Pseudoarthrosis is defined as the failure to achieve a rigid osseous union between two bony segments one year after the index surgery. It is the most common complication following surgery for lumbar spinal fusion and occurs more frequently when done for spondylolisthesis than lumbar disc degeneration [3]. The rate of pseudoarthrosis is higher in patients with a higher-grade spondylolisthesis and in adults [1]. The significance of a pseudoarthrosis varies between patients. While achieving a successful fusion does not always correlate with resolution of the patient’s symptoms, there is a high statistical correlation between pseudoarthrosis and poor clinical outcome [4]. Historically, the rates for pseudoarthrosis of the lumbar spine are up to 56 % [5]. However, there have been no recent studies demonstrating pseudoarthrosis rates with the use of newer implant and bone grafting technology. In addition, this may not be truly representative of the actual pseudoarthrosis rate given that a certain percentage is asymptomatic and therefore underreported. Hence an asymptomatic pseudoarthrosis does not necessarily require treatment. On the other hand, persistent motion and instability would be surgical indications for a nonunion.

Factors contributing to the development of a pseudoarthrosis include excessive motion at the fusion site, poor surgical technique (insufficient decorticated host bone surface area and graft material), metabolic abnormalities (osteoporosis, Vit D deficiency, excessive alcohol use, malabsorption syndromes), smoking, trauma, and infection. The role that these risk factors play in leading to a nonunion is unclear and its etiology may likely be multifactorial [6].

By knowing some of the risk factors for a pseudoarthrosis, tactics aimed at decreasing the incidence include immobilization of the spondylotic segment, meticulous fusion bed preparation, optimal bone grafting, addressing nutritional status and smoking cessation.

Diagnosis of a pseudoarthrosis is complex and remains a difficult condition to assess. The presence of a nonunion is determined by combining the clinical symptoms and radiographic evidence, after ruling out other possible etiologies of the pain [7]. Symptoms suggestive of a pseudoarthrosis include back pain, signs or symptoms of neurologic irritation, progressive deformity, gait disturbance, and hamstring tightness. Symptoms beyond 1 year in duration are especially worrisome, as a solid fusion should be achieved in that time frame [8]. Evaluation of a patient with persistent symptoms should always start with a through history and physical examination. Plain radiographs (AP, lateral, flexion, and extension) of the lumbar spine should always be obtained prior to advanced imaging. While diagnosing a nonunion on plain radiographs can be difficult due to overlapping bony structures, they provide very valuable information. The absence of a fusion mass, gaps in the fusion mass, persistent motion on flexion and extension films, lucency around the implants, or implant failure are key radiographic findings suggestive of a pseudoarthrosis. Thin-cut CT scan may provide benefit when assessing interbody fusion mass and diagnosing a “locked pseudoarthrosis” [9]. They are less helpful in settings where plain radiographs show a large fusion mass or a pseudoarthrosis is evident. The risks of radiation exposure make the routine use of CT scans not possible despite superior results compared to other imaging sources. No single imaging modality should be used to confer the diagnosis of a pseudoarthrosis. Other signs that a pseudoarthrosis may be present include lucency around pedicle screws, broken implants, and slip progression.

Treatment of a pseudoarthrosis requires the surgeon to address the underlying pathology. If there is any broken hardware present, it should be removed and those levels re-instrumented, typically upsizing the pedicle screw by 1 mm. Intra-operative assessment of the fusion mass and localizing the pseudoarthrosis is required. Once localized, this area is meticulously prepped for bone grafting. Removal of soft tissue, adequate decortication to bleeding bone and packing the site with ample iliac crest bone graft. The use of bone morphogenic protein (BMP) can also be considered in this setting. It may improve local biology and improve the likelihood for successful bony healing. If anterior support was not used in the index procedure, the addition of an interbody in the lumbar region is indicated [3]. In the case of an uninstrumented pseudoarthrosis, the addition of pedicle screw and/or interbody should be considered. Lastly, the use of implanted or external bone stimulators to promote healing can be used, although their effectiveness is unclear in the literature. The successful treatment of a pseudoarthrosis depends on correction of the host factors and optimizing surgical technique.

Neurologic complications that may be encountered with surgery for spondylolisthesis include nerve root injury, cauda equina syndrome, autonomic dysfunction, dural tears, and chronic pain. Prior to surgery it is imperative to document a through neurological evaluation, noting any deficits. In addition, an upright lateral radiograph is required to assess the degree of spondylolisthesis. Advanced imaging of the spine, consisting of an MRI or CT myelogram, is useful to assess any compression on the spinal cord and exiting nerve roots.

Nerve root injury may result from compression, traction, or direct trauma mechanism. Compression on the exiting nerve roots can occur either at the intraforaminal or extraforaminal location. Soft tissue impingement caused by soft disc protrusions or fibrocartilaginous debris from attempted healing of a pars defect may cause compression [10]. Other sources of compression include post-operative epidural hematoma formation, which usually presents in a delayed fashion. In order to adequately assess these sources of potential compression, it is important to evaluate the exiting nerve root from the lateral recess and out the foramen. Complete exposure is particularly important when spondylolisthesis reduction is planned, as reduction into a narrowed foramen could exacerbate neural compression. Reduction maneuvers can lead to neurologic deficits via manipulation and/or nerve root traction. A cadaveric study showed a nonlinear relationship between nerve strain and reduction. In this model, 71 % of the strain experienced by the L5 nerve root occurred during the last 50 % of the reduction [11]. In support of this, recent studies have shown a decrease in neurologic complications using “partial reduction” techniques.

Traction injury has been described while placing the patient in the prone position on the Jackson table. It is theorized that the slip acutely worsens when the muscles are relaxed; the spine displaces dorsally causing traction on the nerve roots [10].

Direct trauma from manipulation of the dural sac may injure the lumbar or sacral nerve roots producing a multidermatomal deficit. Additionally, direct trauma may result during instrumentation of the vertebrae or decompression of the nerve.