Using external magnet guidance and endoscopically placed magnets to create suture-free...
-
Upload
christopher-myers -
Category
Documents
-
view
213 -
download
1
Transcript of Using external magnet guidance and endoscopically placed magnets to create suture-free...
Using external magnet guidance and endoscopically placedmagnets to create suture-free gastro-enteral anastomoses
Christopher Myers • Benjamin Yellen •
John Evans • Eric DeMaria • Aurora Pryor
Received: 13 May 2009 / Accepted: 2 October 2009 / Published online: 24 December 2009
� Springer Science+Business Media, LLC 2009
Abstract
Background To facilitate endolumenal and natural orifice
procedures, this study evaluated a novel technique using
external and endoscopically placed magnets to create
suture-free gastroenteral anastomoses.
Methods Seven anesthetized adult swine underwent
endoscopic placement of magnets into the small bowel and
stomach. Using external magnets, the endoscopically placed
internal magnets were brought into opposition under endo-
scopic view. After 1–2 weeks, the pigs were killed and
analyzed. At laparotomy and under sterile conditions, peri-
toneal cultures were obtained. The anastomoses were
evaluated endoscopically and tested using an air insufflation
test. Finally, the anastomoses were resected and evaluated
microscopically.
Results The average operative time for endoscopic
placement of the magnets was 34.3 ± 14.8 min. Successful
placement and creation of anastomoses occurred in six of
the pigs. One pig did not form an anastomosis because the
magnets were too large to pass through the pylorus at the
time of attempted magnet placement. Six swine experi-
enced uncomplicated postoperative courses. One pig’s
postoperative course involved constipation for several
days, requiring additional fluids and fiber supplementation.
The findings at endoscopy showed that the magnets were
adhered to the anastomosis, which were easily freed, or
within the stomach. The air insufflation test results were
negative for all the pigs. At laparotomy, there was no
evidence of infection, abscess, or leak, but two peritoneal
culture results were positive with scant growth of Staphy-
lococcus aureus and coagulase-negative staphylococcus,
presumably contaminants. Microscopically, the anastomo-
ses illustrated granulation and fibrous connective tissue
without evidence of infection or leak.
Conclusion Endoscopically placed magnets with external
magnet guidance is a feasible and novel approach to cre-
ating patent gastroenteral anastomoses without abdominal
incisions or sutures.
Keywords Endoscopy � Gastrojejunal anastomosis �Magnetic instrumentation � NOTES � Stapleless
anastomosis
General surgery has transitioned from maximal to minimal
invasiveness with the development of natural orifice trans-
lumenal endoscopic surgery (NOTES) and endolumenal
C. Myers (&)
Department of Surgery, Duke University Medical Center,
Durham, NC, USA
e-mail: [email protected]
B. Yellen
Department of Mechanical Engineering and Material
Science, Pratt School of Engineering, Duke University,
Durham, NC, USA
J. Evans
Department of Gastroenterology, Wake Forest University
Baptist Medical Center, Winston-Salem, NC, USA
E. DeMaria
Department of Endosurgery, Duke University,
Durham, NC, USA
E. DeMaria � A. Pryor
Department of Surgery, Duke University,
Durham, NC, USA
Present Address:C. Myers
UPMC St. Margaret’s, University of Pittsburgh Medical Center,
Pittsburgh, PA, USA
123
Surg Endosc (2010) 24:1104–1109
DOI 10.1007/s00464-009-0735-5
procedures. We sought to develop a completely endoscopic
procedure to avoid the risk of contaminating the peritoneal
cavity or leak from the enterotomy, which can occur when
the gut wall is crossed during NOTES.
The idea of using magnets and compression necrosis to
create anastomoses within the bowel developed from two
previously published concepts. First, in 1892, Dr. John B.
Murphy [1] developed the ‘‘Murphy button,’’ which con-
sisted of two metal mushroom-shaped devices sutured in
two ends of the bowel. The stems from the two ends of the
bowel were brought together, leading to compression
necrosis of the soft tissues between the metal devices,
creating an anastomosis [1, 2].
Second, young children experienced the development
patent fistulas within the bowel after consuming multiple
magnets in one setting. These fistulas were the result of
compression necrosis between attracting magnets as they
passed through the gastrointestinal tract [3–6].
Combining these two concepts led us to believe that a
patent sutureless anastomosis could be created via com-
pression necrosis between specifically placed magnets
within the bowel. In a preliminary study, we tested this idea
through a porcine model by placing and attracting differ-
ent-shaped magnets in the stomach and the jejunum via
laparotomy to determine feasibility. After 1 week of sur-
vival, patent anastomoses were created using ring-shaped
magnets [7]. In the current experiment, we sought to
expand this idea to an entirely endolumenal approach by
attempting to place the magnets endoscopically and using
external magnets to attract the internal magnets, avoiding
other visceral involvement.
Materials and methods
Sutureless gastroenteral anastomoses were created using
endoscopically placed ring-shaped magnets in swine under
institutional animal care and use committee (IACUC)
approval.
Magnets
Two types of neodymium (NdFeB) grade N42, nickel-
plated magnets (K & J Magnetics, Jamison, PA, USA)
were placed endoscopically into the small bowel and
stomach (Fig. 1). The first type was a large-ring magnet (3/
4-in. outer diameter [OD] 9 1/2-in. inner diameter
[ID] 9 3/8-in.-thick ring; Fig. 1A). The second type was a
small-ring magnet (1/2-in. OD 9 1/4-in. ID 9 1/8-in.
thick ring; Fig. 1B). The large magnets were used for better
attractive force and more successful anastomoses than in
the preliminary experiments. The smaller magnets were
used for facile traversing of the pyloris.
The small-ring magnets were placed in four swine and
the thick-ring magnets in three swine. Two large-block
NdFeB magnets (1 in. wide 9 1/2 in. thick 9 2 in. long)
were used externally.
Operative procedure
Stage 1: magnet placement
Seven Yorkshire crossbred adult swine weighing 25–
35 kg underwent placement of magnets. The animals were
fasted with access to water 12 h before induction of
anesthesia. For perioperative pain control, a fentanyl
patch (50 lg) was placed 24 h preoperatively. A combi-
nation of acepromazine (1.1 mg/kg) and ketamine
(22 mg/kg) via intramuscular (IM) injection and isoflu-
rane (1–3%) via face mask were used for induction and
intubation.
Once the animals were intubated, anesthesia and venti-
lation were maintained using isoflurane (1–3%) and oxygen
with tidal volumes of 10 ml/kg and respiratory rates of 15
to 20 breaths per minute (bpm). Ampicillin (1 g, IM) and
Fig. 1 A Thick ring (3/4-in. outer diameter [OD] 9 1/2-in. inner
diameter [ID] 9 3/8-in. B Small ring (1/2-in. OD 9 1/4-in. ID 9 1/
8-in.). Neodymium magnets were placed endoscopically in the
stomach and the jejunum of swine to form sutureless anastomoses
Surg Endosc (2010) 24:1104–1109 1105
123
Baytril (2.5 mg/kg, IM) were administered preoperatively
and daily throughout the study, per IACUC standards.
Once the animals had been anesthetized in supine
position, a single-channel endoscope (Pentax, Montvale,
NJ, USA) was introduced transorally into the pig’s jeju-
num. Under direct visualization, a 0.035-in. Jagwire Super
Stiff Guidewire (Boston Scientific, Natick, MA, USA) was
passed deep into the small bowel. The endoscope was
removed from the pig, with the guidewire left in place.
Outside the pig, the guidewire was passed through a ring
magnet and eventually through the channel of the endo-
scope. The endoscope was used to push the magnet into the
stomach by passing it over the guidewire.
Once the magnet was in the stomach, a 0.035-in. DU-
RAglide stone removal balloon catheter (BARD, Cov-
ington, GA, USA) was passed over the guidewire and
through the endoscope and ring magnet. Inflation of the
balloon within the ring of the magnet established control
of the magnet for its passage it over the guidewire and
into the small bowel (Fig. 2). Desufflation of the balloon
allowed visualization through the magnet and into the
small bowel.
Once the magnet was properly positioned in the jeju-
num, one large-block magnet was placed on the pig’s
abdomen, attracting the ring magnet and pulling the small
bowel up to the anterior abdominal wall (Fig. 3). After
placement and attraction were confirmed with the endo-
scope, the scope and balloon were removed. A second
magnet was placed into the stomach the same way and
attracted to the anterior abdominal wall with a second
large-block magnet. The two external magnets then were
brought together, attracting the internal magnets together
as well. The scope, guidewire, and balloon then were
removed.
The animals were allowed food and water immediately
after the operation and recovery from anesthesia. All the
pigs survived for at least 7 days (range, 7–14 days). Eating,
drinking, bowel, and behavior habits were observed.
Stage 2: reexploration and killing
The swine were killed 7 to 14 days after the initial oper-
ation using an intravenous lethal injection of Euthasol
(Virbac, AH, Inc., Fort Worth, TX) (175 mg/kg). Imme-
diately after the animals were killed, an endoscope was
inserted into the stomach to locate the magnets and
examine the anastomosis and surrounding mucosa.
Once the anastomosis was evaluated endoscopically, the
animal was prepped, draped, and explored via laparotomy.
Anaerobic and aerobic peritoneal cultures were obtained
immediately at the entrance to the abdominal cavity.
Attention was particularly guided toward the anastomosis
and any evidence of obstruction, leak, or infection. The
stomach was submerged in saline, and a leak test was
performed using unregulated endoscopic insufflation.
Finally, the gastrojejunostomy was excised, with preser-
vation of the involved parts of the stomach and the afferent
and efferent limbs of the jejunum.
Gross and microscopic examinations of the specimens
were performed. Masson trichrome stain was used because
it allows easy identification of mucosal injury and inflam-
matory response.
Fig. 2 The small-ring magnet was controlled and placed into the
small bowel using the Duraglide stone removal balloon catheter over
a guidewire
Fig. 3 External magnets used to attract internal magnets and bowel
to the anterior abdominal wall, avoiding other intraabdominal viscera.
Once attracted to the anterior abdominal wall, the external magnets
are brought together, also bringing together the two internal magnets
1106 Surg Endosc (2010) 24:1104–1109
123
Results
The average time for endoscopic magnet placement was
34.3 ± 14.8 min (range, 20–60 min). All the pigs survived
the postoperative course. One pig experienced a delay in
bowel function despite normal eating, drinking, and
behavioral activities. An abdominal X-ray on postoperative
day 6 showed the entire colon with mild distention and
stool, suggesting ileus/constipation. With additional
hydration and fiber supplementation, this pig resumed
normal bowel function.
Six pigs experienced successful placement of the mag-
nets and creation of gastrojejunal anastomoses. One pig
failed to produce an anastomosis because it could not pass
the thick-ring magnet through the pylorus. At endoscopy,
the magnets were found either within the anastomosis and
easily freed with minimal pressure or within the stomach.
All anastomoses were easily intubated by the endoscope
(Figs. 4, 5) and found to be negative for leak via an
unregulated air insufflation test. Laparotomy showed no
evidence of leak, infection, or obstruction. Two peritoneal
cultures tested positive with scant growth of Staphylococ-
cus aureus or coagulase-negative staphylococcus. Due to a
lack of evidence of inflammation/infection at laparotomy
and the absence of enteric organisms, these cultures were
presumed to be contaminants. Microscopically, the anas-
tomoses illustrated granulation and fibrous connective tis-
sue without evidence of infection (Fig. 6).
Discussion
This study is the first in the English literature to suggest
that by using endoscopically placed magnets via com-
pression necrosis, a sutureless patent anastomosis can be
created in living animals. As discussed in our preliminary
study, this is not the initial study describing compression
necrosis as a means to create an anastomosis. Murphy [1]
reported the ‘‘Murphy button’’ in 1892. Swain and Mills
used plastic rings to create enterocutaneous anastomoses in
dogs [7]. Multiple published studies have described a sta-
pling device, the biodegradable anastomotic ring (BAR),
which placed biodegradable rings in the colon, creating an
anastomosis [8].
Enteric magnet use and technology were first described
by Cope and colleagues, who created anastomoses by
compression necrosis in dogs, pigs, and eventually humans
[7]. Erdmann et al. [9] used magnet technology to create
sutureless vascular anastomosis. Magnets have been used
to treat malignant and biliary duct strictures after liver
transplantation [10, 11].
Fig. 4 Endoscopic view of a patent gastrojejunostomy created by
compression necrosis between magnets
Fig. 5 A patent gastrojejunostomy created by small magnets in a
gross specimen extracted from a pig. The Debakey ‘‘pickups’’ pass
through the jejunum
Fig. 6 Microscopic examination using Masson trichrome dye
preparation showing granulation and fibrous connective tissue at a
patent sutureless anastomosis without evidence of leak or infection
Surg Endosc (2010) 24:1104–1109 1107
123
In our preliminary study, we were able to place magnets
in the jejunum and stomach via laparotomy, creating patent
sutureless anastomoses [7]. In this study, we proved that
patent sutureless anastomoses can be created via a com-
pletely endolumenal approach.
The materials used and the technique for placement of
the magnets underwent change throughout our experiment.
The magnets used were based on our preliminary study, in
which the small-ring and thick-ring magnets produced
large patent anastomoses. The small-ring magnets exhib-
ited an attracting force of 636 mmHg (assuming a constant
intervening tissue layer of 2 mm) and a patent but small
anastomosis after 14 days. They were not found after the
animals were killed, suggesting passage through the gas-
trointestinal tract [12]. When an additional magnet was
added (total of two magnets in the jejunum and two in the
stomach), the resultant anastomoses appeared to be larger,
and the magnets were found in the stomach. Additional
studies are needed to determine whether a difference really
exists in the size of the anastomoses given that the diameter
of the ring did not change and that only the attractive force
changed by placement of an additional magnet on each
side. The size of two magnets together versus four magnets
would imply that the smaller the size of magnets, the easier
they pass through the gastrointestinal tract.
Thick-ring magnets produced a large patent anastomosis
but were difficult to place. The thick-ring magnets showed
an attractive force of 940 mmHg (assuming a constant
intervening tissue layer of 2 mm), produced a large patent
anastomosis, and were found in the stomach [12]. Due to
its size, transporting the magnet past the endotracheal tube
and through the pylorus was difficult. The thick-ring
magnet would not go through the pylorus in one pig and
therefore failed to create an anastomosis. A hole was
drilled in the sides of the magnet to allow passage of the
wire with better control, but difficulties still were experi-
enced due to a lack of change in diameter. After creation of
the anastomosis, the magnets were always found in the
stomach. Longer-term studies are needed to evaluate
alternative magnet configurations, passage of attracted
thick-ring magnets through the gastrointestinal tract, and
consideration of an endoscopic retrieval procedure.
Our technique of using a guidewire to place enteric
magnets was previously reported by Cope [13]. He
described blind transoral wire placement of magnets,
which resulted in intraabdominal visceral trapping
between magnets and a lack of anastomosis creation. His
technique was modified in several ways to identify the
location of our anastomosis and to avoid visceral entrap-
ment. We used a balloon catheter to control the magnet
placement and an endoscope to place the guidewire and
magnets visually into proper position. Once the magnet
was placed into the small bowel, external magnets were
used to draw the magnet and small bowel to the anterior
abdominal wall. When the magnet was placed in the
stomach, again, an external magnet was used to attract the
magnet and stomach to the anterior abdominal wall.
Bringing the two external magnets together brought the
two internal magnets together, with the idea of avoiding
other intraabdominal viscera. Because there was no evi-
dence of impinging other intraabdominal viscera at lapa-
rotomy, our technique was successful.
Due to the endoscopic approach to this procedure,
postoperative care and recovery are minimal. All the pigs
were allowed to eat and drink immediately after recovery
from anesthesia. One pig did experience a delay in bowel
function despite normal eating and behavior. On postop-
erative day 6, due to concerns for involvement of other
intraabdominal viscera, an abdominal X-ray was obtained,
which showed stool throughout a mildly dilated colon
without evidence of free air or obstruction. With addi-
tional water and fiber, bowel function returned later that
day. The remainder of the pig’s postoperative course was
uncomplicated.
At endoscopy, four small-ring magnets were found in
the stomach and a large patent anastomosis created
between the stomach and the small bowel. At laparotomy,
an anastomosis was found between the stomach and the
jejunum just beyond the ligament of Treitz. There was no
evidence of involvement of other bowel, obstruction,
infection, abscess, or leak. Peritoneal culture results were
negative. Despite this occurrence, all the other pigs had a
completely uneventful postoperative course.
Two peritoneal cultures were positive with contami-
nants. Scant growth of Staphylococcus aureus and coagu-
lase-negative staphylococcus was obtained from the pigs
with small- and thick-ring magnets. Because these bacteria
are not of enteric origin, all air insufflation test results were
negative for leak, and there was no evidence of infection at
exploration and microscopic examination. These positive
cultures are considered contaminants.
Daily antibiotics given throughout the entire postoper-
ative course may interfere with culture results, but there
would be evidence of infection if it truly were present. The
antibiotics were administered in our study based on local
IACUC standards. Because there was no intraabdominal
evidence of infection or leak, contamination of these cul-
tures was presumed to occur at the time of killing.
Microscopic examination of the anastomoses suggests
granulation and fibrous connective tissue without evidence
of infection.
These findings are similar to those from a sutured or
stapled anastomosis. As discussed in our preliminary study,
the long-term patency of the anastomoses needs to be
tested. Cope [13] reported anastomotic stricturing and
closure within 30 days of magnet placement in swine.
1108 Surg Endosc (2010) 24:1104–1109
123
In our study, all anastomoses were widely patent at 7
and 14 days. It is possible that the size of the anastomosis
and hence the magnet may contribute to stricture, just as
seen with current circular staplers. Also, the swine stomach
is prone to remodeling of itself. Therefore, studies with
other species, such as canines, may be indicated.
This study suggests that sutureless patent anastomoses
may be created via a completely endolumenal approach
using endoscopically placed enteric magnets with external
magnetic guidance. With the future of surgery focusing on
NOTES and endolumenal procedures, this approach may
be used in any operation that requires an anastomosis as
long as the magnets can be placed safely. Because the
procedure is completely endoscopic and may require only
conscious sedation, possibilities of performing the opera-
tion on an outpatient basis in an endoscopy suite for high-
risk surgical candidates may be entertained.
Disclosures Christopher Myers, John Evans, and Benjamin Yellen
have no conflicts of interest or financial ties to disclose. Eric DeMaria
receives Covidien, Stryker, and Ethicon educational grants as well as
Affinergy and Covidien honorariums. Dr. Pryor has ownership
interests in Transenterix and Barosense. Aurora Pryor also is a
speaker for Covidien and Olympus and a consultant for Covidien,
Olympus, Gore, Transenterix, and Immersion.
References
1. Murphy J (1892) Cholecysto-intestinal, gastrointestinal, entero-
intestinal anastomosis and approximation without sutures (origi-
nal research). Med Rec N Y 42:665–676
2. Dawbarn RH (1895) VI. The relative value of the murphy button
and absorbable plates in intestinal anastomosis. Ann Surg 21:
166–172
3. Alzahem AM, Soundappan SS, Jefferies H, Cass DT (2007)
Ingested magnets and gastrointestinal complications. J Paediatr
Child Health 43:497–498
4. Palanivelu C, Rangarajan M, Rajapandian S, Vittal SK, Mah-
eshkumaar GS (2007) Laparoscopic retrieval of ‘‘stubborn’’ for-
eign bodies in the foregut: a case report and literature survey.
Surg Laparosc Endosc Perctan Tech 17:528–531
5. Dutta S, Barzin A (2008) Multiple magnet ingestion as a source
of severe gastrointestinal complications requiring surgical inter-
vention. Arch Pediatr Adolesc Med 162:123–125
6. Robinson AJ, Bingham J, Thompson RL (2009) Magnet-induced
perforated appendicitis and ileocaecal fistula formation. Ulster
Med J 78:4–6
7. Myers CJ, DeMaria EJ, Mutafyan GA, Bauer MS, Evans JA,
Yellen B, Pryor AD (2009) Sutureless endolumenal gastro-jejunal
anastomosis creation using magnets. Surg Innov (in press)
8. Kaidar-Person O, Rosenthal RJ, Wexner SD, Szomstein S, Person
B (2008) Compression anastomosis: history and clinical consid-
erations. Am J Surg 195:827–828
9. Erdmann D, Sweis R, Heitmann C, Yasui K, Olbrich KC, Levin
LS, Sharkawy AA, Klitzman B (2004) Side-to-side sutureless
vascular anastomosis with magnets. J Vascular Surg 40:505–511
10. Avaliani M, Chigogidze N, Nechipai A, Dolgushin B (2009)
Magnet compression biliary-enteric anastomosis for palliation of
obstructive jaundice: initial clinical results. J Vasc Interv Radiol
20:614–623
11. Mita A, Hashikura Y, Masuda Y, Ohno Y, Urata K, Nakazawa Y,
Ikegami T, Terada M, Yamamoto H, Miyagawa S (2008) Non-
surgical policy for treatment of bilioenteric anastomotic stricture
after living donor liver transplantation. Transpl Int 21:320–327
12. Panofsky WKH, Philips M (1955) Classical electricity and
magnetism. Addison Wesley, New York
13. Cope C (1995) Creation of compression gastroenterostomy by
means of the oral, percutaneous, or surgical introduction of mag-
nets: feasibility study in swine. J Vasc Interv Radiol 6:539–545
Surg Endosc (2010) 24:1104–1109 1109
123