Keywords Endoscopy Small Bowel - Small intestinal bleeding - Capsule endoscopy - Small bowel
endoscopy
Introduction
HemoPill is a swallowable electronic capsule ([Fig. 1 ]
a ) that detects hematin and blood through photometric measurements, which are wirelessly
transmitted every 4 seconds to a proprietary receiver ([Fig. 1 ]
b ) that is put in a single-use bag worn by the patient during the examination. The
photometric measurements are plotted over time and displayed as the “HemoPill Indicator”
(HI) with a HI reaching the red marked zone on the receiver display indicating detection
of intraluminal blood or hematin ([Fig. 1 ]
b and [Fig. 2 ]
a ). Preclinical studies showed that it diagnoses bleeding in the gastric lumen with
high sensitivity and specificity [1 ], with measurements being robust to the effects of gastric content [1 ]
[2 ]. Aside from intact deglutition, examination with HemoPill requires no special preparations
other than a glass of water and can be done bedside. Its major advantage is noninvasive
luminal blood detection in multimorbid and/or geriatric patients not primarily fit
for endoscopic diagnostic, allowing rapid and uncomplicated bleeding detection. It
has been commercially available in Germany since 2019.
Fig. 1
a The swallowable bleeding sensor capsule, HemoPill, measures 26.3 mm x 7 mm with a
measuring gap for photometric measurements. b Proprietary HemoPill receiver, which plots the photometric measurement results over
time. This figure displays an example of a positive blood detection, showed by a HI
within the red zone. Source: Ovesco Endoscopy AG, Tübingen, Germany
Fig. 2
a Positive HI indicating that blood has been detected 5 hours after swallowing the
capsule with a maximum HI reached after 8 hours. b Plain abdominal X-ray shortly after the positive HI. The Hemopill was approximately
located in projection of the cecum. c Colonoscopy revealed a nontraversable stenosis, proven to be cancer in the histology.
Source: Ovesco Endoscopy AG, Tübingen, Germany
Recent clinical reports have shown that it is useful for stratifying patients with
anemia and bleeding in the setting of an Emergency Department [3 ]. This study utilized the first capsule generation with a battery capacity of 2 to
3 hours, allowing blood detection only in the upper gastrointestinal tract. A recent
update brought an increased battery capacity of 9 hours, opening the possibility of
detecting blood in the middle and lower gastrointestinal tract, as shown by Brunk
et al [4 ].
Herein we report another advantage of the extended battery capacity allowing noninvasive
bleeding detection in the middle and lower gastrointestinal tract, leading to early
and expedited endoscopic diagnosis of gastrointestinal malignancies in two patients
with chronic bleeding leading to severe anemia.
First case
A 56-year-old female patient was referred due to recurrent anemia and melena with
a hemoglobin (Hb) concentration of 5 g/dL. A month before presenting to our unit,
she had been admitted to another hospital, also due to severe anemia with a Hb of
3 g/dL. There, a colonoscopy was performed and revealed no sign of bleeding but a
sigmoid tubular adenoma, which was thought to be the potential bleeding source and
thus resected without complication. After transfusion of 3 units of packed red blood
cells (pRBCs), she was discharged with stable Hb and no further signs of bleeding.
An upper gastrointestinal study (esophagogastroduodenoscopy [EGD]) was not performed
for unknown reasons. The patient’s medication was notable for apixaban as stroke prophylaxis
due to atrial fibrillation.
In our unit, an EGD was performed after transfusing 2 units of pRBCs, revealing no
intraluminal blood or other bleeding sign. The patient was scheduled for a repeat
colonoscopy followed by standard capsule endoscopy but she did not tolerate bowel
preparation due to nausea and recurrent vomiting. Thus, only sigmoidoscopy after an
enema was performed, which revealed no signs of bleeding or abnormality at the site
of the previous polypectomy. At that point, no further clinical signs of bleeding
or melena occurred and the Hb remained stable. The patient remained in the hospital
for observation, during which she suffered a syncope due to profuse melena. An urgent
EGD again revealed no blood or bleeding in the upper gastrointestinal tract. Because
the previous bowel preparation had not been tolerated and was deemed impossible after
syncope, HemoPill was utilized and placed directly into the duodenum using a snare
during the urgent EGD. After 1 hour, the receiver indicated a positive signal ([Video 1 ]). Because the receiver has no position indicator or other geographical function,
a plain abdominal x-ray was taken shortly afterwards to approximate the location of
the capsule, and thus, the source of bleeding. It was identified to be in the proximity
of the Treitz ligament, suggesting bleeding in the proximal jejunum. Urgent push enteroscopy
was performed and revealed a sponge-like dark solid lesion with a large blood clot
and active diffuse bleeding upon contact approximately 90 cm from the incisors. The
lesion was tattooed and hemoclips were placed for markings. Urgent computed tomography
(CT) failed to identify the tumor in the jejunum and only the clips were seen. A surgical
consultation was requested and jejunum resection was performed. Intraoperatively a
tumor was confirmed, which the pathology showed to be an angiosarcoma ([Video 1 ]).
The video shows direct placement of the capsule into the duodenum during the urgent
EGD to expedite the diagnostic pathway and bypass the gastric passage. After approximately
1 hour, the HI was positive indicating positive blood detection. Soon after, a plain
abdominal x-ray was performed showing the capsule to be approximately at the level
of the ligament of Treitz. Urgent push enteroscopy revealed a bleeding tumor in the
jejunum, histologically confirmed as an angiosarcoma.Video 1
Second case
A 77-year-old female patient was referred due to severe anemia with a Hb of 5 g/dL
without any clinical sign of bleeding. Due to obesity with a body mass index of 37
kg/m2 and her advanced age, we opted to postpone endoscopic diagnosis and utilized HemoPill
after first transfusing 2 pRBCs. After approximately 5 hours, a positive signal was
recorded ([Fig. 2 ]
a ). A plain abdominal x-ray shortly afterward revealed the capsule to be in the projection
of the cecum. Consequently, the patient was scheduled for a timely colonoscopy and
bowel preparation was initiated ([Fig. 2 ]
b ). Colonoscopy showed stenosis in the ascending colon ([Fig. 2 ]
c ), which proved to be malignant on histology and a malignant polyp in the sigmoid,
which was resected curatively during a second colonoscopy. No further tumor manifestation
was seen during the patient’s staging, allowing her to proceed to surgery, which confirmed
the diagnosis of colon carcinoma.
Discussion
These two cases showcased HemoPill’s potential as a useful novel adjunct in our arsenal
of noninvasive bleeding diagnostic tool in the upper, middle, and lower gastrointestinal
tract. Because of its ease of use, not requiring any preparation other than intact
swallowing and a glass of water, it is very appropriate for use in geriatric or comorbid
patients for whom endoscopic examinations are not suitable. In patients without intact
swallowing and without any signs of bleeding on EGD, HemoPill can be placed directly
in the duodenum, allowing rapid blood detection in the middle and lower gastrointestinal
tract without the necessity for bowel preparation. In contrast to a radiological study,
such as CT or magnetic resonance imaging (MRI), it is immediately available and requires
no contrast agent, which is beneficial in patients with chronic kidney failure. Because
of its small size (approximately half the thickness of a standard capsule endoscopy),
the chance of capsule retention is significantly reduced.
A major drawback of HemoPill compared with contrast-enhanced CT angiography, is lack
of exact bleeding localization. Furthermore, photometric results are only sent to
a local receiver within a limited range of 50 cm, which requires that the patient
wear the receiver during the examination. Thus, multiple manual checks are needed
to see if results turn positive and to schedule a timely plain abdomen study, which
is currently necessary to estimate the location of the HemoPill (and hence, the bleeding).
In our cases, the plain abdomen studies were taken approximately 30 to 60 minutes
after the positive result. Had they been taken at a later time, it may have been more
difficult to identify the site of the hemorrhage. Nevertheless, the x-ray results
would still have been helpful, because the "hemorrhage search area" would be narrowed
down to the capsule site and the area proximal thereof. Performing a CT scan is an
alternative to a plain abdomen study, which could provide the exact capsule location
but can be logistically challenging in a lot of units due to acute availability, especially
on weekends and night. Furthermore, patients would also experience greater radiation
exposure. In the first patient, the actively bleeding tumor was not visible on the
urgent CT scan, despite the use of contrast agent.
The multiple manual checks represent an enormous challenge for staff, especially if
a patient is not able to check the result independently. In our experience, these
challenges can only be tackled when the nursing staff is highly motivated, the patient
and family are empowered to understand the importance of multiple receiver checks
during the 9-hour study period, and abdominal imaging is performed timely to estimate
capsule location and with it, the bleeding source. Future technical improvements are
needed to improve and facilitate capsule localization. A longer transmitting distance
to the receiver or another alerting system for the nursing staff in case of a positive
HI could also improve practicality, avoiding the need for multiple direct checks by
patients.
Compared with conventional pan-enteric video capsule endoscopy (PCE), the HemoPill
is smaller at 7 × 26.3 mm (vs. PillCam Colon or Pillcam Crohn at 11.6 × 31.5 mm),
offering a more comfortable swallowing experience and theoretically reduced risk of
capsule retention. The biggest disadvantage of the device is that it cannot offer
any image from the gastrointestinal tract, but it is also more practical, because
it requires no fasting or bowel preparation. Although results from PCE can be unreliable
if the view is obscured by bowel contents, HemoPill measurements are principally unaffected
by recent meals or other intestinal or gastric content. In addition, PCE recordings
are typically reviewed a day after the capsule is swallowed, causing a delay in diagnosis,
albeit not clinically significant in most patients. Although some PCE devices offer
a live feed on the receiver or through an external device connected to the receiver,
this also demands significant staff involvement for monitoring similar to HemoPill,
if a patient cannot do the monitoring independently. In contrast, HemoPill provides
real-time results plotted directly on the receiver without the need for an extra device.
Moreover, HemoPill is less expensive than standard PCE in Germany. Recently, a study
with another commercially available blood-detecting swallowable capsule has been published,
showing promising results for detection of active bleeding in the upper gastrointestinal
tract [5 ]. The availability of this second blood-detecting capsule heralds a new dawn of noninvasive
bleeding detection in clinical practice.
Conclusions
In summary, HemoPill has the potential to serve as a noninvasive diagnostic adjunct
tool for detection of gastrointestinal bleeding, especially in patients for whom extensive
endoscopic diagnostics are not appropriate (e.g., comorbid or geriatric patients).
Further studies (i.e., the “Rabbit”-Study, Study-ID: DRKS00026103) aim to clarify
the potential of HemoPill as an alternative to video capsule endoscopy in diagnosing
bleeding in the middle gastrointestinal tract. We believe that in the future, the
role of noninvasive diagnostics in detection of upper, middle, and lower gastrointestinal
bleeding will increase.
Bibliographical Record
