Добавил:
Sekretar
kiopkiopkiop18@yandex.ru
t.me/Prokururor I Вовсе не секретарь, но почту проверяю
Опубликованный материал нарушает ваши авторские права? Сообщите нам.
Вуз:
Предмет:
Файл:Ординатура / Хирургия / Библиотека им академика М.И. Перельмана / Книга_6034_Библиотеки_им_академика_М_И_Перельмана.pdf
CHAPTER 47
Microscopic Lumbar Discectomy
Robert Kraemer, Alexander Wild, Holger Haak, Joerg Herdmann,
and Juergen Kraemer
MICRODISCECTOMY OR
MACRODISCECTOMY
Open lumbar discectomy is still the most frequent and
most important intervention in spine. The question is ho w
to perform open disc surgery—with a small or wide
exposure: microdiscectomy or macrodiscectomy? The
choice of procedure depends on the expected pathology:
• Is it multisegmental?
• Is there a concomitant spinal stenosis?
• Is there a tumor or postoperative fibrosis?
A monosegmental disc prolapse is the most frequent
cause for lumbar disc surgery. The surgical approach
should be as small as possible and wide as necessary.
Besides less trauma, less postoperative pain, and more
rapid mobilization, a small approach reduces the amount
of scaring which is important to prevent perineural fibrosis
which is a cause of the failed back surgery syndrome. The
interlaminar approach is similar for both microdiscectomy
and macrodiscectomy, but the wound and scar formation
might be more extensive in a wide approach. Using a
microscope provides better illumination and better threedimensional visualization, and is a better teaching tool.
Several studies report good outcome with shorter hospital
stay and earlier return to work rates compared to conventional disc surgery (1–3). On the other hand, there are
reports of comparable outcome with both procedures, as
well as concerns about specific complications (4,5).
There are potential pitfalls and errors with microdis-
cectomy. The limited exposure makes it easier to operate
at the wrong level, to overlook free fragments, and to
decompress inadequately. Technical difficulties and inexperience with the microscope can result in inadvertent
injury to neural structures and vessels in the spinal canal.
Because of these issues, many surgeons prefer a wide
exposure for open disc surgery that permits better intra-
operative orientation and visualization. It is a challenge
for a spine surgeon to abandon a wide exposure for
microdiscectomy without a loss of quality. Optimal intraoperative orientation with a small exposure can be
achieved by precise preoperative planning of the incision
and X-ray localization with a needle. A small incision
provides a better illumination of the operative f ield than
can be achieved by a headlamp. Surgery through a small
approach can be facilitated by special bayonet-shaped
instruments.
Microscopic lumbar discectomy is a four-step procedure starting from the skin to the anterior epidural space
that provides clear visualization of each layer and segment. The overlapping of the lamina over the disc space
varies by level and must be recognized. The disc space at
L5-S1 is interlaminar in location; at L4-5 the disc is partially co vered b y the lamina of L4; and at L3-4 and higher
the disc space is completely covered by the superior lamina. The skin must therefore be appropriate: for an
infradiscal L4-5 herniation the disc prolapse is below the
spinal process, for supradiscal L5-S1 herniation the disc
prolapse is at the level of the spinal process. For microdiscectomy, it is extremely important to visualize the
pathology in the center of the w ound. In microdiscectom y
the surgeon must be oriented as to the precise location of
the foramen, nerve roots, and pedicles.
For foraminal and lateral disc herniation the skin incision is placed 3 cm lateral to the midline and the approach is between the transverse process.
In conclusion, the surgical incision and approach
should be as small as possible and as wide as necessary.
One way to progress to microdiscectomy is to start with a
headlamp and to use microinstruments. The surgeon
should use the procedure with which he or she is most
comfortable. It is better for a patient to be operated upon
by an experienced macrosurgeon than by an inexperienced microsurgeon.
453
454 /SECTION V/SPECIFIC CLINICAL ENTITIES
CLASSIFICATION OF LUMBAR MOTION
SEGMENTS FOR MICRODISCECTOMY
For a better correlation between preoperati v e and intraoperative findings in microscopic discectomy it is helpful
to have anatomic landmarks that can be easily identif ied
both on radiographic images as well as intraoperatively.
In most classifications, posterior elements such as f acet
joints, laminae, and pedicles are key structures for the
surgical approach to the lumbar spine (7–11). McCulloch
(9) related his classification to the pedicles and compared
the lumbar segments to the stories of a house. The disc
level was the first story, the infrapedicle level the second
story, and the pedicle level the third story of the house.
Wiltse (11) added a suprapedicle level, which could be
considered the upper part of McCulloch’s third story, and
laterally (horizontally), a subarticular zone. For orientation during interlaminar lumbar disc microsurgery, the
surgeon has only a few anatomic landmarks. Exposure of
the facet joint should be avoided in order to maintain the
vascularization and innervation of its capsule and to
avoid damage to the joint itself.
For the routine interlaminar approach to the lower lumbar spine, disc-related orientation is more useful, especially for less experienced surgeons. After identifying the
inferior border of the lamina and the disc space by its
relation to an intraoperative localizing needle on an X-ra y
film, it is easy to f ind the pedicles, foramen, and nerve
roots.
The main difference from the classifications of
McCulloch (9) and Wiltse (11) is that the intraoperative
orientation is related only to the disc, not the foramen,
pedicles and facet joints, which cannot be seen in most
microsurgical approaches.
DISC-RELATED CLASSIFICATION
A disc and the adjacent vertebrae form a segment (the
Junghans motion segment) that is divided vertically into
levels and horizontally into zones (Fig. 47-1).
At the center of a segment is the disc level, with the
supradiscal level above and the infradiscal level below.
Supradiscal and infradiscal levels border on the middle of
the vertebra, which is identical with a line between the
inferior borders of the pedicles. Protruded disc material
can stay at the disc level or dislocate in a supradiscal or
infradiscal direction.
From the midline of the segment in a lateral direction,
there are three zones: medial, paramedial, and lateral. The
medial zone has a right and a left part. The middle of the
paramedial zone is identical with the center of the interlaminar approach at L5-S1 and, after removing parts of
the upper lamina, also at L4-5 and higher segments. Most
contained or noncontained disc herniations are in this
area. They lie under or close to the traversing nerve root
at the disc level or the supradiscal or infradiscal level. If
pathology medial to the traversing root is closer to the
segment midline, it lies in the medial zone.
The lateral zone begins at the medial border of the
pedicle and includes the foraminal area and the extraforaminal (far out) area. Disc herniations that lie lateral to
the traversing root usually have contact with the exiting
root in the foramen, causing a double-root syndrome. All
levels have the same zones except the infradiscal level,
which does not really have a lateral zone due to presence
of the pedicles.
Myelogram and anteroposterior reconstructions on
magnetic resonance imaging (MRI) show the dural sac
and nerve roots with bony structures.
In the lower lumbar segments, the nerve roots traverse
the disc and infradiscal area before they exit the spinal
canal through the intervertebral foramen of the segment
below. The vertical part of the nerve root that passes the
paramedial zone is called the traversing root until it enters
the lateral zone at the medial border of the pedicle. From
there on it is the exiting root.
The interlaminar approach to the lumbar spine always
exposes the traversing root centrally and the exiting root
cranially and laterally within the intervertebral foramen.
The pedicle is caudally and laterally located. The travers-
FIG. 47-1. Disc-related classification of the segment.
Above the disc level is the supradiscal level, and
below it is the infradiscal level; these border on the
middle portion of the vertebra. In the lateral direction
are the medial, paramedial, and lateral zones.
ing roots are intrathecal until they leave the dural sac and
enter the nerve root sheath at the axilla of the root. The
main part of the lumbar traversing root lies intrathecally.
The lumbar traversing roots course past the disc and
supradiscal levels intrathecally. The sheath-surrounded
part of the L3-4 and L5 traversing roots is very short. The
entrance point into the nerve root sheath (axilla point) for
the L5 root is infradiscal medial to the L5 pedicle, and for
the L3 and L4 roots it is caudal to the pedicle. For the S1
root, it is just below the L5-S1 disc. This means that in
microscopic lumbar discectomy an approach to the discal
and supradiscal level in L3-4 and L4-5 segments and to
the supradiscal level at L5-S1 exposes only the lateral
part of the dural sac with the traversing root inside, not
surrounded by a nerve root sheath. A medial part of traversing root does not exist at these levels of the spinal
canal and separation should therefore not be attempted.
Only an interlaminar approach to the disc and infradiscal
levels of L5-S1 and sometimes the intradiscal levels of
L4-5 show sheath-surrounded traversing nerve roots. Traversing nerve roots are most sensitive to mechanical
strain in their sheath-surrounded part, especially at the
entrance point, because they can be manipulated as easily
as in their intrathecal part.
Exiting roots have a nerve root sheath that naturally
adheres to the posterior surface of the vertebra and the
pedicle. The y are also sensiti v e to an y kind of mechanical
strain because they cannot move. Exiting roots are
located craniolaterally to the interlaminar approach and
the disc. They pass around the pedicle into the superior
(upper) part of the intervertebral foramen. The exiting
nerve root and the pedicle adjacent to it have the same
name: the L5 root passes beneath the L5 pedicle, the L4
root around the L4 pedicle, and so forth.
Fragment dislocation in the supradiscal direction often
causes double-root involvement, with simultaneous compression of the traversing and the exiting root. This
involves the intrathecal traversing root and the exiting
root from the segment above.
APPROACH T O THE LAMINA
After incision of the skin and fascia, contact with the
spinous process with a bone rasp leads to the upper corner of the interlaminar window. The bone rasp passes
along the inferior part of the spinous process to the inferior part of the lamina and then to the upper interlaminar
corner. W ith the 30° oblique view the surgeon looks at the
lower part of the spinous process and the medial par t of
the inferior lamina which form the upper interlaminar
corner (Fig. 47-2). The highest point of this corner has
specific relation to the disc space of the segment. At L5S1 this point lies above the disc, at L4-5 it lies at the disc
level, and at L3-4 and higher it lies below the disc space.
The inferior part of the lamina in the upper interlaminar corner is not covered by the ligamentum flavum. The
CHAPTER 47/MICROSCOPIC LUMBAR DISCECTOMY / 455
FIG. 47-2. A surgeon’s 30° oblique view of the upper areas
of a left-sided interlaminar window. Bony contact with the
inferior part of the spinous process leads directly to the
upper interlaminar corner.
superior part of the lamina in the lower interlaminar corner is much thinner and is partially covered by the ligamentum flavum. Before entering the spinal canal by
flavectomy it is useful to take another X-ray with a dissector in the upper interlaminar corner, which is much
closer to the disc than the needle used to radiographically
localize the proper level for skin incision.
SURGICAL VIEW THROUGH INTERLAMINAR
WINDOW AND DISTRIBUTION OF FRAGMENT
LOCATION
After removal of the ligamentum flavum from the lateral interlaminar area, epidural fat and the lateral portion
of the dura and the traversing root appear. It is not necessary to remove the ligamentum flavum in the medial
interlaminar area in order to expose nerve roots. The L5S1 disc lies in the middle of the interlaminar approach to
L5-S1, directly under the traversing S1 root. In the middle of the interlaminar approach to L4-5, directly under
the traversing L5 root, lies the infradiscal area of the L5
vertebral bone surface.
Exiting roots cannot be seen from this approach, they
can only be estimated. On a left-sided approach, the foramen and exiting root are on the surgeon’s left, and on a
right-sided approach they are on the surgeon’s right. The
segmental pedicle is on the opposite side (Fig. 47-3A,B).
In addition to the four main directions—cranial, caudal,
medial, and lateral—there are intermediary planes (craniomedial, craniolateral (foraminal), and caudomedial)
for describing fragment migration. Caudolateral position
of a fragment is not possible because of the presence of
the pedicles. Even if a fragment lies lateral to the traversing root in the infradiscal area, it is still in the paramedial
zone because the lateral border to this zone is the pedicle.
Infradiscal caudal herniations are the most frequent indi-
between bone and the traversing root, which is immobile
because of the presence of the pedicle laterally. Hernia-
456 /SECTION V/SPECIFIC CLINICAL ENTITIES
AB
FIG. 47-3. A, B: Left-sided microscopic interlaminar approach to the lumbar spine. In front of the
approach is the dural sac with the traversing root inside .Only at the disc and infradiscal levels of L5-S1,
and sometimes at the infradiscal level of L4-5, is there a traversing nerve root surrounded by a nerve
root sheath.
tions in other directions less frequently come to surgery.
The distribution of fragment migration in patients who
have been treated conservatively is completely different
from that in surgically treated patients (12).
FOUR STEPS TO THE DISC
First Step: Skin
Needle localization with X-ray is generally more accurate than localization by palpation. The needle should be
placed at a 90°-angle to the skin, approximately 2 to 3 cm
paramedically on the contralateral side (Fig. 47-4).
Palpation of iliac crest and spinous processes is often
misleading because of anatomic variations and difficulty
in palpating bony landmarks in obese patients. The 3 cm
skin incision is placed centrally at the appropriate level.
For supradiscal and discal herniations of L3-4, L4-5, or
L5-S1 the incision is placed at the level of the spinous
process. For infradiscal herniations, the incision should
be slightly below the spinous process.
Second Step: Ligamentum Flavum
After incision of the fascia and stripping aside the back
muscles, palpation along the inferior part of the spinous
process leads to the upper interlaminar corner (Fig. 47-2).
The highest point in this upper interlaminar corner is the
best place to enter the spinal canal. When the skin incision is correctly centered over the L5-S1 disc, the upper
interlaminar corner is the cranial part of the approach; for
the L4-5 disc it is in the middle of the approach. For L34 discs and higher the upper interlaminar corner is in the
caudal part of the approach (Fig. 47-5). A modified
Casper retractor is inserted to maintain exposure.
Before opening the spinal canal the upper interlaminar
corner of the inferior part of the lamina has to be identi-
fied in relationship to the needle landmark in order to
know where to find roots pedicle and discs in microscopic discectomy without further exposure.
Third Step: Posterior Epidural Space
The spinal canal should be opened to reach the parent
disc even if fragments have migrated. After removal of
ligamentum flavum and parts of the lamina the posterior
epidural space can be visualized. In the lateral part of the
interlaminar window at L5-S1 the surgeon has a direct
view on the transversing S1 root that is sometimes covered or surrounded by epidural fat. At L4-5, and in higher
segments, the interlaminar window is more medial and
the dural sac wider so that the transversing root is cov ered
by the lateral dural sac.
Fourth Step:Anterior Epidural Space
Medialization of the transversing root with a lone
nerve root retractor allows a direct view of the disc
level at L5-S1 after flavectomy and of the L4-5 and
more proximal levels after additional bone removal of
the lamina in the upper interlaminar corner. At L4-5
and higher levels a flavectomy without laminotomy
exposes the anterior epidural space on the vertebral
bony surface of the infradiscal zone. This area has
many epidural veins and should not be exposed if the
pathology is at the disc or supradiscal level. Medialization of the dura and nerve root, extraction of the disc
prolapse, and wound closure are the same as with conventional discectomy. A specialized disc extractor with
a depth block is recommended in order to avoid anterior perforation of the annulus fibrosus with possible
injury to the abdominal vessels (Fig. 47-6).
CHAPTER 47/MICROSCOPIC LUMBAR DISCECTOMY / 457
A B
FIG. 47-4. A–C: The needle is placed in a 90°angle paramedian on the contralateral side. The
disc level is marked and so the skin incision in correlation to the pathology shown on the magnetic
C
resonance image.
FIG. 47-5. Second step: lamina flavum at the disc level.
When the skin incision is centered over the disc at L5-S1
the approach exposes the ligamentum flavum with the
upper interlaminar corner cranially located. At L4-5 the
upper interlaminar corner is in the middle and at L3-4 it is
caudal and medial.
458 /SECTION V/SPECIFIC CLINICAL ENTITIES
A
FIG. 47-6. A, B: Rongeur with a depth guard for intradiscal maneuvers to avoid anterior vessel injur y (Aesculap).
B
LIGHT SOURCES
Good lighting is mandatory for a microsurgical approach. If there is no microscope available, a headlamp
can be useful, although only the surgeon will ha ve an adequate view. The use of a microscope is excellent for
teaching conditions—the assistant has the same view as
the surgeon —and it also provides optimal illumination
and magnification of the operating field. Performing
surgery with an operating microscope (Fig. 47-7)
demands a certain level of training and the learning curve
may vary between surgeons. The adapting time can be
individually dif ferent. Ne vertheless, a surgeon w ho uses a
microscope should be able to switch over to the head
lamp if technical problems with the microscope occur.
FIG. 47-7. Operating microscope. Surgeon and assistant
have the same view.
INSTRUMENTS
The standard instruments used for the microsurgical
approach to the lumbar disc are designed for manipulation in the spinal canal and are thin with a special angular shape.
RESULTS
The reported results of microscopic lumbar discectomy
vary. These variations are the result of different indications for surgery and different outcome measures. Results
after microsurgical procedures are reported by Caspar
(13), Kahanovitz (14), Krämer (15,16), McCulloch (6),
Silvers (2), W illiams (17), Wilson (18), and Zahrawi (19).
Two trials were conducted comparing microdiscectomy
with standard discectomy, both included clinical outcomes
that were similar (20,21). Because of different outcome
measures, metaanalysis is generally not possible (22).
EARLY COMPLICATIONS IN MICROSCOPIC
LUMBAR DISCECTOMY
The complications of open disc surgery and microscopic discectomy are discussed mainl y in books of experienced spine surgeons (6,23).
In the European Spine Society questionnaire (24) to
evaluate a risk and value score for different diagnostic
and therapeutic procedures in the spine, open discectomy
had the highest effectiveness for pain relief, but a negative overall risk value score because of complications and
poor results. Our recent studies evaluated different factors that influenced the outcome of open lumbar disc
surgery (24,25).
CHAPTER 47/MICROSCOPIC LUMBAR DISCECTOMY / 459
Classification
Complications of open lumbar disc surgery may be
classified as intraoperative, immediate postoperative, and
late postoperative according to when they are apparent
rather than when they occur. Complications of lumbar
spine surgery can either be general, and therefore
common to any type of surgery, such as thrombosis,
embolism, and anesthetic problems, or they may be specific for spine surgery.
Intraoperative complications are recognized immediately by the surgeon and should be recorded. Operative
reports are not always complete so their true frequency
remains uncertain. Many of these problems can be
avoided by meticulous preoperative planning. Some intraoperative complications are common, such as epidural
bleeding and durotomy, and can be managed quite easily.
Other complications, such as anterior vessel and visceral
injury, are severe, but fortunately extremely rare.
Immediate postoperative general complications such
as vomiting, thrombosis, and circulatory problems can
occur after any kind of surgery. Some of the specific
spine complications occurring during the operation may
be initially unrecognized by the surgeon and become
symptomatic and obvious in the days following surgery.
These include complications secondary to patient positioning, abdominal symptoms, and bladder disturbances.
Late postoperative complications after lumbar disc
surgery may become obvious after the patient leaves the
hospital. These include general complications like thromboembolism as well as specific complications such as
recurrent disc herniation, spondylodiscitis, and the failed
back surgery syndrome due to peridural fibrosis and
instability. Late complications can only be evaluated by
questionnaires or follow-up studies with patient examination since not all patients consult their surgeon when
these complications arise.
Intraoperative Complications
Missed Preoperative Checklist
The surgeon performing a lumbar disc operation
should examine the patient just prior to the surgical procedure to verify any recent change in symptoms or new
findings. Symptoms can change in a shor t time because
of fragment migration or resorption. A difference in visualization between MRI and intraoperative X-ray findings
must be recognized. Sometimes a lumbosacral segment
can be seen on MRI but not on X-ray. General anesthesia
should commence only when all preoperative imaging is
complete including an X-ray with the needle localization.
Wrong Level Exploration
Precise preoperative planning is one of the main
prerequisites for successful microsurgery and avoiding
wrong level exploration. For McCulloch (6) it is the most
important prerequisite. In our comparative study (25),
wrong level exploration occurred 1.2% of the time in a
group of very experienced surgeons and 3.3% of the time
in a group of less experienced surgeons.
Wrong level exploration is more likely to occur at L45 and higher segments than at L5-S1. In all cases the correct segment was ultimatel y identified intraoperatively b y
X-ray. As described previously it is useful to take a second X-ray with a dissector in the upper interlaminar corner before flavectomy.
Missed Pathology
Missed pathology means that the compressive pathology causing the clinical symptoms was not adequately
addressed. This can happen w hen wrong lev el exploration
is not recognized and other nonsignificant pathology is
removed. Under such circumstances the patient awakes
with the same pain or worse than before surgery.
If the suspected intraoperative pathology is not found,
an intraoperative myelogram can be considered.
Other Pathology
Other pathology means the surgeon finds a different
pathologic entity than what w as expected but w hich could
have caused the clinical symptoms. This could be an
undiagnosed neurinoma or a synovial cyst from the facet
joint. In these cases a closer look at the imaging pictures
should be undertaken. In some cases it might be necessary to perform an intraoperative X-ray or myelogram to
identify missed pathology.
Bleeding or Epidural Hematoma
Epidural hematoma causing symptomatic neurologic
compression or cauda equina syndrome is one of the most
feared complications of spine surgery. During a posterior
approach, lumbar spinal canal arterial bleeding from the
back muscles and epidural venous bleeding are the most
important causes for such bleeding. Intraoperative bleeding can be minimized by positioning the patient prone
with the abdomen hanging freely.
Arterial bleedings from the back muscles should be
identified and coagulated carefully.
At the end of the surgical procedure, after the muscle
retractor is removed, the muscle walls should be checked
for bleeders because prolonged muscle retraction may
temporarily occlude potentially significant muscle bleeders, which could begin bleeding after muscle layer closure. When an epidural hematoma is identified, surgical
intervention must be performed as soon as possible to
evacuate the hematoma.
Epidural vein bleedings do not cause compression of
the dural sac, but do cause cauda equina syndrome. Some
460 /SECTION V/SPECIFIC CLINICAL ENTITIES
experienced spine surgeons (26–28) believe that epidural
vein bleeding often stops when the disc fragment is
removed and after wound closure. We prefer to tamponade as long as possible before using bipolar cautery.
Excessive cautery of epidural veins may inhibit the nutrition of the nerve roots and may be the cause of epidural
fibrosis and postdiscectomy syndrome (failed back syndrome).
An epidural hematoma, even if it does not compress
the dural sac, can also cause epidural fibrosis. The main
reason to prevent and stop bleeding from epidural v eins is
that they obscure the visual field. Because of the limited
approach in microdiscectomy, even a small amount of
bleeding may appear as a major hemorrhage under the
microscope and make it difficult to perform a safe and
adequate discectomy. Therefore the following precautions should be followed to prevent intraoperative bleeding:
• positioning of the patient with abdomen hanging freely
•avoid exploring the posterior surface of the vertebra if
it is not necessary
• retract epidural veins with the retractor before entering
the disc space
• cauterize veins if they are in the way.
If epidural vein bleeding occurs it may be better to
remove as much of the protruded disc material as possible before taking care of the bleeding. The bleeding during this maneuver could be managed by continuous suction and the use of cotton tamponades. After removal of
the disc prolapse, it is easier to expose the bleeding vein
and cauterize it if necessary.
For continuous bleeding from cancellous bone we use
a small amount of bone wax.
In our series excessive bleeding occurred in 7.1% of
patients treated by the group of experienced surgeons and
in 3.5% of the patients treated by the group of very experienced surgeons. In all cases, excessive epidural vein
bleeding did not cause intraoperative or immediate postoperative complications (25).
Durotomy
Injuries to the dura with loss of cerebrospinal fluid
(CSF) occur in many types of spine surgery. Clear fluid
in the wound should not automatically mean dural tear. It
could also arise from a puncture hole from previous
myelography, from a spinal anesthetic, or from inadvertent dural puncture from an epidural injection days before
surgery. Other causes of intraoperative fluid include synovial fluid from facet joints (6) or from a w et cotton patty.
Unfortunately presence of clear fluid usually means
CSF from inadvertent durotomy by surgical instruments. Most commonly this occurs during opening of
the dura by incision of the ligamentum flavum. This can
happen when the ligamentum flavum is v ery thin, which
occurs with lumbosacral anomalies (6), or when a big
disc herniation displaces the dural sac posteriorly under
the ligamentum flavum. This is why we prefer a twostep flavotomy with a special semi-sharp dissector.
Under special conditions, intentional durotomy is necessary to deal with intradural pathology, which is rare in
lumbar disc surgery (6).
When a CSF leak is recognized, localization and
assessment of the injury must be determined: Is it medial
or lateral, caused by incision or punch, are nerve roots
involved? When the durotomy is localized, it is better to
avoid it in order not to inadvertently enlarge the hole.
After the disc herniation is removed, there is more space
and less tension on the dura and suture repair is easier. A
head down/back up position reduces dural tension and
empties the dural sac. Tears of more than 3 mm in length
should be closed with 6-0 sutures. Usually, the microsurgical exposure must be extended. Small punctures can be
left alone. We prefer to put a small free f at graft from the
subcutaneous fat to the dural repair. The patient should
have intra v enous antibiotics and be kept in bed for 3 da ys.
Complications of dural tears include headache due
to CSF loss, CSF fistula, and postoperative pseudomeningocele, which can be seen by MRI. Our own experience with a follow-up study comparing patients who
had intraoperative dural tears with a control group
showed better results in the control group (25). With a
two-step blunt perforation of the ligamentum flavum,
appropriate instrumentation, and good visualization of
the lateral dura and nerve root border it should be possible to reduce the number and extent of dural tears in lumbar microdiscectomy to a minimum. In conclusion, durotomies are a matter of experience. They occurred in the
group of less experienced surgeons 7.2% of the time and
among the very experienced surgeons 0.8% (p < .001) of
the time (25).
Nerve Root Lesion
The incidence of nerve root lesions after lumbar spine
surgery has been estimated at 0.2% (29). Such injury may
be suspected postoperatively by the presence of a new or
increased neurologic deficit. Iatrogenic intraoperative
nerve root injuries are classif ied by the site where they
occur, proximal to the foramen or extraforaminal, and by
the way in which the injury occurs: open by sharp instrumentation or closed by excessive traction, compression,
or heat from electrocautery. Poor visibility, perineural
adhesions, and congenital neural anomalies such as conjoined nerve roots are the most common causes of damage to the nerve roots. Therefore, it is absolutely necessary to def ine the lateral border of the root and dural sac
before removing an y material from the spinal canal. Ev en
when the neural elements are safely retracted by a nerve
root retractor, the tissue in the anterior epidural space
should be identified by the 2 mm dissector. The bright
CHAPTER 47/MICROSCOPIC LUMBAR DISCECTOMY / 461
white of disc material should not be mistaken for epidural
fat and vessels.
Rootlets may herniate through a durotomy. After reduction of the rootlets the dura must be repaired. Small
defects may be covered by a free fat graft especially if a
suture could strangle the nerve root. The most vulnerable
area for an open nerve root legion is the axilla of the exiting nerve root. Thus effects to remove intradiscal fragments should not take place medial to the nerve root in
the axilla. One of the principles of microdiscectomy is to
stay lateral to the nerve root in order to avoid axillary
injury.
Anterior Vessel Injury,Visceral Injuries
When a rongeur penetrates the anterior annulus fibrosus, it may contact a major vessel that lies immediately in
front of the lower lumbar discs. Grasping maneuvers in
order to clean out the disc material may rupture the vessels. The most frequent lesion is an isolated injury to the
left common iliac artery (30) caused by surgery of the
L4-5 disc. The overall complication rate for anterior vessel injury is 0.045% (31). Only 50% of such injuries are
immediately apparent with a dramatic unexplained fall in
blood pressure and excessive hemorrhage from the disc.
In these cases disc surgery has to be stopped immediately,
the wound closed, and the patient turned over for a
laparotomy and repair of the injured vessel.
In 50% of the patients the symptoms of anterior vessel
injury and other abdominal injuries are recognized later
in the recovery room with extreme hypotension and
painful abdominal swelling. In these cases, laparotomy
must be performed immediately. Even with prompt
action, the mortality of this complication is approximately 50% (6). Prevention of this major complication is
possible if intradiscal maneuvers are performed only with
rongeurs that cannot be inserted deeper than 25 mm. This
leaves an adequate safety margin since the anteroposterior disc diameter is 35 to 40 mm on average (Fig. 47-6).
Immediate Postoperative Complications
Postoperative Leg Pain and Neurologic Deficits (Table
47-1)
Although not usually considered a complication in
most series, persistent or residual leg pain after nerve root
decompression surgery of the lumbar spine can be considered a complication. If the correct level was operated
upon, and if neurologic symptoms are not severe or progressive, one can wait.
It is important, however, to consider the possibility of a
residual disc fragment or a recurrent herniation. Indications
for a careful postoperative neurologic examination and a
repeat computed tomography (CT) or MRI study are:
•severe leg pain lasting more than 2 to 3 days
• progressing neurologic deficit
• cauda equina syndrome.
When a nerve root has been compressed for a long
time by a disc herniation or an osteophyte it may not
become asymptomatic immediately after decompression.
The reasons for residual symptoms are not completely
understood, but include the duration of compression, the
presence of comorbidities such as diabetes, intraoperative
nerve injury, compression from a hematoma, inadequate
postoperative pain medication, and individual pain sensitivity.
In addition to wrong level exposure, missed additional
pathology at the current level is another cause of failed
lumbar disc surgery. Additional disc fragments can be
missed or concomitant bony stenosis may not be appreciated and thus decompressed. The patient may awake from
surgery with the same pain or it might even be worse
because of additional operative trauma and postoperative
hematoma. Repeat CT or MRI should be performed to
detect the missed fragment or other pathology.
Continued postoperative leg pain and neurologic
symptoms of the same or increased intensity as before
surgery can be caused by either missed pathology or by
an early recurrent disc herniation. This could be caused
by abdominal pressure such as by coughing during the
immediate postoperative period.
Usually symptoms from a recurrent disc prolapse
occur after a pain-free interval. It may occur when the
patient begins to stand and axially load the spine with
more frequency. In our series we had 0.2% rate of recurrent disc herniation in the f irst week after surgery. Once
a new herniation is verified by repeat CT or MRI conservative management or revision surgery at the same level
may be considered. The risk of recurrent disc her niation
cannot be eliminated by extensive disc curettage
(6,29,30). It is generally recommended that all disc material under the annulus perforation that could lead into a
recurrent herniation be removed at the time of the initial
surgery.
All variations of the kneeling position that are used in
lumbar disc surgery can produce compression on the skin
and neurologic structures. Brachial plexus stretch injuries
and compression of the radial and ulnar nerve can
develop by the hyperabduction of the arm. Bernsmann
(32) observed two cases of slight brachial plexus dysfunction in our series, all of which disappeared in the first
few days following surgery . Se v ere lesions from position-
TABLE 47-1. Causes for postoperative leg pain
Residual symptoms from original condition
Intraoperative nerve root injury
Residual fragment disc, foreign body retention
Early recurrent disc herniation
Nerve irritation secondary to intraoperative positioning
462 /SECTION V/SPECIFIC CLINICAL ENTITIES
ing, such as cervical myelopathy from hyperextension of
the neck, or visual disturbances (6,33,34) from failure to
protect the eyes during surgery and in the prone position
are very rare and should be avoided by proper head positioning.
There is a wide range of possible cauda equina symptoms, from slight bladder disturbances to the fully developed cauda equina syndrome with perineal anesthesia,
urinary incontinence and decreased rectal tone, and bilateral progressive leg weakness.
The absence of any of these clinical features does not
rule out a developing cauda equina syndrome.
Injury to the cauda equina can occur at surgery from
direct damage to the nerves or postoperatively from
hematoma. We have not seen a cauda equina syndrome
from fat graft compression in our large prospective randomized study of free fat graft versus no fat graft for
epidural scarring (9), although this condition has been
reported by others.
Concern about a possible cauda equina syndrome mandates a thorough neurologic examination and immediate
CT or MRI. If a compression lesion is found, immediate
surgery to decompress the cauda equina is necessary,
although a study showed that there is no statistically significant difference in outcome between patients who had
decompressive surgery within the first 20 hours after the
onset of cauda equina syndrome and those who had
surgery 24 to 48 hours after onset (35).
The reported incidence of disc space infection ranges
from 0.13% to 0.9% (6,30,36–38). Most studies recommend infection prophylaxis with antibiotics.
It has been claimed that microdiscectomy has a higher
infection rate than standard disc surgery because of contamination by the microscope. However, publications on
microdiscectomy surgery (30) and our own experience
show that the deep wound infection rate in microdiscectomy surgery is not significantly higher than with traditional discectomy.
CONCLUSIONS
It is impossible to avoid all complications in any
surgery , including lumbar microdiscectomy . According to
McCulloch (6), the two major criticisms of microdiscectomy are wrong level exploration and missed pathology.
If wrong level exposure is not recognized, pathology will
be missed. The risk of complications with lumbar
microdisc surgery can be minimized if meticulous attention is given to preoperative, intraoperative, and postoperative details (24). It is helpful for spine surgeons to
master the microsurgery learning curve by working with
other experienced spine microsurgeons and to read the
literature about how to a void intraoperati ve complications
and how to manage them if they occur.
The outcome of lumbar disc surgery depends heavily
upon proper patient selection (Table 47-2). The right
TABLE 47-2. Ways to avoid complications in lumbar
microdisc surgery
Proper patient selection
Surgeon training
Preoperative planning
Systematic 4-step surgical approach
Infection prophylaxis
Postoperative care
patient with the right indication for microscopic disc
surgery will have a good result if a well-trained surgeon
removes the disc fragment using a standard approach.
The learning curve in micro-decompression surgery can
be improved upon in special training courses that provide
instruction for working with the microscope on cadaver
spines. Infection can be decreased by careful draping of
the microscope and by the use of prophylactic antibiotics.
Recurrent disc herniation is an uncommon but impor tant
complication in lumbar disc surgery that may result in
another operation with a greater risk for complications
and scar formation. In many cases it is the beginning of a
failed back surgery syndrome.
REFERENCES
1. Findlay G. 10 year follow study after lumbar microdiscectomy: a comparative study. Paper presented at ISSLS Meeting; 1997; Singapore.
2. Silvers R. Microsurgical versus standard lumbar discectomy. Neurosurg 1988;22: 837–841
3. Wilson D, Harbaugh R. Microsurgical and standard removal of protruded lumbar disc: a comparative study. Neurosurgery 1981;8:
422–427
4. Caspi I. Percutaneous discectomy of lumbar spine. Int Orthop 1987;2:
331.
5. Fager C. Comments on microsurgical and standard removal of the
protrudes lumbar disc. A comparative study. Neurosurgery 1987;8:
426–427.
6. McCulloch J, Young PH. Essentials of spinal microsurgery. Philadelphia: Lippincott-Raven, 1998.
7. Akkerveeken PF. A taxonomy of lumbar stenosis with emphasis on
clinical applicability. Eur Spine J 1994;3:130.
8. Finneson B. Low back pain. Philadelphia: JB Lippincott, 1980.
9. McCulloch J. Principles of microsurgery for lumbar disc disease. New
York: Raven, 1990.
10. White A, Watkins R, Ray C. Surgical anatomy and nomenclature. In:
White A, Rothmann R, Ray C, eds. Lumbar spine surgery. St. Louis:
Mosby, 1987.
11. Wiltse 1992
12. Krämer J. A new classification of lumbar motion segments for
microdiscotomy. Eur Spine J 1995;4:327–334.
13. Caspar W. A new surgical procedure for lumbar disc herniation causing less tissue damage though a microsurgical approach. Adv Neurosurg 1977;4:74–80.
14. Kahanovich N, Viola K, McCulloch JA. Limited surgical discectomy
and microdiscectomy: a clinical comparison. Spine 1989;14:79– 81.
15. Krämer J, Ludwig J. Die operative Behandlung des lumbalen bandscheibenvorfalls. Der Orthopädie 1999;28:579–584.
16. Krämer J (1990) Intervertebral disc disease. Causes, diagnosis, treatment and prophylaxis, 2nd ed. Stuttgart/New York: Georg Thieme/Verlag.
17. Williams R. Microlumbar discectomy: a conservative surgical approach to the virgin herniated lumbar disc. Spine 1978;3:175–182.
18. Wilson DH 1979.
19. Zahrawi F. Microlumbar discectomy. Is it safe as an outpatient procedure? Spine 1994;9:1070–1074.
20. Henrikson L, Schmidt V, Eskesen V et al. A controlled study of micro-