Introduction
This chapter provides an overview of the contemporary literature on lumbar interbody fusion (LIF) instrumentation based on the Spine Patient Outcome Research Trial, Swedish Spinal Stenosis Study, and a recent New England Journal of Medicine article on clinical outcomes. Preoperative factors influencing the surgical outcome are discussed, along with five basic tenets of LIF based on: (1) presence and extent of concurrent listhesis at the level of fusion, (2) need for unilateral versus bilateral foraminal decompression, (3) presence of central canal stenosis, (4) loss of coronal and sagittal balance, and (5) the history of prior surgery at the same level or adjacent levels with or without instrumentations. We also discuss the complications of some original LIF approaches with relevant illustrations depicting the successful use of alternate LIF approaches to correct them. The chapter also portrays the synergistic role of novel techniques and technologies that can make modern LIF procedures safer, more feasible, and more efficacious. These LIF techniques require a lot of expertise and can often be hard to do well, especially in reoperations. These operations are very equipment dependent, and it is important to be familiar with all the common LIF techniques in clinical practice and their individual benefits and complications. Clear understanding about the various LIF approaches can equip the spine surgeon especially when dealing with a complication needing implant retrieval from a distinct approach that was performed by another surgeon.
Background
The first recorded surgical attempt at fusion of the spinal column was in 1891, when Hadra attempted cervical interspinous wiring to treat subluxation caused by Potts spine. However, it took another two decades before the first reports of surgery in the lumbosacral spine emerged in 1911, when Russell Hibbs and Fred Albee reported on their techniques of spinal fusion to treat tuberculosis. Hibbs used “feathered” (morselized) laminae and spinal processes, which were placed into decorticated facet joints to create the world’s first dynamic stabilization. Albee, on the other hand, used tibial grafts between the spinous processes to stimulate fusion. The rationale behind the “posterior fusion” surgery was to prevent deformity, improve stability, and reduce pain. The next major step in development of spinal surgery occurred when Watkins reported the posterolateral intertransverse fusion in 1953.
In 1962 Harrington reported on his series of scoliosis surgery using sublaminar hooks and rods, and the era of spinal instrumentation began. Advances in metallurgy and surgical techniques led to the development of transpedicular, translaminar, corticopedicular, and facet screw systems as well as myriad types of interbody cages made of titanium, polyetheretherketone, and so forth, with variations such as trabecular mesh. Spinal technology grew closely following the prosthetic joint technology; for example, the lessons of enhanced biomechanical pull-out strength and migration resistance offered by porous coating of hip implant (first application of Plasmapore coating of titanium hip prosthesis) in 1986 slowly made its way to the lumbar spine market in 2012 (as the first Plasmapore-coated polyetheretherketone lumbar implant). Continuous improvisation of novel technologies, designs, navigation, and robotics make LIF an ever-evolving area of spine surgery.
Other revolutionizing factors included various osteoinductive and osteoconductive materials being used in bone fusion. A significant step forward was made with the development of recombinant human bone morphogenetic protein (rhBMP). BMPs comprise a group of osteoinductive cytokines that belong to the transforming growth factor beta (TGF-β) superfamily. BMP-2 had been approved by the US Food and Drug Administration (FDA) in 2002 for anterior lumbar interbody fusion (ALIF) based on a pivotal study by Burkus et al. Since its introduction into clinical use, BMP had an immense surge in popularity as spinal surgeons started using osteobiologicals in large numbers to avoid the graft site complications associated with iliac crest grafts. This, in turn, led to reports of many serious complications following off-label use in posterior surgeries, as well as in ALIF. Carragee et al. reported a higher reoperation rate in patients treated with rhBMP-2, mainly to correct graft subsidence. They, among other researchers, found that as many as 20% to 70% of patients had suffered some complications that could be attributed to BMP, including endplate resorption, retrograde ejaculation, seroma formation, bone overgrowth, osteolysis, and an increased risk of cancer. The Yale University Open Data Access study was conducted against this background to assess the safety and utility of BMP-2 and found that the incidence of retrograde ejaculation and neurologic complications were equal in both autograft and BMP-augmented ALIF surgeries. It also demonstrated a small increased relative risk of malignancy with the use of rhBMP-2 in posterolateral lumbar surgeries. However, the absolute risk was very low and therefore clinically insignificant. No difference was found between rhBMP-2 and iliac crest graft, but there was a higher rate of ectopic bone formation in these procedures. Based on these findings, judicious use of BMP is now advocated in posterior lumbar surgeries. In transforaminal lumbar interbody fusion (TLIF), a high risk of postoperative radiculitis has been reported; hence, the use of BMP in these cases is not encouraged. The use of bone marrow aspiration from the exposed lumbar vertebral bodies during the surgery, and then using this aspirate as graft material has been recently reported. This overcomes the graft site complications as well as the problems associated with the use of BMP-2. Further research is ongoing about the use of growth differentiation factor 5, also known as BMP-14, as an osteogenetic material.
Interbody Fusion: A Primer and Recent Literature
Over the years a number of approaches have been developed for LIF, namely posterior, anterior, axial, transforaminal, lateral, extreme lateral, and oblique lateral. In keeping with the trend toward minimally invasive surgeries, reports of percutaneous attempts at surgical stabilization of the lumbar spine first appeared more than two decades ago. Although posterior lumbar fusion via minimally invasive techniques has become commonplace, endoscopic surgeries for TLIF require special training. Even though the exact procedure that is chosen for a particular patient may depend on a number of factors, such as the exact pathology and surgical anatomy of a particular patient and the surgeon’s preference, the pathologies that need surgical fusion of the lumbar spine remain broadly the same. These include degenerative diseases, spinal trauma, deformity correction, infections, and tumors.
Interbody fusion is indicated in a subgroup of patients in whom the surgical approach to treat a pathology results in spinal instability or if preexisting instability is present. Spondylolisthesis, the most common indication for interbody fusion, is defined as the horizontal translation of a vertebral body over an adjacent one and was divided into five groups by Newman and Stone, namely, congenital, degenerative, spondylolytic, traumatic, and pathologic. Spondylolisthesis is graded depending on the length of the vertebral body that is not in contact with the adjacent vertebra (extent of slippage). In grade I spondylolisthesis, the area of noncontact is less than 25% of the anteroposterior diameter of the vertebral body on a lateral x-ray study, whereas in grade II, the slip is between 26% and 50%. When the area of noncontact is between 51% and 75%, it is called grade III; in grade IV, the slip is between 76% and 100%. A greater than 100% slip, where the adjacent vertebral bodies are lying totally separated from each other, is designated grade V, or spondyloptosis. Grades I and II are considered low grade, whereas the rest are designated as high-grade spondylolisthesis. The degenerative variant is usually seen in women over the age of 50 years. Low-grade lesions are commonly treated conservatively, and surgery is reserved for those patients who fail to respond or for those who have neurologic deterioration.
The benefit of surgery has been demonstrated repeatedly in various trials, with the Spine Patient Outcomes Research Trial being the most significant study to support surgery for these patients. The best surgery indicated in each case of degenerative spondylolisthesis and whether these patients need spinal fusion are still open to debate. Presence of spondylolisthesis in patients with lumbar canal stenosis was considered an indication for fusion surgery, even in stable cases where the slip is less than 3 mm. Recent studies have shown that in the United States approximately half the patients with lumbar spinal stenosis and 96% of those with degenerative spondylolisthesis undergo spinal fusion. This view has been challenged by recent studies from Sweden and the United States, which found that the benefit of fusion in patients with stable spondylolisthesis and lumbar spinal stenosis was marginal at best. The Swedish Spinal Stenosis Study was a randomized controlled trial of 247 patients who were divided into fusion and nonfusion groups, with each group containing at least 40 patients with and without degenerative spondylolisthesis (at least 3 mm). At 2- and 5-year follow-ups, no significant difference in outcomes were found in the two groups. The rates of reoperation were also remarkably similar, raising a question about the need for fusion in degenerative spondylolisthesis. In this study, the preoperative evaluation did not include flexion-extension x-ray studies; if this had been done and patients with demonstrable instability were assigned to the fusion group, the results in the nonfusion group vis à vis repeated surgery may have been even better.
However, another study, albeit smaller, published in the same issue of New England Journal Medicine mentioned above, found a minimally improved physical outcome in patients who had undergone fusion surgery at 2, 3, and 4 years. This was not considered sufficiently significant to support the higher cost in terms of financial burden, blood loss, operative time, and hospital stay in these patients. Counter-intuitively the reoperation rate was higher in patients who did not undergo fusion even though this study had excluded patients with instability as demonstrated by flexion-extension x-ray study. This apparent confounding factor may be related to the physician approach in the two countries where the studies were done, with physicians in the United States tending to offer revision-with-fusion to those patients who had pain after decompression alone, whereas the threshold for offering revision surgery to a patient with pain, who had already undergone spinal stabilization, may be much higher. The current evidence seems to point to the need for fusion only in those with unstable degenerative spondylolisthesis as shown on flexion-extension x-ray films, patients with destruction of vertebral bodies owing to trauma, infection or tumors, and spinal deformities such as other variants of spondylolisthesis or scoliosis. The need for fusion in neural-foraminal stenosis owing to postsurgical disk prolapse is another contentious area, with no evidence to support improved outcome with fusion.
In isthmic spondylolisthesis, there is a fracture of the pars interarticularis or isthmus, which is the area of the vertebra where the lamina and inferior articular process join the pedicle and the superior articular process. These cases often occur in a patient population that is younger than the typical patient with degenerative spondylolisthesis, which is common in the third to fifth decades. The management strategy is similar, with a 3-month trial of conservative therapy before opting for surgical management, even though some studies have shown a better outcome for surgery in these patients. Even in this case, multiple surgical techniques are described to treat isthmic spondylolisthesis, depending on many factors such as lateral foraminal compression, fusion of facets, grade of listhesis, and expertise of surgeon.
Preoperative Factors Influencing Outcome of Spinal Fusion
Although appropriate patient selection and an impeccable technique go far in ensuring the success of spinal stabilization surgeries, a number of comorbid conditions or extraneous factors, such as diabetes mellitus, osteoporosis, and smoking, may affect a good outcome. Patients with diabetes mellitus have a much higher rate of complications following any spinal surgery, with surgical site infections accounting for a majority of problems. A study in 2003 by Glassman et al. showed that the overall complication rates in diabetic patients were over 50%, whereas it was only 21% in controls. Nonunion rates in the diabetic patients ranged between 22% and 26%, whereas it was 5% in controls. A more recent study by Guzman et al. showed that for diabetic patients the mean length of stay increased (∼2.5 d), costs were greater (1.3-fold), and there was a greater risk of inpatient mortality (odds ratio = 2.6, P < .0009). The ability of cigarette smoke to inhibit fusion was demonstrated in animal studies and fusion rates following surgery have also been found to be lower in patients who smoked. Cessation of smoking at least 6 months prior to a planned surgery may overcome this risk. Concomitant rheumatoid arthritis can also increase the risk of complications, such as surgical site infections and implant failure, but fusion rates in patients with rheumatoid arthritis have been reported to be comparable to that of controls. Osteoporosis is known to increase the risk of implant failure and fractures and should be medically managed prior to, or concurrent with, surgery. Bone density index (bone densitometry) prior to an elective surgery in a patient at high risk can assess the chance of graft failure and vertebral body osteoporotic collapse. High risk patients undergoing elective surgery can be assessed by an endocrinologist, as the management strategies of these patients are complex and include not only the use of calcium and vitamin D replacement, but also administration of alendronate, parathyroid hormone, calcitonin and raloxifene, with use of a post-operative external bone stimulator.
Tenets of Interbody Fusion
The five basic tenets that govern the type of interbody fusion are (1) the presence and extent of spondylolisthesis; (2) the need for unilateral or bilateral neural foraminal decompression; (3) the presence of coexistent central canal stenosis requiring decompression; (4) the loss of coronal and sagittal balance in relation to the level of disease; and (5) the presence of prior surgery at the same level or adjacent levels with or without instrumentation and/or interbody grafts.
Symptomatic low-grade spondylolisthesis is by far the most common indication for interbody fusion in the lumbar spine. Careful selection of approaches must be directed by goal, lateralization of clinical signs, loss of curvature, and prior surgery, and these must be in relation to the age, gender, and medical condition of the patient. With significant spondylolisthesis and both neural foramina at lower lumbar levels needing to be decompressed, an ALIF can be used, especially if there is no canal stenosis. ALIF is useful for correcting listhesis, especially if the slippage is the cause of central, lateral recess, or foraminal stenosis, as against significant ligamentum flavum hypertrophy with associated large hypertrophic facets. If the patient has circumferential soft tissue canal stenosis resulting in neurogenic claudication rather than radiculopathic symptoms, a TLIF would be a better option.
AxiaLIF can be used if central canal stenosis is not significant, and the foraminal compression does not cause symptomatic radiculopathy. With a predominantly unilateral radiculopathy, a TLIF with wide facetectomy at the side of radiculopathy can be used, with facetal decompression along the symptomatic side. If the surgeon feels that an indirect foraminal decompression is sufficient to treat radiculopathy, a direct lateral (DLIF/lateral lumbar interbody fusion [XLIF]) can be used. There is always a concern about using stand-alone techniques with ALIF, AxiaLIF, and direct lateral approaches, which then would require adjunct instrumentation posteriorly with pedicle screws, facet screws, or cortical/corticopedicular screw placement at those levels.
In reoperations, the following factors must be considered while planning the surgery. Avoid dissecting through the old surgical scar if possible; for example, if there is recurrence after multiple posterior approaches, an ALIF or DLIF can be used, unless the old hardware needs to be revised owing to fracture. A fractured/displaced L4-5 DLIF graft can removed by repeat DLIF or ALIF as the cage is large, whereas a combined TLIF or posterior lumbar interbody fusion (PLIF) might be needed to get the fragmented cage if it has slipped below the level of disk space or is compressing the axilla of nerve root. Always anticipate cerebrospinal fluid leak from a dural tear owing to severe epidural fibrosis from prior surgery, in which case an open approach is preferred over minimally invasive transtubular retractors. It is easier to follow the normal dura mater with an open or mini-open approach compared to transtubular vision. Patients with failed back syndrome are advised to have an electromyography (EMG) to evaluate residual deficits from prior surgery to prognosticate on expected neurologic recovery. Always review the existing hardware using a computed tomography scan, rather than a magnetic resonance image of the lumbosacral spine to rule for fractured implants or haloing around screws (nonunion) or graft dislodgement. Always verify the sagittal/coronal balance (using a full scoliosis film, if needed) and the levels adjacent to the symptomatic one (dynamic x-ray study of flexion and extension).
Continuous EMG and somatosensory evoked potential (SSEP) monitoring during the surgery may be useful in reducing the risk of complications caused by overzealous manipulation. The latest published guidelines on the use of intra-operative monitoring has focused attention on the absence of level I evidence regarding the ability of intraoperative monitoring to prevent (as opposed to diagnosing) injury to the spinal cord during surgery. We would however, advise a set of electrophysiologic monitoring before and after the patient is positioned prone or lateral, especially the latter after breaking/bending the operating table which causes stretching of the psoas muscle. It may be necessary to monitor the upper lumbar plexus in selected cases.
Each case should be individually assessed for safety and feasibility of each approach; for example, if ALIF in a young male patient runs the risk of retrograde ejaculation, XLIF graft, which migrated into the central canal, can be retrieved only by an XLIF approach because of the larger footprint of the cage. In osteopenic cases with fractured pedicles, an adjunct posterior support can be provided by facet screws, laminar clamps, or even interspinous clamps, depending on the presence of canal stenosis and features of spinal anatomy on imaging. It is important for surgeons to be familiar with these multiple interbody fusion techniques and specific implant retrieval methods in graft failures, as one could potentially encounter a complication from any of these approaches (e.g., graft migration, nonunion, osteomyelitic collapse) in the years to come ( Figs. 1.1–1.3 ).
