Endoscopic treatment of gastrointestinal (GI) bleeding is considered the first line of therapy. Although standard techniques, such as epinephrine injection, through-the-scope hemoclips, bipolar coagulation, argon plasma coagulation, and band ligation are routinely used, some GI bleeds are refractory to these therapies. Newer technologies have emerged to assist with the treatment of GI bleeding. This article highlights endoscopic and endoscopic ultrasound-guided therapies that may be used by experienced endoscopists for the primary control of GI bleeding or for cases refractory to standard hemostatic techniques.
Key points
- •
Emerging approaches for endoscopic hemostasis include over-the-scope clips, endoscopic suturing, mucosal ablation devices, fibrin glue injection, hemostatic spray, and endoscopic ultrasound-guided angiotherapy.
- •
These novel techniques may be applied as initial treatment or as rescue therapy for refractory bleeding.
- •
Given the experimental nature of some of these new devices for hemostasis, adequate informed consent is essential.
- •
The successful application of these technologies depends on proper lesion selection and operator experience in the use of these devices.
Introduction
Gastrointestinal (GI) bleeding is a common problem encountered by all gastroenterologists. Established endoscopic techniques to assist in the treatment of GI bleeding include epinephrine injection, through-the-scope clips, monopolar or bipolar coagulation, and band ligation. Because refractory GI bleeding may occur despite these therapies, new technologies are emerging to assist in the treatment algorithm. These include endoscopic methods (ie, over-the-scope clips [OTSC; Ovesco Endoscopy AG, Tubingen, Germany], endoscopic suturing, hemostatic sprays, mucosal ablation devices, stent placement, fibrin glue injection) and endoscopic ultrasound (EUS)-guided angiotherapy (ie, coil and/or glue injection). This article highlights the technique and clinical application of these new technologies. Hemostatic sprays, stent placement for hemostasis, and injection therapy are discussed elsewhere in this issue.
Introduction
Gastrointestinal (GI) bleeding is a common problem encountered by all gastroenterologists. Established endoscopic techniques to assist in the treatment of GI bleeding include epinephrine injection, through-the-scope clips, monopolar or bipolar coagulation, and band ligation. Because refractory GI bleeding may occur despite these therapies, new technologies are emerging to assist in the treatment algorithm. These include endoscopic methods (ie, over-the-scope clips [OTSC; Ovesco Endoscopy AG, Tubingen, Germany], endoscopic suturing, hemostatic sprays, mucosal ablation devices, stent placement, fibrin glue injection) and endoscopic ultrasound (EUS)-guided angiotherapy (ie, coil and/or glue injection). This article highlights the technique and clinical application of these new technologies. Hemostatic sprays, stent placement for hemostasis, and injection therapy are discussed elsewhere in this issue.
Emerging endoscopic therapies for gastrointestinal bleeding
Endoscopic Over-the-Scope Clip
Technique
An OTSC has been approved in the United States since 2011 for endoscopic therapy of GI defects. The caps are available in three diameters (11, 12, and 14 mm) and two working depths (3 and 6 mm), whereas the clip itself has three types of teeth (atraumatic or blunt-toothed, traumatic or sharp-toothed, and gastrostomy closure). Typically the atraumatic or traumatic clip is used for hemostasis ( Fig. 1 ). The setup and deployment of the OTSC system is similar to a band ligator. The OTSC cap is affixed at the tip of the endoscope, with a string wire that runs through the scope channel connected to the deployment system that sits at the entrance port of the suction channel. Once the targeted lesion is identified, suction is applied to bring the entire lesion into the cap, followed by clip release by rotating the hand wheel of the deployment system. For fibrotic or indurated lesions, such as chronic ulcers, a dedicated tripronged anchoring device can be used to help retract the targeted lesion into the cap.
A Padlock clip (Aponos Medical Corp, Kingston, NH) is another OTSC. With this clip, the wire runs alongside the shaft of the endoscope, leaving the suction channel free to allow for better suction capability and passage of other accessories. Deployment is achieved by squeezing a handheld device. The clip has six circumferential prongs that provide radial compression on all sides and has been shown to be effective for closure of defects made in porcine stomachs and colons. However, there are no clinical publications to date regarding its application for GI hemostasis.
Clinical applications
A randomized trial compared the OTSC with two standard hemoclips (Resolution Clip, Boston Scientific, Natick, MA; and QuickClip2, Olympus, Tokyo, Japan) on spurting vessels created at several different locations in an ex vivo porcine stomach. All 45 sites (15 for each clip) were successfully treated with the assigned clip. The OTSC required significantly less time and number of clips to achieve hemostasis compared with the other clips. In the fundus, the OTSC was also thought to be more effective in changing the pressure measured within the vessel before and after clip placement. Similarly, a study measuring the pressure difference among clips found a significant increase in mean pressure favoring OTSC and resulting in a significant decrease in mean vessel diameter compared with the Resolution clip, QuickClip, and TriClip (Cook Medical, Limerick, Ireland). The ability of the OTSC to grasp more tissue and provide greater compressive force likely translated to the observed study findings.
Chan and colleagues recently described a case series on the use of the OTSC in patients with refractory (N = 6) or major bleeding caused by gastric ulcers (N = 2), duodenal ulcers (N = 5), gastric GI stromal tumor (GIST; N = 1), and ulcerative pancreatic adenocarcinoma (N = 1). The ulcers all had visible vessels and were a median of 2.5 cm (range, 1–4) in size. The technical success of achieving hemostasis during the procedure was 100%, whereas the efficacy of preventing rebleeding was achieved in seven patients (78%). The two patients who rebled had duodenal ulcers.
In a larger multicenter case series, 30 patients with refractory GI bleeding who failed conventional endoscopic therapies underwent OTSC placement. Bleeding was attributed to duodenal ulcers (N = 12), gastric ulcers (N = 6), colonic endoscopic mucosal resection (N = 5), Mallory-Weiss tear (N = 2), Dieulafoy lesions (N = 2), surgical anastomosis (N = 1), colonic endoscopic submucosal dissection (ESD; N = 1), and colonic diverticulum (N = 1). Primary hemostasis immediately following OTSC placement occurred in 29 patients (97%); one patient with a posterior duodenal bulb ulcer required interventional radiology (IR) embolization for failed hemostasis. Two patients (one gastric and one duodenal bulb ulcer) rebled at 12 and 24 hours after the procedure, and they were successfully retreated with epinephrine injection to the surrounding mucosa.
The disadvantages of the OTSC are that it requires the scope to be withdrawn to load the device, the OTSC cap may make traversing the cricopharyngeus or luminal stenoses difficult, the challenge in accessing lesions in the posterior-inferior duodenal wall or the proximal lesser curvature of the stomach, and its cost. Prospective randomized trials comparing the OTSC with standard through-the-scope clips are awaited to determine its clinical role in the treatment algorithm of GI bleeding.
Endoscopic Suturing
Technique
Only one endoscopic suturing device (OverStitch; Apollo Endosurgery, Austin, TX) is currently available for clinical use. This device is mounted on a double-channel endoscope (GIF-2T160 or GIF-2T180; Olympus Corporation) and consists of a suture anchor with a detachable needle tip carrying an absorbable (2-0 or 3-0 polydioxanone) or nonabsorbable (2-0 or 3-0 polypropylene) suture that is passed through one accessory channel and coupled to the curved suturing arm of the device. The device is attached by a wire that runs alongside the scope shaft to the handle portion of the system, which is affixed to the entrance port of the working channel. Squeezing the handle component activates transfer of the needle tip, movement of the suture arm, and enables passage and exit of the suture through tissue. A dedicated corkscrew retracting device (Helix; Apollo Endosurgery) or grasping forceps can be advanced through the other working channel to facilitate tissue access to the needle. A dedicated suture-cinching tool is used to tighten, secure, and cut the suture, which can be placed in an interrupted or running fashion.
Clinical applications
Currently the use of the OverStitch device in the setting of GI bleeding is limited because of the lack of widespread availability of the device and accessories, technical complexity of the suturing system requiring specific training, impaired visibility during active hemorrhage, and restricted maneuverability of the device to access all areas of the GI tract. A bench study using porcine stomachs with submucosal splenic arteries connected to a pulsatile pump containing red ink evaluated the use of a prior generation of the endoscopic suturing device (Eagle Claw II; Apollo Endosurgery and Olympus Medical Systems Corp). Of the 25 total sutures placed, 17 (68%) were successful in achieving hemostasis. The reasons behind the remaining eight that failed to treat the bleeding artery included penetration through the vessel wall (N = 4), knots that were too loose to provide hemostasis (N = 2), incorrect positioning of the suture (N = 1), and failure to penetrate an edematous gastric wall (N = 1).
One potential application of endoscopic suturing for GI bleeding may be in the setting of marginal ulcerations at an anastomosis. Three patients with chronic marginal ulcers (two presented with recurrent transfusion-dependent hemorrhage) were treated with endoscopic suturing to close the ulcer bed, followed by fibrin glue (Fibrin Sealant; Baxter, Deerfield, IL) application to the sutured area. Technical success with complete ulcer closure was achieved in all patients by placing one to three 2-0 nonabsorbable stitches in an interrupted fashion. During the second case, bleeding occurred during the initial suture placement, which was successfully treated with epinephrine injection. Thereafter, the authors prophylactically injected epinephrine to the area before suturing. In the two patients with recurrent GI bleeding, repeat endoscopy at 6 weeks revealed complete ulcer resolution and they remained symptom free at 6 weeks and 1 year after the procedure.
Another possible use of endoscopic suturing is to prevent GI bleeding after ESD ( Fig. 2 ). At one center, 12 patients who underwent ESD (four gastric and eight colonic lesions) had their post-ESD defects completely closed with the OverStitch device. The mean size of the lesion was 42.5 mm (standard deviation [SD], 14.8) and a mean of 1.6 (SD, 1) suture was required for closure. There were no immediate or delayed adverse events, including bleeding. At surveillance endoscopy 3 months later, there was complete healing of the ESD sites.
Radiofrequency Ablation
Technique
Focal radiofrequency ablation (RFA) catheters (Barrx; Covidien, Mansfield, MA) have been used to treat GI bleeding in the setting of gastric antral vascular ectasia (GAVE) ( Fig. 3 ) and radiation proctitis. The focal RFA catheters have a tilting platform that contains an array of electrodes that either is mounted at the tip or passed through the working channel of the endoscope. The active electrodes range from 15 to 40 mm in length and 7.5 to 13 mm in width. The catheters are placed in direct contact with the target tissue and the penetration of thermal energy is superficial but uniform. The typical treatment protocol is two to four applications per site using an energy density of 12 J/cm 2 and power density of 40 W/cm 2 . After one site is treated, the catheter is repositioned so the probe lies at the next targeted site and the process is continued until all areas have been treated.
Clinical application
Four studies have evaluated the use of RFA in the treatment of refractory GAVE and are summarized in Table 1 . In the lower GI tract, RFA has been used in the treatment algorithm for radiation proctopathy following pelvic radiotherapy. Four case series reporting on a total of 27 patients who underwent RFA using the HALO 90 (Barrx, Covidien, Mansfield) (all four studies), HALO 90 ULTRA (one study), and the HALO 60 (one study) device for radiation proctopathy have been described. A total of nine (33%) patients failed prior argon plasma coagulation (APC) therapy and one (7%) failed prior bipolar coagulation. Combining the results of the four studies, a mean of 1.9 (SD, 0.9) RFA sessions were performed per patient following either a full colonoscopy (N = 18) or enema-only preparation (N = 6). Clinical success with a decrease in symptoms occurred in 25 patients (93%). One patient had recurrent bleeding while on warfarin that was treated successfully with another RFA session and one patient had no change in symptoms. Mild adverse events with the development of rectal or perianal ulcers occurred in four patients (15%); only one patient with an ulcer presented with bleeding that did not require further therapy, whereas the remaining three patients with ulcers were asymptomatic.
| Author, Year | Study Type | Definition of Refractory | Total Number of Patients | Device Used; Average a # Sessions | Clinical Success N (%), Definition | Average a Hgb (g/dL) Before/After Treatment | Adverse Events |
|---|---|---|---|---|---|---|---|
| Gross et al, 2008 | Prospective case series | All transfusion dependent | 6 | HALO 90 1.5 (1–3) | 5 (83%) no longer transfusion dependent | 8.9 (12–15.5)/10.1 (9.4–11.5) | Ulcer b (N = 1) |
| McGorisk et al, 2013 | Prospective cohort | All failed APC and transfusion dependent | 21 | HALO 90 ULTRA 2 (1–3) | 18 (86%) no longer transfusion dependent | 7.8 (1)/10.2 (1.4) | Ulcer b (N = 2; 1 superficial, 1 bleeding) |
| Dray et al, 2014 | Retrospective case series | 21 failed prior therapy; 23 transfusion dependent | 24 | HALO 90 , HALO 90 ULTRA 1.8 (0.8) | 15 (65%) no longer transfusion dependent | 6.8 (1.4)/9.8 (1.8) | None |
| Raza & Diehl, 2015 | Retrospective case series | All failed APC and transfusion dependent | 9 | HALO 90 3 (2–6) | 7 (78%) no longer transfusion dependent | 7.3 (1.7)/10.5 (1) | Mild abdominal pain (N = 1) |
Related posts:
Advances and Improvements in the Management of Upper Gastrointestinal Bleeding
Epidemiology and Risk Factors for Upper Gastrointestinal Bleeding
Role of Medical Therapy for Nonvariceal Upper Gastrointestinal Bleeding
Endoscopic Diagnosis and Therapy in Gastroesophageal Variceal Bleeding
Unusual Causes of Upper Gastrointestinal Bleeding
Tips and Tricks on How to Optimally Manage Patients with Upper Gastrointestinal Bleeding
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
