22.1 Introduction

With the development of medical technology, the era of minimally invasive techniques is coming in spine surgery. As one important branch of minimally invasive surgery, spine microscopic surgery aims at minimizing surgical trauma and achieving effective outcomes, using microscopy, endoscopy and special apparatus during surgical process. At present, spine microscopic surgery has been carried out in all areas of the spinal axis, including discectomy, laminectomy, introspinal tumor resection, spinal disc fusion, vertebral fusion, anterior release operation for scoliosis or kyphosis malformation, etc.

In the middle 1970s, microscopy has been applied in spine surgery. Harold and Williams were the pioneers in microsurgical operation for treatment of lumbar disc diseases. Afterwards, the microscope became one of the most important tools in spine surgery. In 1990 McCulloch proposed use of microscopy and special instruments to accomplish lumbar laminoplasty for spinal stenosis, with lateral and central and intervertebral foramina on the decompression side. In 1997 Foley and Smith launched a new posterior approach microendoscopic discectomy (MED). This new concept lumbar discectomy is known as an important breakthrough in minimally invasive surgery and endoscopic spine surgery. With innovations in MED, its indications have gradually expanded to complicated lumbar disc herniation and lumbar spinal stenosis. In 2001 Adamson used posterior endoscopic system to conduct foramen decompression laminectomy and discectomy for unilateral cervical spondylotic radiculopathy, resulting in satisfactory surgical outcomes. In 2005 Isaacs firstly tried endoscopic thoracic discectomy in two cadavers. He concluded that compared with other decompression operations, the endoscopic thoracic discectomy could achieve similar therapeutic efficacy but decreased trauma obviously. In 2002 Yeung introduced Yeung’s endoscopic spine system (YESS) and in 2006 Hoogland proposed transforaminal endoscopic spine system (TESSYS), the two methods of percutaneous transforaminal endoscopic discectomy. Compared with YESS, TESSYS has wider indications, as a more direct and thorough decompression technique. In a word, endoscopy can obtain clear images, making all kinds of operations simple and easy, and reducing trauma and operation cost.

In Europe, cervical spondylosis, lumbar disc herniation and other degenerative diseases are treated under microscopy mainly by neurosurgeons at a department of neurosurgery. However, surgical management of spinal diseases is basically undertaken by departments of orthopedics in China. Most surgeons make open operations, which greatly hinders the application of microsurgical techniques in spinal surgery. Although large-scale prospective clinical trials on microsurgical operation are far from enough, it is sure that with the development of science and technology, microsurgical operations will be greatly improved. More and more spine surgeons gradually accept the concept of minimally invasive spine operation, and endoscopic technology will become the mainstream surgical procedure in the future.

22.2 The Microsurgery Treatment for Lumbar Disc Herniation

With the development of microsurgery, microsurgical techniques are more and more widely applied in the field of spine surgery, especially for a common disease—lumbar intervertebral disc herniation. As early as in 1977, Caspar and Yasargil first described lumbar discectomy under a surgical microscope. Clinical case-control studies also confirmed shorter hospital-stay and rehabilitation time, and ultimately better clinical outcomes by the microscopic discectomy than by conventional surgery.

Classic microsurgery for lumbar disc herniation includes lumbar microdiscectomy and minimally invasive transforaminal lumbar interbody fusion and instrumentation (Micro-TLIF).

22.2.1 Introduction

Microsurgery treatment for lumbar disc herniation includes traditional surgical decompression with a surgical microscope (Fig. 22.1) or a pair of operation loupes (Fig. 22.2). The goals of these procedures are to reduce the morbidity associated with traditional posterior lumbar approaches, and at the same time to achieve effectiveness and safety. The microsurgery treatment has following features:

  1. 1.

    More accurately identify anatomical structures.

  2. 2.

    Hopefully minimize the risk of damage to dura and nerve root.

  3. 3.

    Epidural veins can be identified, protected, and coagulated preemptively, thus minimizing blood loss.

Fig. 22.1
figure 1

A surgical microscope

Fig. 22.2
figure 2

Operation loupes

However, a surgical microscope is different from a pair of operation loupes (Table 22.1). Different kinds of surgical microscopes have been applied in spine surgery, neurosurgery, hand surgery and gynaecological surgery and so on. The spine surgical microscope must have a wide-field and a stand high enough to allow lens with at least a 300-mm focal length to provide enough operating space between the microscope and the patient.

Table 22.1 Advantages and disadvantages of surgical loupes and microscope

22.2.2 Lumbar Microdisectomy

22.2.2.1 Preparation, Positioning, Setup

General endotracheal anesthesia or epidural anesthesia is used. The patient should be placed in the prone position on an appropriately sized Wilson or Andrews spinal frame, with care that the abdomen hangs free. The frame should be slightly flexed to diminish the lordotic curve and widen the lumbar interlaminar spaces.

Surgeons generally stands on the same side of the disc herniation, unless there is associated symptomatic foraminal and/or extraforaminal root compression, whereupon they will be on the opposite side. The surgical microscope is positioned on the same side or opposite side of the surgeon.

When surgeons use operation loupes and head light during surgery, the height of the operation table should be adjusted according to the working distance of the loupes (Fig. 22.3).

Fig. 22.3
figure 3

Surgeons wearing operation loupes

22.2.2.2 Exposure

After localizing radiographically or with anatomical landmark, a skin incision is made in the midline. The incision is carried down through skin and subcutaneous tissue and fascia next to the spinous processes. A subperiosteal dissection of the paraspinous musculotendinous attachments is performed. Bleeding is controlled strictly. The facet joint capsule should be kept intact during exposure. Before entry into the spine canal, the surgeon should confirm the level again with the C-arm to avoid unnecessary removal of the bone and exposure of the spinal canal.

22.2.2.3 Exposure and Disectomy

From this point on, the operation is performed with the aid of an operation microscope, which can avoid excessive bone removal during laminotomy. Also, the flavum ligament should be preserved as much as possible to reduce adherence of the dura sac to its surrounding tissues. Once the laminotomy has been performed, the nerve root should be mobilized to expose the disc herniation (Fig. 22.4) before disectomy is performed. Dissection through the epidural fat with bipolar forceps and coagulation of veins with bipolar cautery are frequently necessary. Preventing and controlling epidural bleeding is one of the key aspects of a discectomy operation which is expeditiously and safely performed. If on the other hand, it is difficult to control bleeding by preemptive bipolar electrocautery, then small thrombin-soaked pledgelets and/or cotton paddings above and below the disc herniation can be used to tamponade the bleeding.

Fig. 22.4
figure 4

A nerve root is clearly identified under a microscope

22.2.2.4 Closure and Postoperative Management

The area of decompression is irrigated with normal saline. The most common complication is inadvertent dural tear (when possible, a watertight repair should be performed). Negative pressure drainage tube is put into the wound. The deep and superficial fasciae are closed in layers. The skin is closed with staples or Dermabond.

Antibiotics are routinely used postoperatively. The patients receive NSAID for postoperative pain relief and controlling inflammation around the nerve root. The drainage tube is removed on the first day postoperatively. Then the patient is encouraged to ambulate.

22.2.3 Transforaminal Lumbar Interbody Fusion and Instrumentation

22.2.3.1 Preparation, Positioning

General endotracheal anesthesia or epidural anesthesia is used. The patient is placed prone on a Jackson table (Fig. 22.5). It is important to keep the abdomen free to decrease venous bleeding. On the other hand, it is easier to restore the lumbar lordosis.

Fig. 22.5
figure 5

Jackson table

22.2.3.2 Exposure

The exposure method has been illustrated before, but difference also exists. In order to place the pedicle screws and perform foraminal decompression in the open surgery, the facet joint capsules of the rostral level and the lateral margins of the facet joint need to be revealed.

During the percutaneous placement of pedicle screws, a skin incision approximately 4–5 cm lateral to the midline is made, and then the entry point is exposed through bluntly muscle-splitting approach, which can decrease the damage to the surrounding soft tissues. However, this technique depends too much on the intraoperative radiography, so it increases intraoperative fluoroscopy time and radiation exposure of both the patient and surgeons.

22.2.3.3 Exposure and Disectomy

From this point on, the operation is performed with the aid of an operation microscope. To gain access to the posterior disc space, a unilateral facetectomy is performed and the flavum ligament is partly removed. The herniated disc is exposed after the nerve root and dura sac are retracted and protected. Dissection through the epidural fat with bipolar forceps and coagulation of veins with bipolar cautery are frequently necessary. Discectomy includes excision of the posterior annulus and disc material in order to expose a large surface area of contact for bone graft placement while preserving the osseous end plates. These maneuvers must be performed carefully in order to prevent violation to the subchondral bone of the end plates, especially in old female patients with osteoporosis. All bone resected is cleaned off soft tissues and saved for use as autologous bone graft.

In addition, care must be taken to avoid damage to anterior vascular structures during disc space preparation. The average depth of the disc space is 35 mm in the center of the disc and 25 mm at the level of the facet. Disc preparation instruments are commonly marked so that the depth they are placed into the disc space can be judged in order to minimize the risk of injury with repeated placement and removal of instruments.

22.2.3.4 Cage and Graft Placement

Once the disc space has been properly decorticated, trial spacers are placed in the intervertebral space prior to final placement of graft material. The determination of size is made based on preoperative and intraoperative radiographs, as well as direct visualization and trial implant tension. Local autologous bone graft may be placed anterior to or packed within the interbody device. Implant size is verified and one interbody cage/graft is placed. Once grafts are inserted, the neural elements are inspected to ensure that any direct compression or excessive traction from overdistraction is not present. The distraction instruments are then removed to allow the end plates to compress against the graft material and to prevent graft extrusion.

22.2.3.5 Pedicle Screw Fixation

Pedicle screw instrumentation may be performed before or after discectomy and placement of interbody grafts according to the surgeon’s preference. Care must be taken to avoid injury to the rostral facet joint capsule, decreasing the risk of the adjacent segment degeneration. The C-arm is used to confirm the right position of the pedicle screw to avoid injury to the dura or nerves.

Care must be taken to avoid overdistraction against screws as this may lead to loosening of the screw-bone interface and jeopardize ultimate fixation. If distraction forces are placed on the pedicle screws, it may be advisable to exchange the screws for ones with a larger diameter following the distraction maneuvers to increase the fusion rate.

With the continuous development of minimally invasive techniques, percutaneous placement of pedicle screw has been widely used in clinics. Percutaneous placement of pedicle screws can reduce stripping of the paraspinal muscle, blood loss and soft tissues injury. Tag pedicle projection position in the skin with the aid of intraoperative fluoroscopy devices, then determine the lateral distance to the midline of the skin incision according to the thickness of the soft tissue, and insert Jamshidi needle through the skin incision. Finally the pedicle screw is placed with the guidance of the needle. In the process of placement, care is taken to ensure that there is always a bony “backstop” of residual vertebral body distal to the K-wire tip and it is monitored with frequent live fluoroscopic images to guard against inadvertent advancement.

22.2.3.6 Closure and Postoperative Management

The key points about closure and postoperative management have been described in the lumbar discectomy section. The drainage tube is removed on the second day postoperatively. Then the patient is encouraged to ambulate.

22.3 Endoscopic Surgery for Treatment of Lumbar Disc Herniation

Endoscopic surgery to treat lumbar disc herniation mainly includes microendoscopic discectomy (MED) and percutaneous endoscopic lumbar discectomy (PELD). Discectomy, decompression and releasing of a nerve root can be performed with the help of endoscopy through a minimal incision. The main advantages of this technique are less invasive procedures, minimized trauma of the paravertebral muscles and soft tissue, and magnified spinal anatomy to clearly show the nerve root and disc herniation. With new techniques and new materials constantly emerging, the indications for endoscopic surgery become wider and wider. Previously, it was used only to treat lumbar disc herniation, but now also lumbar spinal stenosis; only discectomy was done under endoscopy, but now also the spinal internal fixation. Minimal invasion is the future of the spinal operation.

22.3.1 Application of Microendoscopic Discectomy

Microendoscopic discectomy is performed under the guidance of endoscopy. First of all, the working channel is established, with a diameter of about 1.6–1.8 cm. Then remove part of the lamina, and resect disc herniation using endoscopic instruments. The whole procedure is carried on with the help of a special endoscopic imaging system, which enlarges vision several times, and transmits the images to the video monitor by the optical fiber. So the nerve root is distinguished clearly, avoiding the risk of nerve injury (Fig. 22.6).

Fig. 22.6
figure 6

Microendoscopic discectomy

22.3.2 Microendoscopic Discectomy

Operative procedures

  1. 1.

    The patient should be placed in the prone position on an open-frame operating table throughout surgery. The abdomen should hang free to reduce the intervertebral venous plexus pressure and reduce intraoperative hemorrhage.

  2. 2.

    The lumbar segment of interest is determined by preoperative imaging studies. The skin entry point is typically approximately 1.5–2 cm parallel to the midline to the desired surgical level. The guide wire is placed through the incision and directed toward the inferior aspect of superior lamina, which can be confirmed by C-arm. An incision about 1.8 cm is made just along the guide wire. Some surgeons believe the incision should cling to the spinous process, that is along the midline, which is chosen when bilateral discectomy needs to be performed.

  3. 3.

    Insert an initial cannulated soft tissue dilator over the guide wire. Once the fascia is penetrated, remove the guide wire and advance the dilator down to the inferior edge of superior lamina. The second, third and fourth dilators are inserted over the initial dilator down to the lamina and the tubular retractor is then inserted. Sequential dilators are removed to establish an operative corridor to the lamina and interlaminar space (Fig. 22.7). Remove soft tissue over the lamina and interlaminar space. A hemilaminotomy is performed and the ligamentum flavum is resected. Identify and separate the dural sac and nerve root. The herniated disc is demonstrated.

  4. 4.

    Remove the posterior longitudinal ligament and annulus fibrosus with a sharp knife. Remove herniated disc as much as possible.

  5. 5.

    When necessary, decompression of lateral recess and nerve root canal should be performed.

  6. 6.

    A drainage tube is placed before closure.

Fig. 22.7
figure 7

Tubular retractor and endoscope

Postoperative management: Antibiotics are used for 48 h. The tube is removed at 12–24 h, and patients are then encouraged to ambulate.

22.3.3 Microendoscopic Decompression, Interbody Fusion, and Percutaneous Pedicle Screw Implantation of the Lumbar Spine

Due to the development of endoscopic instruments and operative skills, microendoscopy is used to treat not only disc herniation, but also the lumbar disc herniation complicated with degenerative instability. Microendoscopy is reported to treat lumbar stenosis mainly using expandable channels (such as X-Tube) and B-Twin expandable spinal spacer. Since the latter is used rarely in clinic due to its more complications, it is not discussed here.

Unlike microendoscopic discectomy, an expandable introducer is used instead of a tubular retractor during posterior lumbar interbody fusion. After the expandable introducer is docked on the lamina, its diameter can be expanded to 4 cm by clamps. When the introducer is inserted, it should be inclined backward 15°, in order to facilitate the pedicle screw implantation.

Operative procedures

  1. 1.

    The patient should be placed in the prone position on a Jackson table to keep lumbar lordosis.

  2. 2.

    Locate and mark the desired surgical level and bilateral pedicles under C-arm fluoroscopy. Then draw a line along the ipsilateral pedicles. An incision about 3 cm is made just along the line. Insert a guide wire into the incision, and confirm the guide wire is on the desired level and facet joints.

  3. 3.

    Serial dilatation then proceeds. Use an expandable retractor to provide greater access through the incision. Using microendoscopy, remove any excess soft tissue so as to expose the interlamina and facet joints.

  4. 4.

    Tilt the retractor laterally about 15°. Percutaneous pedicle screw instrumentation is performed under C-arm fluoroscopy guidance.

  5. 5.

    Transforaminal lumbar interbody fusion (TLIF) or posterior lumbar interbody fusion (PLIF) can be selected according to the surgeon’s preference. It is not necessary to expose exiting nerve roots, but the exposure of traversing one deserves extreme carefulness. Endplate preparation is performed with an appropriately sized reamer fitting snugly between the endplates. Reaming is performed under microendoscopy guidance until the subchondral bone is revealed. After completion of the endplate preparation, implant an appropriately sized cage mixed with bone chips into the disc space. Rods are placed via the minimally invasive tubular retractor.

  6. 6.

    The wound is irrigated and closed in a standard fashion.

Postoperative management: Antibiotics are used for 48 h. The tube is removed at 12–24 h, and patients are then encouraged to ambulate.

22.3.4 Percutaneous Endoscopic Lumbar Discectomy (PLED)

In 1997, a major improvement in minimally invasive surgery was achieved by introduction of the “Yeung Endoscopic Spine System (YESS)” by Yeung. Kambin’s anatomic description of the neural foramen as the endoscopic access is also known as “safe triangle”. The disc herniation can be excised directly through the access, not via the epidural space. However, because of facet joints, YESS could not be widely used for lumbar disc herniation. In 2003, Hoogland from Germany improved YESS and devised transforaminal endoscopic spine system (TESSYS) which enlarges neural foramen to make access amplified (Fig. 22.8). Transforaminal endoscopic discectomy has become an important and popular alternative in the management of lumbar disc herniation.

Operative procedures

  1. 1.

    The accesses of the procedure mainly include interlaminar dorsal access, far lateral or horizontal access and transforaminal access. Transforaminal access is typical and popular for PELC in the management of lumbar disc herniations. The skin point is typically approximately 8–11 cm from the midline and is projected to have 25° (from coronal plane) for L4–5, 35° for L3–4, 40° for L2–3, and 45° for L1–2. The ideal access is that the needle is positioned on the midpedicular line in the anteroposterior projection and on the posterior vertebral line in the lateral projection (Fig. 22.9).

  2. 2.

    After the skin point is confirmed, a transforaminal epidural infiltration through the spinal needle with 0.5% lidocaine is performed to prevent the approach-related pain and discomfort. After insertion of the needle, an intraoperative discography is performed with a mixture of 6 mL of contrast media and 1 mL of indigo carmine. The pathologic nucleus and annular fissure can then be stained for easy discrimination through endoscopy.

  3. 3.

    Insert a guide wire through the needle into the annulus, and then withdraw the needle and make a tapered cannulated obturator (with an outer diameter of 6.9 mm) sliding over the guide wire and introduced gently into the foramen. Remove the guide wire. A surgical sheath (with an outer diameter of 7.9 mm) with a beveled opening is placed over the dilator. The opening of the bevel-ended working sheath should be directed toward the undersurface of the superior facet and toward the caudal direction.

  4. 4.

    Then using a direct endoscopic view and continuous irrigation, decompression is performed. The working field is constantly irrigated with antibiotic-containing saline during the entire endoscopic procedure (Fig. 22.10). Make sure that the dural sac and the traversing nerve root are in free mobilization and released.

  5. 5.

    After full-scale lateral recess decompression is confirmed, inject a mixture of 7.5 mg of Limethason and 2 mL of indigo carmine into the foramen. And then remove the endoscope and the working sheath. A sterile dressing is applied with a one-point subcutaneous suture (Fig. 22.11). If there are no postoperative problems, the patient is permitted to mobilize three hours after operation, and be discharged within 24 h.

Fig. 22.8
figure 8

An endoscope

Fig. 22.9
figure 9

A spinal needle is placed

Fig. 22.10
figure 10

Percutaneous endoscopic lumbar discectomy

Fig. 22.11
figure 11

Postoperative wound and herniated disc

22.4 Functional Reconstruction of Bladder After Spinal Cord Injury via Neural Approaches

22.4.1 Introduction

Spinal cord injury (SCI), caused by trauma or disease, is one of the most serious clinical disabilities. It not only seriously damages somatic motor and sensory functions of patients, but also causes Defecation dysfunction. According to reports about 12–15 years follow-up of spinal cord injury, urinary tract infection and renal failure are the main causes of late death in SCI patients. Besides, 49–66% paraplegia patients in Tangshan earthquake died in uremia. Therefore, the reconstruction of the bladder urination function of patients with spinal cord injury is very important to improve the quality of life and reduce the mortality of paraplegia patients. In the absence of major breakthroughs in spinal cord regeneration, functional bladder reconstruction via neural approaches in paraplegic patients has been a hot topic in the past 20 years.

The reconstruction method differs due to the spinal cord injury location. Functional reconstruction of spastic bladder patients, whose bladder storage and voiding function are impaired by injuries above the spinal cone, requires improvement of both urinary storage and voiding function. Flaccid bladder, caused by spinal cord cone injury, has normal urinary function but voiding dysfunction. Therefore, improving micturition function is the main purpose.

22.4.2 Selective Sacral Rhizotomy

Spastic bladder is caused by spinal cord injury above the cone. Reduced bladder capacity due to the detrusor hyperreflexia, and increased urethral outlet resistance caused by spasm of urethral sphincter, can lead to increased bladder pressure, urine reflux, and further lead to renal function damage. Treatment is based on the specific situation of bladder spasm, to cut off the corresponding nerve roots selectively, by blocking part of the nerve pathway, reducing malignant afferent, and ultimately improve the function of the bladder.

22.4.2.1 Indications

It is suitable for spastic bladder with high tension and high reflexes caused by spinal cord injury above the spinal cone.

22.4.2.2 Surgical Techniques

Cutting from the incision of lumbosacral posterior median, the S1–S4 nerve roots are exposed (Fig. 22.12), and then separate the anterior and posterior roots of them. The S2–S4 nerve anterior roots are stimulated by electromyograph with the same intensity (20 mV, 30 Hz, 5–10 s), and the reaction of bladder detrusor urethral sphincter are observed. The most sensitive nerve roots of the musculocutaneous sphincter are insured and cut off.

Fig. 22.12
figure 12

Intraoperative photographs showing the surgical procedure to reconstruct the bladder reflex pathway in a 20-year-old man. (a) Standard laminectomies from L5 to S3 were performed with the patient lying in the prone position. The dura mater was opened up through a paramedian incision, exposing the dorsal and ventral roots of the S1, S2, S3, and S4 nerves. (b) The ventral roots of S1, S2, and S3 were identified and separated from their respective dorsal roots by microdissection. (c) The S1 and S2/3 ventral roots were anastomosed

22.4.3 Electrical Stimulated Micturition: Brindley Procedure

The bladder micturition function can be reconstructed by electrical stimulation of the anterior sacral nerve roots, including skin electrode in vitro and electrode implantation in vivo, as well as different implantation sites, such as the bladder wall, pelvic nerve, sacral nerve root and conus medullaris. The most widely used and effective method is Brindley Sacral Anterior Root Stimulator (SARS) combined with posterior sacral rhizotomy, can meet the clinical needs of bladder function reconstruction after spinal cord injury. The Brindley procedure includes anterior root stimulation and cut of the dorsal root to abolish neurogenic detrusor overactivity. Stimulation of the sacral anterior roots enables controlled micturition, defecation, and erection, while dorsal root rhizotomy (sacral de-afferatation) enables a good reservoir function. The Brindley procedure has been used in more than 3000 patients.

22.4.3.1 Indications

Electric stimulation of anterior sacral nerve root is only suitable for spastic bladder caused by spinal cord injury above the cone (above the T12 level). Since the flaccid bladder is caused by spinal cord injury at the conus medullaris, stimulating of sacral nerve cannot be used to micturate due to the injury of the spinal cord which innervates the bladder.

22.4.3.2 Surgical Operations

The patient is placed in prone position. The surgery includes two parts: dorsal root rhizotomy and electrode implantation.

A laminectomy from L3–4 to S1–2 is done for an intradural rhizotomy and intradural implantation of the electrode cuff. The dura and arachnoid are opened at the midline to expose the sacral nerve roots. The anterior and dorsal components of the roots, especially relevant anterior roots for micturition, can be identified intradurally by electrical stimulation of these components while monitoring the effects on detrusor activity, blood pressure, and somatomotor responses. A rhizotomy of the identified dorsal sacral roots is done. The anterior sacral roots are positioned into the electrode cuff. The connecting cables are subcutaneously tunnelled to a subcutaneous pocket (lateral thoracic) for the receiver.

Implantation of extradural electrodes requires a laminectomy from L5–S1 to S3–4. The dorsal rhizotomy is done at the level of the posterior ganglia of S2–5. Electrical stimulation tests are used to identify the anterior and dorsal components of the sacral roots. The extradural electrode is implanted and fixated to the nerve using a strip of silicone rubber sheet which is sewn to itself and surrounds the nerve. The connecting cables and the receiver are implanted the same way as in the intradural procedure.

22.4.4 Reconstruction of Bladder Innervation Below the Level of Spinal Cord Injury to Induce Urination by Achilles Tendon–to–Bladder Reflex Contractions

According to the different types of bladder dysfunction caused by spinal cord injury, different artificial reflex arcs can be used to reconstruct the urination.

22.4.4.1 Indications

It is suitable for spastic bladder caused by injury above conus medullaris. The lower limb tendon reflex exists in the patient. The artificial reflex arc of “tendon-spinal cord-bladder” is established by innervating the nerve root of tendon reflex to reconstruct bladder function.

22.4.4.2 Surgical Techniques

Preoperative catheterization should be given to patients. The catheter is connected with the infusion tube and the piezometric tube through a three way tube, so as to detect the bladder pressure during the operation. The patients are prone position for the L5–S2 posterior median incision to expose the S2–S4 nerve. Electrical stimulation is performed on the anterior roots of S2–S4 on both sides of the bladder. Pressure changes during bladder contraction are observed by using a manometer. The most robust nerve root that dominates the bladder is the one whose pressure rises quickly and finally the highest. The S1 anterior root and the strongest nerve root (usually S2 or S3) that innervated the bladder were cut off on the same plane, and the S1 anterior root and S2/S3 anterior root were anastomosed with a 9-0 line (Fig. 22.12).

22.4.5 Reconstruction of Bladder Innervation Above the Level of Spinal Cord Injury to Induce Urination by Abdomen-to-Bladder Reflex Contractions

In our previous experiments, we established a new skin–CNS–bladder reflex (abdominal reflex) pathway to restore controlled micturition in the atonic bladder. The new pathway was established in a rat model of SCI by intradural microanastomosis of the right T13 ventral root to the S2 ventral root with autogenous nerve grafting. After the new reflex pathway was reestablished, long-term function of the reflex arc was evaluated by electrophysiological, detrusor, electromyographic, and urodynamic studies over 8 months postoperatively. We found that the normal somatic reflex superior to the spinal injury level can be used to establish a reflex pathway by spinal ventral root anastomosis between the T13 and S2 nerve roots. The somatic motor root can reinnervate the bladder with a reconstructed efferent branch, and somatic motor impulses can be sent to the bladder through the reconstructed efferent branch and induce contraction of the bladder detrusor. Similar results were obtained in a canine model of SCI. Based on the results of these preclinical experiments, we attempted bladder reinnervation in patients with conus medullaris injury through functional suprasacral nerve transfer.

22.4.5.1 Indications

Flaccid bladder patients caused by complete conus medullaris injury can establish the artificial reflex arc of “abdominal wall-spinal-bladder” by using normal abdominal wall reflex above the injury plane and through the anastomosis between normal spinal nerve anterior root and the sacral nerve anterior root that innervates the bladder, to re-establish the nerve reinnervation of the bladder.

22.4.5.2 Surgical Techniques

After general anesthesia, the patient is placed in the prone position. The posterior median incision of T10–T12 is made, and the nerve roots of T10 (or T11) are separated. Meanwhile, L5–S2 posterior median incision is made to locate and separate the S1–4 nerve roots and its anterior and posterior roots. According to the distance between T10 (or T11) and S2, the sural nerve with a length of about 30 cm was cut intraoperatively, and the anterior root central terminal (proximal end) of T10 (or T11) and the sural nerve were anastomosed with a 9-0 noninvasive needle line. The distal end of the anterior root of S2 anastomoses with the other end of the sural nerve (Figs. 22.13 and 22.14).

Fig. 22.13
figure 13

Images of the surgical procedure to reconstruct the bladder reflex pathway in a 43-year-old woman. (a) Standard laminectomies from L5 to S3 and from T10 to T12 were performed with the patient lying in the prone position. (b) After the ventral root of T10 was identified and found to be functional, it was transected. (c) The dura was then closed, leaving the T10 ventral root outside. (d) A sural cutaneous nerve, about 30 cm in length, was taken for the nerve graft. (e) The T10 and S2 ventral roots were then anastomosed through a nerve graft

Fig. 22.14
figure 14

Urine pressure and flow diagram from the urodynamic test. (a) Preoperative study showed that the detrusor had no reflection but the external sphincter was denervated. (b) Postoperative study showed that the detrusor had regained nerve reflex when the external sphincter was denervated: intravesical pressure increased quickly while abdominal pressure did not increase significantly

22.4.6 Transfer of Normal Lumbosacral Nerve Roots to Reinnervate Atonic Bladder

22.4.6.1 Indications

Patients present with bladder atonia after conus medullaris injury and whose motor function of the lower extremities is preserved. Some patients with atonic bladder due to spinal conus injury had normal or partial lower limb motor function. The normal lumbosacral nerve root can be used as the power nerve to reconstruct the bladder urination.

22.4.6.2 Surgical Techniques

After general anesthesia, the patient is in the prone position for laminectomy from L5 to S3. The dura is opened through a paramedian incision, exposing the dorsal and ventral roots of the S1–4 nerves. The S1 nerve root is located using the L5/S1 intervertebral space as a marker; the S2–4 nerve roots are located in a descending order. The ventral and dorsal roots at the dural incision are identified according to their anatomical characteristics. The VRs of S1, S2 and S3 are identified and separated from their respective dorsal roots by microdissection. An electric stimulator is used to stimulate the S1 VR to observe muscle contractions of the lower limb in order to verify that the root is indeed the S1 root. If the function of the S1 root is normal, the unilateral S1 VR and the S2/3 VRs on the same side are transected using microsurgery and anastomosed with 9-0 non-absorbable sutures (Fig. 22.15). The wound is sutured in three layers with an external drain.

Fig. 22.15
figure 15

Surgical procedure to reconstruct the bladder reflex pathway in a 25-year-old man. (a) Standard laminectomies from L5 to S3 were performed with the patient lying in the prone position. The dura was opened through a paramedian incision, exposing the dorsal and ventral roots of the S1, S2 and S3 nerves. (b) The ventral roots of S1, S2 and S3 were identified and separated from their respective dorsal roots by microdissection. (c) The S1 and S2/3 ventral roots were anastomosed

22.5 Microsurgical Repair of Injury to Spinal Cord and Cauda Equine

22.5.1 Introduction

The spinal cord, located in the middle of the vertebral canal, is flat cylindrical in shape, with a length of 40–45 cm. It can be divided into cervical, thoracic, lumbar, sacral and coccygeal regions. Its upper end is large and connected with medulla oblongata; its lower end becomes sharper, forming the conus medullaris. A strip from the conus medullaris, called filum terminal, travels down the spinal canal to end at the back of the second sacral vertebra. The adult spinal cord ends in the equivalent plane of the first lumbar vertebral lower edge or the second lumbar vertebral upper edge. The Spinal cord varies in thicknesses, with two enlargements, namely cervical and lumbar enlargements. The cervical enlargement is located at C4–T1, with its thickest part at C6. The lumbar enlargement is found at T11–L1; its thickest part is at T12. The spinal cord results from 31 couples of spinal nerves, including 8, 12, 5 and 1 couples of cervical, thoracic, lumbar and sacral nerves, respectively.

The cauda equina count is roughly 40, every stripe is divided into two bundles, connected with each other by loose connective tissues, surrounded by a loose capsule. Cauda equina has ventral and dorsal roots. Ventral roots start from the ventral anterolateral ridge of the lumbosacral spinal cord, as the motor; dorsal roots start from the dorsal posterolateral ridge of the lumbosacral spinal cord, as the sense. Ventral and dorsal roots are parallel to each other and drop vertically in the subarachnoid space, through arachnoid into a bundle, slightly in an oblique line. Cauda equina does not connect with each other in the subarachnoid space. The capsule of cauda equina is called spinal pia mater. The anatomical characteristics of cauda equina facilitate release of fiber tissues (Fig. 22.16).

Fig. 22.16
figure 16

Anatomy showing cauda equina

Vertebral fractures and dislocations below the second lumbar vertebra are often combined with cauda equina damage. Some cases present with intolerable burning pain of lower limbs caused by adhesion in the subarachnoid space or around the cauda equina. Conventional surgical methods have several disadvantages, such as limited vision, bulky surgical instruments, difficulty in accurately and completely releasing the adhesion, and surgical injuries. Therefore, surgery outcomes are not satisfactory, and the treatments are often abandoned. With recent developments in microsurgical techniques, decompressing the cauda equina under a microscope can release the adhesions accurately, precisely and completely. This also reduces cauda equina damage, markedly improving the operation outcomes.

22.5.2 Microscopic Surgical Repair of Spinal Cord Injury

22.5.2.1 Early Microscopic Surgical Treatment of Spinal Cord Injury

Spinal cord injury is typically accompanied by spinal trauma, such as spinal dislocation or fracture. Early treatment aims at relieving spinal compression and/or maintaining spinal stability by means of reduction, internal fixation and decompression.

The early treatment of spinal cord injury with microscopic surgical repair includes myelotomy, durotomy, piotomy, etc.

22.5.2.1.1 Myelotomy

The indications are as follows: (1) neurologically identified complete paraplegia; (2) non-transection injury evaluated by radiography and clinical signs; MRI showing hemorrhage and edema of the spinal cord; (3) intact dura mater under surgical exploration, and subarachnoid space disappearance after the dura mater is incised, due to spinal edema; existence of the vasculature on spinal cord surface, with hardened parenchyma and elevated tension; also, myelotomy can be applied when cysts form inside the spinal cord, days to weeks after injury.

As for timing, it has been proposed that myelotomy should be performed as early as possible, since spinal function can only be restored within 10 h after the injury, that is, when intraspinal cysts are yet to form.

Spinal cord injury is commonly observed with the injury to the lumbosacral enlargement, due to thoracic and lumbar spine dislocations or fractures. The detailed surgical procedures are as follows: (1) the dura mater is exposed with a longitudinal posteromedian incision under local anesthesia; (2) the incision should surpass the swollen part of the spinal cord in length, ideally accompanied by outflow of the cerebrospinal fluid; (3) the pia mater is incised, avoiding damage of the longitudinal blood vessels; (4) the surgical site is gently irrigated with Ringer’s solution or saline; (5) suture of the incision is not likely to be achieved due to spinal edema; thus, a paravertebral fascia or artificial dura mater is covered and fixated with sutures at the incision.

22.5.2.1.2 Durotomy and Piotomy

The spinal cord is rapidly distended after injury, followed by disappearance of the subarachnoid space due to restraint from the dura mater, and a subsequent increase of intraspinal pressure. A longitudinal incision of the dura mater can release the tension and help improve blood flow. Under severe conditions, the pia mater should also be incised to achieve decompression. This presents the advantage of no extra damage to the spinal cord; for those without incision of the pia mater, the spinal tissue is not exposed to the surrounding tissues. The range of incision of the dura mater should be slightly beyond the swollen spinal edema, and the best case scenario is with cerebrospinal fluid outflow on both ends. Inadequate incision may increase the risk of herniation, thus worsening the spinal cord damage. The most severe edema is observed 3 days–8 weeks after the injury, and only completely disappears in 4–8 weeks, depending on edema severity. Young and his collaborators performed durotomy or myelotomy in 30 ASIA A or B patients 3–80 days after the injury, and observed improved physical function in more than half of their cases. Therefore, surgery timing, range and depth should be precisely designed after careful inclusion of cases.

22.5.2.2 Microscopic Surgical Treatment of Incomplete and Old Spinal Cord Injury

22.5.2.2.1 Dural Decompression

Objective: Dural decompression is beneficial for the patients with incomplete paraplegia who have obtained partial recovery of the neural function 3 months after traumatic injury but no further improvement following physical therapy 3 months thereafter.

Indications: It is widely acknowledged that decompression and deformity correction may further improve the neurologic function in patients with spinal fracture combined with spinal cord injury. Treatment of traumatic incomplete paraplegia is typically determined according to CT or MRI scans. We have observed, however, that MRI and CT scans fail to provide radiological support for patients with halted spinal function improvement, probably due to scars in the dural sac that compress the spinal cord. Thus, for the aforementioned patients in whom MRI or CT scans fail to reveal bony compression, spinal stenosis or spinal cord instability, preoperative MRI may reveal approximation of the spinal cord with the dura mater on multiple planes. Surgical exploration of the indicated units is recommended; more importantly, although the compressions are so thin and small that MRI and CT may fail to uncover, they are more direct and harmful, with no buffering of the adipose tissue or cerebrospinal fluid. Therefore, attention should be paid to the reconstruction of the “soft lumen” (i.e. restoration of the dural sac with dura decompression).

Surgical methods: General anesthesia is administered at lateral or prone position; a posteromedian incision is adopted to expose the dura mater. After removal of the internal fixation, the injured laminas with proximal and distal laminae are revealed. While closely looking for pulsatile movement of the dura, small incisions of 1 cm are made intermittently on hardened and thickened areas with scalpel, at a length that penetrates the entire dura while preserving the arachnoid membrane (Fig. 22.17), if no pulsatile movement is observed in 3 minutes, until a small amount of cerebrospinal fluid is leaking with restored pulse. The wound is irrigated after surgery with drainage input for 48 h, and covered with anti-adhesive membrane or a muscular flap. All procedures are carried out under a microscope. A marked restoration of dura’s pulsatile movement should be observed. Compression and adhesion of the dura mater is a major factor that affects neurological restoration. Hence, this microscopic surgical technique effectively promotes the above procedure.

Fig. 22.17
figure 17

The entire dura mater is exposed to the median approach

22.5.2.2.2 Intradural Lysis and Autologous Peripheral Nerve Implantation

Objective: Decompression and lysis of the spinal cord; treatment for incomplete rupture in old spinal cord injury.

Indications: (1) incomplete paraplegia, with early restoration of the neurological function after injury but no further improvement after 3 additional months of conservative treatment; (2) no marked bony compression, spinal stenosis or instability observed with MRI and CT scans, but adhesion of the spinal cord and the dura in at least one plane, with cerebrospinal fluid basically blocked. Surgical exploration should be performed accordingly for complete lysis of the scar tissue, removal of compression for improvement of blood supply and spinal function; (3) preoperative MRI shows approximation of the dura and spine, and no restoration of pulsatile movement after dural lysis.

Methods: A lateral position is adopted, with general anesthesia. A posteromedian incision is made to expose the dural sac. The arachnoid membrane, pia mater, nerve root, anterior and posterior branches are carefully explored, to look for bands, cords, or scars, as well as adhesion between structures, spinal cord deformity, thickened and adhered pia mater, and compression to the spinal cord. Then, a complete lysis is conducted in the arachnoid membrane, pia mater, denticulate ligament, nerve root, and surrounding adhesions, using microscopic surgical techniques. An autologous sural nerve should be harvested and trimmed to be longitudinally implanted in the corresponding areas or the lumen of cysts.

22.5.2.2.3 Microscopic Lysis of the Lacerated Ends of the Spinal Cord and the Nerve Root

Objective: To salvage the nerve root function at C5–C7 and T12–L2 in complete spinal injury patients, and improve quality of life in patients with paraplegia and quadriplegia.

Indications: Complete paraplegia in C5, C6, C7, and T12–L2, clinically classified as ASIA A or Francle A. Surgery is only performed in patients younger than 50 years old, considering better restoration of the nerve function and perseverance in rehabilitation in younger patients.

Methods: A lateral position is adopted under general anesthesia and a posterior incision made. Partial or complete removal of the internal fixation is made accordingly. The scar tissue is dissected and removed; generally, the proximal end is extended instead of the distal end. The proximal end is incised to form a cystic or sac-like dura in patients with completely ruptured spinal cord and dura, while a longitudinal incision is made in patients with an intact dura. The anatomically intact nerve roots are radically loosened from the beginning to the intervertebral foramen, with no suture of the thickened and scarred dura. An artificial dura or muscular flap is covered. Administration of low-dose corticosteroid should be applied for 3 days in addition to regular postoperative treatment (Fig. 22.18).

Fig. 22.18
figure 18

Exposure of adherent nerve roots and venous plexus

22.5.3 Microsurgical Repair of Cauda Equina Injury

22.5.3.1 Cauda Equina Nerve Neurolysis

Damaged or oppressed cauda equina nerve is prone to adhere, which often occurs between the cauda equina or between the cauda equina and arachnoid. The adhesion could eradicate the subarachnoid, and block the cerebrospinal fluid flow. The above changes affect cauda equina repair. Held in prone position, local or general anesthesia is performed, and segmental vertebral plates are removed. The catheter probe is outside, in front of the oppression, even below the L3 vertebral fracture in front of the ward. As long as the lamina resection is done, oppression may also be lifted, especially the segmental nerve root canal. Then, one could open the catheter removal decompression, with the bone in front of the brain membrane covering the wound bleeding. Previous cut should be from head to tail ends to the middle of adhesion, especially from the far side. The cauda equina regiment and arachnoid boundaries could be distinguished after the surgery; cauda equina adhesion is separated from arachnoid blunt from bottom to top. Till the end, the cerebrospinal fluid flows out based on gradual separation between nerve roots into the intervertebral foramen. The cauda equina is separated from the second half of the subarachnoid; the cauda equina between adhesions is not easy to separate, and forced separation could damage the cauda equina. In the absence of dural defects, which can be stitched, saline solution is administered through catheter in order to maintain postoperative cerebrospinal fluid flowing. The scaring cauda equina and that with great tension cannot be sutured directly because of the risk of causing local stenosis. Compressed cauda equina can be repaired by artificial muscle flap or autologous fat tissue flap, and the wounds can be sutured conventionally. A low negative pressure drainage tube can be placed in the wound, but the scale and quantity of cerebrospinal fluid should be closely observed. Generally the drainage tube should be removed no more than 48 h. After the wound is oppressed for 24 h in the recumbent position, the head and shoulder should be gradually raised after around 3 weeks to the sitting and standing positions. When releasing the cauda equina, we adopt the hydraulic expansion method, from near and far end clearance of the adhesion horsetail, using preoperative bending TB syringe needle with appropriate thrust and speed. This is done in the center of the film adhesion cluster to the damage under the injection of 0.25% lidocaine hydrochloride encroach on progressive sharpness and blunt separation. Precisely, the adhesion degree of mild cauda equina shed gentle separation with hydraulic adhesion heavier using microscopic scissors and cut knife.

22.5.3.2 The Cauda Equina Suture/Bonding Technique

Since the 1960s, multiple studies assessing cauda equina injury have suggested the possibility of cauda equina nerve regeneration. Indeed, many experiments and studies have confirmed that cauda equina repair plays a major role in restoring the motor function. Therefore, it is particularly important for the repair of broken cauda equina with no obvious defects to be an early artificial suture. First, the injured cauda equina should be exposed and the bleeding cleared. Second, according to the thickness of different cauda equina fibers, nerve stumps should be aligned and the injured cauda equine carefully sutured. Lastly, the dura should be sutured. If the dura cannot be sutured easily, artificial dura, muscular flap or autologous adipose tissue flap can be employed to repair the dura, in order to avoid partial stenosis and cauda equina compression.

22.5.3.3 Nerve Transplantation for the Cauda Equina Defects

When the cauda equina is injured seriously, it is impossible to conduct normal direct suture without tension. Another nerve must be transplanted to bridge the injured cauda equina. We use a 4×-magnification microscope to explore cauda equina during operation. The motor nerve, which is thicker, is located in the front of the canalis vertebralis. If the nerve stumps shows obvious contusion, they can be decompressed and removed by microsurgery instruments. After the length of the defect is measured, a segment of sural nerve of the same length, whose perilemma has been removed, is taken under the microscope to bridge the cauda equina. We also use 2–3 sensory nerves to bridge one motor nerve by the same method. Anatomical repair should not be strived for too much, because expanding too many surgical fields increases the surgical trauma. The function of both quadriceps and hip muscle can be recovered in patients with fresh lumbar spine fracture and dislocation after cauda equina repair, indicating that recovery of the muscle function is related to nerve regeneration. But all the muscles below the knee do not recover, demonstrating the distance between muscles and injured area has an adverse effect on muscle recovery. No recovery in patients with obsolete injury shows that the curative effect is time based. It is pointless to repair the sensory nerve.

22.5.4 Microsurgical Management and Functional Restoration of Patients with Obsolete Injury to Spinal Cord and Cauda Equine

Bladder and bowel dysfunctions and spastic paralysis caused by complete high position spinal cord injury are difficult to restore effectively only by non-surgical treatment, such as the drug or physical therapy. It has been confirmed that some of nerve function of the brachial plexus with root injured can improve by transferring the normal peripheral nerve to it. This mature technique based on decades of the clinical practice of surgeons in many countries aroused the authors to improve part neurological functions of patients with complete high position spinal cord injury by using above theory, which connected normal peripheral nerves with nonfunctional nerves. As the patients suffer from limb spastic paralysis and their degeneration of peripheral nerves and their dominating effectors happens later and more mildly after spinal cord injury than those with brachial plexus root injury, the excellent rate of surgery is still higher than that of similar surgery for brachial plexus root injury. Moreover, after the donor nerve grows into the muscles innervated by the recipient, the nerve impulses causing the muscle to contract can also stimulate the synchronous contraction of the high tension synergistic muscle that can assist the completion of the function after rehabilitation training. However, the amount of nerve function that needs to be reconstructed in patients with paraplegia and quadriplegia is much greater than that in patients with brachial plexus injury, and the number of donor nerve fibers is relatively small, so the nerve function that can be regained is limited. How to accurately match the donor nerve fibers with the target nerve is critical. Another question to concern is how to prevent the recipient innervation muscles from atrophying before the new nerve fibers grow into the muscles during postoperative period. In view of the above two problems, we have cut the lateral adventitia and part of the perineurium of the receptor nerve, and selectively cut off some of the nerve fibers while preserving the proper muscle tension. Finally, the donor nerve is inserted in the incision of the receptor nerve, and sewed the outer sheath of the two nerves together. We call the procedure “nerve grafting” (Fig. 22.19) and the clinical results are quite satisfactory.

Fig. 22.19
figure 19

Schematic diagram of end to side neurorrhaphy

Next, we introduce the operative technique of peripheral nerve side-to-side interfascicular anastomosis. Take C2–C4 injury as an example: the nerve branch of accessory nerve is connected with the phrenic nerve.

C2–C4 injury: The nerve branch of accessory nerve is connected with the phrenic nerve.

Indications: Patients with C2–C4 injury who show no spontaneous breathing and require ventilator support and whose trapezius muscle is paralyzed at one side at the least.

Surgical objective: To restore partial function of diaphragmatic breathing, allowing the patient to breathe through the shrug movement without ventilator support in the awaken state.

Anatomy: Accessory nerve is composed of cranial nerve root and spinal cord root (mainly C1–C4), so the function of the accessory nerve is basically intact below C5 of spinal cord injury plane. The accessory nerve mainly dominated trapezius and sternocleidomastoid muscles and most of the nerve branches are dispersed in the dominant muscles. Therefore, cutting the accessory nerve in the plane above the clavicle will only affect part of the trapezius muscle strength without affecting other important functions.

Surgical procedure: Cutting the accessory nerve at the proximal end, and then “grafted” the nerve to the phrenic nerve in the manner of the side-to-side stitching (Fig. 22.20). The detailed steps are as follows: The normal donor nerves in the paralytic area are fully exposed and properly released to close to the receptor nerves at the same site. The two nerve outer membrane of the adjacent area was cut from 1 to 2 cm. Then the fasciculus of some nerve bundles was cut and the nerve fibers were exposed. Cut the two nerves’ epineurium on the adjacent segments, the length is 1 to 2 cm, and then cut the perineuriums of some nerve bundles separately to expose the nerve fibers. The donor nerve is inserted laterally into the incision at the side of the recipient nerve and the epineurium and perineurium are sutured to each other (Fig. 22.21). For example, if the tibial nerve is damaged and the common sacral nerve is normal, their adjacent segments are approximately 5 cm proximal to the branch, and the nerves can be displaced and stitched together by the above method.

Fig. 22.20
figure 20

Surgical procedure showing the accessory nerve is connected to the phrenic nerve in the neck

Fig. 22.21
figure 21

Schematic diagram of side-to-side neurorrhaphy: (a) incise the epineurium of the two neighboring nerves; (b) Reveal two side of the perineurium; (c) suture one side of the epineurium; (d) suture another side of the epineurium

According to our clinical observations, most patients with chronic spinal cord injury have achieved partial functional recovery after autologous peripheral nerve grafting and implantation. For patients with general paralysis, even bladder and bowel function is restored to some extent in addition to partial sensory and motor restoration, bringing about much convenience, reducing complications and greatly improving their quality of life. Since only a small number of nerves can be used for grafting, this series of microsurgery methods can only recover limited but important functions. More training is necessary for effective motor function after muscle contraction caused by pathological reflex and grafted nerve. Our follow-ups of 226 cases for 3–28 years revealed that effect active movement function (M3) was restored in 37%, sensation (S2–S3) in 76%, and reflection in 81%. Therefore, satisfactory results can be hardly achieved in patients who cannot persist in standard rehabilitation training. Such surgery is not recommended either for senior patients in poor general condition, or patients with financial difficulties.