Extract
Video-assisted thoracoscopic surgery has become an accepted alternative
approach to the thoracic spine for release of disc space contents with or
without fusion. The technique can be used for anterior spinal release,
bone-grafting of the intervertebral space, biopsy, removal of benign tumors,
or insertion of
implants1-7.
Thoracoscopy involves surgical access to the spine through several small
incisions, each 1.0 to 2.5 cm in length. Through these incisions, a
muscle-splitting technique allows placement of soft or rigid trocars (ports)
through which instruments can be inserted. While this approach is being
utilized for release and instrumentation at a few centers, the technique has
not replaced traditional thoracotomy. Traditional thoracotomy requires a long
incision with extensive dissection of muscles. It involves rib resection with
substantial spreading of the intercostal spaces, and there is tissue
desiccation. There is also limited visibility of the cephalad vertebrae. In
addition, the extensive thoracotomy incision may alter pulmonary and shoulder
girdle function. The goal of endoscopic surgery is to perform the same
operations as the classic open procedures but with less invasive methods.
Video-assisted thoracoscopic surgery has become an accepted alternative
approach to the thoracic spine for release of disc space contents with or
without fusion. The technique can be used for anterior spinal release,
bone-grafting of the intervertebral space, biopsy, removal of benign tumors,
or insertion of
implants1-7.
Thoracoscopy involves surgical access to the spine through several small
incisions, each 1.0 to 2.5 cm in length. Through these incisions, a
muscle-splitting technique allows placement of soft or rigid trocars (ports)
through which instruments can be inserted. While this approach is being
utilized for release and instrumentation at a few centers, the technique has
not replaced traditional thoracotomy. Traditional thoracotomy requires a long
incision with extensive dissection of muscles. It involves rib resection with
substantial spreading of the intercostal spaces, and there is tissue
desiccation. There is also limited visibility of the cephalad vertebrae. In
addition, the extensive thoracotomy incision may alter pulmonary and shoulder
girdle function. The goal of endoscopic surgery is to perform the same
operations as the classic open procedures but with less invasive methods.
Look for this and other related articles inInstructional
Course Lectures, Volume 54, which will be published by the American
Academy of Orthopaedic Surgeons in February 2005:
• "Indications for Anterior Only and Combined (Anterior and
Posterior) Approaches (Thoracic and Lumbar Spine) and Techniques in Pediatric
and Adult Patients," by Keith H. Bridwell, MD
Video-assisted thoracoscopic surgery is a minimal-incision procedure with
less morbidity than a thoracotomy. While it is considered in some centers to
be minimally invasive, it is as invasive as a thoracotomy. In fact, it is
maximally invasive but minimally incisional and is therefore appealing both
cosmetically and physiologically because of limited skin incisions and muscle
destruction. Thoracoscopy provides the entire surgical team with much better
visualization and magnification of the exposure. The exposure of individual
disc spaces at all levels of the spine is superior to that provided by
thoracotomy, and it can be extended to L1, or to L2 by releasing the lateral
part of the diaphragm. When one enters the endothoracic cavity with a standard
thoracotomy at T10, it is sometimes difficult to perform an adequate disc
space excision at the T3-T4 level with conventional instruments without
utilizing a second thoracotomy
incision8. With
selective placement of portals, video-assisted thoracoscopic surgery allows
each disc space to be approached coronally with full exposure, regardless of
its level. The procedure also involves less dissection, which reduces blood
loss. The time for opening and closing that is needed with a thoracotomy is
eliminated, thus decreasing operative time. Also, respiratory mechanics and
shoulder function are less compromised. While the postoperative residual
volume:total lung capacity ratio is only marginally better after
video-assisted thoracoscopic surgery, this difference becomes more substantial
when severely affected patients are
treated9.
Postoperative incisional pain is reduced, leading to a speedier
rehabilitation. In our experience, video-assisted thoracoscopic surgery has
reduced the stays in the intensive care unit and in the hospital. Healed scars
from video-assisted thoracoscopic surgery in the axilla are much more
cosmetically acceptable than is a long thoracotomy scar, leading to a higher
degree of patient satisfaction. Video-assisted thoracoscopic surgery has been
shown to be as effective as open thoracotomy in increasing spinal flexibility
in the porcine
model10-12.
Finally, if instrumentation and correction are carried out, they can be
performed over fewer levels in appropriate curves in selected patients.
The procedure is technically demanding and requires considerable training
of the surgeon and the anesthesiologist. The anesthetic considerations are
more challenging than are those for thoracotomy, and they require a longer
preparation time. Double-lumen intubation is used for the lateral approach,
and it requires single-lung ventilation, which can have respiratory
complications. Fiberoptic bronchoscopy is used to position the double-lumen
tube, and the tube should be rechecked for distal migration following lateral
positioning. Double-lumen intubation may create high airway pressures within
the bronchial tree, leading to air leakage or rupture and
pneumothorax13.
Single-lumen intubation and double-lung ventilation are used when the patient
is treated in the prone position, and it is associated with fewer anesthetic
challenges; however, vascular complications may be more difficult to treat
with the patient in this position.
Video-assisted thoracoscopic surgery has a relatively steep learning
curve14. It is
thought that a surgeon must perform an average of thirty procedures before he
or she can carry out an adequate anular release and end-plate removal of the
intervertebral disc space. On the other hand, this learning curve may be
lessened by convening a team consisting of a spinal surgeon and an experienced
access surgeon4. It
may be difficult to treat complications, specifically bleeding, and thus a
thoracotomy tray should be available and open, with the surgeon prepared to
perform an emergency thoracotomy if necessary. There is a lack of
three-dimensional perception with use of a thoracoscope, and tactile feedback
is missing when the instruments are manipulated. Technologically
sophisticated, specially designed, rather expensive instruments are mandatory
to perform this
procedure15.
Adequate release of the disc space contents is time-consuming because there is
no efficient removal device. The use of technologically sophisticated
equipment can be costly, approximately 28% more expensive than the equipment
required for a classic open
thoracotomy15, and
it is considerably more time-consuming. The placement of anterior
instrumentation is also difficult to learn. Endoscopic correction and
instrumentation have been complicated by rod breakage and a less-than-robust
fusion mass. Finally, there is danger to the great vessels, lungs, and
esophagus, especially as one ascends the learning curve.
Our indications for video-assisted thoracoscopic surgery in children and
adolescents are the same as those for open
thoracotomy6,15,16.
They include a rigid idiopathic scoliosis of >75° that does not correct
to <50° on side-bending and a rigid sagittal plane (kyphotic) deformity
of >80° that is inflexible to the point of not decreasing to 55°
when the patient is hyper-extended over a bolster. With video-assisted
thoracoscopic surgery, it is easier to access a spine with a kyphotic
deformity than one with scoliosis, and fewer portals are used. The kyphotic
spine presents as a semicircle, and occasionally only three portals are
necessary to approach it in a centripetal fashion. Other indications for
video-assisted thoracoscopic surgery include a young patient at risk for the
crankshaft phenomenon with posterior instrumentation, internal
costoplasty17,18
neuromuscular deformity, metabolic disease with progressive deformity, a
patient with Type-I neurofibromatosis with or without intercostal or
paravertebral tumors, incisional or excisional biopsy of rib tumors, anterior
hemiepiphysiodesis for congenital scoliosis, repair of anterior pseudarthrosis
with augmentation of the fusion in a patient who has undergone posterior
instrumentation, thoracic outlet
syndrome19 and
selective endoscopic instrumentation for idiopathic
scoliosis1,20,21.
From 1993 through 2003, we performed or assisted in more than 160
video-assisted thoracoscopic procedures. Seventy-eight of the patients had
idiopathic scoliosis; twenty-five, Scheuermann kyphosis; eighteen,
neuromuscular scoliosis; fourteen, scoliosis secondary to neurofibromatosis;
four, thoracic outlet syndrome; three, myelomeningocele spinal deformities;
three, congenital scoliosis; two, spinal dysplasia; three, juvenile idiopathic
scoliosis; and one each, vertebral fracture, infantile scoliosis, Marfan
syndrome, pseudarthrosis, rib tumor, postradiation deformity, a malignant
tumor, and vertebral osteomyelitis requiring incision and drainage.
A primary contraindication to video-assisted thoracoscopic surgery
performed with the patient in the lateral position is the inability of the
patient to tolerate single-lung
ventilation22.
Other contraindications include severe respiratory insufficiency, high airway
pressures, pleural symphysis, and
empyema23,24.
Relative contraindications include previous thoracotomies (which are not
uncommon in patients with scoliosis, who may have had perinatal cardiac
disease) and severe spinal deformity with the spine touching the chest wall.
As our experience has increased, most of these contraindications have become
relative as opposed to absolute, and we usually perform the approach from the
convex side of the greater deformity.
Video-assisted thoracoscopic surgery offers the opportunity to develop a
true team, which includes the anesthesiologist, an access (either thoracic,
general, or spine) surgeon, and the primary spine surgeon. The scrub nurse is
a fundamental part of the team and should be experienced and familiar with the
instruments and
techniques25. The
use of the thoracoscopic mini-cam (thoracoscope) as well as multiple monitors
allows all members of the team the same visual access to the pathological
entities. Most often, double-lumen intubation is carried out, with bronchial
blockers being necessary for smaller children. We routinely use spinal cord
monitoring, and we currently employ transcranial motor-evoked-potential
monitoring.
There is a need for specialized endoscopic instruments, which have improved
substantially over the past ten years. The optical component includes variably
angled high-quality thoracoscopes of 10 mm in diameter, most often with a
0°, 30°, or 45° viewing angle. The thoracoscope should be kept in
warm saline solution prior to use to prevent fogging when it is placed in the
chest. The monitors may be coaxial, and they allow comparable views from both
sides of the table. The magnification capabilities, starting at fifteen power,
may be enhanced with the use of ever-improving optical instruments, such as
zoom lenses.
The thoracoscopic mechanical instruments are all specialized adaptations of
conventional instruments. In the early years of spinal endoscopy, the shafts
of the proportionately longer instruments (elevators, rongeurs, and curets)
were wider as it was thought that insufflation would be necessary both in the
chest and in the abdomen and the surgeon had to achieve a seal around the
trocars. It is now well known that, with retraction of the lung, there is
deflation and resorption atelectasis, which obviates the need for insufflation
and allows use of an instrument with a smaller shank, or shaft. The current
ports are either flexible or rigid and are available in multiple sizes. The
thoracoscope should always be inserted through rigid ports, but these ports
have the potential of causing neurapraxia of the intercostal nerves. It is
important to alternate trocars and the position of the thoracoscope as much as
possible. The flexible ports are somewhat broader and allow the surgeon to
insert several instruments such as a Kitner retractor and a suction device.
Variously sized rongeurs, curets, and elevators are necessary, but they are
used in a conventional fashion. Specialized funnels have been developed to
allow placement of bone graft into the intervertebral disc space. The lengths
of the electrocautery units, both monopolar and bipolar, are adapted to the
thoracoscopic dimensions. The ultrasonic, or harmonic, scalpel has been
developed for endoscopic use and includes a forceps model that allows one to
grasp each particular vessel and a hook model that is commonly used to paint,
or feather-stroke, the vessel to coagulate it prior to incising it. Once the
vessel has turned brown it is safe to incise it. The suction and irrigation
systems are also of endoscopic length. Equipment is available to allow direct
application of bone wax, Surgicel (oxidized, regenerated cellulose), and
Avitene (microfibrillar collagen) as hemostatic agents. It is extremely
important that a thoracotomy tray be available during all thorascopic
procedures.
The patient is placed on a radiolucent table in the lateral or prone
position according to the physician's preference
(Fig. 1-A). When the lateral
position is used, the approach to the convex side of the deformity is
preferable to bring the spine closer into the field. An axillary roll is
positioned, and continuous brachial plexus monitoring is performed. The effect
of gravity is used to assist in retraction of the lung and vessels away from
the spine. When using the lateral position, we usually flex the table. By
doing this, we prevent the shoulder and pelvis from impeding the horizontal
movement of the thoracoscope in the most proximal and distal regions of the
field.
The prone position (Fig.
1-B) allows adequate release of the spine comparable with that
possible with the patient in the lateral position. However, with the patient
in the prone position, the operation requires less time because there is less
preparation for anesthesia as a regular endotracheal tube is used. Another
time-saver is that there is no need to reposition the patient between
sequential anterior and posterior
procedures5,26.
The lung tends to fall anteriorly, out of the way, and requires less
retraction. Some surgeons have operated with the patient in the prone position
so that two teams can perform simultaneous discectomy and release and
posterior exposure for
instrumentation27.
A downside is that an emergency thoracotomy is much more difficult to perform
with the patient prone.
When a patient is treated through the lateral approach, it is important to
pad the upper and lower extremities properly with careful attention paid to
the pelvis and greater trochanter on the down side. The patient is prepared
and draped in the lateral decubitus position from just above the axilla to
below the iliac crest, anteriorly to the midsternal level and posteriorly
across the spine to the opposite chest wall. Following appropriate draping of
the patient, the T6-T7 interspace is usually entered just posterior to the
midaxillary line. This level usually allows one to avoid the diaphragm and
liver when approaching from the right side. An incision is made through the
skin and subcutaneous tissue, and electrocautery is used to dissect through
the intercostal muscles into the intrathoracic space. A Kelly clamp is then
employed to widen this incision so that one can see directly whether the lung
is deflated. A rigid trocar is then placed into the incision, and the
thoracoscope is inserted in a posterior direction (or aiming posteriorly) to
avoid injuring the lungs or major vessels. Once the thoracoscope has been
placed into the chest, the levels that were previously selected to undergo
release are identified (Fig.
2). The thoracoscope is directed into the upper part of the chest
above the lung. The first rib is usually covered by muscle and soft tissue,
and, as a result, the second rib is the most cephalad rib that can be easily
identified. The superior intercostal vein usually drains the T3-T4 interspace
directly into the azygous vein; however, there may be anatomical variations.
The assistant taps the chest with a finger in the midaxillary line, diagonally
over the area corresponding to the selected interspace, to determine the most
direct approach to the rostral disc to be excised. A skin incision and portal
are then placed in this area. The ports should be adequately positioned to
achieve a panoramic view, and usually three to four incisions are made. The
ports should be spaced to avoid the instruments bumping against each other.
Alternatively, a Steinmann pin can be inserted percutaneously in this region
under direct vision toward the disc to be excised. An incision can then be
made directly over the pin. One should consider using this technique when
first performing the procedure.
It is important to maintain proper camera orientation at all times.
Instruments should be placed into the chest only under direct visualization
and should always be directed out toward the posterior chest wall, rather than
directly down toward the operating table. It is dangerous to insert or move an
instrument without direct observation.
Depending on the procedure, the technique is divided into three or four
stages: exposure, disc excision, bone-grafting, and (possibly)
instrumentation. The strategy most often involves the use of four portals, for
the thoracoscope, a retractor, suction, and the dissecting instrument. Lung
retraction is necessary at the beginning of the procedure. It can be performed
with a variable-angle retractor, which can be attached to the table, or with
Kitner retractors. The correct level is identified by either counting ribs or
using an intraoperative radiograph. The contour of the anterior spinal column
is one of mounds (the disc spaces) and valleys (the vertebral bodies). The
segmental vessels are in the valleys. We prefer to incise the pleura
longitudinally over the disc spaces and the vertebral bodies close to the rib
heads. The rib heads are an important landmark. The spinal canal is always
posterior to the rib heads, and dissection posterior to them should be
avoided. The pleura is retracted, and the vessels should be coagulated and
incised at each level with the harmonic scalpel. Some surgeons maintain the
vessels and dissect around them. Others prefer to clamp them and observe the
spinal cord monitors for signal changes prior to coagulating them. The pleura
and the vessel stumps can then be elevated, dissected, and retracted
anteriorly from the vertebral bodies. A sponge is then packed beneath the
pleural sleeve and anterior to the vertebral bodies. This protects the vessels
and assists in retraction beyond the anterior longitudinal ligament.
On occasion, an anterior release of the anulus and the anterior
longitudinal ligament is all that is done. The surgeon should use whatever
technique he or she is comfortable with to achieve incision of the anulus,
elevation of the cartilaginous end plate, and removal of the disc. A Cobb
elevator or similar instrument should be used to confirm motion. If fusion is
to be done, a rongeur and curet are used to remove all end-plate fragments
down to bleeding bone from anterior to posterior, but not including the
posterior longitudinal ligament. One may elect either to pack the disc space
with Surgicel or Gelfoam and continue dissection or to insert bone graft for
fusion at this time. The use of native rib, allograft, iliac crest autograft,
structural cages, or bone substitute for disc space fusion is optional when a
posterior fusion with instrumentation is to be carried out. We prefer to use
fibular structural allograft struts for anterior fusion when a posterior
fusion is to be performed (Fig.
3) but only milled autograft from either the rib or the iliac
crest when only anterior instrumentation is to be used. Following completion
of the release of the disc space and fusion, one or two chest tubes, depending
on local bleeding, are inserted under direct vision. We prefer to tunnel the
chest tube between the lower two ports, such that it goes over the rib above
the lower port and then into the chest through the second port. This tunneling
effect creates a valve to prevent an air leak when the chest tube is removed.
Pleural closure is optional. Lung inflation and mediastinal hydrostatic
pressure have maintained bone graft position in all of our patients treated in
the lateral position. It is important at this time to have the
anesthesiologist suction the dependent inflated lung to prevent the so-called
down lung syndrome or mucus plug formation. The posterior instrumentation and
fusion is usually carried out the same day, except in patients with severe
deformity, in whom interval traction may be used (Figs.
4-A, 4-B,
4-C,
4-D).
The learning curve for anterior instrumentation is quite steep, and the
indications should be extremely
rigid1,5,20,21.
The primary indication is a single structural thoracic idiopathic curve
between 40° and 70° as measured with the Cobb method, preferably in a
tall person. The end vertebrae should be between T4-T5 and T12-L1. The curve
should be extremely flexible on side-bending radiographs. It is desirable for
the patient to be tall because a longer chest with larger vertebral bodies
facilitates screw
insertion6. A
patient with hypokyphosis is an ideal candidate for this procedure, as
kyphosis generally increases after discectomy and anterior segmental
compression. Anterior instrumentation also offers the theoretical advantage of
saving levels from fusion. We prefer the patient to be in a true lateral
position, and we use the image-intensifier to identify levels. A marking
pencil is used on the skin to identify the midvertebral body position in the
sagittal plane at each level. A Steinmann pin is used over the skin to
identify the posterosuperior aspect of the vertebra. This is where the rib
head typically articulates with the vertebral body. An "x" is
marked on the skin. The image-intensifier is then rotated and, with a
Steinmann pin, a line is drawn onto the posterior chest wall to identify the
coronal direction in which the screw should be placed to allow it to be
positioned in the midportion of the vertebral body. The crossing of these
lines represents the direction of screw insertion.
The skin incisions, usually four or five, are made directly over a rib.
Placement of the incisions obliquely over a rib allows the trocar to be placed
above or below the rib to achieve accurate screw placement. The surgeon can
stand anterior or posterior to the spine for screw insertion. We prefer to
stand posterior to the spine. Following release of the anulus across the
anterior longitudinal ligament, a thorough discectomy is performed with
rongeurs, elevators, curets, and rasps. The disc space is then packed with
Surgicel. The image-intensifier is brought into the field, and an awl with a
staple is used to develop a starter hole for the rostral screw. The awl is
placed immediately anterior to the rib head in the midlateral portion of the
vertebral body and is directed somewhat anteriorly (Figs.
5-A,
5-B,
5-C,
5-D). One should never direct
the awl posteriorly for fear of guiding the screw into the spinal canal. The
screw is 6.5 or 7.5 mm in diameter. The length varies depending on the level,
with a length of 25 to 30 mm used at T4-T5, approximately 35 mm used in the
midthoracic spine (T6-T9), and 40 to 45 mm used in the lower thoracic spine
(T10-T12) in most
adolescents6. The
correct length can be determined with a depth gauge or by using the laser
etchings on the tap.
Depending on the instrumentation system, we prefer staples for the most
rostral screws (usually one or two and possibly three), with washers used for
more caudal screws. The rostral vertebrae are smaller, and removal of the rib
heads facilitates posterior screw placement. The body is tapped to the
opposite cortex, and a screw of appropriate length is placed. The screws are
placed anterior to the rib head and directed straight across. An attempt is
made to direct the screw across the vertebra from convex to concave pedicle
shadows on the c-arm image. The pedicle shadows are usually in the middle to
upper quadrant on the coronal image. We tend to leapfrog the placement of the
screws—i.e., we place them in every other vertebra—in an effort to
maintain the best alignment possible. The screw in between is then usually
easier to align. An endoscopic guide can be used to assist in attaining this
alignment. Also, by superimposing a rod horizontally over the posterior chest
wall and viewing it on the c-arm, one can get an idea of the height alignment
of the screw heads and whether the rod will fit comfortably in the screw
heads.
Once all screws are in place, the Surgicel is removed from the disc space,
and milled bone from either the iliac crest or a rib is inserted through the
bone funnel and packed into the disc space. A rod is then cut to the
appropriate length and is inserted from an inferior portal onto the screw
heads. There are various mechanisms for capturing the screws to the rod,
depending on the system that is used. We prefer to capture the screws from a
rostral-to-caudal direction. By capturing the top three screws and applying
compression, as opposed to performing sequential rod capture at each level
followed by compression, the rod can be cantilevered into the distal screws
with greater resistance to pull-out. Often the rostral screw head is not in
the correct plane, and when that is the case the options are to elevate the
top screws by adding a washer or two or prebending the rod.
Compression can be carried out with a rack-and-pinion device, cable
compressor, or compression forceps, depending on the surgeon's preference. If
the positions of the screws and rod are unacceptable, they can be readjusted
prior to torquing. Pleural closure is optional, and a chest tube is inserted
under direct vision. We then have the anesthesiologist suction the dependent
lung to prevent mucus plugging (Figs.
6-A and
6-B).
So-called down lung syndrome is seen most frequently with lengthy
operations and is characterized by absorption atelectasis, accumulation of
secretions, and formation of transudate in the dependent
lung5,24.
Mucus plugs can contribute to this condition and, if present, should be
cleared by the anesthesiologist. Airway leaks producing a pneumothorax can
occur as a result of the high pressure occurring with double-lumen
ventilation13,23.
An attempt is made to ligate or coagulate the segmental vessels as far away
from the great vessels as possible—i.e., immediately in front of the rib
heads. This leaves a long stump to grasp if the vessel bleeds. If bleeding
does occur, one should remain calm and remember that the field is magnified
fifteen times. One should immediately retract the thoracoscope so that a blood
splash does not obscure vision once bleeding has occurred. A large Kitner
retractor is used to apply pressure directly on the vessel toward the spine,
and then, with the use of a bipolar cautery, harmonic scalpel, or clip
applier, the vessel is secured and bleeding is controlled. Occasionally,
removal of the periosteum produces a large, active bleeding vessel. A dollop
of bone wax firmly applied with a Kitner retractor will usually stop it. For
all other bone-bleeding, we apply a putty (ground Gelfoam, topical thrombin,
and Avitene) as well as bone wax. The thoracotomy tray is always on the back
table and should be ready for immediate use if bleeding cannot be
controlled.
One should always be prepared to encounter lung adhesions, even when the
patient has not had previous intrathoracic intervention. It may be necessary
to release incidental adhesions to access the spine. It is important to
observe the pathway of instruments into the field. Instruments should always
be inserted toward the posterolateral chest wall and never directly downward.
Movement of instruments outside one's field of direct vision should be
avoided. A lung-stapling device should always be available to seal an air leak
in the lung if necessary.
A tension pneumothorax may occur if a sharp instrument enters into the
opposite lung28.
One should observe all guide-pins and other sharp instruments during their
insertion into the chest. The use of cannulated screws should be avoided if at
all possible. The vertebra should be tapped only to the opposite cortex in an
effort to not tap into the opposite chest cavity or damage the lung or great
vessels. The opposite lung fields should always be observed on the
image-intensifier during application of instrumentation to look for the
appearance of free air in the opposite chest cavity. If the pulmonary status
deterio-rates and there is evidence of free air in the opposite lung field, a
posterior thoracotomy can be performed on the opposite side.
Prevention of a dural tear is critical. One should define the dissection
plane of the anulus and limit it to the rib head posteriorly. The posterior
longitudinal ligament should never be released. Each disc space should be
observed for leakage of clear fluid. If a leakage of spinal fluid is seen
initially, the disc space should be packed with Gelfoam, thrombin, and Avitene
(the three are mixed as a putty). This may be all that is necessary. If the
leak continues, it can be sealed with Tis-Seal (fibrin gel). If there
continues to be a leak, a neurosurgical consultation may be beneficial.
Postoperatively, a patient with a dural leak should be well hydrated and kept
supine for two to three days.
The appearance of cloudy fluid after irrigation and suction suggests a
thoracic duct laceration. A diligent search should be carried out for the
thoracic duct leak, which is most often on the right side near the diaphragm,
and, if one is found, one should attempt to cauterize it. If a chylothorax
occurs, one should initiate hyperalimentation immediately.
If the spinal cord monitors show decreasing potentials, the
anesthesiologist should immediately administer steroids. Any instruments that
are in the chest cavity should be removed, and repeat monitoring should be
done. If there is an indication of spinal cord injury, a neurosurgery
consultation should be obtained immediately. Any spinal instrumentation should
be removed, the chest should be immediately closed, and urgent contrast
magnetic resonance imaging or computed tomography myelography should be
performed to rule out an epidural hematoma or a more severe spinal cord
injury.
Both the long thoracic nerve and the phrenic nerve may be at risk for
injury during portal placement. A thorough review of the anatomy is mandatory
when considering more rostral approaches.
It is important to be proactive and to warn the child and parents of
possible sympathetic release (i.e., cold leg or warm leg) following this
procedure. Sympathetic release is probably the rule rather than the exception
because, invariably, the sympathetic nerves cross over the vertebral bodies at
the level of the disc spaces and they are often incised. The family should be
assured that there will be a complete return of normal temperature response
within six months to a year.
It is important that adequate padding be placed on all extremities at the
time of patient positioning. Most often, skin pressure problems are associated
with longer operations, and they are more common when only rigid ports are
used. One should use a rigid port only for the thoracoscope and should make an
effort to alternate thoracoscope placement to improve visibility and prevent
neurapraxia.
The benefits of video-assisted thoracoscopic surgery for spine surgeons are
not readily evident. They must learn an unfamiliar technique, must acquire
specific skills, often require the assistance of a thoracic surgeon, possibly
need more operating room time, and must have a sufficient volume of patients
to acquire the expertise to make the procedure
worthwhile29.
Our learning curve was decreased by convening a team consisting of an
access (cardiothoracic) surgeon who had performed over 300 thoracoscopic
procedures. One can acquire experience by using animate or inanimate models,
proctoring, or perhaps performing hybrid procedures using both open and
endoscopic techniques. It is strongly recommended that an animate model (with
the potential for bleeding) be used so that the surgeon gains experience
manipulating the lungs and the great vessels. This experience has allowed some
surgeons to make an informed decision about whether to utilize this
technique.
Video-assisted thoracoscopic surgery for release of the disc space and
anterior fusion has been found to be as effective as thoracotomy at our
institution. The technique is time-consuming because of the need for multiple
passes to remove disc and end-plate fragments. The time could be decreased
with the development of instrumentation to allow expeditious removal of the
disc and end-plate fragments. Image-guided navigational systems could assist
in anatomical localization.
Use of endoscopic instrumentation is technically demanding with a steep
learning curve and may not be appropriate for all surgeons who treat spinal
deformity. Fear of injuring the aorta by the screw tips during use of
endoscopic instruments has diminished the enthusiasm for this method of
correction. Recent imaging studies have revealed screw tips penetrating the
vertebral cortex within millimeters of the
aorta30. Caution
must be used regarding the extent of the screw-tip penetration, which should
be kept to a minimum. Preoperative assessment of the diameter of the vertebral
body and multiple sizing of the screws is
mandatory31. An
intrinsically stable unicortical purchase screw might solve this problem in
the future.
We have investigated fusionless correction of scoliosis with use of staple
epiphysiodesis and have created a scoliosis model, but we have not performed
the procedure in
humans32. Other
centers have begun to use endoscopic stapling for progressive curves in young
patients33.
Endoscopic access will facilitate this technology. Finally, the current
problems of uncertain fusions and rod breakage could be decreased once
biologics such as genetically engineered bone morphogenetic proteins have been
found to be safe to enhance intervertebral disc space fusion.
Note: The authors thank Randal K. Wolf, MD, Atiq Durrani, MD,
and Jose Herrera-Soto, MD, for their review of the manuscript.
Blackman RG, Picetti GD, O'Neal K, et
al.Surgical technique for endoscopic anterior correction of
idiopathic scoliosis. Presented as a poster presentation at the Annual
Meeting of the Scoliosis Research Society; 1998 Sep 16-20; New
York, NY.
1998
Blackman R.Multiple level
anterior thoracic discectomy using an endoscopic exposure. Presented
as a poster presentation at the Annual Meeting of the Scoliosis Research
Society; 1993 Sep 18-23; Dublin, Ireland.
1993
Crawford AH, Wolf RK, Wall EJ,
Picetti GD, Blackman RG, O'Neal K. Pediatric spinal deformity. Alternative
technique for thoracoscopic release in pediatric patients. In: Regan JJ,
McAfee PC, Mack MJ, editors. Atlas of endoscopic spine
surgery. St. Louis: Quality Medical Publishing;
1995. p 225-31.225
1995
Crawford AH, Wall EJ, Wolf R.
Video-assisted thoracoscopy. Orthop Clin North Am.1999; 30: 367-85,
viii.30367
1999
[PubMed][CrossRef]
Sucato DJ. Thoracoscopic anterior
instrumentation and fusion for idiopathic scoliosis. J Am Acad
Orthop Surg.2003;11:
221-7.11221
2003
Lenke LG. Anterior endoscopic
discectomy and fusion for adolescent idiopathic scoliosis.
Spine.2003;28(15
Suppl): S36-43.28S36
2003
[PubMed]
Mack MJ, Regan JJ, McAfee PC, Picetti
G, Ben-Yishay A, Acuff TE. Video-assisted thoracic surgery for the
anterior approach to the thoracic spine. Ann Thorac
Surg.1995;59:
1100-6.591100
1995
[CrossRef]
Mehlman CT, Crawford AH.
Video-assisted thoracoscopic surgery: pediatric orthopaedic applications.
Endoscopic approaches to the anterior thoracic spine. In: Zdeblick TA, editor.
Anterior approaches to the spine. St. Louis: Quality
Medical Publishing; 1999. p
150-5.150
1999
Pollock ME, O'Neal K, Picetti G,
Blackman R. Results of video-assisted exposure of the anterior thoracic
spine in idiopathic scoliosis. Ann Thorac Surg.1996;62:
818-23.62818
1996
[PubMed][CrossRef]
Wall EJ, Bylski-Austrow DI, Shelton
FS, Crawford AH, Kolata RJ, Baum DS. Endoscopic discectomy increases
thoracic spine flexibility as effectively as open discectomy. A mechanical
study in a porcine model. Spine.1998;23:
9-16.239
1998
[PubMed][CrossRef]
Newton PO, Cardelia JM, Farnsworth
CL, Baker KJ, Bronson DG. A biomechanical comparison of open and
thoracoscopic anterior spinal release in a goat model.
Spine.1998;23:
530-6.23530
1998
[PubMed][CrossRef]
Connolly PJ, Ordway NR, Sacks T,
Kolata R, Yuan HA. Video-assisted thoracic diskectomy and anterior
release: a biomechanical analysis of an endoscopic technique.
Orthopedics.1999;22:
923-6.22923
1999
[PubMed]
Sucato DJ, Girgis M. Bilateral
pneumothoraces, pneumomediastinum, pneumoperitoneum, pneumoretroperitoneum,
and subcutaneous emphysema following intubation with a double-lumen
endotracheal tube for thoracoscopic anterior spinal release and fusion in a
patient with idiopathic scoliosis. J Spinal Disord
Tech.2002;15:
133-8.15133
2002
[CrossRef]
Newton PO, Shea KG, Granlund KF.
Defining the pediatric spinal thoracoscopy learning curve: sixty-five
consecutive cases. Spine.2000;
25: 1028-35.251028
2000
[PubMed][CrossRef]
Newton PO, Wenger DR, Mubarak SJ,
Meyer RS. Anterior release and fusion in pediatric spinal deformity. A
comparison of early outcome and cost of thoracoscopic and open thoracotomy
approaches. Spine.1997;22:
1398-406.221398
1997
[PubMed][CrossRef]
Regan JJ, Mack MJ, Picetti GD
3rd.A comparison of VATS to open thoracotomy in thoracic spinal
surgery. Read at the Annual Meeting of the Scoliosis Research Society;
1993 Sep 23; Dublin, Ireland.
1993
Mehlman CT, Crawford AH, Wolf RK.
Video-assisted thoracoscopic surgery (VATS). Endoscopic thoracoplasty
technique. Spine.1997;
22: 2178-82.222178
1997
[PubMed][CrossRef]
Schwab FJ, Smith V, Farcy JP.
Endoscopic thoracoplasty and anterior spinal release in scoliotic deformity.
Bull Hosp Jt Dis.2000;59:
27-32.5927
2000
[PubMed]
Wolf RK, Crawford AH, Hahn B.
Thoracoscopic first rib resection for thoracic outlet syndrome. In: Yim APC,
Hazelrigg SR, Izzat MB, Landreneau RJ, Mack MJ, Naunheim KS, editors.
Minimal access cardiothoracic surgery. Philadelphia:
WB Saunders; 1999. p 328-34.328
1999
Picetti GD 3rd, Ertl JP, Bueff
HU. Endoscopic instrumentation, correction, and fusion of idiopathic
scoliosis. Spine J.2001;1:
190-7.1190
2001
[PubMed][CrossRef]
Desiderio DP, Burt M, Kolker AC,
Fischer ME, Reinsel R, Wilson RS. The effects of endobronchial cuff
inflation on double-lumen endobronchial tube movement after lateral decubitus
positioning. J Cardiothorac Vasc Anesth.1997;11:
595-8.11595
1997
[PubMed][CrossRef]
Szegedi LL, Bardoczky GI, Engelman
EE, d'Hollander AA. Airway pressure changes during one-lung ventilation.
Anesth Analg.1997;84:
1034-7.841034
1997
[PubMed]
Dieter RA Jr, Kuzycz GB.
Complications and contraindications of thoracoscopy. Int
Surg.1997;82:
232-9.82232
1997
Nymberg SM, Crawford AH.
Video-assisted thoracoscopic releases of scoliotic anterior spines.
AORN J.1996;63:
561-76.63561
1996
[PubMed]
King AG, Mills TE, Loe WA Jr, Chutkan
NB, Revels TS. Video-assisted thoracoscopic surgery in the prone position.
Spine.2000; 25:
2403-6.252403
2000
[PubMed][CrossRef]
Lieberman IH, Salo PT, Orr RD,
Kraetschmer B. Prone position endoscopic transthoracic release with
simultaneous posterior instrumentation for spinal deformity: a description of
the technique. Spine.2000;25:
2251-7.252251
2000
[PubMed][CrossRef]
Roush TF, Crawford AH, Berlin RE,
Wolf RK. Tension pneumothorax as a complication of video-assisted
thorascopic surgery for anterior correction of idiopathic scoliosis in an
adolescent female. Spine.2001;26:
448-50.26448
2001
[PubMed][CrossRef]
Arlet V. Anterior thoracoscopic
spine release in deformity surgery: a meta-analysis and review. Eur
Spine J.2000;9 Suppl 1:
S17-23.9S17
2000
[CrossRef]
Sucato DJ, Kassab F, Dempsey M.Thoracoscopic anterior spinal instrumentation and fusion for idiopathic
scoliosis: a CT analysis of screw placement and completeness of
discectomy. Read at the Annual Meeting of the Annual Scoliosis
Research Society; 2001 Sep 21; Cleveland, OH.
2001
Sucato DJ, Duchene C. The
position of the aorta relative to the spine: a comparison of patients with and
without idiopathic scoliosis. J Bone Joint Surg Am.2003;85:
1461-9.851461
2003
[PubMed]
Wall EJ, Bylski-Austrow DI, Kolata R,
Crawford AH.Endoscopic mechanical spinal hemi epiphysiodesis
modifies spine growth. Read at the Annual Meeting of the Scoliosis
Research Society; 2003 Sep 12; Quebec City, Canada.
2003
Betz RR, Kim J, D'Andrea LP, Mulcahey
MJ, Balsara RK, Clements DH. An innovative technique of vertebral body
stapling for the treatment of patients with adolescent idiopathic scoliosis: a
feasibility, safety, and utility study. Spine.2003;28:
S255-65.28S255
2003
[PubMed][CrossRef]