Introduction
The concept of creating a fusion between two vertebral bodies as a solution to pain arising from lumbar spinal segments dates back to the 1930s when the first anterior lumbar interbody fusion (ALIF) was performed, using an autogenous tibial peg for the graft. The presumed mechanism involves both removing potentially pain-generating disc tissue and stabilizing the involved spine segments. In the 1940s, a posterior approach to remove the lumbar disc and insert the graft between the two vertebral bodies was proposed and successfully performed by Dr. Cloward with the advent of the posterior lumbar interbody fusion (PLIF). Years later, the transforaminal interbody fusion (TLIF) was introduced by Harms and Rolinger, who reported on a unilateral variation on the theme established by Dr. Cloward. Much weight is given to their emphasis on a unilateral approach, but perhaps more significant was their use of pedicle screw fixation, which allowed for the removal of the pars articularis and a complete facetectomy. In contrast, the PLIF, as originally conceived, required some facet joint preservation to maintain adequate stability, as pedicle screw fixation was not used. It is worth mentioning that in the current era, PLIF is always performed with pedicle screw fixation and therefore there is no restriction of facet resection. Keep in mind that Harms and Rolinger achieved bilateral interbody support even while the approach was unilateral.
The posterior approach brings distinct advantages, such as the opportunity to provide decompression of the neural structures while avoiding the abdominal organs and large vessels. Retrograde ejaculation is also avoided. However, it presents an increased risk of a dural tear with cerebrospinal fluid leak and damage to the neural structures, in particular the traversing nerve root and dorsal root ganglia. These risks are even higher in the revision cases that this chapter focuses on, related to the presence of scar tissue and distorted anatomy complicating the procedure.
Both the implants and techniques have evolved in a continuing effort to increase the fusion rate and reduce complications, with the end goal of improving clinical outcomes. The development of minimally invasive surgical (MIS) approaches has been an important part of this effort. The modern implementation of the TLIF involves a single cage and often results in asymmetric interbody support despite best efforts. This contrasts with the TLIF as described by Harms and Rolinger, which accomplished symmetric interbody support by rolling a second cylindrical titanium cage to the contralateral side with a “pickle-fork” impactor. As a result, in the current era the decision to insert a single cage via a unilateral approach rather than two cages using a bilateral approach can be critically important, and unfortunately the significance is often underappreciated.
To achieve the goals of surgery and consistently secure the best outcomes, a combination of data points needs to be accounted for, including the patient’s symptoms, the biomechanical properties of the spine, and the peculiarities of the individual patient affecting the rate of fusion. In this chapter, we will evaluate the pros and cons of unilateral compared with bilateral interbody struts in the context of revision cases.
Complications in Posterior and Transforaminal Lumbar Interbody Fusions
PLIF and TLIF are well-established procedures to provide symptom relief and spine column stability, in particular when used in conjunction with pedicle screw fixation, which currently constitutes the state of the art when decompression and stabilization are required.
These operations present risks specifically related to the proximity of the neural structures and their dural covering to the working field and trajectory. The cauda equina and the single nerve roots leaving the spinal canal along with their dural covering sit on top of the disc space that needs to be exposed, cleaned, and eventually filled with bone graft using sharp metallic tools that must be inserted and removed from the surgical field multiple times. These delicate structures reduce the working corridor, and therefore need to be mobilized to facilitate the surgery itself. The operative constraints are well defined in the description of Kambin triangle, which has recently been extended to the concept of Kambin’s area. Examined from this perspective, TLIF operations start more laterally, proceeding medially until the dorsal root ganglia are safely out of harm’s way, whereas a PLIF procedure will start more medially and proceed laterally until the traversing root is safe. In both cases a safe working corridor with access to the disc space can almost always be developed that requires minimal, if any, nerve root retraction.
The other important advance is the development of impacted, in distinction to threaded, cages. Early on, threaded cages were developed because they afforded greater biomechanical stability. As adjunct pedicle screw fixation became a standard component of both PLIF and TLIF, the additional stability afforded by threaded cages was unnecessary. Because of their circular cross-section—a requirement to be able to screw them into the interspace—the amount of interbody distraction dictated by the height was equaled by the degree of neural retraction dictated by the width. Impacted cages have the distinct advantage of the height and width being independent of each other because the cross-section is rectangular, and as a result adequate distraction could be achieved while limiting the dural retraction required ( Fig. 16.1 ). As a result, complications such as intraoperative dural tear and postoperative radiculopathy, which have been reported with a frequency as high as 15%, are reliably less than 1% to 2% in modern series.
Previous lumbar surgeries are a risk factor for the occurrence of intraoperative complications in both PLIF and TLIF cases. The presence of more than two surgical decompressive procedures is a significant predictor for an increased incidence of both dural tear and neural injury. In our experience, a chief advantage of unilateral strut placement is the opportunity it affords to avoid the most difficult scar tissue and thereby decrease the intraoperative complication rate.
Reoperation in Lumbar Spine Surgery
A considerable increase in the number of patients requiring revision surgery of the lumbar spine has become evident in the last decade. Rates of reoperation in lumbar spine surgery vary between 5% and 20% as an overall estimate, including both decompression and fusion procedures. According to a cohort study of 4718 patients who underwent spine reoperation, the significant majority of these reoperations occur more than 1 year after the index surgery (72.1%). The indication for the new operation is more often the development of new or recurrent symptoms, pseudarthrosis, or hardware failure, as opposed to acute postoperative complications such as infection or bleeding. The overall rate of reoperation was found to be higher in patients who underwent fusion as opposed to decompression alone, apart from the subgroup of patients treated for spondylolisthesis. These results are in line with the evidence provided by randomized controlled trials on the surgical treatment of spondylolisthesis. The evidence of Kao et al. suggests that a reoperation performed within the first year most commonly follows a microdiscectomy, whereas beyond the first year the rates following a microdiscectomy or a laminectomy tend to converge.
Multiple factors are involved in the risk of reoperation; being of a younger age (<60 years), for example, seems to be consistently associated with this risk after any type of back surgery. Younger patients rely on their spine for longer periods of time and are, in general, more active. Consequently, they exert more stress on the lumbar segment of their spine, with increased risk of causing further harm.
In general, when approaching a patient who has already had lumbar surgery, it is important to keep in mind that the probability of improvement from the second surgery is lower than from the first surgery and the rate of complications in fusion surgery tend to be higher, regardless of whether a fusion was performed in the first surgery or a subsequent one. For this reason it is incumbent on the surgeon to identify and clearly articulate the goals of the revision surgery.
Essential steps in the evaluation of patients with prior spine surgery who present with recurrent or new symptoms are to critically review the previous surgery and identify the reasons for its failure. These may relate to an error in the biomechanical construct or in the execution of the surgery, or it may simply be the result of further degenerative changes.
In terms of surgical complications, it has been shown that revision surgery of the spine presents an increased risk of dural tear and neural injury. This is caused by the presence of epidural fibrosis and scar tissue in general, the consequent increased stiffness of the structures involved, and the loss of clear anatomical landmarks. The overarching principle in the avoidance of complications in revision cases such as these is to secure the exposure of normal anatomy both above and below before engaging the area of significant scar tissue.
Single Versus Double Cage
The evolution of the cages used in PLIF and TLIF followed the need to reduce complications such as graft collapse, displacement, subsidence, or migration, providing an efficient support for the fusion. The use of cages for fusion, as opposed to posterolateral fusion alone, presents the theoretical advantage of capitalizing on Wolff’s law. This allows for better bone remodeling owing to the compressive forces to which they are subjected, leading to a stronger fusion as well as a higher fusion rate.
The differences between single and double cage placement in spine reoperation can be explored through two main perspectives: the risk/benefit of the surgical approach and the biomechanical properties implied. With regards to the surgical techniques, the advantages of using a single cage relates to the use of a unilateral approach and therefore the possible opportunity to avoid scar tissue. The surgical dissection of nonvirgin anatomy is always challenging because of anatomical distortion of the different tissues and compartments, as well as the presence of dense adhesions. In spine surgery specifically, the main risks relate to the production of dural tears, damage to the neural structures, and residual compression. Choosing a unilateral approach to place an intervertebral cage facilitates the reduction of these risks, particularly when the previously operated side of the spine can be avoided altogether.
From a biomechanical standpoint, the positioning of a single cage presents technical challenges in comparison to using two cages, which can be mitigated by the position and shape of the cage implanted ( Figs. 16.2 and 16.3 ). The main advantage of positioning two cages relates to the greater surface available, and to the symmetric positioning of two struts that allows for a more equal distribution of the load. The ideal positioning of a single cage, to provide a symmetric load-bearing capacity, requires traversing the central section of the end-plate, which is less dense, and securing solid loading on the apophyseal ring. In the case of a linear cage, the cage is placed obliquely and contact with the apophyseal ring is secured both proximally and distally (see Fig. 16.2 ). In the case of a curved or banana-shaped cage, the desired location is on the anterior portion of the apophyseal ring ( Fig. 16.4 ). For the revision surgeon the important point is to make sure that the appropriate balance is achieved between minimizing the difficulty of the dissection and maximizing the biomechanical strength of the construct.
