Keywords
end to side - IJV - venous anastomosis - head and neck reconstruction - re-exploration
Introduction
Successful venous anastomosis plays a major role in microvascular reconstruction of
the head and neck as flap failure is mostly attributed to venous thrombosis.[1] An end-to-side (ES) anastomosis to internal jugular vein (IJV) serves several benefits
such as constant anatomy, large-caliber venous outflow, provision of accommodating
multiple anastomoses, suitable match for any size and length of flap vein(s), and
availability after neck dissection in redo cases. Improved flap survival and reduced
venous thrombosis have been reported with the ES anastomosis to IJV in several studies.[2]
[3]
[4] Despite its many advantages, the ES venous anastomosis is still not routinely used
by all, probably due to technical preferences for an end-to-end (EE) anastomosis by
many surgeons, particularly whenever a branch of the IJV is available. We present
here our experience with ES venous anastomosis for head and neck oncosurgical reconstructions
over the past 20 years of microvascular surgical practice.
Materials and Methods
This retrospective cohort study includes all consecutive patients reconstructed with
free flaps for head and neck malignancies with our “routine (ES) IJV anastomosis”
approach in two separate affiliated institutes between 2015 and 2023. The preferential
(ES) venous anastomosis was done to an IVC stump on either side of the neck, whenever
feasible. The arterial anastomosis was preferably made with the facial artery stump
on either side of the neck, or to an alternate stump (superior thyroid artery, lingual
artery, external carotid artery stump, etc.) of the same side whenever facial artery
stump was not available. In the cases where due to insufficient pedicle reach an ES
with IJV was not feasible, an IJV tributary had to be used for an EE anastomosis and
these cases are excluded from the present study. The oncologic clearance and reconstruction
were done by two separate teams of oncosurgeons and plastic surgeons. Patient-related
variables were collected from the hospital database and analyzed. Informed written
consent was obtained for unanonymized data publications, and standard ethical guidelines
were followed. Institutional ethical clearance was taken for publication.
The oncosurgical clearance and neck dissection were performed by the oncosurgeons.
The facial artery and IJV on one side have been preserved ([Fig. 1]).
Fig. 1 Surgical anatomy of head and neck microvascular reconstruction. (A,B) Anatomical location and orientation of the recipient vessels. The anatomical structures
are marked in the inset of the clinical photograph. a, facial artery stump; b, internal
jugular vein; c, posterior belly of digastric; d, sternomastoid; e, mylohyoid.
Three different free flaps, radial forearm flap (RFF), osteocutaneous fibula flap
(OCFF), or anterolateral thigh (ALT) flap, were used for reconstruction according
to the size of the defect and component of tissue loss. The RFF was elevated with
a segment of the cephalic vein and communicating vein as the authors' preferred approach.[5] The artery was dissected up to the origin to obtain maximum length and diameter
in all cases. The anastomosis was performed on the opposite side of the neck, where
a suitable vascular stump was not available on the same side. All the anastomoses
were performed under 4.5X loupe magnification.
Anastomosis
The flap was transferred to the defect. The neck vessels were prepared on the selected
side for anastomosis. The preserved facial artery stump was delivered under the digastric
muscles and trimmed until a spurt of blood flow was seen. The arterial anastomosis
was completed with 7–0 Prolene interrupted sutures. The arterial anastomotic clamp
was left in place and the venous pedicle was oriented for anastomosis. The venous
anastomosis site was selected along the length of the IJV according to the available
length of the venous pedicle while avoiding any kinking or tension on the flap vein.
A Satinsky clamp was applied along the chosen IJV site ([Fig. 2A, B]). A transverse venotomy was made on the IJV to match the diameter of the flap vein
and stretched along the length of the vein to produce a “fish mouth.” An ES anastomosis
was then performed from the posterior wall to the anterior wall with the prior placement
of corner sutures and interrupted sutures in between, resulting in a rhomboid-shaped
anastomosis ([Fig. 3]). After completion of both anastomoses, the arterial anastomosis, venous anastomosis
on the IJV, and confluence of venous and arterial pedicle remain in a triangular orientation
without any acute angulation ([Fig. 4]). When additional flap veins were available or if the operating surgeon felt the
need for a second ES venous anastomosis, it was carried out along the same lines.
The flap inset was completed and the neck wound was closed over a suction drain.
Fig. 2 End-to-side (ES) anastomosis with a radial forearm flap (RFF). (A, B) Orientation of the vascular pedicles, and method of venous elevation-occlusion with
Satinsky's clamp. (C) RFF harvested with the communicating vein and cephalic vein. a, flap pedicle with
radial artery and VC; b, communicating vein; c, cephalic vein; d, facial artery; e,
internal jugular vein.
Fig. 3 End-to-side (ES) anastomosis technique with internal jugular vein (IJV). (A) ES anastomosis technique. A1, posterior wall fixed with the apex of the fish mouth;
A2, upper and lower borders of the vein fixed with the center of the upper and lower
margins of venotomy; A3, posterior wall suture completed with interrupted sutures
in between the fixed points; A4, anterior margin of the vein fixed with the anterior
apical point of venotomy and anastomosis completed with interrupted sutures in between;
A5, diamond-shaped anastomosis. (B,C) Venotomy and venous anastomosis.
Fig. 4 (A, B) Postanastomotic “triangular orientation” of pedicles. a, vascular pedicle of flap;
c, venous pedicle; d, arterial pedicle; e, internal jugular vein; f, sternomastoid
muscle; x, confluence of venous and arterial pedicles; y, venous anastomosis; z, arterial
anastomosis.
When re-exploration and redo of the venous anastomosis was required, the ES anastomosis
was divided and sealed off. A second ES anastomosis using the same flap vein (or another
if available) was carried out at a separate site on the IJV depending upon the health
of the venous pedicle. The patients were followed up for flap monitoring and survival
for 5 days in the hospital.
Results
Operative Data
A total of 585 free flaps including 303 RFFs, 143 OCFFs, and 139 ALT flaps were used
for reconstruction within the study period. In 10 cases, anastomosis was performed
on the contralateral side with 6 of them being done for prior irradiated neck and
4 for previously failed free flaps. The harvested venous pedicle length varied between10
and 18 cm. Single venous anastomosis was performed in 576 cases and dual anastomosis
was done in 9 cases, of which 6 were OCFFs and 3 were ALT flaps.
Outcome
Re-exploration was done in 45 cases (7.69%). Re-exploration was done within the first
12 hours in 31 cases and after that period in 14 cases. In 30 cases of re explorations,
the cause of flap compromise was found to be other than anastomosis patency. The anastomosis
was maintained at the same site of the IJV with management of responsible causes such
as position change, evacuation of clots, reorientation of drains, improvement of circulatory
status, etc. In the remaining 15 cases, a redo anastomosis was done to a separate
site of the IJV. Redo anastomosis with the same venous pedicle to a separate site
was done in nine cases. Anastomosis with the second pedicle to a separate site was
done in six cases. The flap revision survival rate was 38/45.
Complete flap failure was observed in 12 cases (6 RFFs, 3 OCFFs, and 3 ALT flaps),
resulting in a flap survival rate of 573/585 (97.94%). Partial flap loss was seen
in one case due to skin paddle necrosis in the OCFF case. The skin defect was reconstructed
with a free RFF with anastomosis to the opposite side in this case successfully. Among
the 12 failed flaps, 3 were due to arterial thrombosis in a previously irradiated
neck. The other nine were due to venous thrombosis, out of which five were noted in
the previously irradiated neck.
Discussion
Flap failure after routine microvascular oncosurgical head and neck reconstructions
is mainly attributed to venous failure. A range of choices for venous anastomosis
including external jugular vein (EJV), anterior jugular vein, transverse cervical
vein, and branches of the IJV such as the anterior and common facial vein, lingual
vein, and superior thyroid vein remain available for an EE anastomosis.[4] According to availability, superficial location, familiarity with the anatomy, and
surgeons' preferences, an EE anastomosis has been practiced widely for head and neck
reconstruction. However, unavailability, traction, torsion to the pedicle, and compression
during neck movement remain major concerns for such anastomosis. An ES anastomosis
with the main IJV or remnant IJV stump is reserved for such cases.[6]
[7] Moreover, an ES anastomosis is preferred when there is a significant size discrepancy
between the anastomotic vessels.[8]
The IJV is the preferred venous draining avenue over the EJV in many centers due to
its anatomical configurations and availability. The unavailability of the IJV within
the operative field is very rare in routine oncosurgical head and neck cases. Large
caliber, constant location, availability after neck dissection, provision of multiple
anastomoses, and accommodation of vascular discrepancies are major factors influencing
the choice of the IJV for venous anastomosis.[2]
[3]
[4] Besides being subjected to negative pressure during respiration, large-caliber anastomosis
and protection from traction and torsion are believed to diminish venous thrombosis
and improve flap survival with the IJV compared to the EJV.[9]
[10]
[11]
However, a technical difficulty in creating anastomosis deep in the jugular groove,
and unfamiliarity with the anatomy and ES anastomosis, particularly in small-caliber
pedicles, may restrain the use of ES IJV anastomosis routinely as the first choice.
As per the senior author's (AG) 20 years' experience of microvascular anastomosis,
after a few technical adaptations such as use of a Satinsky clamp for venous occlusion
and elevation of the IJV during anastomosis, orientation of the anastomotic pedicles,
a posterior to anterior diamond-shaped anastomosis, use of 4.5X loupe magnification
and 7–0 sutures, etc., a technical familiarity was obtained and a reduced re-exploration
rate was noted with ES anastomosis. This led to the preference for the IJV and ES
anastomosis, and this has been used routinely subsequently.
Frequent use of microvascular reconstruction for oncosurgical defects by the oncoplastic
team has improved recipient vessel preservation after radical neck dissection. Routine
use of facial artery and ES anastomosis with the IJV saves the effort of preserving
and searching for suitable vessels for anastomosis. The available pedicle length is
adjusted quite well with the provision of venous anastomosis at any location along
the vertical length of the IJV. With the triangular geometric orientation, traction
and torsion to the pedicle are avoided. After completion of both anastomoses, the
pedicle recoils back deep to the sternomastoid groove, so it remains well protected.
Our routine ES venous anastomosis approach suits quite well with free ALT flap and
OCFF where the venous diameter is usually 2 to 4 mm. For the RFF cases, the diameter
of the venae comitantes is usually less than 1.5 mm. However, the incorporation of
a segment of the cephalic vein with the communicating vein in the venous pedicle facilitates
venous drainage by a reliable deep venous system through the avenue of a large-caliber
cephalic vein, and increases the venous pedicle length[5]
[12]
[13] ([Fig. 2A, C]). ([Fig. 5]) Routine use of the cephalic vein and communicating vein facilitates the ES approach
in RFFs. A vein graft has not been not required in any of our cases. In 10 postradiotherapy
neck and redo cases, reconstruction is done with the RFF, where the ipsilateral facial
artery stump is not available and anastomosis is made on the contralateral neck. With
routine use of these three flaps, vessel diameter, discrepancies, use of vein graft,
and operating microscope remain a lesser concern with our approach.
Fig. 5 The possible long pedicle length of a free radial forearm flap (RFF) harvested with
communicating vein and cephalic vein.
For a vertically oriented anastomosis, it is easy to construct as well as re-explore.
It is convenient to re-explore and redo venous anastomosis without using a vein graft.
The revision could be done with a higher-up anastomosis and sealing off the previous
anastomotic rent on the IJV.
The most obvious difficulty in constructing an ES anastomosis with the IJV is the
deeper location of the IJV. The use of a Satinsky vascular clamp facilitates the elevation
of the selected segment of the vein and provides venous occlusion ([Fig. 2B]). This enables us to construct anastomosis at a superficial plane without any added
maneuver.
We have used a horizontal elliptical “fish mouth” venotomy and a rhomboid- or “diamond”-shaped
venous anastomosis with the ES approach ([Fig. 3]). A similar technique has been described earlier by Sen et al for arterial ES anastomosis.[14] According to them, a “diamond”-shaped anastomosis is responsible for improved patency
due to increased anastomotic surface area. We believe this may be responsible for
better anastomotic patency for venous anastomosis as well. Although we do not have
any objective data for this apart from improved re-exploration rate from our earlier
cases, we believe the technique is helpful to avoid posterior wall bite, collapse
of fish mouth venotomy, and wider anastomosis, which may be responsible for the reduced
re-exploration rate in our cases.
An 8–0 or 9–0 suture is commonly used for head and neck reconstruction and done best
under the operating microscope. Although this norm is practiced at the beginning of
the learning curve, not much data are available on the use of 7–0 sutures or loupe
magnification for microvascular anastomosis. Although an operating microscope was
available, the cases were performed under “loupe only” and the microscope was not
required in any cases in this series. We have observed that in the most commonly performed
flaps, RFF, OCFF, or ALT flap, the vessel diameter remains greater than 2.5 mm, which
could be comfortably handled with a 4.5X magnifying loupe with placement of a 7–0
Prolene suture. The use of loupe improves operative time by reducing the time to set
up and move the operative platform, better visualization of the operative field, easy
movement and coordination between surgeon and assistants, comfortable operative position,
freedom for access to difficult locations and positions, and reducing surgeons' fatigue.
Moreover, the use of a loupe has a quicker learning curve.[15]
[16] The use of a 7–0 suture further improves the operative time by placement of lesser
number of approximating anastomotic sutures. Only in a few cases with relatively small-caliber
anastomosis were 8–0 sutures used under loupe magnification. Thus, we believe that
at the centers where an operating microscope and 8–0 and smaller sutures are not available
or not routinely used, an ES approach still have good feasibility of microvascular
head and neck reconstruction.
Finally, we have experienced a decrease in the re-exploration rate than in our earlier
practice. A decreased re-exploration rate has been reported after overall improved
microvascular suturing and flap survival.[17] The described re-exploration rate and salvage rate in the literature vary from 6
to 14% and 36 to 72%, respectively.[1]
[18] Re-exploration and salvage rates of the present study were 45/585 (7.69%) and 38/45
(84.44%). Out of 45 re-explorations, 31 were early within 12 hours of primary surgery
and all of them were successfully salvaged, improving the overall success of re-exploration
as well. Poor salvage rate after re-exploration (8/12) was noted mainly in patients
with previously irradiated necks.
The study is limited by the lack of comparative outcome data. As the present study
involves a single operating surgeon and a routine ES approach has been practiced exclusively
within the study period, a comparative analysis is beyond the scope of this study.
However, the study may be considered as a feasibility study of the routine ES approach
and results may be used for future comparative analysis of various technical aspects.
Conclusion
Routine use of ES anastomosis simplifies microvascular anastomosis by avoiding efforts
related to the selection of recipient vessels, providing the benefit of performing
a single large-caliber venous outlet, the favorable geometric orientation of the pedicle,
and ease of re-exploration. Vascular anastomosis for RFF, OCFF, and ALT flap can be
performed quite conveniently with this approach using 7–0 sutures and under 4.5X loupe
magnification, with similar outcomes to the reported flap survival rate in the contemporary
practice.
