Minimally invasive transforaminal lumbar interbody fusion (MITLIF) allows a surgeon to restore intervertebral disk height, lumbar lordosis, and achieve an indirect decompression of the spinal canal and neural foramen while preserving vital posterior soft tissues. Adequate exposure of anatomic landmarks in open transforaminal lumbar interbody fusion (OTLIF) mandates a longer muscle-splitting incision and use of large self-retaining retractors at high tension that can generate pressures in the erector spinae musculature from 61 to 158 mm Hg. The resultant prolonged retraction induces ischemic changes resulting in reduced muscle fiber diameter, fibrosis, and fatty infiltration. These degenerative changes, in combination with disruption of stabilizing ligamentous structures, have been implicated as a major source of chronic back pain, failed back syndrome, and reductions in patient-reported functional outcome scores following open lumbar spine surgery.
Foley et al. introduced MITLIF in 2003 as a means of achieving the same objectives as OTLIF while minimizing iatrogenic soft tissue injury. Use of a tubular retractor placed via sequential dilation maintains continuity of muscle fibers and allows equal distribution of pressure around the wound edges. Critics of the MITLIF will cite technical limitations including loss of surface area for fusion, as the posterolateral recess is typically not exposed, and a heavy reliance on fluoroscopic imaging that increases radiation exposure to the surgeon, patient, and operating room staff. Lack of exposure of anatomic landmarks and limited working space also contribute to a significant learning curve for a surgeon looking to adopt MITLIF. To limit exposing the patient and surgical staff to risk, it is imperative that the surgeon follows a reproducible process of patient evaluation and perioperative care. The goal of this chapter is to describe in detail the MITLIF focusing on preoperative evaluation, intraoperative technical pearls, patient outcomes, and complication avoidance.
The MITLIF carries the same indications as OTLIF; both allow restoration of disk height with a structural interbody graft resting in the anterior column, which is responsible for bearing 80% of the load transmitted through the spine. This recreates the normal sagittal alignment between the two vertebrae and opens the facet joints to their native apposition, thereby achieving an indirect decompression of the spinal canal and contralateral nerve root. An MITLIF is an effective treatment for symptomatic spondylolisthesis, lumbar stenosis with instability, recurrent disk herniations, as well as instability secondary to trauma, pseudarthrosis, or iatrogenic sources.
Contraindications and limitations with respect to MITLIF can be divided into those that are absolute and those that are relative. Absolute contraindications to both OTLIF and MITLIF include conjoined nerve roots, acute trauma, or active infection. Aberrant location and connections seen with conjoined roots make safe performance of a TLIF nearly impossible, as limited mobility of the nerve root restricts access to the intervertebral disk ( Fig. 9.1 ). With acute trauma to the vertebral endplate there is no stable foundation to distract upon or to support the interbody cage, greatly increasing the risk of nonunion or cage migration. Active systemic infection is a contraindication to TLIF as well as other elective orthopedic surgeries where metal hardware is implanted into the body. However, it should be noted that implantation of titanium interbody cages with posterior spinal fixation has been shown to be safe and effective in the treatment of discitis or vertebral osteomyelitis as it provides the necessary stability for healing to occur. Treatment includes focused antibiotic therapy, typically for 12 weeks, and radical debridement of infected tissue mandating an open approach. Relative contraindications would include severe epidural scarring, severe osteoporosis, or grade III and IV spondylolisthesis. MITLIF may be preferable to OTLIF in the setting of morbid obesity or soft tissue compromise owing to burns, trauma, or cutaneous lesions.
Perceived complexity of the MITLIF increases when there is severe collapse of the disk space or significant osteophyte formation, especially over the posterior edge of the disk space ( Fig. 9.2 ). When encountering these cases, the preoperative images should be scrutinized for the presence of a mobile spondylolisthesis or vacuum phenomenon within the disk. Both of these signs indicate laxity of the soft tissues around the disk space, which should allow for distraction and restoration of height. When osteophyte overgrowth covers the disk, the experienced surgeon can use an osteotome or bur to debride the lip and gain entry to the disk space.
In general, no specific situations exist in which OTLIF is overtly better suited than MITLIF; however, this depends on the surgeon’s experience with each technique. Notably, the progressive narrowing of the interpedicular distance at each cranial level in the lumbar spine makes performing an MITLIF much more difficult above the L3-4 disk. Performing a facetectomy in this region reveals the dural sac and a small Kambin’s triangle, which increases the risk of dural tear or postoperative radiculopathy from retraction of neural structures.
Laterality of radicular symptoms should be noted and, in the setting of bilateral symptoms, it should be determined which side is more symptomatic via focused questioning and provocative maneuvers. Typically, the facetectomy is performed on the more symptomatic side, whereas the contralateral side may be addressed via indirect decompression or direct decompression by undercutting the spinous process and contralateral facet. Laterality of a previous decompression or diskectomy should be taken into account especially when first gaining experience with the procedure.
When using a minimally invasive approach, the posterolateral recesses are not exposed and, as a result, the surface area available for fusion is reduced compared with the open procedure. Therefore, the surgeon may elect to use bone morphogenetic protein (BMP-2) anterior to or within the interbody cage to help achieve a solid fusion. It should be noted that use of BMP-2 in this fashion is considered off-label and therefore consideration of patient-related risk factors is necessary along with informed patient consent. Any history of previous or active cancer should be explored; in women of childbearing age, a pregnancy test should be conducted if indicated. These patients should also be advised to avoid becoming pregnant for one year following surgery. The BMP-2 is typically delivered in an absorbable collagen sponge and may be deployed anterior to the cage or as a strip of sponge placed inside the center of the cage. When utilizing BMP-2, the author prefers to place a strip inside the cage, in order to limit contact with the neural structures, which may cause a postoperative radiculitis. Some surgeons also utilize fibrin glue or other barriers to seal the annulotomy site to prevent egress of BMP-2 into the neural foramen where it may lead to heterotopic ossification.
Baseline preoperative anteroposterior (AP) and lateral radiographs allow for the evaluation of disk height, sagittal alignment, and presence of osteophytes, whereas flexion-extension films will reveal the presence of instability. In the absence of neurologic symptoms, preoperative magnetic resonance imaging (MRI) may not be necessary. If the patient has contraindications to MRI or previous hardware that would obscure visualization, then computed tomography (CT) myelography should be considered. These examinations allow the surgeon to identify central, lateral recess, or foraminal stenosis to be correlated with patient symptomatology and physical examination findings. Facet hypertrophy can be assessed on AP radiographs; however, a CT scan can be helpful in surgical planning as severe osteophyte formation from either facet may obscure visualization of the joint space, which is an important intraoperative landmark. Compressive osteophytes from the contralateral joint may demand the use of a foraminotomy rongeur to aid in undercutting the superior facet when performing contralateral decompression. Computed tomography may also be compared with MR images to determine if the source of compression is primarily caused by bony or soft tissue structures. Ultimately, selection of which side to approach and how much of a decompression to perform is made based on patient symptomatology rather than CT/MRI findings. These scans are primarily used to evaluate for the presence of aberrant anatomy and create a safe plan for execution of the surgical procedure.
Bone mineral density testing is not routinely ordered; however, it should be considered in patients at risk because the incidence of subsidence or cage migration is far greater in osteopenic and osteoporotic patients. In patients with poor bone quality, opting for a posterolateral fusion may provide a more equitable risk-to-benefit ratio.
The procedure begins with the induction of general anesthesia and turning the patient onto the operating table. Two commonly used table options are available to the surgeon. A Jackson table with radiolucent posts and a chest pad is appropriate; however, in the authors’ opinion, this conformation may create lumbar lordosis in excess of the typical anatomic alignment. The authors prefer to use a Wilson frame with the pads lowered down completely and with the pelvis and the knees well supported. This results in a more anatomic lumbar lordosis, which facilitates access to the disk space under the exiting nerve root. Care must be taken, however, when using a Wilson frame. Failure to position as outlined above may result in loss of lordosis and the patient fused in a relative flat back.
Ideally, the stand of the C-arm and monitor should be placed opposite the surgeon for easy visualization and to afford working space. A T-bar attachment should also be fixed to the table opposite the surgeon to support the flexible arm that will hold the tubular retractor in place ( Figs. 9.3 and 9.4 ). Neuromonitoring remains controversial in routine degenerative spine operations. The authors do utilize somatosensory-evoked potentials as well as free-running electromyography (EMG). These modalities alert the surgeon to excessive nerve root retraction during diskectomy or cage implantation. If used, it is crucial that the appropriate dermatomes and myotomes are monitored corresponding to the instrumented levels. Motor-evoked potentials may also be monitored; however, the authors do not routinely use this modality.
Percutaneous Pedicle Screws
Depending on surgeon preference, percutaneous instrumentation of the spine may be conducted with a single or double C-arm technique. In the double C-arm technique, the major hurdle is proper preoperative positioning. First, a C-arm should be moved into position perpendicular to the table for a lateral image and then the ‘C’ should be tilted toward the head of the bed so that the arm sits close to the undersurface of the table. The second C-arm for AP fluoroscopy will sit at an angle to the table and should approach from distally ( Fig. 9.5 ). In the single C-arm technique, the critical step is to align the image intensifier perpendicular to the superior endplate with the vertebral body in neutral rotation, identified by the spinous process in the middle of the body and pedicles equal in size and relationship to the lateral walls ( Fig. 9.6 ).
Some surgeons prefer to draw out the incisions for each pedicle screw before starting the procedure by first identifying the center of each pedicle then making a mark 1 cm lateral to this point. In the authors’ experience, these skin markings will tend to deviate as each k-wire is inserted, which may result in suboptimal placement of the incisions. Rather, the entry points should be identified and marked sequentially for each Jamshidi needle ( Fig. 9.7 ). Skin should be incised using a #10 scalpel, with the incision carried down through the fascia. Fluoroscopic guidance and tactile feedback from the needle tip should be utilized to align the Jamshidi needle at the junction of the transverse process and superior articular process. The Jamshidi needle should then be advanced under fluoroscopic guidance so that the tip of the needle reaches the pedicle–vertebral body junction on the lateral image before it reaches the medial border of the pedicle on the AP ( Figs. 9.8–9.10 ). This typically corresponds to depth of insertion of approximately 2.0 cm. A guidewire may then be passed, feeling for impaction of cancellous bone. The authors prefer the use of nitinol wires as opposed to stainless steel. Nitinol allows the wires to be bent and held away from the operative field without the formation of a permanent kink. Such a kink may create difficulty when placing instruments over the wire and may even result in driving the wire deeper into or beyond the vertebral body during tapping or screw placement. Once all wires have been placed, they may be held out of the way using a towel clamp to create room for placement of the tubular retractor. Alternatively at this point, screws may be placed on the side contralateral of the planned TLIF (see below). If utilized in this manner, a rod may be placed and used to aid in distraction of the intervertebral space.