Mechanical

Spinal nerve roots, although well protected by the vertebral column from outside compression, are particularly susceptible to compression from objects that arise from within the vertebral column. They lack the perineurium and epineurium that are seen in peripheral nerves. Their vascular supply is also more easily compromised. Two studies were done before Olmarker’s series of studies that showed that spinal nerve roots were more susceptible to compression than peripheral nerve roots. Olmarker and colleagues showed over several publications that spinal nerve roots, when compressed, exhibited intraneural edema, deprived nutritional supply, and loss of amplitude of nerve conduction. The nerve conduction study also demonstrated that more rapid compression and higher pressures have a deleterious effect on recovery of nerve function. These findings were confirmed in awake surgical patients. In the study by Kuslich and colleagues, noncompressed nerve roots did not reproduce the patient’s pain when stimulated intraoperatively. However, stimulation of compressed nerve roots consistently reproduced the patient’s preoperative symptoms.

Chemical

Olmarker followed his mechanical studies on porcine cauda equina with further studies on the effects of autologous nucleus pulposus on spinal nerve roots. His group first showed that autologous nucleus pulposus applied to nerve roots without mechanical compression caused statistically significant changes in nerve conduction velocity when compared with nerve roots that had epidural fat applied to them. Electron microscopic examination of the nerves revealed that the changes were not limited to nerve conduction but also caused axonal injury and Schwann-cell damage. Follow-up studies demonstrated inflammatory changes induced by the nucleus pulposus, including leukotaxis and increased vascular permeability. This inflammation was then attributed in part to tumor necrosis factor α. Inhibition of this substance reversed the effects in pig models, but human studies have failed to show effectiveness of treatment with antibody to tumor necrosis factor α.

Open Diskectomy Technique

Although Mixter and Barr described the first open approach in 1934, this approach required a complete laminectomy and was transdural. Love and Semmes modified this approach by performing a hemilaminectomy with an extradural approach.

The patient is placed prone on a radiolucent table and the lumbar spine is flexed to open the interlaminar space. Using a spinal needle and intraoperative fluoroscopy, the correct level is identified. Incision is made in the midline, centered over the spinous processes, and is approximately 4 to 6 cm long. The incision can be performed as far proximally and distally as is required to remove disk material and obtain good visualization. Dissection is continued through the fat and fascia down to the spinous processes. Using a Cobb elevator and electrocautery, the multifidus (and longissimus as necessary) is dissected subperiosteally on the affected side, exposing the lamina and facet joint as dissection continues laterally. Retractors are placed to hold the paraspinal muscles out of the operative field. The ligamentum flavum is identified in the interlaminar space and sharply incised. Using Kerrison rongeur, the ligamentum flavum and hemilamina are removed at the level of the pathology. Alternatively, the bone removal can be done with a high-speed burr. In the depth of the incision, the dural sac is identified and carefully dissected off the floor of the spinal canal. The dural sac and nerve root are retracted medially with a right-angled nerve retractor to expose the intervertebral disk. The posterior longitudinal ligament is incised with a long-handled blade, and herniated disk material is removed.

Conventional Microdiskectomy

Yasargil and Caspar published articles outlining the microdiskectomy approach in 1977. Shortly thereafter, Williams published his series of patients, many of whom were Las Vegas showgirls in search of a more cosmetic procedure for lumbar disk disease. His satisfactory results in 91% of patients helped to popularize the procedure.

This procedure was originally described for a patient to be placed in the lateral decubitus position with the affected side up but is also performed with the patient in the prone position as previously described. In this approach, the skin incision is made just lateral to the spinous process instead of directly over the spinous processes, as in the open procedure. Incision is typically only 2 cm long for a single-level diskectomy. The fascia is incised just a few millimeters lateral to the spinous process, and dissection is then carried subperiosteally to identify the inferior third of the superior lamina at the 12 o’clock position, the medial aspect of the facet joint laterally, and the superior third of the inferior lamina at the 6 o’clock position, with the interlaminar space and ligamentum flavum in the center of the field. The operative microscope is typically used to develop the surgical interval through this small incision, and various retractors have been developed along with coaxial lighting to assist with visualization. Depending on the positioning of the patient and the level of surgery, a portion of the lamina and facet are removed with high-speed burr or Kerrison rongeur. Numerous studies have shown that up to 50% of the facet joint can be removed without destabilizing the spine. The ligamentum flavum is carefully released from the superior aspect of the inferior lamina and removed with Kerrison rongeurs, as needed, to expose the compressed nerve root. As is the case with standard open diskectomy, a right-angled nerve retractor is used to protect the dural sac and nerve root, and the posterior longitudinal ligament is incised sharply with a blade. Disk material is then removed in the standard manner.

Microendoscopic Diskectomy/Tubular Diskectomy

Foley and colleagues described the endoscopic technique for microdiskectomy in 1997. They used a muscle-splitting approach as opposed to the previously described subperiosteal dissection to further reduce the trauma to the paraspinal musculature and improve results. Arthroscopic techniques are increasingly desired by patients for other orthopedic procedures, and it was only a matter of time before this was applied to the spine. Specialized instruments were required, and as these were fine-tuned, results and technical barriers improved. Those instruments and retractors have also been used for standard microdiskectomy.

For this procedure, although positioning is the same, the incision is typically made 1.5 cm lateral to the spinous process on the side of the disk herniation. The incision length is also around 1.5 cm long, depending on the size of the tubular retractors. A k-wire is inserted into the medial facet, and a series of increasingly larger diameter dilators are placed over the k-wire. The final endoscopic tube is placed around the largest dilator and secured to the bed with a flexible arm. Fluoroscopy confirms the final positioning of the tube and the inner dilators are removed. An endoscope is placed into the working portal, with other retractors and tools capable of being placed through the same portal to perform the dissection and diskectomy. The dissection and diskectomy are performed in the same fashion as described in the microdiskectomy section. Many of today’s surgeons use the dilator tube as described but look down the tube with an operative microscope as opposed to an endoscope. This is sometimes referred to as tubular microdiskectomy or tubular diskectomy.

Comparison of Techniques

When multiple treatments for the same disease are developed, questions revolve around which treatment is best. Many studies have been performed to date comparing the previously mentioned surgical approaches, but the results have been mixed.

Several studies have investigated the difference between conventional open diskectomy with minimally invasive techniques. Yasargil and Foley and colleagues in their initial articles each reported on their techniques and on their results. They established the safety of their procedures, and subsequent articles validated the safety of the procedure and showed significant improvement in outcomes over the traditional open diskectomy. These studies demonstrated a wide variety of improvements, including smaller incision, less blood loss, and quicker return to work, with similar clinical outcomes for improvement in symptoms.

Nakagawa, and colleagues, published their results on 30 consecutive patients treated with Love’s open method and 30 consecutive patients treated with Foley and Smith’s microendoscopic method (MED). The MED group showed statistically significant improvement over the Love group for recovery rate through the first 16 weeks, creatine phosphokinase levels, and return-to-work rate (49.2 days vs 85.9 days). Recovery rate was not significantly different at 24 weeks or final follow-up. Blood loss trended toward lower and length of surgery, longer (109.1 minutes vs 79.3 minutes) with MED, but neither were significant. The increased length of surgery is potentially related to experience, because the 30 patients were very early in the investigators’ adoption of the MED technique. A review of the individual cases shows a significant improvement in operative time from the first case (more than 180 minutes) to the thirtieth case (<60 minutes). Brayda-Bruno and colleagues also noted the improvement in operative time with experience in their series of 68 patients, from 150 minutes to 35 minutes (average of 1 hour).

Muramatsu and colleagues evaluated the microendoscopic diskectomy in a different way. His group looked at postoperative magnetic resonance imaging images of 25 patients who underwent MED and 15 who underwent standard open diskectomy (Love technique). They also reported their outcomes on 70 patients with MED procedures and 15 with open diskectomy procedures. Their results showed statistical improvement in the MED group for blood loss, analgesic requirement, number of days before ambulation, earlier return to low demand jobs, and duration of hospitalization. Radiographically, the MED group had muscular edema limited to the small incision (typically <2 cm), compared with the edema along the larger incision with the Love technique (4–5 cm). Some of the patients in the Love group demonstrated extensive edema throughout the paraspinous musculature to large areas outside of the incision area; however, no significant difference was noted in enhancement of nerve or muscle tissue between the 2 groups.

After proving the efficacy and improved morbidity of minimally invasive procedures, the 2 types of minimally invasive techniques needed to be compared and evaluated against each other. Righesso and colleagues performed a prospective, randomized controlled trial on 40 patients to evaluate microdiskectomy (Caspar’s technique) and MED (with an endoscope). Statistically significant improvement of the MED group over the microdiskectomy group was noted for length of hospital stay (24 hours vs 26 hours) and size of incision (2.1 cm vs 2.6 cm). The microdiskectomy group had statistically significant improvement over the MED group in operative time (63.7 minutes vs 82.6 minutes) and immediate (12 hours) postoperative pain. No statistical significance was demonstrated for blood loss, neurologic status (trend toward earlier recovery in microdiskectomy), visual analog scale (VAS) pain score after the immediate postoperative period, or Oswestry Disability Index score.

In 2008, Brock and colleagues published their results on 125 patients who were prospectively randomized to a transmuscular or subperiosteal approach. This study was unable to show statistical significance for leg or back pain on the VAS, but the authors were able to show a difference favoring the transmuscular group in the postoperative Oswestry Disability Index score, as well as requirement for postoperative analgesics.

Arts and colleagues performed the first double-blind randomized controlled trial in 2009 on 328 patients, comparing conventional microdiskectomy (Caspar’s technique) with tubular diskectomy. The study was interesting because of its significantly larger size compared with previous studies and the double-blind randomized manner in which it was undertaken. Statistical significance was noted in favor of conventional microdiskectomy for the Roland-Morris Disability Questionnaire score at 1-year follow-up (but not earlier), operative time (36 minutes vs 47 minutes), leg pain, back pain, and general health on the VAS. The median recovery rate was 2 weeks in both groups, and rates of reherniation and blood loss were also similar. There was no significant increase in complications seen in the tubular diskectomy patients.

After a thorough review of the literature, a few recommendations can be made. Microscopic techniques seem to be superior to the traditional open technique as first described by Love. Patients tend to return to work faster, have less blood loss, and require less postoperative analgesia. Microdiskectomy, as described by Yasargil and Caspar, seems to provide similar outcomes to the more minimally invasive, transmuscular approach described by Foley and Smith. There is a significant learning curve associated with endoscopic and tubular techniques, which may provide a barrier to its incorporation into many surgeons’ practice. All studies listed earlier were limited to single-level disk protrusion with sciatica, and care should be taken when applying these results to multilevel or bilateral disease.