Mechanical Hemostasis Techniques in Nonvariceal Upper Gastrointestinal Bleeding




One of the most important advances in gastroenterology has been the use of endoscopic hemostasis techniques to control nonvariceal upper gastrointestinal bleeding, particularly when high-risk stigmata are present. Several options are available, including injection therapy, sprays/topical agents, electrocautery, and mechanical methods. The method chosen depends on the nature of the lesion and experience of the endoscopist. This article reviews the available mechanical hemostatic modalities.


Key points







  • •

    Mechanical hemostasis is safe and effective for the treatment of nonvariceal upper gastrointestinal (GI) bleeding associated with high-risk stigmata.


  • •

    Hemoclips remain the most commonly used modality, owing to the weight of evidence supporting their use, ease of application, widespread availability, and familiarity among endoscopists.


  • •

    Although other mechanical techniques are less evidence based, their use is conceptually logical and they may be applied at the discretion of the endoscopist.






Introduction


Mechanical hemostatic techniques have gained popularity over time owing to their safety, efficacy, ease of application, and widespread availability. The mainstay of mechanical hemostasis involves placement of through-the-scope hemoclips, but newer modalities (eg, over-the-scope clips [OTSCs]) and methods borrowed from other endoscopic therapies (eg, band ligation) have emerged as alternatives. These therapies can be applied to a variety of lesions with much overlap in terms of which device is chosen. This review highlights mechanical techniques used to control bleeding specifically in patients with nonvariceal upper GI bleeding, typically ulcer bleeding.




Introduction


Mechanical hemostatic techniques have gained popularity over time owing to their safety, efficacy, ease of application, and widespread availability. The mainstay of mechanical hemostasis involves placement of through-the-scope hemoclips, but newer modalities (eg, over-the-scope clips [OTSCs]) and methods borrowed from other endoscopic therapies (eg, band ligation) have emerged as alternatives. These therapies can be applied to a variety of lesions with much overlap in terms of which device is chosen. This review highlights mechanical techniques used to control bleeding specifically in patients with nonvariceal upper GI bleeding, typically ulcer bleeding.




Mechanical hemostatic techniques


Hemoclips


Endoscopic clipping devices were first described in 1975 but were initially abandoned because of their complexity. They were reintroduced for use in hemostasis in 1988. Further modifications to clip design resulted in improved ease of use and higher levels of clinical efficacy. In the early 1990s, endoclips were reported to have a 100% success rate in 88 patients with ulcer bleeding.


Initial devices required preloading onto the applicator and subsequent sheathing, making the delivery slow and cumbersome, which was problematic in the setting of acute bleeding. However, multiple advances in endoclip design have taken place, resulting in clips that come preloaded and are easier to position and deploy. Furthermore, currently available endoclips have longer-lasting retention rates. In aggregate, these advances and the ever-increasing familiarity of endoclips have made them one of the most commonly used and effective endoscopic therapeutic techniques.


Mechanism


Hemoclips are metallic devices that can be applied directly to a blood vessel or by apposition of tissue on either side of a vessel or bleeding lesion, resulting in direct vessel occlusion or tamponade. Initial hemostasis of nonvariceal lesions is achieved in 85% to 98% of patients, with rebleeding rates in the 10% to 20% range.


When hemoclips fail, the reasons are often technical or due to operator error, although patient factors may on occasion be responsible. For example, advanced patient age may play a role in the failure to control bleeding. The most common technical reason for hemoclip failure is because the lesion is found in a position that makes en-face application technically difficult (eg, a lesion on the posterior wall of the antrum, lesser curve side of the duodenal bulb, or posterior wall of duodenal bulb). Additional causes of failure include operator error, often with tangential approaches and placing hemoclips improperly or in a fibrotic ulcer base where they are likely to fall off and thus fail to provide optimal vessel compression. It is the authors’ practice to visualize the target lesion with aggressive water lavage so as to map out clip deployment, because blood may obscure the lesion. The clip is then positioned as close to the lesion as possible and firmly abut the clip to the mucosa with each prong on either side of the vessel before deploying; if the scope is too far from the target lesion, one loses mechanical advantage and may miss the target. It is important not to scrape the vessel or to deploy around it rather than directly on it. Finally, more than 1 clip may be necessary to fully ligate a vessel, particularly if of large caliber.


Difficulty in applying hemoclips may be overcome by repositioning the lesion with the use of a distal attachment or by using an endoscope with a therapeutic channel at the 5-o’clock position. Distal attachments are clear caps that extend off the end of the scope. Use of a clear cap can be particularly helpful if the lesion is tangential; in this situation, the clear cap can push the mucosa so as to put the target area in a more en-face position for the clip to grasp. Because clips reach only the mucosa and submucosa, there is minimal risk of deep or surrounding tissue injury, although rare reports of causing delayed hemorrhage when applied to normal tissue exist. In addition, they can assist with ulcer healing if normal tissue on either side can be approximated.


Although hemoclips are generally thought to fall off after approximately 2 weeks, there are reports of clip retention for up to 2 years. Whether there is a relationship between hemoclips retention and control of bleeding is unknown.


Types of hemoclips


The first generation of hemoclips had a reusable loading device, but reloading was cumbersome and slow; these were largely abandoned once disposable clips were introduced in the early 2000s. Many variations of the single-use devices have been developed, with variations in opening diameter, ease of rotation, and presence of a sheath. A theme across the field is that there have been many advances in clip design. For example, the TriClip (Cook Endoscopy, Winston Salem, NC, USA), introduced in 2003, is a 3-pronged design, which attempted to obviate rotation. The Instinct hemoclip (Cook Endoscopy, Winston Salem, NC) is a sheathless, wide-opening (16 mm opening diameter), 2-pronged device that is easy to rotate and advance and can be opened and closed, a marked improvement over older devices.


The QuickClip was introduced in 2002 and the QuickClip2 in 2005 (both Olympus America, Center Valley, PA, USA). Versions of the QuickClip2 have opening diameters of 9 to 11 mm and working lengths of 165 to 270 cm, with the longer being designed to accommodate enteroscopes. The QuickClip Pro has an opening width of 11 mm and a working length of 1650 to 2300 mm. It was designed to improve ease of rotation to orient the clip to a target lesion and can be opened and closed.


The Resolution clip (Boston Scientific, Natick, MA, USA) can also be opened and closed and has an 11-mm opening diameter and a working length of 155 to 235 cm, thus also allowing accommodation of an enteroscope.


Applications


Hemoclips have been used to treat and are effective for the treatment of many lesions, including ulcers (see later discussion; Figs. 1–4 ), Dieulafoy lesions ( Fig. 5 ), Mallory-Weiss tears, angioectasia, diverticula, endoscopic resection sites, and mucosal lesions such as tumors and polyps. Hemoclips are most commonly used and were largely developed for use in patients with peptic ulcer lesions. However, they have been shown to be effective in patients with all of the lesions highlighted earlier.




Fig. 1


Peptic ulcers for which mechanical endoscopic therapy is not indicated. In ( A ) is shown an example of a clean-based antral ulcer, with a flat ulcer base and well-demarcated margins that appear to be healing. This lesion does not have any evidence of recent bleeding and represents a lesion with essentially no risk of rebleeding (and thus should not be treated with endoscopic therapy). In ( B ) is shown a large, deep, duodenal bulb ulcer that has only flat spot stigmata of recent bleeding. Not only does this lesion not have a readily identifiable vessel but also because of its depth, large size, and fibrotic base, it would be difficult to approach with a hemoclip (or other mechanical approach).



Fig. 2


Hemoclipping of a gastric ulcer with visible vessel. In ( A ) is shown a gastric ulcer, with an exudative ulcer base and a clearly visible protuberant purple punctum (representing an artery). This example represents an ideal candidate for placement of hemoclips. In ( B ) is shown the same lesion after placement of 4 hemoclips. The patient had no further bleeding.



Fig. 3


Hemoclipping of a gastric ulcer with oozing visible vessel. In ( A ) is shown a complex gastric ulcer with readily visible areas of exudate in the ulcer bed along with a visible vessel that is actively oozing blood (fresh blood is seen pooling on the right side of the image). In ( B ) is shown the same lesion after washing, injection of epinephrine, and a single hemoclip was placed on the vessel. No fresh blood can be seen. The patient had no further bleeding.



Fig. 4


Hemoclipping of an actively spurting gastrojejunal anastomotic ulcer in a patient who has undergone Roux-en-Y gastric bypass. In ( A ) is shown a stream of blood emanating from the artery in the ulcer bed (the ulcer is poorly visualized because of the orientation of the ulcer and the large amount of blood present). In ( B ) is shown the same lesion after 2 hemoclips were placed. The ulcer bed is now also more clearly seen in the retroflexed position. The patient had no further bleeding.



Fig. 5


Hemoclipping of a Dieulafoy lesion in the duodenal bulb. In ( A ) is shown a typical, nonspurting, Dieulafoy lesion in the duodenal bulb. In ( B ) is shown the same lesion after 2 hemoclips were placed. The patient had no further bleeding.


Effectiveness


Several randomized controlled trials have examined hemoclips, comparing them to other single modalities such as injection or heater probe or combined approaches (ie, injection plus hemoclip). Results from these studies have been mixed and difficult to interpret given the nuances in study design. Evidence in this area is confounded by different comparison groups. For example, there are multiple comparison groups such as hemoclip versus injection, hemoclip versus thermocoagulation, injection hemoclip versus injection plus thermocoagulation, and so on. Some studies have shown superior hemostatic control with hemoclips compared with other modalities, whereas others have found that they are no different or even less effective than injection therapy or thermocoagulation therapy. A meta-analysis suggested that hemoclips were not superior to other endoscopic modalities in terms of initial hemostasis, rebleeding rate, and the need for emergency surgery. In addition, there was no difference in all-cause mortality irrespective of the modalities of endoscopic treatment. Thus, rigorous data comparing hemoclips with other modalities do not yet exist. Although the authors prefer combination therapy in certain circumstances (ie, injection plus hemoclips for actively bleeding lesions), whether this is better than hemoclip alone is somewhat controversial, and small studies have failed to demonstrate an advantage of epinephrine plus hemoclip over hemoclip alone. The authors also inject epinephrine into large clots before removal if they are adherent after washing and to large vessels where the risk of rupture could result in substantial bleeding, as with the gastroduodenal or left gastric arteries, although data to guide this practice are limited.


Limited comparative data exist among the various types of clips, and most studies have examined older clips, which had conflicting retention and hemostasis rates. One study that compared the hemoclip device with the TriClip device in patients with high-risk bleeding peptic ulcers found that initial hemostasis was obtained in 47 patients (94%) of the Hemoclip group and in 38 patients (76%) of the TriClip group ( P = .011). Rebleeding episodes, volume of blood transfusion, hospital stay, numbers of patients requiring urgent operation, and mortality were not statistically different between the 2 groups. It was speculated that the mechanism for the difference in initial hemostasis was that the TriClip device was more difficult to maneuver in sites where it was difficult to apply. In aggregate, no real conclusions can be drawn, and there is no clear evidence that one type of clip is better than another. Table 1 highlights specifications of currently available through-the-scope clips.


Tags:
Sep 10, 2017 | Posted by in GASTOINESTINAL SURGERY | Comments Off on Mechanical Hemostasis Techniques in Nonvariceal Upper Gastrointestinal Bleeding

Full access? Get Clinical Tree
Get Clinical Tree app for offline access