New Diagnostic Imaging Technologies in Nonvariceal Upper Gastrointestinal Bleeding

This article covers new endoscopic imaging modalities in nonvariceal upper gastrointestinal bleeding, such as Doppler ultrasound probe technology, endoscopic ultrasonography, color Doppler optical coherence tomography, and magnification endoscopy. A more in-depth discussion of these modalities and the published evidence supporting their use are included. Furthermore, the shift in focus from identification of conventional visual surface stigmata of recent hemorrhage to an assessment and understanding of subsurface blood flow as it relates to the bleeding lesion is discussed.

Ever since gastrointestinal (GI) endoscopy was first practiced, generations of endoscopists have been carefully looking, inspecting, and documenting findings on the surface of the GI tract. Researchers have sought to correlate the information obtained from visual inspection of the surface with underlying physiologic and pathophysiologic states. Over the last several decades, using this approach, tremendous strides and advances have been made in the practice of GI endoscopy. Looking at the surface of the GI tract still remains a foundational cornerstone in the practice of GI endoscopy.

In terms of nonvariceal upper GI (UGI) bleeding, and specifically peptic ulcer bleeding, the management decision of whether or not to endoscopically treat an acutely bleeding ulcer resides entirely on its surface appearance and finding of stigmata of recent hemorrhage (SRH). Because it relates to peptic ulcer bleeding, the so-called Forrest classification of SRH was first published by Forrest and colleagues almost 4 decades ago. Since then, many studies have shown that looking at visual stigmata alone can be quite subjective with significant interobserver variability. For instance, when endoscopists were shown color images or videoclips of acutely bleeding peptic ulcers and asked to identify whether there was any SRH, endoscopists disagreed more than 25% of the time. Furthermore, even among international experts in GI endoscopy, there was only good agreement when there was obvious spurting blood. In addition, aside from the generally accepted variability of simple visual inspection, some studies have shown that high-risk stigmata, such as nonbleeding visible vessel (NBVV), may have uncharacteristic or atypical visual appearances (for example, nonpigmented, pale, translucent, or pearl-colored protuberances) and thus may be easily misinterpreted by endoscopists as low-risk stigmata that need not be treated. Misguided application of endoscopic therapy (either underuse or overuse) can potentially lead to adverse patient outcomes and inappropriate use of health care resources. It is therefore important to evaluate other, more objective approaches in deciding whether or not to perform endoscopic hemostasis for an acutely bleeding peptic ulcer.

This article discusses new diagnostic imaging technologies in nonvariceal UGI bleeding, with a focus on technologies that allow the endoscopist to determine subsurface blood flow, such as endoscopic Doppler ultrasound (DopUS) probe technology, endoscopic ultrasonography (EUS), and color Doppler optical coherence tomography (CDOCT). In addition, magnification endoscopy is also discussed, which is a new technique for enhancing surface visual stigmata but does not permit evaluation of subsurface blood flow.

Determination of subsurface blood flow beneath a bleeding lesion

Conceptually, any bleeding lesion must have an active arterial blood supply for it to bleed. On the other hand, a bleeding lesion can no longer continue to bleed if its major arterial blood supply has ceased. It is therefore important to know whether a bleeding lesion still has active arterial blood flow supplying it or whether the blood flow may have ceased as a result of spontaneous intravascular thrombosis (about 80% of UGI bleeding ceases spontaneously without intervention). Therefore, for a lesion that has stopped bleeding, the risk of recurrent bleeding would theoretically be reduced if there is no longer any major blood supply to the lesion. Therefore, it is important to develop new technologies that allow endoscopists to determine subsurface blood flow beneath a bleeding lesion because these new technologies can theoretically permit

1.

Assessment of presence or absence of blood flow in subsurface blood vessels
2.

Evaluation of uncertain or indeterminate visual surface stigmata based on subsurface blood flow ( Figs. 1 and 2 )

Fig. 1

DopUS probe. ( A ) Uncertain visual SRH in bleeding gastric ulcer in antrum (flat spot vs NBVV). ( B ) DopUS probe examination: Doppler positive. ( C ) DopUS probe examination immediately after endoclipping: Doppler negative. Note: the ulcer started to bleed briskly immediately after placement of the first endoclip. Bleeding only ceased after the third endoclip was placed.

Fig. 2

DopUS probe. ( A ) Clean-based gastric ulcer in the fundus in a patient with severe recurrent UGI bleeding. DopUS probe examination demonstrated strong arterial Doppler signal at 11-o’clock position on ulcer base where there was no visible bleeding stigmata ( arrow ). ( B ) Targeted dual combination endoscopic therapy with epinephrine injection and endoclipping at 11-o’clock position on ulcer base. ( C ) DopUS probe examination immediately after endoscopic therapy: Doppler negative.
3.

Differentiation between subsurface artery and vein
4.

Assessment of size and depth of subsurface blood vessels
5.

Subsurface mapping of direction, route, and course of blood vessels ( Fig. 3 )

Fig. 3

DopUS probe. ( A ) Duodenal ulcer in bulb with actively bleeding visible vessel. ( B ) DopUS probe examination of ulcer immediately after dual combination therapy with epinephrine injection and endoclipping of visible vessel demonstrated a subsurface artery coursing linearly along the right lateral edge of the ulcer ( arrows ) (acoustic Doppler map). ( C ) DopUS probe examination immediately after placement of 2 additional endoclips along the course of the subsurface artery: Doppler negative.
6.

Assessment of relative blood flow velocity
7.

Ability to titrate the extent of endoscopic therapy based on blood flow signal
8.

Precise targeted application of endoscopic therapy based on location and depth of blood flow (see Figs. 2 and 3 )
9.

Assessment of anatomic and topographic aspects of tissue structure in the vicinity of a bleeding lesion.

Endoscopic DopUS probe

Although the use of an endoscopic DopUS probe to assess a bleeding lesion is not new, it has recently taken on heightened interest with the availability of a portable DopUS system that uses disposable single-use probes and has received the US Food and Drug Administration (FDA) clearance for use in GI endoscopy in the United States (20-MHz DopUS system, Vascular Technology Inc, Nashua, NH, USA). The other DopUS system that has been used in published studies that has not received FDA clearance uses reusable DopUS probes that are reprocessed by high-level liquid disinfection or gas sterilization (16-MHz DopUS system, “Endo-Dop,” DWL GmbH, Singen, Germany).

In distinct contrast to standard EUS, DopUS is a nonimaging technique that was originally developed in England in the early 1980s for the evaluation of bleeding lesions in the GI tract. Again, in marked contrast to standard EUS, the use of DopUS probe does not require endoscopists to have knowledge of EUS and, furthermore, does not require advanced endoscopic training in EUS. The technique for using the DopUS probe can be readily acquired by most general GI endoscopists in a relatively short period. Output signals from DopUS systems are either audible alone (VTI system) or audible plus graphical display (DWL system) systems ( Fig. 4 ).

Current DopUS probe technology uses a small, flexible, pulsed-wave, 16- or 20-MHz DopUS probe that is passed down the accessory channel of a standard diagnostic or therapeutic forward-viewing endoscope. In addition, the DopUS probe can also be passed through a diagnostic or therapeutic duodenoscope to evaluate lesions that require a side-viewing endoscope. The ultrasound beam exits the distal tip of the DopUS probe in a linear fashion. In contrast to miniature probes used in traditional EUS, DopUS probe technology does not require the use of a water-filled balloon for acoustic coupling. The through-the-scope DopUS probe only has to make direct physical contact with the lesion of interest, for example, the base of an ulcer for which no balloon is required ( Fig. 5 ).

Various preset scanning depths ranging from less than a millimeter to many millimeters can be selected based on the subsurface blood vessel of interest. For evaluating bleeding peptic ulcers, published studies have used a shallow scanning depth of 1.5 mm or less. Sometimes a middepth setting (eg, 0–4 mm) can be used to confirm a faint or weakly positive Doppler signal that is detected at shallow depth. It is important for endoscopists to appreciate that the selection of scanning depth depends on whether there is increased physical distance between the surface and the subsurface blood vessels. Correct selection of scanning depth is important because if it is too deep, then innocent “bystander” blood vessels uninvolved in the bleeding process may be detected (a false-positive Doppler signal). On the other hand, if there is increased physical distance between the surface and the subsurface blood vessel, for instance, with the presence of a firmly adherent clot or immediately after injection of epinephrine when there is a subsurface cushion (or bleb) of fluid, then too shallow a scanning depth may fail to detect the culprit subsurface artery (a false-negative Doppler signal). The following DopUS scanning depth settings can be used:

•

Shallow (0–1.5 mm): untreated peptic ulcer, injection therapy followed by thermal contact therapy, thermal contact therapy alone, endoclip therapy alone
•

Mid (0–4 mm): adherent clot, injection therapy alone, injection followed by endoclip, confirm a weakly positive Doppler signal detected at shallow depth
•

Deep (0–7 mm): usually not needed.

The output signal from the DopUS system is based on the Doppler effect, in which blood cells contained within the subsurface blood vessel act as moving targets reflecting ultrasound waves back to a stationary transducer (DopUS probe). The resultant Doppler shift is automatically calculated by the system, and the output (or result) is immediately expressed (in real time) either as an audible Doppler signal (VTI system) or audible plus graphic Doppler signals (DWL system). The Doppler shift equation is