip), and firmness (soft or hard) are available for this technique. (Olympus America, Center Valley, Pennsylvania)

Ligation-assisted EMR is another technique utilized to perform EMR. There are several single-use band ligation devices that are available, including the Duette Multi-Band Mucosectomy device (Cook Medical Inc., Winston-Salem, North Carolina) and the Captivator EMR device (Boston Scientific, Natick, Massachusetts). Both of these devices involve attaching the ligation device to the end of the upper endoscope (very similar in structure and function to standard banding device as would be used to treat esophageal varices). The lesion is then suctioned into the banding cap (typically without prior submucosal injection) and then a band is deployed around the lesion circumferentially. The result of this process is the creation of a pseudopolyp. The included snare can then be advanced though the working channel of the endoscope through the attached device (without having to remove the device), the snare placed around the pseudopolyp (either above or below the band, whichever is technically easiest in a given situation), and then the electrocautery can be applied to remove the lesion. If necessary, for larger lesions or additional lesions, multiple bands can be utilized and the lesion can be removed in a piecemeal fashion (Fig. 5.3).

Possible complications from EMR include bleeding, perforation, and esophageal stricture formation (which are often delayed in presentation). Rates of bleeding after EMR in the literature vary widely, partially dependent on how bleeding is defined by the individual study and how aggressive the EMR procedure under evaluation is. Bleeding after esophageal EMR was evaluated in a large single-center study including 681 patients who underwent 2513 EMR procedures [11]. Clinically significant bleeding, defined in this study as any bleeding requiring endoscopic intervention, blood transfusion, or hospitalization, was only reported in 1.2% of patients.

Perforation rates after esophageal EMR are overall low with rates <0.5% for endoscopists experienced in performing EMR. The perforation risk increases when piecemeal resection is required [12–14, 23].

Stricture formation has been reported to occur in as few as 6% of patients and in as many as 88% of patients undergoing esophageal EMR for Barrett’s esophagus with HGD or intramucosal carcinoma in various studies [15–19]. The higher rates of stenosis are associated with patients who have undergone EMR with more extensive resection. A study of 73 patients undergoing EMR (for Barrett’s esophagus with HGD or intramucosal carcinoma) found symptomatic strictures in 25% of patients, with strictures more common if the resection area involved more than 50% of the esophageal lumen (odds ratio 4.2, 95% CI 1.3–14) [20].

The strictures caused by EMR are typically able to be effectively managed with endoscopic dilation. In a study of 136 patients undergoing esophageal EMR, a total of 37 patients (27%) developed an esophageal stricture [21]. Of note, 65% of the patients who developed a stricture also had a history of RFA treatment, so the cause of the stricture was likely multifactorial. In the group of patients that did not develop stricture, 56% had history of RFA treatment, suggesting that even EMR combined with RFA does not always lead to stricture formation. The authors note that all of the patients who developed stricture had resolution of dysphagia with endoscopic dilation. A median number of 2 dilations were needed per patient. Another study examining esophageal stricture post-EMR demonstrated similar findings with an average of 2.3 dilations required per patient [22].

ESD is a technique that utilizes submucosal injection and then needle-knife for en bloc removal of larger (and possibly deeper) lesions. Many different types of needle-knife catheters are available for performing ESD. Overall complication rates, including perforation, are higher with ESD than with EMR. Bleeding during an ESD procedure is common and is typically able to be treated intra-procedurally with coagulation. Delayed bleeding is less common with esophageal ESD than with gastric ESD, in which rates up to 15.6% have been reported [23]. In a series of patients treated with esophageal ESD, delayed bleeding rates were reported in between 0 and 5.2% in the seven studies (with 568 cases) that provided this information [24].

Review of data from multiple series of esophageal EMR demonstrates a pooled perforation rate of 2.3% (19 of 816 cases), recognizing that most of these cases were performed by experts [25]. Almost all of these perforations were recognized during the procedure and were treated with placement of endoscopic clips. Strictures develop in approximately 12–17% of patients after esophageal ESD [26–29]. As with EMR, the stricture rate increases when more extensive and circumferential lesions are resected.

Since ESD is a technically difficult procedure with higher rates of adverse events than EMR, the utilization of ESD in the USA is limited to specialized centers with endoscopic expertise at performing this technique.

Radiofrequency ablation (RFA) is an endoscopic ablative therapy that delivers energy via a balloon (or catheter) with a series of closely spaced electrodes that generate a thermal injury with controlled depth and uniformity. Circumferential ablation and focal ablation are the two primary methods of performing RFA. Circumferential ablation (with an electrode-laden balloon) is typically performed in settings of more extensive areas to treat (such as long-segment Barrett’s esophagus), while focal ablation (with an ablation catheter placed on the tip of the endoscope) is used to treat smaller areas. A smaller through-the-scope probe is also available for very small areas of Barrett’s esophagus (Video 5.1).

Prior to performing ablation, the esophageal wall should first be irrigated with water to remove any mucus or other debris. Cleansing of the esophagus has traditionally performed using acetylcysteine; however, it has been demonstrated that water is just as effective at cleaning the esophagus [30]. The next step is careful identification of the esophageal-gastric landmarks, including the top of the gastric folds and the proximal extent of the Barrett’s esophagus.

Prior to performing circumferential ablation, as the endoscope is positioned in the stomach, a stiff guidewire is placed through the working channel of the endoscope, and the endoscope is withdrawn as the wire is kept in place. The Barrx^TM 360 soft sizing balloon is then advanced over the wire and connected to the Barrx FLEX generator (Medtronic, Minneapolis, Minnesota). This sizing balloon is utilized to measure the inner diameter of the esophagus prior to performing ablation. Based on the measurements from the sizing balloon, an appropriate ablation balloon catheter is selected. The BARRX^TM 360 RFA balloon catheters (Medtronic, Minneapolis, Minnesota) are all 3 cm in length and are available in size diameters ranging from 18 to 31 mm.

The RFA balloon catheter is advanced over the wire and then the endoscope can be advanced adjacent to the wire and positioned proximal to the ablation balloon. With direct endoscopic visualization, the proximal edge of the balloon is positioned approximately 1 cm above the proximal extent of the Barrett’s esophagus. The balloon is then inflated, and then radiofrequency energy (typically 12 J/cm²) is activated by depressing a foot pedal attached to the generator. After the energy has been delivered, the balloon is repositioned more distally (allowing approximately 5–10 mm of overlap with the prior ablation area) and the same process repeated until the entire segment of Barrett’s esophagus has been treated.

After the entire segment has been treated, the balloon catheter, wire, and endoscope are removed from the patient. A soft cap is attached to the end of the endoscope and the esophagus is then cleansed by removal of the coagulum with the soft cap combined with irrigation of the esophagus with water. After this is complete, the entire process is repeated (placement of wire, insertion of balloon catheter, and then ablation using the same settings as previously performed) as needed to treat the entire area of Barrett’s esophagus.

A variety of different RFA catheters is commercially available and can be utilized to ablate smaller segments of Barrett’s esophagus when non-circumferential disease is encountered. Several of the catheters (Barrx60, Barrx90, Barrx Ultra Long) can be attached to the end of the endoscope and one of the catheters (Barrx Channel) is a through-the-scope device for treatment of focal areas of Barrett’s esophagus. When utilizing the attachments made to be affixed to the endoscope tip, the device is positioned at 12 o’clock on the endoscopic image. The endoscope and ablation catheter are advanced into the esophagus under direct visualization for use. The through-the-scope RFA ablation catheter is rotatable and usable under direct endoscopic visualization as well.

Once the endoscope has been advanced to the target tissue, ablation is performed by using the wheels of the endoscope to bring the ablation catheter into close contact with the mucosa in the desired treatment area. RFA energy (typically 15 J/cm²) is then delivered by depressing a foot pedal attached to the generator. Prior to moving the electrode away from the mucosa, a second delivery of energy (at the same setting) is applied. All of the remaining areas of Barrett’s esophagus are then treated in a similar fashion. As with circumferential ablation, the coagulum should then be cleansed from the esophageal wall after each treatment. This can be performed by using the tip of the electrode catheter to scrape off the coagulum. The endoscope should then be completely removed from the patient and the catheter cleansed with water. The endoscope and catheter are then reinserted and another treatment is performed in the exact same manner as previously (another two pulses of ablation at each treatment station) (Fig. 5.4).

Post-RFA treatment care typically includes high-dose PPI treatment. All patients with Barrett’s esophagus should already be taking a PPI agent; however, increased acid suppression therapy may help improve esophageal healing after an ablation session. A prospective study demonstrated that effective esophageal pH control (24-h pH monitoring was utilized) was associated with improved outcomes, including reduction in Barrett’s esophagus surface area and complete eradication rate, after RFA treatment [31].

As patients may experience chest pain and/or dysphagia immediately after treatment, alteration in the diet for several days after treatment is generally recommended. Dietary recommendations after RFA typically include liquids only for the first day after the procedure, a soft-consistency diet on the second day, and slow advancement as tolerated after that time. Other medications that can be considered include sucralfate suspension and pain medications if needed.

RFA treatment is generally well tolerated. There are a multitude of studies describing complication rates after RFA for Barrett’s esophagus. Overall stricture rates from RFA range between 0 and 6%, depending on the study. A multi-centered community-based study including 429 patients treated with RFA for Barrett’s esophagus demonstrated a stricture rate of 1.1% of cases (2.1% of patients), with no serious adverse events (including no bleeding or perforation) [32]. In this study, the strictures resolved with a median of three endoscopic dilations. A large meta-analysis of 18 studies demonstrated that the most frequent complications from RFA include esophageal stricture (5%), chest pain (3%), and bleeding (1%) [33].

Cryotherapy is a technique that has been utilized in many different fields in medicine; however, this technology has only recently been adapted for use in endoscopy in general and Barrett’s esophagus specifically. At this time, it is most commonly used for patients with refractory Barrett’s esophagus who have failed or developed complications from RFA treatment (such as chest pain or stricture), or who are not candidates for RFA, or in patients who do not want to undergo RFA. Cryotherapy can also be utilized as a primary therapy for Barrett’s esophagus treatment and can be used to treat esophageal cancer locally in nonsurgical candidates.

The two currently commercially available cryogens are liquid nitrogen and carbon dioxide. The destruction of the Barrett’s epithelia is caused by freeze-thaw cycles using either of the cryogens. The available endoscopic systems for cryotherapy treatment include the CryoSpray Ablation system (CSA Medical, Baltimore, Maryland), Polar Wand cryotherapy (GI Supply, Camp Hill, Pennsylvania), and the Coldplay Focal Cryoballoon Ablation System (C2 Therapeutics, Redwood City, California).