Lumbar degenerative disease can present with a multitude of symptoms, so it is important to distinguish between various pathologic processes. An appropriate trial of nonoperative therapy should always be attempted before consideration of surgical intervention. If the patient’s symptoms persist, imaging studies as well as other diagnostic testing must be carefully reviewed prior to formulating a surgical plan.

Lumbar degenerative disease clinically manifests in five basic forms: axial low back pain, spinal stenosis, radiculopathy, degenerative spondylolisthesis, and degenerative scoliosis. These conditions may occur alone or in combination with one another. Surgical decision making, with regard to the need for fusion, for any of these entities depends upon multiple factors, which will be discussed in relation to each disease process.

Chronic axial low back pain has a high socioeconomic impact, with a lifetime prevalence of 30–50 % for moderate and severe back pain [7]. Pure axial low back pain without deformity is successfully treated nonsurgically in the vast majority of cases [8]. Nonoperative modalities include medications, short-term bed rest, physical therapy, and chiropractic therapy. Medications include nonsteroidal anti-inflammatories (NSAIDs), acetaminophen, muscle relaxants, and short durations of narcotics.

After an initial trial with one of the above treatments, epidural and/or facet injections may be useful. In patients with symptoms refractory to nonoperative management for greater than 6 months, fusion may be indicated. However, there is a lack of consensus regarding the efficacy of fusion for discogenic axial back pain [9]. Multiple studies have demonstrated only marginal improvement in back pain following fusion [10–14]. Because of these less than optimal results, many surgeons advocate diagnostic testing to attempt to isolate the pain generator prior to surgical intervention [15]. Provocative discography is the most common test utilized in this scenario, although there is debate in the literature over whether it actually predicts outcomes of spinal fusion [16–19].

In a 2006 comparison of single-level fusion patients with axial low back pain and concordant discogram to a matched control group of patients with unstable spondylolisthesis, Carragee et al. reported a dismal 43 % satisfactory outcome in the discography group at 2 years compared to 92 % in the controls [19].

In a recent systematic review, no subset of patient with chronic back pain could be identified for whom spinal fusion is a predictable and effective treatment. They concluded best evidence does not support the use of provocative discography for patient selection in clinical practice [20]. Clearly, there is lack of agreement upon whether concordant discography can predict the outcome of a fusion for axial low back pain. Due to this fact, it is important to keep in mind that discography is merely one diagnostic component to consider when developing a treatment plan for this challenging problem.

We consider all the factors in the axial low back pain patient’s case and, in this subset of patients, would only recommend fusion after failure of a minimum of 6 months nonoperative therapy. We utilize provocative discography, performed by an experienced pain management specialist, in patient who have failed this regimen and have radiographic evidence of single-level degenerative disease. We offer surgery only if a concordant disk correlates with MRI findings and adjacent disks are non-concordant [21].

Unlike axial low back pain, the question of whether to fuse in cases of lumbar spinal stenosis (LSS) is somewhat more straightforward. If nonoperative measures fail to relieve symptoms in this subset of patients, most obtain meaningful relief of symptoms from surgical decompression [22–24]. Patients with pure stenosis in the absence of degenerative spondylolisthesis or scoliosis may be treated with decompression alone. Numerous studies have shown excellent outcomes in this patient population with regard to relief of leg pain and paresthesias, lower extremity weakness, and walking tolerance [22–24].

LSS patients who demonstrate radiographic evidence of instability should be strongly considered for a fusion procedure [25, 26]. Gross instability may be detected on plain lateral radiographs in the form of spondylolisthesis that translates with active flexion or extension. Additionally, lateral listhesis seen on AP images may also indicate instability. Finally, excessive facet joint fluid seen on T2-weighted trans-axial MRI images is a sign of a more subtle, but clinically significant, instability. Patients with such instability, who undergo decompression alone, are more likely to experience worsening instability and back pain secondary to the loss of posterior stabilizing elements including the supraspinous ligament, intra-spinous ligament, ligamentum flavum, and varying amounts of the facet joints and pars interarticularis. Even in patients without preoperative instability, removal of greater than 50 % of the facet joint or pars interarticularis can lead to iatrogenic instability resulting in poor outcomes and potential need for further surgery [27].

One recent study evaluated nationwide trends in the surgical management of patients with lumbar spinal stenosis (LSS) with and without coexisting spondylolisthesis and scoliosis from 2004 to 2009. They demonstrated simple fusion surgery has increased for treatment of LSS compared with decompression only [4].

As previously mentioned, a clinician must take into account multiple patient factors, including obesity, which has been shown to negatively impact lumbar surgery. One study demonstrated inferior results of surgery for lumbar stenosis, [28] while another demonstrated longer operative times and an increased rate of infection [29].

In today’s era of advanced imaging studies, it is important to remember to obtain good anteroposterior, lateral, and flexion/extension lumbar spine radiographs to rule out degenerative scoliosis or spondylolisthesis. Both of these entities can easily be missed if only an MRI is examined during preoperative planning. If scoliosis or spondylolisthesis is identified in a patient with symptomatic spinal stenosis who has failed nonoperative treatment, it is our typical practice to recommend decompression and fusion in addition to at the scoliotic or unstable levels. Patients undergoing this procedure should be extensively counseled that the purpose of the procedure is to relieve lower extremity symptoms and that relief of back pain is less predictable.

If nonoperative management fails, radiculopathy secondary to herniated nucleus pulposus with or without coexisting stenosis is reliably treated with decompression through lamino-foraminotomy and excision of the disk fragment [30]. As with symptomatic lumbar stenosis, we recommend fusion for patients with lumbar radiculopathy only in the setting of degenerative scoliosis or spondylolisthesis.

Some spine surgeons believe that fusion is necessary for treating disk reherniation. As repeated diskectomy for either ipsilateral or contralateral recurrence requires the removal of more disk material and posterior elements, such as lamina or facet joint, further invasion at the same surgical level can increase the risk of segmental instability [31]. A recent study evaluated the treatment patterns for recurrent lumbar disk herniation among US spine surgeons. The study highlighted the lack of consensus with regard to treatment plan, as some surgeons preferred repeat microdiskectomy, while others preferred microdiskectomy with posterior lumbar interbody fusion/transforaminal lumbar interbody fusion (PLIF/TLIF) [32].

Rarely, patients with lumbar radiculopathy may also have a significant component of low back pain. In these patients, if there is clear evidence via discography or single-level disk degeneration on MRI that the back pain may be referable to the same level as the source of radiculopathy, concomitant fusion may be contemplated. However, it should be noted that studies have shown that such back pain often resolves with diskectomy alone [30, 33].

Among patients with radiculopathy who have a degenerative spondylolisthesis or degenerative scoliosis on imaging studies, decompression with fusion should be strongly considered, regardless of whether or not a significant back pain component is present. Failure to identify these instability patterns and address them at time of surgery may lead to progressive instability and pain.

Surgical treatment, consisting of decompression and fusion with or without instrumentation, has been shown to be an effective treatment of symptomatic degenerative spondylolisthesis [25, 34, 35]. Numerous retrospective and prospective studies have demonstrated satisfactory results with a variety of fusion techniques [35]. Standing flexion and extension radiographs may be the key images necessary to identify this diagnosis as it is often a dynamic rather than a static process and may be missed with MRI alone. Advanced imaging studies should be obtained to identify any areas of stenosis to ensure an adequate decompression and fusion may then be limited to the levels demonstrating instability.