
Inferior
vena caval thrombectomy without laparotomy
(Portuguese
PDF version)
Arno
von Ristow1, João Marcelo Rocha Loures2,
Ricardo Pires Coelho3, Glória A. Martins4,
José M. Cury5, Marcus H. Gress6,
Robson Chicrala de Abreu7, Paulo Soares3
1.
Head of the Department of Vascular Surgery and Department of Endovascular
Surgery, Centervasc, Rio de Janeiro, Brazil.
2. Graduate student, Vascular Surgery and Resident
doctor, Centervasc, Rio de Janeiro, Brazil.
3. Anesthesiologist, Centervasc, Rio de Janeiro,
Brazil.
4. Intensivist, Hospital Barra D'Or.
5. Head of Clinical Service, Department Of Vascular
Surgery, Centervasc-Rio.
6. Assistant, Department of Vascular Surgery and
Department of Endovascular Surgery, Centervasc-Rio.
7. Head of the Department of Perioperative Care,
Centervasc-Rio.
Correspondence:
Arno von Ristow
Rua Barão de Lucena, 48 - Gr. ½
CEP: 22.260-020 - Rio de Janeiro - RJ
Tel.: +55 21 266.2349 Fax: +55 21 537.2465
ABSTRACT
21-year-old
male patient with cavo-ilio-femoro-distal venous thrombosis presented
with thrombectomy associated with temporary therapeutic arteriovenous
fistula. Caval thrombectomy was performed by endovascular surgery.
The migration of the thrombus was avoided by positioning two occlusion
balloons in the cava proximal to the thrombus and at the right common
iliac artery. The proximal thrombi were removed by balloon catheters
and the distal ones by retrograde compression. Residual stenosis
of the left common iliac vein due to compression exerted by the
right common iliac artery was observed in control venography and
treated with self-expanding stent graft. The outcome was satisfactory.
After seven weeks of anticoagulation therapy, the fistula was closed
by the endovascular technique by means of intra-arterial implantation
of an endoprosthesis. The unblocked veins are totally patent.
Key
words: thrombectomy, venous thrombosis, treatment
Palavras-chave: trombectomia; trombose venosa; tratamento.
J
Vasc Br 2002;1(1):71-78.
The thrombectomy
of greater vein trunks to reestablish vascular patency and to preserve
the valvar function (thus reducing acute morbidity and postthrombotic
syndrome) is still controversial.1 We report a case of inferior vena caval
thrombosis, thrombosis of the common iliac (internal and external) arteries
and of the whole deep venous tree of the left lower limb, in which thrombectomy
was performed without laparatomy.
CASE
REPORT
21-year-old,
university student, formerly healthy male was admitted to the emergency
room of Hospital Barra D'Or on March 10, 2002. He presented pain and
edema of the left lower limb. The symptoms had begun on March 5, 2002,
with pain at the root of the thigh and progressive edema of the lower
extremity. A causal relation could not be established. The patient,
on his own initiative, intensified the physical exercises he was used
to doing and self-medicated with nonsteroid anti-inflammatory drugs,
with progressive deterioration of the symptoms. The patient sought medical
care on the fifth day after the onset of symptoms. Deep vein thrombosis
(DVT) was initially suspected, and was confirmed by vascular ultrasonography.
The patient was admitted to the Intensive Care Unit and the treatment
with unfractionated heparin and coumarin was initiated. After 48hrs
on this treatment scheme, the symptoms persisted, and we were called
in to evaluate the patient. The intensive care unit team, which was
in charge until then, inquired the possibility of a more efficient treatment.
There was no family history of thrombophilia.
On examination,
the patient presented typical symptoms of left iliofemoral thrombosis
of the Plegmasia alba dolens type. Color Doppler ultrasonography
confirmed our clinical suspicion. Since this exam does accurately assess
the intra-abdominal veins, we requested computed tomographic vemography.1
This exam could not be carried through as ideally planned due to the
difficulty in puncturing a vein on the left foot (because of the edema).
Therefore, an iodinated contrast medium was injected into a vein on
the upper limb. Despite the low quality image, the migration of the
thrombus into the vena cava (around 8cm) was evident (Figure 1). We
therefore suspected of compression of the left iliac vein caused by
the left iliac artery.2
Figure
1 - Computed tomographic angiography
of the abdomen showing migration of the thrombus from the left common
iliac vein to the vena cava (bigger arrow). It is possible to suspect
of common iliac vein compression at the crossing level with the left
iliac artery (smaller arrow).

After
the confirmation of the migration of the thrombus into the inferior
vena cava (IVC), thrombectomy was our first option, since the patient
had been formerly healthy and had a large life expectancy. Our decision
was based on our experience and on international literature.1,3-10
The patient
was placed in a supine position and prepared for vena caval thrombectomy
by means of laparotomy and thrombectomy of the ilio-femoral-distal segment
by left inguinal access under general anesthesia. Systemic heparin administration
was not interrupted (we had planned to neutralize heparin administration
after implementing laparotomy). The left common femoral vein and its
tributaries were exposed by an oblique incision on the inguinal crease.
The same was done on the superficial femoral artery within a 3-cm length,
2 cm distal to its origin.
The insertion
of a venous occlusion catheter adjacent to the vena cava thrombus, via
right jugular vein, is a usual procedure. As we have a digital surgical
arc in our operating room, we decided to use a percutaneous catheter,
with an 11F sheath. A lower-extremity venocavography was performed,
and a roadmap image was obtained. The right iliac vein and the vena
cava were carefully probed with Edwards 8/14 catheter as far as the
juxtarenal IVC. After subtotal balloon inflation in the infrarenal position,
a new lower-extremity cavography (via the sheath) was performed, thus
limiting the thrombus (Figure 2). One of the authors (JMRL) suggested
entrapping the caval thrombus with another balloon positioned at the
common iliac vein, thus allowing thrombectomy of the IVC with safety
and reduced blood loss.
Figure
2 - Surgical photograph. Surgical exposure
of left femoral vessels and visualization of two catheters inserted
through the 11F sheaths into the right common femoral vein for venous
occlusion.

A second
11F sheath was also introduced into the same right common femoral vein
2cm away from the first one. An Edwards 8/10 catheter, positioned at
the proximal common iliac vein, was inserted through the sheath.
Figure
3 -Intraoperative ilial cavography:
presence of floating thrombus at the inferior vena cava. Observe the
subtotal balloon inflation to allow for small residual flow.

With both
balloons inflated, a longitudinal venotomy was performed on the left
common femoral vein, and its borders were repaired. Thrombectomy of
the vena cava and of the iliac veins (common and left external) was
performed via another Edwards 8/10 catheter and through an aspiration
probe, as previously published1 ( Figure 4).
Figure
4 - Inflated occlusion balloons at
the vena cava, proximal to the thrombus and at the right common iliac
vein. The thrombectomy catheter is partially inflated (1).

After repeated
maneuvers, we obtained appropriate patency of the IVC and of the iliac
veins. On the first intraoperative venography, with inflated balloons,
venous compression by the iliac artery was evident (Figure 5, arrow).
After making sure that the iliocaval axis was free of thrombi, blood
flow was allowed and a new angiography was performed (Figure 6). The
angiography confirmed the compression and reduction of vein diameter.
We opted for a self-expanding stent graft at the level of the stenosis.
At the time, a Wallstent 10X39, the largest available stent, was accurately
inserted. Distal thrombectomy was performed by distoproximal compression
with Löfqvist tourniquet (insufflated at 180 mmHg) from the foot
up to the root of the thigh, followed by rigorous massage of the limb
muscle masses, taking care to keep the femoral venotomy open during
the compression maneuvers in order to prevent vein ruptures. At the
end, a temporary therapeutic arteriovenous fistula (TAVF) was placed
between the proximal superficial femoral artery, which was previously
prepared, and the common adjacent femoral vein.
The removed
thrombi measured approximately 80 cm. These thrombi showed a light-colored
segment with high platelet count, at the level of the iliocaval junction,
which corresponded to the site of compression and the probable onset
site of the thrombotic events (Figure 7).
The patient
did not receive blood transfusion during the surgery, and was awake
when he was transferred out of the Intensive Care Unit.
Figure
5 - Venography after thrombectomy,
with still inflated balloons. Total removal of thrombi. The mark left
by the venous compression exerted by the artery is clearly visible (arrow).

Figure
6 - Venography after normalized flow:
point of the stenosis (arrow).

Figure
7 - Removed thrombi, with length greater
than 80 cm. Observe the difference in color at the level of the iliocaval
junction (arrow).

Pain relief
was immediate and the edema gradually decreased during three days. Heparin
administration was maintained until the sixth day after the surgery.
Coumarin administration was implemented on the second day after the
surgery. The patient was discharged with INR 2.4, and with elastic compression.
The investigation of thrombophilia associated with non-vitamin K-dependent
coagulation factors yielded negative results.
Ten days
after the surgery, the patient presented with phlebitis of the internal
femoral saphenous vein, which was treated locally. Vascular ultrasonography
revealed patency of the deep venous system.
Anticoagulation
was maintained for seven weeks and discontinued on May 2, 2002. On May
7, the T-AVF was closed by contralateral arterial catheterization. A
Balkin vascular sheath was positioned at the left common femoral vein.
Venous angiography was performed via the fistula. The perfect patency
of venous trunks is evident, as shown in Figure 8.
Figure
8 - Venocavography on the day the T-AVF was closed (arrow). Perfect
patency of all unblocked veins can be observed.

T-AVF
was obliterated by a Jostent Graft , placed in the superficial femoral
artery with a 7x40 angioplasty balloon. Figure 9A shows an active T-AVF,
marked off with Ligaclip LC 400 . Figure 9B shows the covered stent
graft expanded at the site of T-AVF. Figure 9C is the final control
arteriography, which shows the occlusion of the fistula and arterial
patency.
Figure
9 - A) Temporary therapeutic arteriovenous
fistula between the superficial femoral artery and the common femoral
vein, with an 8-mm anastomotic stricture. A metal clip shows where the
fistular orifice begins. B) Endoprosthesis inserted at the level of
the fistula, thus occluding it. C) Final control arteriography after
T-AVF closure.

On the
same day, coumarin administration was implemented, supported by low
molecular weight heparin until the appropriate INR was obtained. The
patient was discharged on the first postoperative day, with elastic
compression. Coumarin therapy is to be maintained for three months and,
after this period, a new color Doppler ultrasonography will be made.
If the patency of proximal and distal venous axes is confirmed, anticoagulation
will be discontinued, and additional investigation of thrombophilia
of vitamin K-dependent factors will be carried out.
DISCUSSION
Läwen
performed the first venous thrombectomy in 1937.4
Since then, this method has been advocated by some and refuted by others.
Most professionals
involved in the treatment of DVT choose the anticoagulation therapy
for all cases. There is factual evidence in literature that proximal
deep venous thrombosis is potentially fatal in a great number of patients.
The sequelae of these extensive thromboses are often devastating to
the patient's quality of life, especially if DVT occurs within the first
decades of life, thus allowing a long time for the development of postthrombotic
syndrome and its dire sequelae.3
Currently,
anticoagulation (with unfractionated heparin or low-molecular-weight
heparin) is the most widely used therapy. Heparin is administered simultaneously
with antagonists of vitamin K. In the last few years, the wide publicity
campaign on the treatment of thromboembolic disease run by the pharmaceutical
industry gave professionals with inadequate knowledge about pathology
access to the diagnosis (by means of color Doppler ultrasonography)
and allowed them to treat DVT independently and uniformly. These professionals
refer the patient to angiographic investigation only after hospital
discharge. When complications are present or when an immediate response
to treatment is not obtained, a specialist is called in. At this point,
treatment options are often unfeasible, since prothrombin activity is
extremely low. It is clearly evident that we should be called in immediately
in order to evaluate the situation of each of these patients. Antagonists
of vitamin K should only be administered when other treatment options
that are immediately more efficient and that produce better results
in the long run, such as thrombectomy and/or fibrinolysis, are not available.
The treatment
of DVT should be considered on a case-by-case basis.
The suspected
diagnosis of DVT is usually confirmed by color Doppler ultrasonography.
This exam, which is unarguably accurate when performed by highly qualified
professionals, does not safely assess the extension of intra-abdominal
thrombi. This occurs because patients with proximal DVT have intense
meteorism, which hinders the transmission of sound waves. In addition,
compression maneuvers on the obstructed venous trunks can produce emboli.
Computed tomographic venography, with iodinated contrast medium diluted
into the veins of the foot, has been our method of choice for the investigation
of proximal thrombosis, as corroborated by other authors.1,3
In this exam, the patient is placed in a supine position; all veins
of the trunk, from the calf to the heart, are visualized. Computed tomographic
venography is quick and easy, especially when helical computed tomography
is used. The exam determines the extension of the thrombotic events,
detects congenital disorders of the pelvic and abdominal veins, and
reveals unsuspected etiologies (a malignant tumor, for instance).
Several
criteria are currently necessary for indication of venous thrombectomy:
thrombosis should involve the proximal venous trunks (caval, iliac and
common femoral veins) in individuals with a good health status and large
life expectancy. The evolution of the thrombotic events should be less
than seven days and the diagnosis must be confirmed by a reliable imaging
method. If ischemic venous thrombosis is present, the indication is
unarguable, and a prompt action should be taken before gangrene sets
in. A surgical ward equipped with a surgical arc facilitates the procedure
and allows the simultaneous correction of disorders, such as iliocaval
compression syndrome (May and Turner),2
which was observed in this case.
Similarly
to what happened with thrombectomy, thrombolysis was discredited after
strong popularity.3 However, just like thrombectomy,
there is a renewed interest in thrombolysis, which has been increasingly
used for the treatment of proximal DVT.7-9
Systemic lysis does not yield satisfactory results in 50% of the cases.3
The thrombolytic agent must be administered inside the thrombus so that
the lysis is effective. Semba and Dake report a 72% success rate with
a jugular catheter.9 Selective catheterization
of abdominal and pelvic veins and of the whole limb is necessary. Retrograde
venous catheterization is hampered by the presence of venous valves.
An option would be to puncture and catheterize the popliteal vein and
to inject additional thrombolytic agents into the veins of the foot.
Another problem is the selection of the lytic agent, since we do not
have urokinase. We only have streptokinase (SK) and recombinant tissue
plasminogen activator (rt-PA). The latter has a high rate of hemorrhagic
complications when used for several days, as is usually necessary to
obtain the extensive thrombolysis in large veins. Elevations of fibrin
degradation products favor hemorrhagic events.3,7,9
Several authors with renowned experience prefer to use thrombectomy
for the treatment of proximal thromboses, but it is important that the
indications listed above be followed.1,3,5,6,10
Therefore, from our standpoint, lysis is not so often indicated for
the treatment of large thrombosis, as is the case of our patient.
The digital
x-ray surgical arc has been an inestimable tool for a countless number
of endovascular or combined (directly surgical and endovascular) procedures,
as the case under study. Ease of use and accuracy allow for the appropriate
positioning of occlusion balloons and also permit us to detect residual
thrombi, and treat disorders shown by angiography (for example, venous
iliac compression). Vascular surgeons should get acquainted with this
method in order to broaden their treatment options. Angioscopy has not
been consolidated as a routine exam.10
In our study, we used intraoperative radiological control instead of
angioscopy.
Since the
introduction of venous thrombectomy catheters by Fogarty in 1966, this
has been the most widely method for this kind of surgery.5 These catheters
are efficient, but they would be safer if we could probe over the guidewires,
using Seldinger technique. If available, latex catheters inserted over
the guidewires can be used to occlude the vena cava, thus saving Fogarty
catheters for thrombectomy proper. A thrombus will be hardly removed
by the smooth passage of a flexible guidewire.
Mechanical
thrombectomy of venous thrombi has been individually and conjointly
used with thrombolysis at the iliocaval level, with proximal balloon
protection.8 Certainly, its use would be
effective in this case, but the devices were not available at the moment.
Mechanical thrombectomy is restricted to abdominal and pelvic veins.
In infrainguinal veins, the valvas do not allow the progression of systems.
The need
for a caval filter after venous thrombectomy is still questioned. Even
though some cases have been reported in literature,1,8,10
in our experience with 49 thrombectomies, we never had pulmonary embolism
after a successful thrombectomy. In this specific case, we decided not
to use the filter because of perfect unblocking and the high quality
of T-AVF.
The stent
graft used to correct the iliac compression syndrome was too small in
our case since we did not have an ideally sized device. We believe a
self-expanding device with a diameter of 14 mm and a length of 40 mm
would have been more appropriate. The common left iliac vein has an
ellipsoid section, caused by the physiological compression exerted by
the right iliac artery. Therefore, a larger stent, which could fit this
conFiguretion, would have been more appropriate.
Thrombectomy
of infrainguinal veins could be performed by means of thrombectomy catheters.
In practice, the presence of valvas hinders the procedure and poses
the risk of destroying these delicate and essential parietal structures.
We prefer to use elastic compression, which could be obtained by the
use of a Löfqvist tourniquet or by the classic Esmarch bandage.1
In the
two last decades, several international studies clearly established
the indication of venous thrombectomy and valued its use for the treatment
of proximal deep venous thrombosis.1,3,5-10
The accurate topographic diagnosis offered by computed tomographic venography,
associated with the improvement of this technique and with the development
of methods that provide immediate efficacy - angioscopy and/or intraoperative
venography, as well as the routine use of T-AVF - contributed towards
better results. The prompt and simultaneous treatment of the iliac compression
syndrome and the treatment of associated thrombophilia can substantially
reduce early and late rethrombosis and positively influence results
in the long run.
REFERENCES
1.
Ristow AV. Tratamento da trombose venosa profunda - Trombectomia.
Cir Vasc Angiol 1998;14:49-56.
2.
May R, Turner J. Ein Gefässsporn in der V. iliaca communis sinistra
als Ursache der vorviegend linksseitigen Beckenvenenthrombose. Z Kreisl-Forsch
1956;45:912-22.
3.
Rutherford R. Trombectomia e Trombólise. In: Maffei FHA &
Ristow AV, (editores). Trombose Venosa Profunda e Embolia Pulmonar.
Rio de Janeiro: Sanofi-Winthrop Farmaceutica; 1996. p. 98-110.
4.
Läwen A. Über Thrombectomie bei Venenthrombose und Arteriospasmus.
61. Tag. der Deutsche Gesellschaft für Chirurgie. Berlin: 1937.
5.
Fogarty TJ, Dennis D, Krippaehne WW. Surgical management of iliofemoral
venous thrombosis. Am J Surg 1966;112:211-7.
6.
Neglén P, Eklöf B. Thrombectomy with Temporary Arterio-Venous
Fistula: The method of choice to treat acute iliofemoral venous thrombosis.
In: Vieth FJ, (editor). Currrent Critical Problems in Vascular Surgery.
St. Louis: Quality Med Publ; 1991. p.142-50.
7.
Morgan R, Belli AM. Percutaneous thrombectomy: a review. Eur Radiol
2002;12:205-17.
8.
Largiadér J, Blatter W, Gloor B. Combinations therapy of venous
thrombosis with local thrombolysis and venous thrombectomy. Kongressbd
Dtsch Ges Chir Kongr 2001;118:479-81.
9.
Semba CP, Dake MD. Iliofemoral deep venous thrombosis: Aggressive
therapy with catheter directed thrombolysis. Radiology 1994;191:487-94.
10.
Wack C, Wolfle KD, Loeprecht H. Früh und spät Ergebnisse der
inferior vena cava transperitoneale thrombectomie. Zentralbl Chir 2001;126:456-60.
|