|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
PERCUTANEOUSLY CONTROLLED HYDRAULIC OCCLUDERS AND THEIR
APPLICATION IN GRADUAL VENOUS OCCLUSION
COLIN WAYNE SEREDA
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
Colin Wayne Sereda, DVM
I would like to thank Nikki and my family for their support and encouragement in
pursuit of this degree. I would also like to thank Dr. Christopher Adin for his support and
guidance throughout the last two years.
TABLE OF CONTENTS
A C K N O W L E D G M E N T S ................................................................................................. iii
LIST OF FIGURES ......... ......................... ...... ........ ............ vi
ABSTRACT .............. ..................... .......... .............. vii
1 METHODS OF GRADUAL VASCULAR OCCLUSION AND THEIR
APPLICATIONS IN TREATMENT OF CONGENITAL PORTOSYSTEMIC
SH U N T S: A R E V IE W ..................................................................... ..................... 1
In tro du ctio n ...................................... ............................. ..... ......... ...... .
S ilk S u tu re ........................................................... ................ 3
A m eroid C onstrictors........ ................................................................ ............... .5
C ello p h an e ................................................................................................... . 9
T hrom bogenic C oils ................. .. ...... ...... ..... .... ........ .............. .... .11
H hydraulic O ccluders .................................................... .................. 14
C onclu sion ............................................................. .. .................... 17
2 A PERCUTANEOUSLY CONTROLLED HYDRAULIC OCCLUDER IN A RAT
MODEL OF GRADUAL VENOUS OCCLUSION. ..............................................18
Intro du action ...................................... ................................................ 18
M materials and M methods ....................................................................... ..................20
A n im a ls .......................................................................................................... 2 0
Im p la n ts ............................................................................................................... 2 0
S u rg e ry ................................................................2 2
B lood F low M easurem ents................................................................... ...............24
N ecropsy ................................................................. ... ..... ........ 25
H istopathologic Exam nation ................................................... ............... 25
Statistical A analysis ........................ ............ ................ ....... 25
R esu lts ............... ..... ... ........ .......................................................... ... 2 6
Volume Injected vs. % Occlusion .......................... ....... .. ............... 26
Surgery ....................... ....................26
W weekly Blood Flow M measurements ............................ .................................26
N ecropsy ................................................................ .... .... ........ 28
H istopathologic Findings.......................................................... ............... 29
D discussion ......... ........ ............................................. ......30
3 MECHANICAL AND PHYSICAL PROPERTIES OF SILICONE HYDRAULIC
OCCLUDERS ................ ........ .......................35
Intro du action .....................................................................................................3 5
M materials and M methods ..............................................................................................36
Evaluation of Size Differences Among HOs........................... .............. ......36
Diffusion of Saline and Sodium Hyaluronate Through the HO..........................37
D iffusion of A ir Through the H O ..................... ..... ........................................37
Changes in Internal Pressure Over Time.........................................................38
Statistical A analysis ........................................ ................... .. .. ... 39
R esu lts............................................ .. ....... ....................... ................ 3 9
Evaluation of Size Differences Among HOs................3......................... ......39
Diffusion of Saline and Sodium Hyaluronate Through the HO..........................40
D iffusion of A ir Through the H O ........................... ................. .......... .........42
Changes in Internal Pressure Over Tim e ........................................ ................43
D isc u ssio n ................................................... ................... ................ 4 4
4 C O N C L U SIO N ......... ......................................................................... ........ .. ..... .. 50
L IST O F R EFE R E N C E S ......... ......................... .......... ........................... ............... 53
B IO G R A PH IC A L SK E T C H ..................................................................... ..................61
LIST OF FIGURES
1. Photograph of a 5mm ameroid constrictor and key ..............................................6
2. Photograph of two intravascular thrombogenic coils. .............................................12
3: Photograph of a 2 mm internal diameter hydraulic occluder............... ..................15
4. Graphic representation of changes in % occlusion associated with volume injected in 2
m m hydraulic occluders.. .............................. ... ........................................ 21
5. Intraoperative view of the instrumented caudal vena cava........................................23
6. Individual weekly vena caval blood flow measurements. ........................................27
7. Radiographic image documenting caval patency. ...................................... .......... 29
8. Schematic diagram representing the experimental setup used for evaluating changes
in internal pressure over time within individual hydraulic occluders ....................39
9. Graphic representation of the variations in luminal area within 3 models of the non-
inflated hydraulic occluder ......................................................... .............. 40
10. Graphic representation of the mean weight loss over time for 2 mm (A), 5 mm (B),
and 20 mm (C) hydraulic occluders filled with either saline (*) or sodium
hyaluronate (m).............................. ...................... 41
11. Graphic representation of the mean % occlusion over time for 2 mm (A), 5 mm (B),
and 20 mm (C) hydraulic occluders filled with either saline (*) or sodium
hyaluronate (m).............................. ...................... 42
12. Graphic representation of the mean change in occlusion over a period of 12 hours for
2 mm (m), 5 mm (e), and 20 mm (A) hydraulic occluders filled with air. ............43
13. Graphic representation of the change in internal pressure over time for 2 mm (m), 5
mm (e), and 20 mm (A) hydraulic occluders filled with saline ..............................44
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
PERCUTANEOUSLY CONTROLLED HYDRAULIC OCCLUDERS AND THEIR
APPLICATION IN GRADUAL VENOUS OCCLUSION
Colin Wayne Sereda
Chair: Christopher A. Adin
Major Department: Veterinary Medicine
Surgical occlusion is widely accepted as the treatment of choice for congenital
portosystemic shunts in dogs. Unfortunately, acute occlusion of portosystemic shunts is
frequently associated with life threatening portal hypertension. Despite years of research
in this area, veterinary surgeons continue to search for a safe and effective method to
produce gradual occlusion of portosystemic shunts in dogs.
The overall goal of this thesis project was to evaluate a novel technique for gradual
venous occlusion using a percutaneously controlled silicone hydraulic occluder. In order
to identify the qualities desired in an ideal surgical device for treatment of congenital
portosystemic shunts, a comprehensive literature review pertaining to congenital
portosystemic shunts in veterinary patients was performed.
Following the literature review, a prospective evaluation of the silicone hydraulic
occluder was performed using a rat model. The vena cava of each rat was instrumented
with a silicone hydraulic occluder, as well as a perivascular ultrasonic flow probe.
Gradual decrease in blood flow was documented during an 8-week period, although
limitations were identified with the experimental model due to the size disparity between
the large implants and the small, compressible rat vena cava. A second problem identified
was a marked difference between the predicted filling volume of the occluder and the
actual filling volume required to cause cessation of blood flow. It was hypothesized that
several factors contributed to this phenomenon, including diffusion of filling solutions,
size differences in the HOs secondary to manufacturing inconsistencies, and plastic
deformation of the materials over time.
Due to questions regarding the reliability of the vascular occluders in chronic
applications, a series of in vitro experiments were performed to evaluate the mechanical
properties of hydraulic occluders immersed in simulated body fluid. Data from these
studies confirmed that air diffused rapidly from the occluders, whereas both saline and
sodium hyaluronate functioned as appropriate filling solutions. Size differences among
the occluders secondary to manufacturing were statistically significant, and could
potentially affect filling volumes and fine adjustment of the level of occlusion in chronic
applications. Changes in internal pressure were also noted to occur over time, suggesting
that mechanical deformation of the silicone occluders may also contribute to changes in
filling volume required. Mechanical deformation appeared to be more significant in the
smaller occluders studied.
METHODS OF GRADUAL VASCULAR OCCLUSION AND THEIR
APPLICATIONS IN TREATMENT OF CONGENITAL PORTOSYSTEMIC SHUNTS:
Gradual vascular occlusion was originally attempted in experimental models of
human coronary artery disease and portal hypertension.1'2'3 While these experimental
studies gave rise to a small number of clinical applications for surgical treatment of
vascular anomalies in human patients,4'5 veterinary surgeons have performed numerous
clinical and experimental studies evaluating methods of gradual vascular occlusion for
congenital portosystemic shunts (CPS). The purpose of this manuscript is to review the
reported clinical and experimental techniques of gradual vascular occlusion and to
discuss their potential for application in the treatment of CPS.
Surgery is generally advocated for the definitive treatment of CPS in dogs and
cats.6-15 Animals that survive the perioperative period have a good long-term prognosis,
while animals treated with medical therapy undergo progressive hepatic failure.6'13
Although surgical attenuation of CPS has been performed in thousands of dogs over the
past 3 decades, refinements in surgical techniques continue to evolve.
For many years, CPS were treated via ligation of the aberrant vessel with suture
material. Unfortunately, acute complete attenuation of the shunt at the time of surgery
often caused intolerable portal hypertension, and as a result, only 17% (4 of 24) to 55%
(8 of 15) of single extrahepatic CPS, and 13% (15 of 114) of intrahepatic CPS could be
completely ligated during a single surgical procedure.16-24 It was felt that the poorly
developed hepatoportal vascular system was unable to cope with the sudden increase in
blood flow. Guidelines, based on measurement of portal pressure16,25,26 and subjective
visual criteria23 during complete occlusion of the shunt were established to determine the
amount of attenuation that could be performed during a single surgical procedure. Partial
ligation was recommended when complete ligation could not be safely achieved during a
Despite causing an early clinical improvement, it became evident that partial
ligation of CPS was associated with a less favorable long-term prognosis than complete
ligation.8'14'16'27 Adverse sequellae manifested by the central nervous system,
gastrointestinal tract, urinary tract, and cardiovascular dysfunction were frequently
associated with partial occlusion of CPS.8 Complete attenuation, as compared to partial
attenuation, was superior in obtaining resolution of clinical signs and improving long-
Accordingly, some authors advocated performing a second surgery to attain
complete occlusion of CPS when portal pressures precluded acute ligation during the
initial procedure.6 It was suggested that if attenuation was performed over a period of 4
to 6 weeks, the liver would adapt to the redirection of normal portal blood flow.6'7'29
Multiple surgeries, however, introduced increased risk to the animal, greater morbidity,
and increased costs.
In an attempt to avoid multiple anesthetic and surgical, most recent efforts have
focused on identification of a surgical implant that could be used to produce gradual, and
complete occlusion of CPS after a single surgical procedure. The ideal implant would
produce reliable progressive venous occlusion, would be biocompatible, inexpensive,
simple to apply, associated with minimal patient morbidity, and would be reversible if
postoperative portal hypertension were to occur. Methods of gradual vascular occlusion
that have been employed in the clinical treatment of CPS include: partial ligation with
silk suture material,6'8 ameroid constrictors (AC),9 cellophane banding,10'29 and
intravascular deployment of thrombogenic coils (TC).12 The silicone hydraulic occluder
(HO) has also been used to produce gradual venous occlusion in experimental models,
but its use in the treatment of CPS has not yet been reported. The purpose of this chapter
is to provide a comprehensive review of the available experimental and clinical studies
related to each method of gradual venous occlusion and to discuss potential advantages
and disadvantages of each technique in the treatment of CPS in dogs and cats.
Silk has long been utilized as a suture material for vascular ligation due to its
excellent handling properties. This natural, multifilament suture is classified as a non-
absorbable material due to its prolonged retention in tissues.30 Silk suture material is
capable of inducing an inflammatory response within a vessel wall.7 Acute
inflammation characterized by chemotaxis and degranulation of neutrophils is followed
by a more chronic response consisting of fibroblast infiltration, proliferation, and
eventual scar formation.7'31-33 It has been suggested that the inflammatory response
initiated after partial vascular occlusion with a silk ligature may eventually lead to
complete vascular occlusion.7 In fact, various investigators have reported successful
complete CPS occlusion, as determined by portal scintigraphy, after partial attenuation
with silk.7'8'22'34 An experimental model using the femoral vein in dogs failed to
corroborate these clinical observations.35 Of the four methods of gradual venous
occlusion examined in that study, partial occlusion with silk was the sole technique that
failed to produce appreciable venous attenuation over a 6-week period.35
These discrepancies associated with the use of silk for progressive vascular
occlusion may be related to several different factors affecting the degree of perivascular
inflammation. The primary factor implicated in the variable rate of occlusion after partial
ligation with silk suture is the degree of inflammation produced by perivascular
dissection. Delayed vascular occlusion may be the result of excessive manipulation of
the vessel during surgery, and the resulting postoperative vasospasm or swelling of the
vessel wall.7 Surgically induced vasculitis may induce a transient narrowing of the
vessel and decrease in blood flow during the immediate perioperative period.7'27'35 Some
authors have postulated that variations in the degree of fibrosis associated with the use of
silk ligatures are directly related to suture diameter.36 Hunt and Hughes, however, found
no significant difference in the proportion of animals experiencing complete occlusion
after partial attenuation with different gauges of silk suture.27 Another potential
explanation for improved clinical status following partial occlusion with silk sutures is
that occlusion is not actually produced by progressive attenuation of the vessel lumen, but
rather by alterations in blood flow that predispose the shunt vessel to thrombosis. For
example, partial re-direction of blood flow through the portal system may encourage
hepatic regeneration and a reduction in portal vascular resistance to the point that
shunting of blood through the attenuated vessel is no longer favored.35 The resulting
decrease in blood flow through the shunt may produce closure or thrombosis of the vessel
secondary to stasis.
Advantages of partial CPS ligation with silk suture include widespread
availability, and low cost of the materials. Silk is adaptable to both extravascular,
transportal, or transcaval approaches to portosystemic shunt ligation,21 and is thus
applicable for both extrahepatic and intrahepatic CPS. Disadvantages include the
requirement for measurement of intraoperative portal pressures during CPS ligation and
the potential need for a second surgical procedure to achieve complete ligation in many
dogs with CPS. Although partial occlusion of CPS with silk suture appears to provide
favorable short-term clinical results in select cases,8 the ability of silk to produce
complete attenuation in a predictable manner has yet to be demonstrated.35 The use of
silk suture material in the treatment of extrahepatic CPS in dogs has, for the most part,
been replaced by the application of ameroid constrictors or cellophane bands that can be
placed during a single surgical procedure without measurement of portal pressures.
Intrahepatic PSS, however, are still treated using partial ligation with silk when they are
not amenable to placement of larger surgical implants.
Ameroid, first described by Berman et al in the mid 1950's, is a hygroscopic
substance consisting of compressed casein that is hardened via a formalin curing
process.1 The material expands after exposure to water. Ameroid constrictors were
originally designed with the intent of producing physical compression of blood vessels,
resulting in gradual vascular occlusion. The first ameroid constrictors consisted of large
rings with eccentrically placed lumens. The eccentricity allowed one side of the ring to
expand faster than the other, effectively filling the lumen and occluding the
circumscribed vessel.2 A second device utilizing ameroid was later described for slow
progressive vascular occlusion, specifically as a method of banding the pulmonary artery
of infants.37 This implant relied on an ameroid cylinder that acted as a piston as it
expanded, pushing an inert plastic anvil against an adjustable polytetrafluoroethylene
band and causing constriction of entrapped tissues.37 The AC being used today for
treatment of CPS contain an inner ring of ameroid within an outer stainless steel sheath
(Research Instruments NW, Inc. Sweet Home, OR) A small cylindrical piece of ameroid
functions as a key, which closes the ring once it has been placed around the vessel.
Figure 1. Photograph of a 5mm ameroid constrictor and key. An outer steel sheath
(arrow) contains an inner ring of casein (dashed arrow) which forms a
complete ring when the cylindrical key (arrow head) is inserted.
As the ameroid absorbs fluid, expansion is directed centrally by the stainless steel
ring, resulting in a decrease in internal luminal area. Although size, shape, and stiffness
of material encasing the ameroid, as well as the type and temperature of the surrounding
fluid alter the rate of closure, a relatively consistent expansion pattern has been
demonstrated, occurring most rapidly in the first 3-14 days following implantation, and
Ameroid constrictors have been used extensively in a variety of research
applications for over 50 years, including experimental models of coronary stenosis,1'2'40
myocardial ischemia,41 limb ischemia,42 development of coronary collateral circulation,43
esophageal varices,44 and portal hypertension.45 In 1996, Vogt et al reported the use of
AC for treatment of single extrahepatic CPS in dogs.9 In Vogt's original study, complete
attenuation of the CPS was documented in 10 of 14 animals using technetium
scintigraphy. Complete occlusion was identified at times ranging from 30 to 210 days.
There was, however, a 14% mortality rate due to portal hypertension, and 17% of animals
developed multiple extrahepatic (MEH) portosystemic shunts, presumably due to portal
hypertension that occurred during occlusion of the CPS. A subsequent study reviewed
the results in 111 dogs with single extrahepatic CPS that were treated with AC.46
Perioperative mortality was only 7%, but persistence of portosystemic shunting was
documented in 17.5% of dogs at 8 weeks after surgery using trans-rectal technetium
scintigraphy.46 AC have also been utilized for treatment of intrahepatic CPS.18'24'47 In
one study employing a jugular venograft technique, extrahepatic portocaval shunts were
created, which allowed complete ligation of the intrahepatic CPS while facilitating AC
placement on the newly created shunt.18 Nine of 10 dogs in that report survived surgery,
although 5 of those experienced postoperative complications.18 Five of 8 dogs continued
to have portosystemic shunting 8 to 10 weeks following surgery, and in 4 of those 5 that
were further examined, evidence of MEH portosystemic shunts was found.
Results of these early clinical and experimental reports suggested that some of the
original assumptions regarding the mechanism of vascular occlusion produced by the AC
were incorrect. First, a number of investigators have noted that venous occlusion occurred
prior to physical closure of the ameroid constrictor lumen.9'35'48'49 Adin et al
demonstrated that the lumen of the AC decreases by only 32% after 6 weeks following
implantation in the peritoneal cavity of rats, suggesting that implant associated
inflammation and fibrosis, rather than hygroscopic expansion of the casein was primarily
responsible for vascular occlusion produced by the AC.39 Second, recent studies have
suggested that there is significant variability in the rate of venous occlusion produced by
the AC.9'10'35'48 Vascular occlusion may be dependent on several factors, including
vessel diameter, AC size, implantation site, and the individual patient's inflammatory
response to the AC. The variable nature of the inflammatory response to the AC in
individual animals may be the cause of inconsistency in the reported venous occlusion
rates, with times to complete occlusion in different models ranging from 6 to 210
The occurrence of MEH shunts in a number of animals after application of the AC
has led to suspicion that the AC may cause vascular occlusion too rapidly, resulting in a
state of subclinical portal hypertension and MEH shunts.9'18 Post-operative development
of MEH portosystemic shunts has been reported in 40% of dogs with intrahepatic CPS18
and 17% of dogs with extrahepatic CPS9 treated with AC. Data obtained in a recent
experimental study in which ACs were applied to the iliac veins of dogs actually
suggested that ACs do not cause a gradual decrease in venous blood flow in this model.48
Rather, delayed acute thrombosis occurred in 3 of 6 veins within 6 days following
application. This phenomenon of acute thrombosis may explain the development of
ascites and MEH shunts in dogs after application of ACs to CPS.9'18'46 A device that
would produce a more gradual venous occlusion may avoid subclinical portal
hypertension and produce complete occlusion without the formation of MEH shunts.
The AC has proven to be extremely beneficial in the treatment of animals with
single extrahepatic CPS. Use of the AC has been shown to decrease overall surgery time,
as well as the overall cost of performing CPS surgery when compared to conventional
ligation procedures.24,50 Application of ACs negates the need to measure portal
pressures during surgery, external manipulation is not required after surgery, and surgery
is limited to one anesthetic event. Use of the AC has also been associated with a
decreased intraoperative and postoperative complication rate, especially when compared
to partial ligation techniques.46'50 Disadvantages of AC use, however, include increased
dissection necessary for application, difficulty in passing the bulky AC around thin-
walled shunting vessels, variable rates of vascular occlusion, difficulty in using the AC
for intrahepatic shunts, development of MEH shunts, and the potential for the AC to kink
the shunting vessel once in place and cause acute portal hypertension.
Extravascular placement of cellophane bands produces a chronic foreign body
reaction similar to that described with the application of ACs, and is thought to be
responsible for the progressive vascular occlusion produced by this technique.35 Similar
to silk and the AC, this inflammatory-mediated method of vascular narrowing would be
expected to cause inter-patient variability. Factors including the initial internal diameter
of the cellophane band, the initial degree of vessel attenuation, the width of the
cellophane band, and the patient's natural inflammatory response may all contribute to
the degree and rate at which occlusion occurs. When the effects of cellophane banding
were examined in a canine femoral vein model, it was found that the bands produced
progressive, but not complete, vascular occlusion in 5 of 6 veins.35 Maximal changes
were seen after 14 days, and were attributed to acute inflammation.
Historically, the use of cellophane to induce fibrosis and gradual vascular occlusion
has been described to study experimentally induced portal hypertension, esophageal
varices, and coarctation of the aorta.51'52,53 Breznock first suggested the use of
cellophane for treatment of CPS in 1979;54 however, it wasn't until 1990 that Harrari et al
first reported its use in a dog with a naturally occurring CPS.10 In that case, a cellophane
band was placed around a portoazygous CPS, initially attenuating the shunt vessel
diameter by approximately 50%. Clinical signs resolved and complete occlusion of the
shunt vessel was confirmed five weeks postoperatively using mesenteric portography.
More recently, Youmans and Hunt have investigated the use of cellophane in a series of
clinical and experimental studies.29'35 In a study in which cellophane bands were used in
11 dogs with extrahepatic CPS, clinical improvement was noted in all dogs, and complete
occlusion was suspected in 10 of dogs based on the results of ultrasonography and liver
function tests.29 Shunts were all initially attenuated to an internal diameter of 2.5 or 3
mm. Variable closure rates suggested that 3 mm may be the maximum internal diameter
that would progress to complete occlusion.29 Cellophane bands have also been used with
variable success in dogs with intrahepatic CPS.55,56 In one report, banding of
intrahepatic CPS in two Irish wolfhounds to an internal diameter of 6 mm and 8 mm was
performed.56 One dog was euthanized postoperatively due to the development of
seizures which were refractory to treatment. Clinical signs resolved in the second dog,
though nuclear scintigraphy performed 6 months after surgery suggested that complete
occlusion was not achieved. 56
It appears that the cellophane band produces slower occlusion than does the AC,
although when used with an internal diameter greater than 3 mm, its ability to produce
complete occlusion is questionable.35 In vessels greater than 3 mm, application of the
cellophane band may only produce partial attenuation. Dissection prior to application of
cellophane bands is less extensive than that required for placement of ACs. 29 The bands
are also less likely to produce mechanical distortion, or kinking of the shunting vessels as
is the AC.29
Advantages of cellophane banding in its application to the treatment of CPS
include its ease of application, low cost, and ready availability. A single surgery is
required, intraoperative portal pressure measurements are not necessary, and no external
manipulation is needed after surgery. Unfortunately, variable rates of vascular occlusion
and questionable efficacy on large vessels may limit its application in some clinical cases.
Transvascular embolization involves the placement of thrombogenic material into a
vessel lumen via catheter access and guide wires. Occlusion of the vessel lumen may
occur as a thrombus develops on and around the embolizing materials.12 Coils are
constructed from a flexible metallic strip (stainless steel or platinum) and multiple
polyester fibers that can be deployed through a catheter to stimulate thrombosis (Figure
Figure 2. Photograph of two intravascular thrombogenic coils.
Use of thrombogenic coils (TC) for acute arterial occlusion in the treatment of
arterial aneurysm57'58 and patent ductus arteriosus59,60 is well established. Advances in
technology and interventional radiology have also stimulated an interest in the use of TC
for gradual venous occlusion. Initial attempts at transvenous coil embolization of
intrahepatic CPS were described in at least 3 individual case reports, 2 canine11'12 and 1
feline.61 All animals were clinically improved, although complications including coil
migration and the need for multiple procedures were noted. Two11 and 412 separate
embolization procedures failed to produce complete shunt occlusion in either of the dogs.
The single cat was treated successfully, although complete occlusion was demonstrated
after only 48 hours following embolization.61 A subsequent case series involving 7 dogs
with extrahepatic CPS and 3 with intrahepatic CPS revealed other problems associated
with the use of TC.62 Although clinical improvement was noted in all 3 dogs with
intrahepatic, and in 4 of 7 dogs with extrahepatic CPS, 3 dogs died due to migration of
the coils or acute portal hypertension and other serious complications such as coil
migration, severe hemolysis, and development of MEH occurred in the surviving dogs.
Complete occlusion, as determined by nuclear scintigraphy, was not evident in any of the
surviving dogs. Based on this study, coil migration was found to be a major problem,
particularly in the treatment of extrahepatic shunts. The authors hypothesized that this
may be due the lack of supporting liver parenchyma around the shunt vessel. 62 Results
of an experimental study in which TC were placed in the femoral veins of dogs further
questioned the efficacy of this technique at producing gradual venous occlusion.35 In this
study, TC produced complete occlusion between 3 and 7 days in only 50% (4 of 8) of
canine femoral veins. Recanalization occurred shortly thereafter in 3 of the 4 (75%).35
Recently the TC technique has been modified to address problems with coil
migration. Two recent manuscripts describe the deployment of a caval wall stent at the
junction of the shunt with the vena cava, preventing the coils placed in the shunt from
migrating out of the shunt vessel.63'64 Using this technique shunt attenuation was
achieved in 13 of 14 dogs with intrahepatic CPS, and 46% (6/13) of the dogs were
ultimately deemed clinically normal.64 Unfortunately, multiple interventions were
typically required to achieve staged occlusion of the CPS. 64
Use of TC in the treatment of CPS has also been described in a small number of
human patients.5'65 Although CPS is diagnosed infrequently in people and limited
clinical data is available, it has been suggested that transcatheter embolization is
preferable to surgical ligation or resection.5
Potential advantages of transvenous coil embolization include a shorter anesthetic
period, a less invasive procedure, faster recovery and shorter hospital stay.12
Disadvantages include the possibility of coil migration, the need for specialized
instrumentation and training, multiple anesthetic episodes, the inability to obtain tissue
samples for histopathological evaluation, the inability to visualize and subjectively
evaluate the abdominal contents during shunt attenuation, and the potential for rapid
thrombosis and development of portal hypertension. 12,35,61
Clearly, initial experiences with TC have generated mixed results. While a number
of dogs have improved clinically following treatment, multiple complications have also
been documented. Unfortunately, the ability to gradually produce complete occlusion
often requires multiple anesthetic episodes and embolization procedures, and the
potential for rapid thrombosis and complete occlusion may limit TC therapy to those
animals that are amenable to complete occlusion at the time of surgery. In such cases, as
determined by pre and post-occlusion portal pressure measurements, TC may prove to be
Hydraulic occluders consist of an inflatable silicone membrane within a polyester-
reinforced, stretch-resistant cuff (Figure 3). The HO is placed around a vessel and the
ring is closed by placing suture material through holes molded at each end of the cuff
Inflation of the HO can be controlled percutaneously through injections of fluid into a
subcutaneous injection port that is attached to the balloon via a length of actuating tubing.
A variety of fluids have been used to fill the HO balloon, including normal saline,66
hypertonic saline,67 water,68 radiographic contrast solutions,69 indocyanine green dye,70
and dextrose solution.71 Despite the ability of some solutions to diffuse across silicone
membranes, all of these solutions have been used successfully, and an optimal filling
solution has yet to be established.
Figure 3: Photograph of a 2 mm internal diameter hydraulic occluder. Identified are the
occluder (arrow head) and attached subcutaneous injection port (arrow).
Hydraulic occluders are most commonly made of medical grade silicone in order
to minimize inflammation and foreign body reaction. After their introduction in the
1960's, a variety of designs were initially reported, although all were based on similar
principles.68,72 Initially, the HO was used in acute applications to completely occlude a
blood vessel, allowing calibration of electromagnetic blood flow probes to zero flow.68
However, in 1969, Bishop and Cole reported use of the HO to maintain partial occlusion
of the pulmonary artery in dogs for extended periods of time in a model of progressive
pulmonic stenosis.73 Subsequently, investigators have reported chronic implantation of
silicone HO for up to 32 months, and documented retained adjustability up to 12 months
Hydraulic occluders are used extensively in a variety of research applications,
including the investigation of blood flow characteristics, organ ischemia and reperfusion,
and as experimental pulmonary arterial bands.66'67'69'70'74'75 Similar devices have also
been used clinically for more than two decades in non-vascular applications such as
artificial urethral sphincters (AMS Sphincter 800 TM Urinary Control System, American
Medical Systems, Minnetonka, MN), artificial anal sphincters (Acticon TM
Neosphincter, American Medical Systems, Minnetonka, MN) and as adjustable gastric
bands (Lap-Band, Inamed Health, Santa Barbara, CA).
Despite being used extensively in models of arterial occlusion, experience with
the use of the HO for venous occlusion is limited.71'76'77 A recent study by Peacock et al
reported the successful occlusion of the caudal vena cava in 7 dogs over a period of two
weeks using a HO in a model for development of collateral circulation prior to
adrenalectomy.71 The HO was easy to place surgically and manipulate postoperatively,
produced and maintained complete occlusion, and was well tolerated by all dogs,.
Inflammation surrounding the HOs was characterized as being consistently mild.71
The HO has several theoretical advantageous attributes for the treatment of CPS
including the ability to be placed at a single surgery without measurement of portal
pressure, to produce gradual and total vascular occlusion, and to be reversed if necessary.
Occlusion is not mediated by inflammation, but by physical compression of the vessel
which may be percutaneously controlled. Cost of the HO is similar to that of the ameroid
constrictor and application of the device may be performed without specialized
Use of the HO as a treatment for CPS has not yet been reported, although research
investigating this application is well warranted. Prior to application of the HO in dogs
with CPS, however, a study investigating the gradual occlusive abilities of the HO and
resulting changes in blood flow through a venous structure is warranted.
Partial ligation with silk suture, ameroid constrictors, cellophane bands,
intravascular thrombogenic coils, and hydraulic occluders have all been investigated as
potential methods of producing gradual vascular occlusion. While all of these methods
have been successful in achieving vascular occlusion, although the predictability, rate of
occlusion, adjustability, and ability to achieve complete occlusion appears to vary widely
amongst techniques. Alhtough research into methods of gradual vascular occlusion has
clearly improved the perioperative management of CPS in dogs, the complication rates
continue to be unacceptably high, and further study is indicated.
A PERCUTANEOUSLY CONTROLLED HYDRAULIC OCCLUDER IN A RAT
MODEL OF GRADUAL VENOUS OCCLUSION.
Congenital portosystemic shunts (CPS) are the second most common congenital
cardiovascular malformation in dogs78, and consist of anomalous vessels that divert blood
from the portal to systemic vasculature, effectively bypassing the liver.13'14'26 As a
result, byproducts that are normally metabolized in the liver accumulate within the
systemic circulation. This results in a variety of clinical abnormalities, including those
associated with hepatic encephalopathy.15 Hepatotrophic substances are similarly
diverted, resulting in liver hypoplasia, atrophy, and eventual hepatic failure.6'15
Complete surgical ligation of CPS has been recommended to achieve resolution of
clinical abnormalities and improve long-term survival;6'13'14'23'25'28 however, acute intra-
operative complete occlusion of CPS may cause intolerable portal hypertension. In
previous reports, complete ligation was achieved during a single surgical procedure in
only 17% (4 of 24)19 to 55% (8 of 15)23 of single extrahepatic CPS and 13% (15 of
114)17-24 of intrahepatic CPS (CIPS). Some investigators have advocated staged,
incremental, partial ligation to achieve complete occlusion if portal hypertension
precludes complete ligation during the initial surgery.6 Sequential ligation requires a
second anesthetic episode and surgery, resulting in increased hospitalization and
increased cost. In response to these concerns, several investigators have sought to
develop a device capable of producing gradual occlusion that can be implanted during a
single surgical episode.
An ideal device would be applied during a single surgical procedure, induce
minimal tissue reaction, allow controlled and gradual vascular attenuation, produce
eventual complete occlusion and allow reversal of attenuation should portal hypertension
occur postoperatively. Clinical application of cellophane bands,10 ameroid constrictors,9
thrombogenic intravascular coils,11'12 and partial ligation with silk suture6-8 have been
reported, though the ideal method of gradual venous occlusion has yet to be identified. In
fact, experimental studies have failed to support the ability of any of these methods to
produce gradual venous occlusion.35'48 In one study using a dog femoral vein model,
cellophane, partial ligation with silk and thrombogenic coils failed to produce reliable
venous occlusion during a 6-week period, while ameroid constrictors produced rapid
occlusion within 14 days following surgery.35 A more recent study that measured blood
flow following application of ameroid constrictors to the common iliac vein in dogs
demonstrated that ameroid constrictors did not actually cause gradual occlusion in this
model; rather, ameroid constrictors caused delayed acute thrombosis at 5-6 days after
application, presumably secondary to an inflammatory response.48 In light of these
findings, we have been seeking to develop a technique that will allow percutaneously-
controlled gradual occlusion of CPS. The purpose of this study was to evaluate the
effects of a percutaneously controlled silicone hydraulic occluder (HO) on vena caval
blood flow in a rat model of gradual venous occlusion.
Materials and Methods
This study was performed with the approval of the University of Florida
Institutional Animal Care and Use Committee. Ten male Sprague-Dawley rats, weighing
between 350 and 374 g, were utilized.
Two-millimeter internal diameter hydraulic occluders (OC2, In Vivo Metric,
Healdsburg, CA) were utilized throughout the study (Figure 3). Initially, the relationship
between fluid volume injected and % occlusion for the 2 mm HO was determined. The
silicone actuating tubing from the HO was cut to 12 cm, and then attached to a
subcutaneous injection port (ROP-3.5, Access Technologies, Skokie, IL) using two 3-0
polypropylene sutures (Surgilene, Davis and Geck, Wayne, N.J.). The incomplete ring of
a non-inflated HO was closed by tying a strand of 3-0 polypropylene suture material
through the pre-existing holes (Figure 3). A 22-gauge, non-coring needle (22-ga x 34"
Huber point needle, Access Technologies, Skokie, IL) was used for all injections to
minimize damage to the silicone membrane of the injection port. Physiologic saline was
injected and aspirated repeatedly to remove air bubbles from the HO. The non-inflated
occluder was placed on a flat bed scanner (HP Scanjet 4470 C, Hewlett-Packard Co. Palo
Alto, CA) and digital images were obtained. The occluder was then inflated
incrementally with 0.01 mL injections of saline. Digital images were obtained after each
0.01 mL injection until the internal lumen of the occluder was completely obliterated by
the inflated balloon. After completion of 3 trials, the internal area of the occluder lumen
was traced for each of the stored digital images and luminal areas (LA) were measured
using a computer software program (Image J 1.27Z, National Institutes of Health, USA).
Percent occlusion after each injection was calculated based on the following formula:
[% occlusion = LA non-inflated occluder (cm2) LA inflated occluder (cm2) x 100]+
[LA of the non-inflated occluder (cm2)].79
Data were graphed for volume injected vs. % occlusion (Figure 4) so that
investigators could achieve planned stages of luminal occlusion by injection of
incremental volumes of fluid into the subcutaneous injection port.
0 0.01 0.02 0.03 0.04 0.05 0.0 0.07 0.08 0.0
Figure 4. Graphic representation of changes in % occlusion associated with volume
injected in 2 mm hydraulic occluders. Included are 3 individual trials and an
overall mean, as well as a regression line (bold). Fluid was injected in 0.01
mL increments, and the corresponding occlusion of the 2 mm hydraulic
occluder was measured. Volume injected was highly correlated with
percentage occlusion (R2= 0.9619).
Transit time ultrasound flow probes (MC 2.5 PSS, Transonic Systems Inc., Ithaca,
NY) were calibrated according to the manufacturer's directions. Briefly, the probes were
submerged in a beaker of distilled water, at which point zero flow was recorded.
Rats were anesthetized using chamber induction with 4% isoflurane (IsoFlo, Abbott
Laboratories, Chicago, IL) and 100% oxygen, and were maintained on 1.5-2% isoflurane
delivered via a mask for the remainder of the surgical procedure. Transit time ultrasound
flow probes and HOs (OC2, In Vivo Metric, Healdsburg, CA) were sterilized with
ethylene oxide prior to use. High molecular weight, isotonic sodium hyaluronate
(Hylartin V, Pharmacia & Upjohn Co., Kalamazoo, MI) was selected as a filling solution
to minimize diffusion of substances across the silicone membrane of the HO during long-
term implantation. The actuating tubing of the HO was cut to 12 cm and filled with
sodium hyaluronate using a 22-ga intravenous catheter (Intracath, Becton Dickinson,
Sandy, UT) in a retrograde manner to minimize air entrapment within the HO. Actuating
tubing was attached to a subcutaneous injection port using two 3-0 polypropylene
The ventral abdomen, left flank, and dorsum of each rat was shaved and aseptically
prepared for surgery. The rats were then placed in dorsal recumbency on a sterile drape.
A 6 cm ventral midline celiotomy was made, extending caudally from the xyphoid
process. The small intestine was retracted, providing exposure of the caudal vena cava
(CVC) and abdominal aorta. The right gonadal vein was ligated and transected near the
CVC. Blunt dissection was used to separate and isolate the CVC from the aorta just
caudal to the left renal vein. All venous branches of the CVC between the left renal vein
and the iliolumbar veins were ligated and divided in an attempt to create a 2 cm segment
of CVC that was free of all collateral vasculature. A flow probe was placed around the
caudal aspect of isolated CVC. The HO was passed around the CVC just cranial to the
flow probe, and the ring of the HO was closed using 3-0 polypropylene suture material.
The HO was then secured to the flow probe and to the hypaxial musculature with several
interrupted 3-0 polypropylene sutures in an attempt to avoid kinking of the vena cava
between the flow probe and the HO (Figure 5).
Figure 5. Intraoperative view of the instrumented caudal vena cava. A midline
celiotomy has been performed and the small intestines have been reflected
with the aid of a sterile cotton-tipped applicator (arrow head) to increase
exposure. The hydraulic occluder (A) is located cranial to the perivascular
flow probe (B) on the caudal vena cava (arrow).
After replacing the intestines into the abdominal cavity, the flow probe cable and
HO actuating tubing were exited through the incision and the linea alba was closed using
4-0 polygalactin 910 (Vicryl, Ethicon Inc., Somerville, NJ) in a simple continuous suture
pattern. The flow probe cable and HO actuating tubing were then tunneled
subcutaneously on the left side of the abdomen and exited at the dorsal midline, just
caudal to the scapulae. An injection port was attached to the actuating tubing and secured
in a subcutaneous location. The dorsal incision was closed over the injection port using
several interrupted sutures, incorporating and exteriorizing the flow probe cable and
connector in the process. The flow probe connector was anchored in a rigid plastic cuff
(Transonic Systems Inc., Ithaca, NY), which was finally secured to the skin using several
interrupted sutures. For the first 24 hours after surgery, rats were monitored every 8
hours for evidence of discomfort, during which time buprenorphine (Buprenex Injectable,
Reckitt Benckiser Pharmaceuticals Inc., Richmond, VA [0.1 mg/kg, SC, q 8-12 hrs]) was
administered for analgesia.
Blood Flow Measurements
Rats were randomly assigned to either group I (n = 6 rats) or group II (n= 4).
Beginning 6 to 8 days after surgery, all rats were anesthetized once weekly using the
protocol described above. With the rats in left lateral recumbency, flow probe connectors
were attached to a flowmeter (TS420, Transonic Systems Inc., Ithaca, N.Y.) and data
acquisition system (IOX 18.104.22.168, EMKA Technologies, Falls Church, VA), allowing data
from the flowmeter to be captured and analyzed on a desktop computer. Rats were
allowed 1 minute to stabilize after induction of anesthesia, then automatic recording of
blood flow every ten seconds for a period of three minutes was initiated. Rats in group II
were weighed, and then recovered without any further intervention. Rats in group I
received 6 weekly injections of 0.02 mL to cause incremental occlusion of the HO, with a
final injection of 0.04 mL performed at 7 weeks after implantation to ensure complete
occlusion. After injections into the subcutaneous injection ports, group I rats were
weighed, and then recovered from anesthesia. Blood flow measurements recorded over
the three-minute time periods were averaged and expressed as mL/min/kg body weight.
Eight weeks after surgery, blood flow measurements were performed as described
above, and rats were euthanized while under anesthesia by intracardiac administration of
pentobarbital (Beuthanasia-D, Schering-Plough Animal Health Corp., Kenilworth, N.J.).
Necropsy was performed immediately and the CVC was grossly inspected for evidence
of occlusion or the development of collateral circulation before being removed en-bloc
with the implants. In the 2 rats with grossly patent vena cavae at the time of necropsy,
catheterization of the caudal vena cava was performed and radiographs were obtained
during infusion of water soluble contrast medium (Omnipaque, Amersham Health Inc.,
Princeton, NJ) to illustrate patency of the vessel. The flow probe was carefully removed
from the CVC of all rats, and the remaining tissue sample and HO were fixed in 10%
formalin for histopathology.
After fixation, all implanted materials and sutures were removed. Representative 5
micron sections of the vena cava were taken at the previous location of the hydraulic
occluder and were routinely processed and stained with hematoxylin and eosin.
Repeated measures analysis of variance was used to compare mean blood flow
measurements over time for group I animals. Probability values < 0.05 were considered
significant. Post hoc analysis using a Fischer's protected least squared difference test
was performed to compare weekly mean blood flow measurements to baseline values in
group I animals.
Volume Injected vs. % Occlusion
Percent occlusion of the 2 mm HO was highly correlated with the volume of saline
injected into the infusion port (R2=0.9619, Figure 4). Complete occlusion of the HO was
achieved after injecting 0.09 mL of fluid in all three trials.
Surgical placement of the implants was technically difficult due to the small size of
the rat CVC and the relatively large, heavy implants. The presence of the flow probe and
HO caused kinking of the vena cava at the time of surgery in several rats, despite the
placement of multiple tacking sutures between the occluder, the flow probe, and the
hypaxial musculature. All 10 rats survived the surgical procedure and eight-week
postoperative period and showed no evidence of major postoperative complications.
Weekly Blood Flow Measurements
Signal quality from the flow probe was adequate to obtain readings in all rats at all
times. Blood flow could not be detected at the 1-week time period in 1 group I and 2
group II rats. Occlusion of the CVC secondary to kinking of the vessel between the flow
probe and HO was confirmed by necropsy examination in each of these 3 rats. Thus,
these 3 rats were removed from subsequent analysis of mean blood flow measurements.
The total volume of sodium hyaluronate required to produce occlusion (0 blood flow)
was 0.06 mL (0.02 mL for 3 weeks) for one rat in group I. The remaining 4 rats required
0.17 mL (0.02 mL for 6 weeks, followed by 0.05 mL at 7 weeks). This exceeded the
initially predicted volume of 0.09 mL as determined during the acute inflation of the 2
mm HO prior to the experiment (Figure 4). Weekly blood flow measurements for
individual group I rats are represented in figure 6A. Zero blood flow was detected at 8
weeks in 4 of the 5 rats in group I. The residual flow recorded in the remaining rat was
later discovered to be due to the presence of a collateral vessel entering the vena cava
between the HO and flow probe. Mean blood flow for group I rats (n = 5) decreased
significantly over time from 40.71 (+/- 24.32) mL/ kg /min at day 1 to 4.68 (+/- 8.41)
mL/kg/min at 8 weeks (Table 1, P = 0.0094, Power = 0.91). Post hoc analysis using a
Fischer's protected least squares difference test was performed and demonstrated a
significant decrease in group I mean blood flow from baseline at 4, 5, 6, 7, and 8 weeks
(P = 0.0352, 0.0342, 0.0434, 0.0050, and 0.0001, respectively). Both group II rats
maintained blood flow at all times during the study period (Figure 6B).
a 70 -
i 20 -
1 2 3 4 5 6 7 8
2 80 -
o 30 -
1 2 3 4 5 6 7 8
Figure 6. Individual weekly vena caval blood flow measurements. Measurements are
normalized to mL/min/kg body weight. A) Group I rats (n=5). B) Group II
Table 1. Summary of weekly mean blood flow for group I rats.*
Mean Blood Standard Standard Error Post Hoc
Flow Deviation Analysis (P-
Week 1 40.705 24.316 10.875 ---
Week 2 30.431 12.737 5.696 0.2172
Week 3 28.949 12.235 5.472 0.1596
Week 4 22.697 16.194 7.242 *0.0352
Week 5 22.582 16.737 7.485 *0.0342
Week 6 23.486 14.262 6.378 *0.0434
Week 7 15.881 10.119 4.561 *0.0050
Week 8 4.680 8.411 3.762 *0.0001
*Weekly mean + SD and SE vena caval blood flow measurements are normalized to
mL/min/kg body weight and are depicted for group I (n = 5 rats) over the 8-
week study period. Blood flow decreased significantly over time (P = 0.0094,
Power = 0.91) during gradual inflation of the hydraulic occluder. Significant
changes (denoted by asterisks) from baseline blood flow were noted at 4, 5, 6,
7, and 8 weeks (P = 0.0352, 0.0342, 0.0434, 0.0050, and 0.0001,
Necropsy examination revealed occlusion of the vena cava secondary to kinking
between the flow probe and the vena cava in 1 group I and 2 group II rats. Complete
occlusion of the CVC at the HO site was confirmed in the 5 rats from group I. Collateral
vascular development caudal to the flow probe, consisting of multiple small vessels
between the CVC and iliolumbar veins, was identified in all group I rats. A small vessel
was identified entering the vena cava between the FP and HO in 1 group I rat, allowing
blood to pass through the FP and bypass the HO. The CVC remained patent in two group
II rats, as documented via contrast radiography (Figure 5). Adhesions between the
momentum and the HO were present in all cases.
Figure 7. Radiographic image documenting caval patency. The image was obtained
after injection of 0.5 mis of water soluble contrast medium (arrow) into the
caudal vena cava of a rat in group II, demonstrating patency at the level of the
hydraulic occluder (arrow head) 8-weeks after surgery. The perivascular flow
probe (A) is seen caudal to the hydraulic occluder.
Sections of vena cava had varying degrees of fibrosis throughout the vessel wall
and contained organizing thrombi leading to variable degrees of occlusion. Areas of
mural fibrosis were infiltrated by small numbers of lymphocytes, plasma cells and
neutrophils. The negative spaces left by the removed materials was surrounded by dense
bands of fibrous connective tissue with an internal layer of small to moderate numbers of
lymphocytes, plasma cells, macrophages and fewer neutrophils and eosinophils. A
proliferation of small vessels surrounded the areas of fibrosis in some sections.
Fragments of suture material were surrounded by granulomatous to pyogranulomatous
To the authors' knowledge this is the first study to successfully document a gradual
decrease in venous blood flow after application of an occlusive device. Previous
investigators have used contrast radiography,9'29 clinical signs,6 portal scintigraphy,9 and
serum biochemical indices9'29 to confirm effective occlusion of CPS. Although these
methods are effective in confirming functional occlusion of CPS, they are unable to
characterize whether vascular occlusion was gradual or acute. In fact, aside from the
current manuscript, only 1 other study has been performed using direct measurement of
venous blood flow after application of a device intended to produce gradual vascular
occlusion.48 Interestingly, the results of that study have called into question the ability of
the AC to produce gradual venous occlusion, showing instead that delayed, acute
occlusion occurred only 6 days after application of an ameroid constrictor to the common
iliac veins in normal dogs.48 The clinical implications of this information are substantial,
suggesting that the methods used for evaluation of vascular occlusive devices may
require further evaluation and that surgical techniques that have been validated using
indirect methods may not actually produce the gradual vascular occlusion that would be
advantageous for the treatment of dogs with CPS.9,29
The hydraulic occluder used in the study consists of three components: an
inflatable hydraulic balloon, a variable length of actuating tubing, and a subcutaneously
placed injection port. The entire HO is made of silicone (Silastic Q7-4849 Biomedical
Grade LSR), making it suitable for long-term implantation. Silicone is widely accepted
as a non-inflammatory, biocompatible polymer and has been used extensively in surgical
implants for several decades.80 The tissue response to silicone implants is similar to that
seen with other foreign bodies, although minimal in comparison. so A fibrous tissue layer
is typically formed, encasing the implant.80 The long-term implantation of the hydraulic
occluder and the flow probe on the caudal vena cava of rats in group II did not result in a
decrease in blood flow during the eight week study period, suggesting that the alterations
in blood flow among group I rats was not due to a chronic inflammatory response, but
was instead directly related to physical compression of the vessel after inflation of the
HO. Despite acute occlusion associated with the combined use of the HO and FP which
occurred in 3 rats in the current study, we did not observe complications that could be
ascribed solely to the HO. Considering the fact that silicone implants are routinely used
in human patients with limited adverse sequelae, we feel that the HO would be suitable
for long-term implantation in dogs with CPS
There was an appreciable variation between individuals with respect to baseline
blood flow using this model. Baseline blood flow was not compared statistically between
groups due to the low numbers of group II rats with available data. When data from the
individual rats is evaluated (Figure 6A and B), it is apparent that the widest variation in
flow rates occurs during the first 3 measurements, whereas blood flow seemed to stabilize
after 3 or 4 weeks. The authors attribute this variation is due to perioperative factors such
as inflammation produced by vessel manipulation or poor early coupling between the
CVC and FP. Unfortunately, the reasons for these variations cannot be determined from
the data obtained through this study.
A variety of fluids have been used to fill HOs during acute and chronic studies,
including normal saline,66 hypertonic saline,67 water,68 radiographic contrast solutions,69
indocyanine green dye,70 and dextrose solution.71 Manufacturer recommendations at the
time of this study indicated that high molecular weight substances would be preferred for
maintenance of chronic inflation with minimal diffusion out of the silicone HO. Based
on this recommendation, sodium hyaluronate (Hylartin V, Pharmacia Animal Health) was
empirically chosen as a filling solution due to its high molecular weight, iso-osmolarity,
and safety in the instance of accidental leakage into the peritoneal cavity.81 The product
is also readily available as a sterile preparation, making it suitable for use in body
cavities. The authors' experience in using the HOs suggested that in vivo filling volumes
required to produce occlusion after chronic implantation may exceed those required for
acute occlusion, in vitro. Our goal in this study was to produce slow occlusion in greater
than 4, but less than 8 weeks. As a result, we elected to use filling volumes determined in
the in vitro study as a starting point for an injection protocol that should not have
produced compete occlusion in less than 5 of the 0.02 mL injections. Weekly injections
of 0.02 mL were performed until 7 weeks, at which time a 0.04 mL injection was
performed to ensure complete occlusion. The total volume of sodium hyaluronate
ultimately required to produce vascular occlusion after chronic implantation was greater
than that required to produce acute occlusion of the HO in 4 group I rats. This may be
explained by several different mechanisms, including diffusion of air, mechanical creep
experienced by the inflated silicone occluder, and HO size variations secondary to lack of
standardization during manufacturing. Despite attempts to eliminate the introduction of
air into the HO at the time of initial filling, the small size of the 2 mm HO and the
viscous nature of the sodium hyaluronate often resulted in the entrapment of a small air
bubble with the cuff of the HO. Unfortunately, silicone is gas permeable and it is likely
that partial deflation of the HO occurred as air bubbles trapped within the HO balloon
were displaced. Due to the fact that very small filling volumes for the 2 mm HO used in
this study were extremely small (0.09 mL), diffusion of small volumes of trapped air may
have led to significant deflation of the occluder. In the larger HOs with increased volume
capacities that would be required for use in the treatment of CPS in dogs, the inclusion of
small air bubbles at the initial filling would be less likely to ultimately affect occlusion to
the same degree. Another contributing factor may be related to creep, which is defined as
plastic deformation of a material after the application of mechanical stress.82 If creep
occurred over the 8-week study period, the silicone used in construction of the HO may
have expanded slightly, creating a larger potential filling volume than those recorded
during acute inflation prior to the study. Lastly, the injection volume to % occlusion
curve was acquired after performing three consecutive filling trials with the same HO.
Subsequent gross examination of the 10 HOs used during this study revealed obvious
differences in size. This lack of standardization is likely due to the manufacturing
process, which is performed by hand and is therefore subject to some variability. If so,
the predicted filling volume may have been slightly different between individual HOs.
Perivascular flow probes have been utilized in both acute and chronic applications
and have proven to be accurate and reliable.83'84 Volumetric flow is measured by
emitting a plane wave of ultrasound through the vessel of interest.84 A loose fit around
the vessel is acceptable, thereby making the flow probe ideal for use on easily collapsible
vessels such as veins. It has been shown that the presence of the flow probe around
venous structures in chronic applications does not appear to affect venous circulation or
hemodynamics when the device is used alone.48'84 However, the adjacent location of the
HO and the large flow probe on the rat CVC in the current study caused an obvious
tendency to kink the vessel between the large implants, predisposing the vessel to
Advantages of the rat model used in this study included low cost, genetic similarity
of animals, ease of maintenance and use of a mammal that is low on the phylogenetic
scale. The size disparity between the rat CVC and the required instrumentation may,
unfortunately, limit the use of this model in future studies. Although the interior diameter
of the HO and flow probe were appropriate for the size of the vessel, the weight and
exterior diameter of the implants were poorly adapted for use on the thin walled,
collapsible CVC. Despite attempts to stabilize the CVC, flow probe, and HO to the
surrounding musculature with several interrupted sutures, the size and close proximity of
the implants caused kinking of the vessel segment postoperatively in three of the rats.
These three rats had no flow detected at any time point during the 8-week study period.
Use of smaller implants or a larger animal model may eliminate similar complications in
subsequent studies. Alternatively, the HO could be evaluated without the presence of a
flow probe, though this would preclude the non-invasive blood flow monitoring that was
required to document gradual venous occlusion in this experiment. Despite these
limitations, results of the current study demonstrated a gradual decrease in venous blood
flow over an 8-week study period, with complete occlusion achieved in all rats.
MECHANICAL AND PHYSICAL PROPERTIES OF SILICONE HYDRAULIC
Hydraulic occluders (HOs), consist of an inflatable silicone membrane within a
polyester-reinforced, stretch-resistant cuff. Inflation of the HO can be controlled
percutaneously by injection of a filling solution into a subcutaneous injection port that is
attached to the occluder via a length of actuating tubing (Figure 3). Chronic implantation
of silicone HOs has been described in experimental models since the 1960's.73 More
recently, investigators have reported the potential for various clinical applications of the
HO in dogs, including use of the HO as an artificial urethral sphincter,79 or as a means of
producing gradual venous occlusion for the surgical treatment of adrenal neoplasia.1
Despite extensive experimental use, there is a paucity of information regarding the
reliability of silicone HOs during chronic implantation. Earlier experiences with the
HOs, (Chapter 1) have raised concerns regarding potential size differences among HOs
secondary to manufacturing. Manufacturing variability may negatively impact the ability
to accurately predict filling volumes and adjustments of the level of occlusion. In
addition, silicone is known to act as a semi-permeable membrane85-90 and the diffusion
properties of various filling solutions may affect long-term maintenance of occlusion.
The purpose of this study was to examine size differences among HOs of the same
model, diffusion of various filling solutions including: saline, sodium hyaluronate, and air
from the HO, and to describe changes in internal pressure over time in three different
models of the HO that were obtained from a single manufacturer.
Materials and Methods
Evaluation of Size Differences Among HOs
Five each of 2, 5, and 20 mm internal lumen diameter HOs (HO2, HO5 and HO20,
respectively) were obtained from a single manufacturer (DOCXS Biomedical Products
and Accessories, Ukiah, CA). The incomplete ring of each non-inflated HO was closed
by tying a strand of 3-0 polypropylene suture material (Surgilene, Davis and Geck,
Wayne, N.J.) through the pre-existing holes molded at either end of the HO cuff. HOs
were all prepared in a similar fashion, in which actuating tubing was cut to a standard
length of 6 cm. HOs were filled in a retrograde manner using a 22 ga. jugular catheter in
an attempt to eliminate air from within the HO. Injection ports (ROPAC-3.5, Access
Technologies, Skokie, IL) were also pre-filled using non-coring huber point needles
(Posi-Grip Huber Point Needle, Access Technologies, Skokie, IL) to eliminate residual
air prior to attachment of the HO actuating tubing. HOs were attached to injection ports
by threading the actuating tubing over the male connector and securing it with two
individual circumferential 3-0 polypropylene ligatures. The closed, non-inflated HOs
were placed on a flat bed scanner (HP Scanjet 4470 C, Hewlett-Packard Co. Palo Alto,
CA) and digital images were obtained and saved to a desktop computer for further
analysis. Using a computer software program, (Image J 1.27Z, National Institutes of
Health, USA) the internal luminal area (LA) of each HO was traced and calculated three
times. The mean LA for each HO was then calculated and used for statistical analysis.
Diffusion of Saline and Sodium Hyaluronate Through the HO
All HOs were properly exercised prior to use, and then assembled and filled as
previously described. Two HOs of each model (HO2, HO5 and HO20) were randomly
assigned to be filled with either sodium hyaluronate (SH) (Hylartin V, Pharmacia and
Upjohn, Kalamazoo, MI) or 0.9% sterile saline. Saline is considered to be the standard
filling solution for inflatable silicone implants in human beings due to it's isotonicity,
ready availability, and low cost. Based on previous studies suggesting that molecular
weight plays a major role in diffusion of substances through the silicone membrane of the
HO,89 we elected to compare saline to SH: an isotonic, non-toxic, high molecular weight
substance. HOs were filled with additional saline or SH through their injection ports
until complete occlusion was noted. The inflated HOs were scanned to document 100%
occlusion and weighed to within 10-3 g using a laboratory scale. (APX-203, Denver
Instrument, Arvada, CO) HOs were then placed in a bath of simulated body fluid (SBF)
kept at a constant 370 C.91 SBF was replaced on a weekly basis. HOs were removed
from the SBF once daily for the first 29 days, then once weekly for a total of 8 weeks to
be dried, scanned, and weighed. The digital images were analyzed using a computer
software program as described above to determine the internal LA of the HO. These
values were then compared to the initial, non-inflated LA for each HO in order to
calculate the % occlusion using the following formula: [% occlusion = LA non-inflated
occluder (cm2) LA inflated occluder (cm2) x 100] + [LA of the non-inflated occluder
Diffusion of Air Through the HO
Two each of HO2, HO5, and HO20 were utilized. All HOs were properly
exercised prior to use. HOs were all assembled and filled as previously described. A
huber point needle was used to fill each of the HOs with air until complete occlusion was
noted. The inflated HOs were scanned to document 100% occlusion. HOs were
submerged into a bath of SBF kept at a constant temperature of 370 C. HOs were
removed from the SBF every 4 hours and digital images were obtained for calculations of
LA as previously described.
Changes in Internal Pressure Over Time
Internal pressures were measured over a 30-day study period in the HO2, HO5, and
HO20 occluder models (1 of each model). All HOs were properly exercised prior to use.
Each occluder was prepared as shown in Figure 8. Actuating tubing was secured to the
injection port, T-port (T-5S-CS-UF, Access Technologies, Skokie, IL), and brass fitting
with 3-0 polypropylene ligatures and additional silicone (Silastic, Dow Coming
Corporation, Midland, MI) to prevent leakage at the tube insertion point. The HO,
injection port, actuating tubing, and brass fitting were filled with 0.9% saline in a
retrograde manner, taking care to eliminate air entrapped within the closed system. The
entire apparatus was attached to a pressure transducer (PX 603-030G5V, Omega
Engineering Inc., Stamford, CT) and process meter (DP25 B-E-A, Omega Engineering
Inc., Stamford, CT), which were in turn connected to a desktop computer for data storage
and analysis. Once assembled, HOs were filled via the injection port with additional
0.9% saline until complete occlusion was achieved. HOs were scanned at this time to
document 100% occlusion. HOs, including the injection port and all actuating tubing
were submerged into a bath of SBF that was kept at a constant temperature of 370 C and
changed weekly.91 Pressure monitoring was initiated when HOs were completely
occluded and submerged in the SBF. Software that allowed 24 hour pressure monitoring
(IOX 22.214.171.124, EMKA Technologies, Falls Church, VA) was employed to record internal
pressure on an hourly basis for a period of 30 days.
3cm 3 cm O
Hydraulic T-Port Injection
Figure 8. Schematic diagram representing the experimental setup used for evaluating
changes in internal pressure over time within individual hydraulic occluders.
Differences in luminal area between individual HOs were compared using an
ANOVA. A 2 way repeated measures ANOVA was used to evaluate differences in
weight and % occlusion over time and between different filling solutions. P < 0.05 was
considered significant. A Bonferroni/Dunn post-hoc test was used to test for interactions
Evaluation of Size Differences Among HOs
Mean closed, non-inflated LA values ranged from 0.038 cm2 to 0.053 cm2 for the
H02, 0.156 cm2 to 0.189 cm2 for the H05, and 2.227 cm2 to 2.855 cm2 for the H020.
Mean LA values, along with their standard deviations for the 5 HOs of each size are
displayed graphically in figure 9. Statistical analysis revealed a significant difference in
the non-inflated internal LA among the 5 HOs in all three sizes tested (P < 0.001).
0.05 HO 2 mm #1
- i.03 -HO 2 mm #
] HO 2 mm #3
0.02 -~ HO 2 rnmm
0.01 Eo HO 2mm #5
mHO 5 mm #1
m 0.15 o HO 5 mm #2
M HO 5 mm #3
IC HO 5 mm #4
0.05 HO 5 mm #5
E 2.5- m HO 20 mm #1
2 o HO 20 rnmm #2
-- 1.5- MHO 20mm #3
[ HO 20 mm #4
0 EB HO 20 mm #5
Figure 9. Graphic representation of the variations in luminal area within 3 models of the
non-inflated hydraulic occluder.
Diffusion of Saline and Sodium Hyaluronate Through the HO
Statistically significant decrease in weight was seen in all HOs over time (P <
0.0001). However, no significant difference was found when comparing weight changes
in HOs filled with saline to those filled with SH. (H02: P = 0.7650, Power = 0.055; HO5:
P = 0.5340, Power = 0.075; H020: P = 0.4742, Power = 0.085). Weight loss over time is
represented graphically in figure 10.
A statistically significant loss of occlusion was seen in all HOs over the 64-day
study period (P < 0.0001), however loss of occlusion was much more dramatic in H02
(mean 26.6%) compared to H05 (mean 4.5%) and H020 (mean 0.4%), regardless of
which filling solution was used. No significant difference was found when comparing
the change in % occlusion in HOs filled with saline to those filled with SH. (H02: P =
0.2704, Power = 0.150; HO5: P = 0.3960, Power 0.102; H020: P= 0.2650, Power =
0.153). Change in % occlusion over time is represented graphically in figure 11.
0 10 20 30 40 50 60 70
0 10 20 30 40 50 60 70
0 10 20 30 40 50 60 70
Figure 10. Graphic representation of the mean weight loss over time for 2 mm (A), 5 mm
(B), and 20 mm (C) hydraulic occluders filled with either saline (e) or sodium
0 10 20 30 40 50 60 70
0 10 20 30 40 50 60 70
100 =- -- -
0 10 20 30 40 50 60 70
Figure 11. Graphic representation of the mean % occlusion over time for 2 mm (A), 5
mm (B), and 20 mm (C) hydraulic occluders filled with either saline (e) or
sodium hyaluronate (m).
Diffusion of Air Through the HO
A significant loss of occlusion was seen in all air-filled HOs over a twelve-hour
period (P < 0.0001) (Figure 12). A significant interaction was also noted between change
in occlusion over time and HO size (P = 0.0014), with a more rapid loss of occlusion
occurring in the smaller HOs.
Changes in Internal Pressure Over Time
Pressure changes over time for H02, H05, and H020 are displayed graphically in
figure 13. The smaller HOs generated a greater internal pressure at filling, and
experienced a more rapid loss of pressure than the larger HOs. All HOs experienced the
most significant pressure losses within the first 5 days, and tended to plateau as time
progressed. Despite losses in pressure, complete occlusion was documented in both the
H02 and H020 at the end of the study using digital imaging. The H05 had undergone a
slight loss of occlusion deemed to be 3.5% using the same image analysis software
0 4 8
Figure 12. Graphic representation of the mean change in occlusion over a period of 12
hours for 2 mm (m), 5 mm (e), and 20 mm (A) hydraulic occluders filled with
air. Note the trend towards a more rapid loss of occlusion in the smaller HOs.
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Figure 13. Graphic representation of the change in internal pressure over time for 2 mm
(m), 5 mm (e), and 20 mm (A) hydraulic occluders filled with saline.
Previous work with the silicone HO in a model of chronic vascular occlusion
(Chapter 2) demonstrated a marked disparity between the predicted filling volume and
the actual volume required to produce vascular occlusion in vivo. Data from the current
mechanical evaluation of silicone HOs has revealed a number of factors that may have
contributed to variations in filling volumes during chronic applications. For example,
gross observation of the HOs had suggested that there was variation in the luminal area
due to inconsistencies in the manufacturing process. Hydraulic occluders are manually
assembled from sheets of medical grade silicone membrane, Dacron backing and silicone
tubing, using silicone gel to "hand weld" the seams. Digital images of the occluders
acquired during the current study did reveal significant variations in LA between
individual HOs in all three sizes that were evaluated (HO2, HO5, and HO20). The
disparity in luminal area was most remarkable for the HO2 size, with a range in LA from
0.038 to 0.053cm2, a range that encompasses nearly 40% of the total area. It is also
important to note that while the occluders are sold in "sizes" that are intended to describe
the internal diameter formed by the closed ring (i.e., the HO2 corresponds to a 2mm
internal diameter), the luminal areas measured in our study do not correspond to the area
of a circle with the diameter denoted by each occluder size. For example, a circular
occluder with a 2mm internal diameter would have a calculated LA ofrr2 or 0.0314cm2
while the measured LA of these occluders actually ranged from 0.038 to 0.053cm2.
Luminal area was measured after "exercising" the occluders as recommended by the
manufacturer, which may have contributed to increasing the LA to a figure that exceeded
the original LA due to acute stretching and deformation of the silicone HO. Thus, the
data from our current in vitro study would suggest that there is significant manufacturing
variability between individual HOs and that the measured luminal area after exercising of
the HOs does not necessarily reflect the advertised luminal diameter. Variations in LA
are particularly apparent in the smaller occluders (HO2) and this fact, combined with the
extremely small filling volumes required for occlusion, may have a profound effect on
fine control of rate of vascular occlusion when these smaller devices are applied in vivo.
A second concern in chronic application of silicone HOs was diffusion of filling
solutions through the thin silicone membrane, leading to deflation of the occluders over
time. Polydimethylsiloxane, generally referred to as silicone, is a highly cross-linked
elastomer that is largely bio-inert and as such has been used extensively in medical
implants since the late 1940's.92,93 It is a semi-permeable material, and diffusion of
various solutions and solutes across silicone membranes has been well documented.85-90
Diffusion across the HO silicone membrane is likely affected by numerous factors,
including hydrostatic pressure within the HO, filling solution osmolarity and molecular
weight, surface area of the silicone membrane, solute lipophilicity, osmotic gradient of
the environment to which the HO is exposed, temperature, and thickness of the silicone
membrane. A variety of solutions have been used to fill silicone HOs in previous short-
term experimental studies, including normal saline,66 hypertonic saline,67 water,68
radiographic contrast solutions,69 indocyanine green dye,70 and dextrose solution.71
Indeed, vast experience with diffusion of filling solutions from similar silicone implants
has been obtained through clinical application of artificial urethral sphincters, breast
implants and tissue expanders in human beings. In these clinical applications, isotonic
saline is currently the most commonly employed filling solution.90 Although these
implants are expected to behave similarly to the HO investigated in the current study, our
primary concerns with deflation of the HO are related to the unique application of the
device for chronic maintenance of complete vascular occlusion. As demonstrated in the
current study, complete inflation of the HO leads to the generation of extremely high
internal hydrostatic pressures, particularly in the smaller HOs. However, despite the
elevated hydrostatic pressure gradient, minimal diffusion of either saline or SH occurred
over the 8 week study period. Diffusion of solutions across a semi-permeable membrane
is also related to the thickness of materials used in membrane construction. Despite the
fact that there is some variation in materials used among different occluder sizes, such
that the membrane and actuating tubing of the HO20 are thicker than those used in the
construction of the HO2 and HO5 (0.020" compared to 0.007", respectively), diffusion of
filling solution did not appear to differ substantially among HOs of the three sizes tested.
In addition, no statistically significant difference was noted when comparing weight loss
or loss of occlusion between saline filled HOs and those filled with SH. Although it is
possible that a significant difference in diffusion would be detected between filling
solutions if a larger number of occluders was used in each experimental group, the
similarity of results in the current study would suggest that a clinically significant
difference between saline and SH would be unlikely.
In order to fully evaluate the results of our diffusion study, we must also analyze
the accuracy of the model. The in vitro model used in our study was based on the
recommendations of a previous group performing a similar evaluation of filling solutions
in silicone implants. 89 In an attempt to simulate the intra-abdominal environment,
implants were submerged in SBF kept at a constant 370C to simulate the temperature,
pH, and ion concentrations found in plasma.89,91 Implants were weighed and the non-
inflated LA was determined prior to submersion in the SBF to provide a baseline. Both
weight and LA determination proved to be highly accurate and reproducible means by
which serial measurements could be obtained, as reflected by the repeatability of the
measurements and small significant differences that both provided.90 Weight loss, as
well as loss of occlusion occurred to some degree in all HOs, suggesting that diffusion of
filling solutions occurs over time. Initial decreases in HO weight preceded any visible
loss of occlusion. This may be explained by the fact that weight change was a more
sensitive test, and that small volumes of filling solution may have been lost before it was
grossly evident as a loss of occlusion. It is also possible that while filling the HOs to
complete occlusion, they were inadvertently over-inflated. In this situation, any excess
filling solution could initially diffuse out of the HO without affecting occlusion.
A third concern regarding deflation of HOs over time was the effect of air bubbles
that may be entrapped within the silicone balloon during retrograde filling with fluid
solutions. Permeation of air and other gases through silicone membranes has been
documented previously and far exceeds that of fluid filling solutions.8 Data obtained in
the current study confirms the rapid nature of air diffusion from the HOs, which led to a
marked loss of occlusion in air-filled occluders over a matter of hours (Figure 12). Based
on this data, we suspect that entrapped air bubbles may have contributed to variability in
filling volumes required to reach complete occlusion in our previous use of the HO.
Based on the rapid nature of air diffusion through the silicone membranes, it is
recommended that investigators attempt to remove any air bubbles from the HO by filling
the actuating tubing in a retrograde manner using a long catheter as described. Small air
bubbles would likely be more significant in the HO2 compared to the HO5 and HO20
due to the smaller initial filling volume.
As a final aspect of this study, we evaluated changes in internal pressure of the
HOs over time. Internal hydrostatic pressures after complete inflation of the HOs were
inversely proportional to the size of the HO; a phenomenon that is likely related to the
total surface area of the occluders. An initial rapid decline in pressures occurred over the
first 5 days, followed by a more gradual rate of decline that continued for the entire 28-
day study period for all three sizes of HO. The initial rapid decrease in internal pressures
is likely to be associated with creep (plastic deformation over time) experienced by the
silicone material after application of tensile stress. Polymers are known to undergo
permanent deformation when subjected to chronic stress.82 It is possible that stretching
of the HO occurred over time, leading to an increase in the potential filling volume of the
HO, a resulting decrease in internal pressure, and ultimately a partial loss of occlusion.
This study has documented that inconsistencies in manufacturing, diffusion of fluid
or gaseous filling solutions through the silicone membrane and mechanical creep may all
affect the reliability of silicone HOs used in chronic applications. Although measurable
loss of occlusion and diffusion of filling solutions occurred in all HOs during the 64-day
period, these changes were most substantial in smaller HOs (HO2 and HO5). The
clinical implications of changes in occlusion of the larger HO20 occluders would likely
be negligible. Detectable differences were not demonstrated between saline and the
higher molecular weight SH with regard to loss of occlusion or diffusion of filling
solutions over time, and either substance would serve as an acceptable filling solution for
long-term maintenance of occlusion.
Despite successful treatment of congenital portosystemic shunts (CPS) via partial
occlusion with silk suture, the application of ameroid constrictors and cellophane bands,
as well as the deployment of intravascular thrombogenic coils, an ideal surgical device
has yet to be identified. An ideal device for CPS therapy would allow its safe
application to the shunting vessel, complete non-invasive adjustability in the
postoperative period, a predictable and adjustable rate of vascular occlusion and the
capability of attaining complete occlusion. Although its application in treating congenital
portosystemic shunts has not yet been described, the silicone hydraulic occluder (HO)
theoretically has many of the qualities listed above. Unfortunately, there is a paucity of
information present in the scientific and medical literature regarding use of the HO at
producing gradual and complete venous attenuation in long term applications.
Prior to use of the HO for treatment of CPS in canine patients, it was felt that an in
vivo study should be performed describing the HO's ability to gradually produce and
maintain complete venous occlusion. In the rat study described in Chapter 2 of this
manuscript, it was shown that the HO was indeed capable of producing a gradual
decrease, and eventual complete cessation of blood flow through the caudal vena cava of
rats. However, several complications with the model were identified, as well as some
potential issues that may affect future use of the HO in a clinical setting. Most notably,
the ability to accurately correlate the level of occlusion with the injected filling volume
appeared to be somewhat inconsistent in vivo. Potential causes for these discrepancies
were thought to include size differences among the HOs secondary to the manufacturing
process, the diffusion of air and/or filling solution from the HO over time, and the
phenomenon common to many elastomers known as creep (defined as plastic
deformation over time).
In Chapter 3 of this manuscript, a series of in vitro studies are described that were
designed to further evaluate the factors mentioned above. It was found that there is a
statistically significant difference in the size of uninflated HOs in each of the 3 size
categories evaluated (2mm, 5mm, and 20mm). It was also evident that all HOs undergo a
slight loss of occlusion over time, regardless of whether or not saline or sodium
hyaluronate was used as a filling solution. This loss of occlusion was much more
apparent in the 2mm HOs, and was almost insignificant in the 5 mm and 20 mm HOs.
Air diffused very rapidly from the HOs in all cases, although again this was more
apparent in the 2mm HOs. All HOs experienced a loss of internal pressure over time,
implying that creep may have occurred.
HOs are an appropriate means of producing controlled, gradual, and complete
venous occlusion. The ability to adjust and maintain occlusion over time appears to be
more difficult in the small (2mm) HOs, and may be related to a combination of factors
including: (A) inconsistencies in the manufacturing process which complicate
extrapolating volume injected : % occlusion curves from one occluder to another, (B) the
unavoidable entrapment of air upon initial filling and its subsequent displacement from a
small potential filling volume, (C) diffusion of filling solution from within the occluder,
and (D) the physical changes that are associated with creep and occur within the HO over
time. Larger HOs, despite being subject to the same factors, do not appear to be as
severely affected, and are much more capable of maintaining occlusion over an extended
period of time.
Future studies examining the use of HOs in the treatment of CPS are warranted.
Use of HOs with sizes ranging from 5 mm to 20 mm would be expected (corresponding
to reported CPS vessel diameters), and as such would be less susceptible to the
complications noted with use of the 2 mm HOs. Repeating the experiments described in
Chapter 3 is also warranted in attempt to generate improved power of the statistical
analyses by increasing the number of experimental subjects.
LIST OF REFERENCES
1. Litvak J, Siderides LE, Vineberg AM: The experimental production of coronary
artery insufficiency and occlusion. Am Heart J 53:505-518, 1957.
2. Berman JK, Fields DC, Judy H, Mori V, Parker RJ: Gradual vascular occlusion.
Surgery 39:399-410, 1956.
3. Brunschwig A, Bigelow R, Nichos S: Elective occlusion and excision of the portal
vein. Surg 17:781-783, 1945.
4. Egun A, Slade D, McCollum N: Gradual huntarian ligation for infected prosthetic
bypass. Eur J Vasc Endovasc Surg 19:437-439, 2000.
5. Nishie A, Yoshimitsu K, Honda H, Kaneko K, Kuroiwa T, Fukuya T, Irie H,
Ninomiya T, Yosimitsu T, Hirakata H, Okuda S, Masuda K: Treatment of hepatic
encephalopathy by retrograde transcaval coil embolization of an ileal vein-to-right
gonadal vein portosystemic shunt. Cardiovasc Intervent Radiol 20:222-224, 1997.
6. Hottinger HA, Walshaw R, Hauptman JG: Long term results of complete and
partial ligation of congenital portosystemic shunts in dogs. Vet Surg 24:331-336,
7. VanVechten BJ, Komtebedde K, Koblik PD: Use of portal scintigraphy to monitor
blood flow and progressive postoperative attenuation of partially ligated
portosystemic shunts in dogs. J Am Vet Med Assoc 204:1770-1774, 1994.
8. Komtebedde J, Koblik PD, Breznoc EM, Harb M, Garrow LA: Long term
outcomes after partial ligation of single extrahepatic portovascular anomalies in 20
dogs. Vet Surg 24:379-383, 1995.
9. Vogt JC, Krahwinkel DJ, Bright RM, Daniel GB, Toal RL, Rohrback B: Gradual
occlusion of extrahepatic portosystemic shunts in dogs and cats using the ameroid
constrictor. Vet Surg 25:495-502, 1996.
10. Harari J, Lincoln J, Alexander J, Miller J: Lateral thoracotomy and cellophane
banding of a congenital portoazygous shunt in a dog. J Small Anim Pract 31:571-
11. Gonzalo-Orden JM, Altonaga JR, Costilla S, Gonzalo Cordero JM, Millan L,
Asuncion Orden Recio M: Transvenous coil embolization of an intrahepatic
portosystemic shunt in a dog. Vet Radiol Ultra 41:516-518, 2000.
12. Partington BP, Partington CR, Biller DS, Toshach K: Transvenous coil
embolization for treatment of patent ductus venosus. J Am Vet Med Assoc
13. Taboada, J: Medical management of animals with portosystemic shunts. Semin Vet
Med Surg Small Anim 5:107-119, 1990.
14. Johnson CA, Armstrong PJ, Hauptman JG: Congenital Portosystemic shunts in
dogs: 46 cases (1979-1986). J Am Vet Med Assoc 191:1478-1483, 1986.
15. Vulgamott JC: Portosystemic shunts. Vet Clin North Am Sm Anim Pract 15:229-
16. Swalec KM, Smeak DD: Partial vs. complete attenuation of single portosystemic
shunts. Vet Surg 19:406-411, 1990.
17. Komtebedde J, Forsyth SF, Breznock EM, Koblik PD: Intrahepatic portosystemic
venous anomaly in the dog. Perioperative management and complications. Vet
Surg 20;37-42, 1991.
18. Kyles AE, Gregory CR, Jackson, Ilkiw JE, Pascoe PJ, Adin C, Samii VF,
Herrgesell E: Evaluation of a portocaval venograft and ameroid ring for the
occlusion of intrahepatic portocaval shunts in dogs. Vet Surg 30:161-169, 2001.
19. Bostwick DR, Twedt, DC: Intraheptaic and extrahepatic portal venous anomalies in
dogs: 52 cases (1982-1992). J Am Vet Med Assoc 206:1181-1185, 1995.
20. Breznock EM, Berger B, Pendray D, Thomas B, Strack D, Bauer T, Koblik P,
Hornof W, Ludders JD: Surgical manipulation of intraheptaic portocaval shunts in
dogs. J Am Vet Med Assoc 182:798-805, 1983.
21. Hunt GB, Bellenger CR, Pearson MRB: Transportal approach for attenuating
intrahepatic portosystemic shunts in dogs. Vet Surg 25:300-308, 1996.
22. White RN, Burton CA, McEvoy FJ: Surgical treatment of intrahepatic
portosystemic shunts in 45 dogs. Vet Rec 142:358-365, 1998.
23. Smith KR, Bauer M, Monet E: Portosystemic communications: Follow-up of 32
cases. J Sm Anim Pract 36:435-440, 1995.
24. Swalec Tobias KM, Seguin B, Johnston G: Surgical approaches to single
extraheptaic portosystemic shunts. Compend Contin Educ Pract Vet 20:593-601,
25. Martin RA, Freeman LE: Identification and surgical management of portosystemic
shunts in the dog and cat. Semin Vet Med Surg 2:302-306, 1987.
26. Martin RA: Congenital portosystemic shunts in the dog and cat. Vet Clin North Am
Sm Anim Pract 23:609-623, 1993.
27. Hunt GB, Hughes J: Outcomes after extrahepatic portosystemic shunt ligation in 49
dogs. Aust Vet J 77:303-307, 1999.
28. Lawrence D, Bellah JR, Diaz R: Results of surgical management of portosystemic
shunts in dogs: 20 cases (1985-1990). J Am Vet Med Assoc 201:1750-1753, 1992.
29. Youmans KR, Hunt GB: Cellophane banding for the gradual attenuation of single
extrahepatic portosystemic shunts in eleven dogs. Aust Vet J 76:531-537, 1998.
30. Fossum TW, Hedlund CS, Hulse DA, Johnson AL, Seim HB, Willard MD, Carroll
GL, Orton CE: Small Animal Surgery (ed 2). St. Louis, MO, Mosby, 2002, pp 43-
31. Bellinger CR: Sutures part 1. The purpose of sutures and available suture materials.
Compend Contin Educ Pract Vet 4:507-515, 1982.
32. Stashack TS, Yturraspe DJ: Considerations for selections of suture materials. Vet
Surg 7:48-55, 1978.
33. Postlethwait RW: Long term comparative study of nonabsorbable sutures. Ann
Surg 171:892-898, 1970.
34. Meyer HP, Rothuizen J, van Sluis FJ, Voorhout G, van den Brom WE: Progressive
remission of portosystemic shunting in 23 dogs after partial closure of congenital
portosystemic shunts. Vet Rec 144:333-337, 1999.
35. Youmans KR, Hunt GB: Experimental evaluation of four methods of progressive
venous attenuation in dogs. Vet Surg 28:38-47, 1999.
36. Peacock EE, Van Winkle W: Wound Repair (Ed. 2). Philadelphia, PA, Saunders,
1976, pp 217-219.
37. Lange PE, Sievers HH, Nurnberg JH, Engler K, Pilarczyk J, Onnasch DGW,
Bernhard A, Heintzen PH: A new device for slow progressive narrowing of vessels.
Basic Res Cardiol 80:430-435, 1985.
38. Elzinga WE: Ameroid constrictor: uniform closure rates and a calibration
procedure. J App Phsiol 37:419-421, 1969.
39. Adin CA, Gregory CR, Kyles AE, Griffey SM, Kendall L: Effect of petrolatum
coating on the rate of occlusion of ameroid constrictors in the peritoneal cavity. Vet
Surg 33:11-16, 2004.
40. Li J, De Leon H, Ueno T, Cui J, Cossement PK, King SB 3rd, Chronos NA,
Robinson KA: Vasomotor function of pig coronary arteries after chronic coronary
occlusion. J Cardiovasc Pharmacol 41:600-608, 2003.
41. Kawamoto A, Tkebuchava T, Yamaguchi J, Nishimura H, Yoon YS, Milliken C,
Uchida S, Masuo O, Iwaguro H, Ma H, Hanley A, Silver M, Kearny M, Losordo
DW, Isner JM, Asahara T: Intramyocardial transplantation of autologous
endothelial progenitor cells for therapeutic neovascularization of myocardial
ischemia. Circulation 107:461-468, 2003.
42. Baffour R, Garb JL, Kaufman J, Berman J, Rhee SW, Norris MA, Friedmann P:
Angiogenic therapy for the chronically ischemic lower limb in a rabbit model. J
Surg Res 36:618-624, 2000.
43. Mills JD, Fischer D, Villanueva FS: Coronary collateral development during
chronic ischemia: serial assessment using harmonic myocardial contrast
echocardiography. J Am Coll Cardiol 36:618-624, 2000.
44. Jensen DM, Machicade GA, Tapia JI, Kauffman G, Franco P, Beilin D: A
reproducible canine model of esophageal varices. Gastroenterology 84:573-579,
45. Canty TG, Jauregizar E, Fernandez-Cruz L: Experimental portal hypertension in
the rat. J Pediatr Surg 15:819-826, 1980.
46. Mehl ML, Kyles AE, Adin CA, Kass P, Gregory CR: Evaluation of ameroid ring
constrictors for extrahepatic portosystemic shunts in 111 dogs (1996-2001). Vet
Surg 30:487, 2002 (Abstract).
47. Sackman CJA, Clarke KM: Use of an ameroid constrictor to occlude a patent
ductus venosus in a dog. Vet Surg 2:212, 1999 (Abstract).
48. Besancon MF, Kyles AE, Griffey SM, Gregory CR: Evaluation of the
characteristics of venous occlusion after placement of an ameroid constrictor. Vet
Surg 32:481-482, 2003 (Abstract).
49. Krahwinkel DJ: Ameroid Constrictors in Portosystemic Shunts. Vet Surg 27:364,
50. Murphy ST, Ellison GW, Long M, Gilder JV: A comparison of the ameroid
constrictor versus ligation in the surgical management of single extrahepatic
portosystemic shunts. J Am Anim Hosp Assoc 37:390-396, 2001.
51. Wiles CE, Worthington GS, Lindenberg J: The experimental production of portal
hypertension. Ann Surg 136:811-817, 1952.
52. Tamiya T, Thal AP: Esophageal varices produced experimentally in the dog. Surg
Gyn Obst 147-154, 1960.
53. Markowitz J, Archibald J, Downie HG: Experimental Surgery Including Surgical
Physiology, 5th Ed., Baltimore, MD, Williams and Wilkins Co, 1964, pp 414.
54. Breznock EM: Surgical manipulation of portosystemic shunts in dogs. J Am Vet
Med Assoc 774:819-825, 1979.
55. Hunt GB, Tisdall PLC, Webb A, MacPherson GC, Brain P, Malik R: Congenital
portosystemic shunts in toy and miniature poodles. Aust Vet J 78:530-532, 2000.
56. Connery NA, McAllister H, Skelly C, Pawson P, Bellenger CR: Cellophane
banding of congenital intrahepatic portosystemic shunts in two Irish wolfhounds. J
Sm Anim Pract 43:345-349, 2002.
57. Bendok BR, Hanel RA, Hopkins LN: Coil embolization of intracranial aneurysms.
Neurosurg 52:1125-1130, 2003.
58. Graves VB, Strother CM, Rappe AH: Treatment of experimental canine carotid
aneurysms with platinum coils. Am J Neuroradiol 14:787-793, 1993.
59. O'Donnell C, Neutze JM, Skinner JR, Wilson NJ: Transcatheter patent ductus
arteriosus occlusion: Evolution of techniques and results from the 1990s. J Paediatr
Child Health 37:451-455, 2001.
60. Fellows CG, Lerche P, King G, Tometzki A: Treatment of patent ductus arteriosus
by placement of two intravascular embolisation coils in a puppy. J Sm Anim Pract
61. Weisse C, Schwartz K, Stronger R, mondschein JI, Solomon JA: Transjugular coil
embolization of an intrahepatic portosystemic shunt in a cat. J Am Vet Med Assoc
62. Leveille R, Johnson SE, Birchard SJ: Transvenous coil embolization of
portosystemic shunt in dogs. Vet Radiol Ultra 44:32-36, 2003.
63. Schneider M, Plassmann KR: A new method for treatment of large intrahepatic
shunts. Proc 18th American College of Veterinary Internal Medicine 2000:545-546
64. Weisse C, Solomon JA, Holt D, Nicholson M, Cope C: Percutaneous transvenous
coil embolization of canine intrahepatic portosystemic shunts: short term results in
14 dogs. Vet Surg 32(5):499-500, 2003 (Abstract).
65. Kim IO, Cheon JE, Kim WS, Chung JW, Yeon KM, Yoo SJ, Seo JK, Choi JH:
Congenital intrahepatic portohepatic venous shunt: treatment with coil
embolisation. Pediatr Radiol 30:336-338, 2000.
66. Edmunds LH, Rudy LW, Heymann MA, Boucher JK: An adjustable pulmonary
arterial band. Trans Amer Soc Artif Int Organs 18:217-223, 1972.
67. Leeuwenburgh BPJ, Schoof PH, Steendijk P, Baan J, Mooi WJ, Helbing WA:
Chronic and adjustable pulmonary artery banding. J Thorac Cardiovasc Surg
68. Jacobson ED, Swan KG: Hydraulic occluder for chronic electromagnetic blood
flow determinations. J Appl Physiol 21:1400-1402, 1966.
69. Solis E, Heck CF, Seward JB, Kaye MP: Percutaneously adjustable pulmonary
artery band. Ann Thorac Surg 41:65-69, 1986.
70. Park SC, Griffith BP, Siewers RD, Hardesty RL, Ladowski J, Zoltun RA, Neches
WH, Zuberbuhler: A percutaneously adjustable device for banding of the
pulmonary trunk. Int J Cardiol 9:477-484, 1985.
71. Peacock JT, Fossum TW, Bahr AM, Miller MW, Edwards JF: Evaluation of
gradual occlusion of the caudal vena cava in clinically normal dogs. Am J Vet Res
72. Henry WL, Ploeg C, Kountz SL, Harrison DC: An improved hydraulic vascular
occluder for chronic electromagnetic blood flow measurements. J Appl Physiol
73. Bishop SP, Cole CR: Production of externally controlled progressive pulmonic
stenosis in the dog. J Appl Physiol 26:659-663, 1969.
74. Solis E, Bell D, Alboliras H, Seward J, Kaye MP: Left ventricular preparation with
an extrathoracically adjustable balloon occluder. Ann Thorac Surg 44:58-61, 1987.
75. Le Bret E, Bonhoeffer P, Folliguet TA, Sidi D, Laborde F, de Leval RM, Vouhe P:
A new percutaneously adjustable, thoracoscopically implantable, pulmonary
arterial banding: an experimental study. Ann Thorac Surg 72:1358-1361, 2001.
76. Ferko A, Lesko M, Hukek P, Krajina A, Pipal R: A new approach to modeling
prehepatic portal hypertnesion. Preliminary experimental results. Rozhl Chir 74:70-
77. Ferko A, Lesko M, Krajina A, Hulek P, Rosch J, Rabkin J: Intrahepatic portal vein
branches after extrahepatic portal vein occlusion. Experimental study.
Heaptogastroenterology 48:475-479, 2001.
78. Buchanan JW: Causes and Prevalence of Cardiovascular Disease, in Kirk RW and
Banagura J (eds.): Current Veterinary Therapy XI. Philadelphia, PA, Saunders,
1992, pp 647.
79. Adin CA, Farese JP, Cross AR, Provitola MK, Davidson JS, Jankunas H:
Urodynamic effects of a percutaneously-controlled static hydraulic urethral
sphincter in a canine cadavers. Am J Vet Res 65:283-288, 2004.
80. Tiffany JS, Petraitis DJ: Silicone Biomaterials, in Encyclopedic Handbook of
Biomaterial and Bioengineering. Wise DL, Trantolo DJ, Altobelli DE, Yaszemski
MJ, Gresser JD, Schwartz ER (eds.): New York, NY, Marcel Dekker Inc., 1995, pp
81. Reijnen MMPJ, Bleichrodt RP, van Goor H: Pathophysiology of intra-abdominal
adhesion and abscess formation, and the effect of hyaluronan. Br J Surg 90:533-
82. Callister WD: Materials Science and Engineering (Ed 5). New York, NY, John
Wiley and Sons Inc., 2000, Sec 16.8. pp. 491-493.
83. D'Almeida MS, Gaudin C, Lebrec D: Validation of 1 and 2-mm transit-time
ultrasound flow probes on mesenteric artery and aorta of rats. Am J Physiol (Heart
Circ Physiol 37) 268:H1368-H1372, 1995.
84. D'almeida MS, Cailmail S, Lebrec D: Validation of transit-time ultrasound flow
probes to directly measure portal blood flow in conscious rats. Am J Physiol (Heart
Circ Physiol 40) 271:H2701-H2709, 1996.
85. Studer UE, Bishop MC, Zingg EJ: How to fill silicone catheter balloon. Urology
86. Tuncali D, Ozgur F: Spontaneous autoinflation of saline-filled mammary implants:
Postoperative volume determination by magnetic resonance imaging. Aesth Plast
Surg 23:437-442, 1999.
87. Schanzer A, Wong GB: Air dissipation in saline breast implants. Plast Reconstr
Surg 109:2238-2241, 2002.
88. White RI, Ursic TA, Kaufman SL, Barth KH, Kim W, Gross GS: Therapeutic
embolization with detachable balloons. Radiology 126:521-523, 1978.
89. Yu LT, Latorre G, Marotta J, Batich C, Sisson Hardt N: In vitro measurement of
silicone bleed from breast implants. Plast Reconstr Surg 97:756-764, 1995.
90. Robinson Jr OG, Benos DJ: Spontaneous autoinflation of saline mammary
implants. Ann Plast Surg 39:114-121, 1997.
91. Oyane A, Kim HM, Furuya T, Kokubo T, Miyazaki T, Nakamura T: Preparation
and assessment of revised simulated body fluids. J Biomed Mater Res 65A: 188-
92. Silver FH, Christiansen DL: Biomaterials Science and Biocompatibility. New
York, NY, Springer-Verlag New York Inc., 1999, pp 279-286.
93. Gabriel SE: Soft Tissue Response to Silicones, In Handbook of Biomaterial
Properties. Black J, Hastings G (eds.): London, UK, Chapman & Hall, 1998, pp.
Colin W. Sereda began his post-graduate education in a two year pre-veterinary
program through the Department of Agriculture at the University of Alberta in
Edmonton, Alberta, Canada, between 1995 and 1997. He was then accepted to the
Western College of Veterinary Medicine at the University of Saskatchewan in Saskatoon,
Saskatchewan, Canada, where he completed a 4 year program in veterinary medicine and
graduated in 2001 with a Doctor of Veterinary Medicine. He then completed a clinical
internship in small animal medicine and surgery at the Virginia-Maryland Regional
College of Veterinary Medicine in Blacksburg, Virginia, between 2001 and 2002. After
completing the internship, Colin moved to Gainesville, Florida, in 2002, where he entered
the Master of Science degree program at the University of Florida's College of
Veterinary Medicine. In combination with the master's degree, Colin is also completing
a residency in small animal surgery, and is scheduled to be done in the summer of 2006.