Authors

1. John Alex Sielatycki MD – Center for Sports Medicine and Orthopaedics, Chattanooga Orthopaedic Group
2. Tyler Metcalf MS IV – The Ohio State University College of Medicine
3. Grant Chudik OMS II – The Ohio University Heritage College of Medicine
4. Clinton J. Devin MD – Department of Orthopaedic Surgery, Vanderbilt University School of Medicine
5. Scott Dean Hodges DO – Center for Sports Medicine and Orthopaedics, Chattanooga Orthopaedic Group
6. Marissa Koscielski, MS – University of Cincinnati College of Medicine 

Abstract

Study Design: Narrative Review

Objective: This narrative review is intended to integrate the current clinical, radiographic, biomechanical, and finite element literature on lumbar instability and to specifically focus on how degenerative changes and surgical interventions may affect lumbar instability.

Methods: A literature search (English Language) was performed via PubMed (Medline) and Google Scholar with the search terms “spinal instability”, “spinal instability biomechanics”, “spinal instability biomechanical”, “spinal instability clinical”, “spinal instability radiographic”, and “spinal instability finite element”. The articles relevant to clinical practice were organized into clinical, radiographic, biomechanical, and finite element study categories.

Results: Facet joint effusion, sagittal facet orientation, disc degeneration, loss of ligamentotaxis, and chronic pars defects are among the most commonly noted reasons a patient may develop instability.  The presence of facet effusion on MRI seems to be most predictive of pathologically increased motion and potential instability at a given lumbar segment.  An intact posterior ligamentous complex appears to be more important in maintaining stability than the health of the facet joints or intervertebral disc. Numerous studies report that fusion is not required in most cases of laminectomy for stenosis even in cases with spondylolisthesis or when complete facetectomy is performed. 

Conclusions:  The concept of instability is vitally important when caring for patients with spinal disorders. This review highlights that there is a need for a more universal understanding of the progression of lumbar degeneration into lumbar instability and the need for advancements in surgical techniques to appropriately treat lumbar instability.    

Keywords: Spine, Instability, Review, Spinal, Disorders, Lumbar, Degeneration, Surgery, Biomechanics, Biomechanical, Radiographic

Introduction
The concept of spinal stability may seem simple; however, a universal agreement on this concept has eluded clinicians and researchers alike for decades. The term dates to the 1940’s when Knutsson reported the association between disc degeneration and radiographic signs of instability.¹ He studied a series of 69 patients with disc degeneration that underwent dynamic flexion-extension radiographs and found both abnormal translation and angulation.  Prior to this report it was believed that disc degeneration caused a loss of the normal disc height and led to the superior vertebra slanting backward along the path of the spinous processes. Knuttson was the first to report on the presence of abnormal lumbar motion as a sign of instability and a potential precursor to disc degeneration.  He also provided a description of his technique for dynamic radiographs to be used as a functional test for disc degeneration. Meschan reported on the anatomic restraints to anterior and posterior translation in the lumbar spine.^2,3 He explained that the vertebral body was restrained in the normal state by the collection of paraspinous ligaments, muscles, and superior and inferior apophyseal joints. The pars interarticularis was defined as the isthmus of bone lying between the superior and inferior articular processes. In cases of defects of the bilateral pars interarticularis, the vertebral body remains stable superiorly but not inferiorly, and with sufficient stress anterior and posterior motion can occur, causing instability. In 1950, Barr first correlated mechanical instability related to degenerative disc lesions with clinical complaints of lower back pain.⁴ Shortly later, Harris and MacNab reported on the relationship between degeneration in the intervertebral discs and clinical complaints of low back pain and sciatica.⁵ In this early report, the term instability was not associated with an increased motion of the vertebra, but instead an irregular movement of the vertebra. As this review paper will illustrate, over the subsequent decades the term instability has had many differing definitions based on clinical, radiographic, biomechanical, and mathematical criteria. We will show that despite extensive research and numerous publications, there remains significant disparity in defining the term instability. Moreover, even when agreement is found in the meaning of instability, little agreement exists on its direct relation to clinical symptoms and outcomes of treatment.  

Translating basic science research into accurate and reproducible clinical results is an important challenge. This review will highlight studies showing significant changes in the quantity or quality of lumbar motion, as well as highlight studies showing a discrepancy regarding the clinical relevance of these findings. Although biomechanical, radiographic, and mathematical models and results provide us with useful information, none of these studies are capable of exactly reproducing the environment found in the living body and are thus limited in their ability to predict clinical results or outcomes.  

The purpose of this paper is to review the extensive literature available regarding lumbar spine instability with an emphasis on the relationships between degenerative changes and postoperative/ iatrogenic changes to lumbar stability. There are many other potential causes of instability including trauma, infection, deformity, and tumor that are beyond the scope of this paper. We will begin with some background information on the biomechanical terms associated with spinal instability and then review relevant clinical, radiographic, biomechanical, and finite element studies.

Methods

A literature search utilized PubMed (Medline) and Google Scholar to identify clinically relevant literature on spinal instability. The search criteria included English language, peer-reviewed articles, and the search terms “spinal instability”, “spinal instability biomechanics”, “spinal instability biomechanical”, “spinal instability clinical”, “spinal instability radiographic”, and “spinal instability finite element”. Due to the limited number of articles with these search terms, no publication year criteria were used. Articles that were applicable to clinical practice were chosen and the references of these articles were screened for additional publications. The relevant spinal instability articles were organized into clinical studies, radiographic studies, biomechanical studies, and finite element studies. The study design, outcome measures, and study findings of the articles in these categories are summarized in Tables 3, 4, 5, and 6.

Review of Biomechanical Concepts

Table 1 provides definitions for many of the biomechanical terms used throughout this paper.^6,7 One of the most commonly used definitions for spinal instability comes from the classic text by White and Panjabi which states: “Clinical instability is the loss of the ability of the spine under physiologic loads to maintain its pattern of displacement so that there is no initial or additional neurologic deficit, no major deformity, and no incapacitating pain.”^6,7 This definition attempts to account for all etiologies of spinal instability and allow for application to the entire spinal column.  White and Panjabi developed a scoring system for diagnosing clinical instability as summarized in Table 2.⁷ The radiographic criteria used in this scoring system are based on measurements from static and dynamic lumbar spine radiographs as shown in Figure 1.

Another key concept to explain at the outset is the “neutral zone.” This refers to the range of motion of a body (e.g., vertebra), starting from the neutral position up to the beginning of some resistance offered by the joint.⁷ As shown in Figure 2 (left), the analogy commonly used to explain this concept is a marble in a soup bowl.⁸  In a normal or stable spinal segment the ligaments, muscles, disc, and facet joints act as constraints to motion so there is a small amount of motion present without the application of an increasing amount of force. Likewise, in a curved bowl a marble will have limited motion without an applied force to move it up the curved surface of a bowl. In an unstable spine the normal constraints to motion are altered, allowing greater motion with minimal applied force. Similarly, the analogy of a flat bowl allows a marble to have a greater available motion with minimal force applied. The neutral zone refers to the initial amount of motion present in the spinal segment (or “marble”) with little to no applied forces. The elastic zone refers to the next portion of motion where an increasing amount of an applied force is necessary to overcome the normal constraints to motion. During this phase, once the applied force is removed the object is able to return to its normal position without any permanent changes. However, as illustrated in Figure 2 (right), if the applied force continues to increase, the normal constraints can be damaged allowing for abnormal motion of the object and an inability to return to its normal state after the applied forces have been removed.^7,9 This range is called the plastic zone. The normal physiologic range of motion, where no permanent damage occurs, consists of the combination of the neutral zone and the elastic zone. As we will discuss later, alterations in the neutral zone due to either degenerative or iatrogenic changes in the normal constraints or application of sufficient force to enter the plastic zone can result in instability of the spinal segment.

Clinical Studies

Patients with lumbar spinal stenosis that have failed nonsurgical options, often undergo decompression with or without spinal fusion. An ongoing debate remains on which patients would benefit from the addition of fusion in an attempt to prevent postoperative instability and the potential need for subsequent surgery.¹⁰ Many clinical studies have investigated the effects of various patient characteristics on developing signs of clinical instability, while others have attempted to identify either preoperative or perioperative risk factors for the development of postoperative symptomatic instability. Preoperative variables consist of patient gender, weight/body mass index, and presence of degenerative changes on imaging studies including static or dynamic spondylolisthesis. Surgical variables have included the extent of decompression performed and the use of minimally invasive surgical techniques. A common goal of these studies is to identify factors that could predispose patients to increased risk of developing instability after decompression-only surgery for lumbar spinal stenosis.  

Lee reported in a retrospective review of 27 patients that underwent an extensive posterior decompression (complete laminectomy, flavectomy, and bilateral partial facetectomy to the level of the pedicles) with a mean follow-up of 2.5 years (range 1-4.5 years).¹¹ The overall incidence of postoperative instability was 1 of 27 patients (3.7%); however, all four patients with preoperative spondylolisthesis had progression postoperatively. While this study showed a low overall risk for developing instability even after an extensive decompression, the presence of preoperative spondylolisthesis appeared to increase that risk.

The degree of disruption of the facet joints has also been postulated to affect postoperative stability. Garrido and Connaughton reported on a respective review of 41 patients (mean age 46) that underwent a complete unilateral facetectomy for foraminal lumbar disc herniation.¹² The mean follow-up time was 22.4 months (range 4-60 months). One patient (2.4%) developed postoperative instability and went on to a fusion for complaints of low back pain. The authors concluded that there was a very low risk of iatrogenic instability following complete unilateral facetectomy for lateral disc herniation and as a result, fusion did not have to be performed routinely. In a similar study, Epstein reported on a retrospective review of 170 patients undergoing surgery for a far-lateral lumbar disc herniation using three different surgical techniques.¹³ Surgery involved either complete facetectomy, laminotomy with medial facetectomy or intertransverse discectomy. The overall rate of patients requiring later fusion for symptomatic instability was 4%. No significant differences were identified in terms of clinical outcome or need for additional surgery based on the type of surgery performed. In cases of far lateral disc herniation, a complete unilateral facetectomy arguably provides the best visualization and decompression of the nerve root. Concerns of potential iatrogenic instability following a complete unilateral facetectomy might prevent surgeons from performing this technique without a fusion, but these concerns did not appear to be justified by either of these reports.

Decompression at multiple levels has also been raised as a concern for developing instability. Ofluoglu et al. conducted a retrospective review of 34 patients following decompression surgery for lumbar spinal stenosis.¹⁴ The surgery consisted of complete laminectomy with preservation of the facet joints at 1-3 levels. The mean follow-up was 23 months (range 12-60 months). One of 34 patients (3%) developed radiographic evidence of postoperative instability with complaints of back and leg pain. This patient was offered revision surgery with fusion but declined further intervention.

More recent studies have evaluated the effects of using minimally invasive surgical techniques on the development of instability. A potential benefit of these techniques minimized disruption of the normal soft tissue envelope surrounding the spine. It is believed that by leaving the posterior ligamentous complex intact as well as minimizing the disruption of the facet joints and facet capsules, there could be less risk for instability. Sasai et al. reported a retrospective study of 48 patients treated with microsurgical bilateral decompression via a unilateral approach for either lumbar stenosis or lumbar degenerative spondylolisthesis.¹⁵ The mean follow-up was 46 months (range 24-71 months). Postoperative results included clinical outcomes measures as well as radiographic measurements of intervertebral angles and slip percentages. The only measure that increased postoperatively was slip percentage. Measures for intervertebral angle, dynamic intervertebral angle, and dynamic slip percentage did not change significantly. Additionally, there were no significant differences in clinical outcomes measures for the two groups, and no patient went on to additional surgery. The authors concluded that patients with both lumbar stenosis and degenerative spondylolisthesis could be effectively treated with this less invasive decompression procedure without fusion and without significant risk for instability leading to additional surgery.

In a similar study, Alimi et al. retrospectively reviewed of 110 patients with lumbar spinal stenosis that received a minimally invasive laminectomy to determine clinical outcomes and reoperation rates.16 There was no preoperative spondylolisthesis in 46 (47.4%) patients, and 51 (52.6%) had preoperative grade I spondylolisthesis. No case of postoperative spondylolisthesis occurred in patients without preoperative spondylolisthesis, and no significant changes in the magnitude of spondylolisthesis occurred in those with preoperative spondylolisthesis. The overall revision surgery rate was 3.5%, but there was no significant difference among patients with or without preoperative spondylolisthesis in terms of either clinical outcomes or need for revision surgery. The authors concluded that the revision surgery rates for iatrogenic instability with their minimally invasive technique were less than those reported for the standard open decompression techniques and that by preserving the posterior elements of the lumbar spine routine fusion may not be needed in all patients with lumbar stenosis and stable grade I spondylolisthesis.  

Guha et al. performed a systematic review of the literature to determine the rates of iatrogenic spondylolisthesis following lumbar decompression.¹⁷ They reviewed 24 studies including 2,496 patients undergoing either open laminectomy or minimally invasive decompression procedures. Overall, the rate of postoperative radiographic instability was 5.5%. Higher rates were found in patients with pre-existing spondylolisthesis (12.6%) and in patients undergoing open laminectomy (12%). The overall rate of additional surgery for instability was 1.8%, but it was higher for patients with pre-existing spondylolisthesis (9.3%) and those undergoing open laminectomy (4.1%). Interestingly, the rate of subsequent surgery for instability (1.8%) was substantially lower than the rate of radiographic evidence of instability (5.5%). This finding highlights the fact that even if there is radiographic evidence suggesting the presence of instability, this does not always correlate with clinical symptoms and need for additional treatment.

In a retrospective study of 163 consecutive patients with lumbar degenerative spondylolisthesis Sato et al. performed either posterior lumbar interbody fusion or laminectomy without fusion.18 The mean follow-up was 5.9 +/- 1.6 years. The overall reoperation rates were 6.1% at 1-year and increased to 23.3% at the final follow-up period. There was a significantly higher rate of additional surgery in the laminectomy group as compared to the fusion group (33.8% versus 14.4%, p=0.01).  Independent risks factors for additional surgery at the same level included obesity and disc height greater than 10 mm. While this study did provide specific variables associated with developing instability, their revision surgery rate was much greater than many previous reports. It is uncertain whether this affected their conclusions.

The use of preoperative dynamic radiographs to assess for instability is frequently performed. In an attempt to assess the need for these images, Donnarumma et al. reported on a retrospective study of 174 patients that underwent either decompression alone or decompression with instrumented fusion for degenerative lumbar stenosis without deformity, spondylolisthesis or instability.¹⁹ They studied the effects of various patient factors as well as the presence of pre- and post-surgical “micro-instability” defined as minor spondylolisthesis detectable on dynamic radiographs but not on static imaging studies.²⁰ The mean follow-up was 38.6 months (range 18-60 months). Pre-surgical micro-instability was identified in 72 (41.4%) patients. Decompression was performed in 82 (47.1%), and fusion performed in 92 (52.9%) patients. At follow-up 20 (24.4%) of the decompression patients and 3 (5.8%) of fusion patients had micro-instability. There was no significant difference in pain or disability scores based on the type of surgery performed (p=0.20 and p=0.83, respectively). There was no significant difference or change of disability scores based on the presence of pre-surgical instability between the two surgical groups. There was a significant effect from pre-surgical instability on change in pain scores regardless of the type of surgery performed (p=0.007). Patients with post-surgical micro-instability had worse outcomes regardless of the type of surgery performed. Decompression patients with pre-surgical micro-instability that did not worsen postoperatively had similar results to decompression patients without pre-surgical and post-surgical micro-instability. The authors concluded that decompression alone was the preferred option for these patients, and the dynamic radiograph played a negligible role in the preoperative evaluation. Additionally, they argued that stabilization was not any more effective than decompression alone for the symptoms of micro-instability. The clinical significance of pre-surgical micro-instability remained unclear, but its presence did not justify fusion for improved clinical benefit in this report.

Ghogawala et al. performed a prospective, randomized, controlled trial of patients with stable degenerative spondylolisthesis and symptomatic lumbar stenosis treated with either open laminectomy alone or with the addition of posterior instrumented fusion.²¹ The study included 66 patients, and the follow-up was 4 years. Patients were included with spondylolisthesis from 3-14 mm, lumbar stenosis, and neurogenic claudication with or without radiculopathy. Patients with a preoperative dynamic instability greater than 3 mm were excluded. The open decompression included a complete laminectomy and partial medial facetectomy. Patients in the fusion group achieved clinically meaningful improvements in overall physical health-related quality of life as compared to the laminectomy group. The patients in the fusion group had a significantly lower rate of revision surgery compared to the decompression group (14% versus 34%, p=0.05). Revision surgery in the decompression group was a fusion for postoperative instability in all cases, while revision surgery in the fusion group was for adjacent level degenerative changes in all cases. With a postoperative revision rate for instability of 34%, this study’s results were similar to those reported by Sato et al, but higher than most other studies. Obesity was not found to be a risk factor for revision surgery (p=0.58). The authors did make note that the use of newer minimally invasive decompression techniques could have had lower risk of instability.^15,22 Importantly, Kuo et al performed a propensity-matched analysis showing lower 5-year re-operation rate following unilateral laminotomy with bilateral decompression as compared to fusion for cases of degenerative spondylolisthesis.²² The dichotomous findings of Kuo’s paper thus may indicate that minimally invasive decompression techniques may protect against iatrogenic instability.

In some contrast to Ghogawala’s paper, Forsth et al reported no difference in outcomes or 6.5-year reoperation rate between decompression alone vs. decompression with fusion for degenerative lumbar stenosis with or without spondylolisthesis.²³ In this study by Forsth, 247 patients were randomly assigned to decompression alone vs. decompression with fusion for degenerative lumbar stenosis with or without spondylolisthesis. All patients undergoing fusion had a central decompression in combination with instrumented fusion (90% posterolateral fusions, 5% interbody fusions, and 5% non-instrumented fusions). Patients undergoing decompression alone had either central decompression (excision of midline structures with partial medial facetectomy – 82%) or bilateral laminotomy with preservation of the midline structures (18%).  Patient-reported outcomes did not differ between the groups at two years (ODI 27 in fusion group, 24 in decompression group) or at 5 years. The decompression group had significantly less blood loss and shorter hospital stay as compared to the fusion group. At 6.5 years, the reoperation rates did not differ between the fusion (22%) and decompression alone (21%) groups. Additionally, the presence of degenerative spondylolisthesis did not impact the results, such that the patient-reported outcomes and reoperation rates decompression alone vs. decompression with fusion did not differ in those patients with spondylolisthesis. The findings of this randomized controlled trial indicate that decompression without fusion can be done without excessive risk for instability and reoperation as compared to fusion.

Ahmad et al. performed a prospective cohort study of 83 consecutive patients with lumbar spinal stenosis without spondylolisthesis treated with decompression alone.²⁴ Patients had preoperative dynamic radiographs to rule out instability. The decompression entailed a spinous process osteotomy along with bilateral laminotomy, flavectomy, partial medial inferior facetectomy, and superior facetectomy to the pedicle. The mean follow-up was 36 months (range 19-48 months). Eight patients (10%) went on to subsequent fusion for either back pain, back and leg pain, or leg pain. This was a relatively low instability rate given the rather extensive decompression performed. Routine postoperative radiographs were not obtained to evaluate for instability, so it was not possible to determine if this was a cause for the additional surgery in any of the eight cases. As a result, it is difficult to compare their results with others.

A retrospective review by Pichelmann et al. of 222 patients undergoing total lumbar facetectomy and hemilaminectomy without fusion by a single surgeon was performed to assess stability and need for subsequent fusion.²⁵  All patients had three-month follow-up, and 187 (84.2%) had long term follow-up of more than a year. At the short-term 86.5% had either no pain or improved pain. Long-term follow-up revealed 71.1% with zero or minimal pain. Of those with continued pain, 13 underwent an additional surgical procedure including 12 fusions and 1 unknown procedure for either instability, salvage procedure, both, or unknown reason. These results showed a 6.95% rate of subsequent surgery for all causes. The authors argued that complete facetectomy could be performed safely with a relatively low risk of clinical instability necessitating revision surgery. These results agreed well with those of Epstein and Garrido providing additional evidence that even with complete facetectomy, fusion is not always required to prevent subsequent instability.^12,13 It is likely that the degree of disc space collapse, intrinsic stiffness and/or spondylosis, and level of patient activity all play a role in whether the addition of fusion would be required; these parameters were not analyzed in the cited studies.

Tye et al. performed a retrospective review of 364 worker’s compensation patients undergoing either decompression alone or decompression with fusion for lumbar spinal stenosis with a minimum 3-year follow-up.²⁶ There was a significantly greater return to work rate among patients with decompression alone versus those with fusion (35% versus 25%). Of the patients in the decompression group, 18 (8.1%) went on to have a subsequent fusion. The authors concluded that the addition of fusion had a significant negative impact over decompression alone in terms of return to work. Even though the decompression patients had better return to work rates compared with the fusion patients, neither group in this study had a high rate of returning to work. However, even in this notoriously difficult patient population the need for subsequent fusion following decompression alone was relatively low at three years.

Wang et al. performed a meta-analysis of studies to determine if the presence of degenerative spondylolisthesis affected outcomes for patients undergoing decompression-only surgery for degenerative lumbar stenosis.²⁷ Their analysis included 11 studies with 1,081 cases. They concluded that the presence of grade I-II degenerative spondylolisthesis did not significantly affect the outcome of decompression surgery without fusion for patients with degenerative stenosis. They argued that the presence of degenerative spondylolisthesis may not be an indication to perform fusion on patients being treated with decompression for degenerative stenosis.  

Performing a decompression adjacent to a fusion is thought by many to increase the risk of instability since not only is the decompression level losing some support, but it also going to be under increased stress due to the greater mechanical forces found adjacent to a fusion.²⁸ Matsumoto et al. performed a retrospective cohort study on 45 consecutive patients who had undergone a L3-L4 decompression and L4-L5 posterior lumbar interbody fusion (PLIF).²⁹ The study group was compared to a control group of 45 patients that underwent an L4-L5 PLIF without L3-L4 decompression. The decompression procedure involved removal of the supraspinous and interspinous ligaments, partial laminotomy, flavectomy, and bilateral medial facetectomies up to the pedicles. The recovery rates of the JOA score were 58% and 61% for the study and control groups, respectively. The reoperation rates were 2.2% and 6.7% for the study and control groups, respectfully. Neither of these differences were statistically significant, suggesting that decompression with partial facetectomy had no negative clinical effect or increased risk of reoperation for iatrogenic instability even when performed adjacent to a fusion. This study suggests that a laminectomy including bilateral partial facetectomies remains stable even when performed in the more stressful environment adjacent to a fusion.

Hiratsuka et al. reported on a retrospective series of 110 patients undergoing single-level lumbar decompression for stenosis with no or minimal spondylolisthesis.³⁰ The decompression procedure consisted of an open resection of the supraspinous and interspinous ligaments and medial facetectomy sufficient to unroof the lateral recess. A total of 6 (5.5%) patients developed symptomatic instability at the operative level during a mean 4-year follow-up period. A multivariate logistic regression analysis revealed chronic glucocorticoid use for autoimmune connective tissue disorders to be a significant risk factor for reoperation due to instability. No significant differences were identified in any radiographic parameters including disc angle, lumbar lordosis, pelvic incidence, or pelvic incidence-lumbar lordosis in the patient’s developing instability. It is believed that the loss of bone strength associated with the chronic glucocorticoids and the ligamentous laxity associated with the connective tissue disorders was responsible for the higher risk of instability.

Kuo et al. performed a multi-center retrospective cohort study comparing unilateral laminotomy for bilateral decompression versus traditional posterior decompression with instrumented fusion in 164 patients with low-grade degenerative spondylolisthesis and spinal stenosis.²² The technique for the unilateral approach involved performing a laminotomy with ipsilateral partial medial facetectomy, undercutting the ventral aspects of the spinous process and lamina and contralateral partial medial facetectomy, while sparing the midline ligamentous and bony structures. At a five-year follow-up the reoperation rates for the unilateral decompression group were 10.4%, and 17.2% for the fusion group. However, this study was limited by the fact that routine postoperative radiographs were not available to assess for any change in spondylolisthesis. Additionally, the study did not randomize patients into the two surgical groups so the chance for selection bias was high.

In summary, the preponderance of the clinical research on instability seems to indicate that an array of decompression-only procedures can safely be performed without excessive risk for iatrogenic instability. When possible, surgeons should strive to minimize disruption of critical stabilizing structures such as the facet joints, supraspinous ligament, laminae, and ligamentum flavum; however, in cases of neurologic compression it is often necessary to sacrifice these structures to achieve adequate neural decompression. The clinical evidence reviewed in this section indicates that these structures may indeed be sacrificed with relatively low risk for revision to fusion. The clinical literature on lumbar instability reviewed in this section is summarized in Table 3.

Radiographic Studies

Radiographic studies have been widely utilized to assess both preoperative and postoperative instability. The goal of much of this research has been to accurately identify patients that are more likely to develop postoperative symptomatic instability following a decompression procedure for lumbar stenosis. Without this knowledge, patients are likely to undergo more invasive and expensive procedures that might not be necessary, while others may have to endure a subsequent fusion operation due to instability following decompression alone. The various criteria under investigation include: the amount motion/translation on static and dynamic sagittal radiographs, level of degeneration of the disc and facet joints, sagittal plane alignment, facet joint orientation, presence of facet joint effusion, and findings from dynamic computer tomography (CT) and magnetic resonance imaging (MRI).  

Despite its common use, Penning and Blickman argued against the use of vertebral translation on dynamic radiographs as an accurate measure of lumbar instability.³¹ Their radiographic study evaluated dynamic flexion-extension radiographs of 24 patients with known lumbar spondylolisthesis. The axis of rotation was identified for each level and for each patient by superimposing the radiographs and labeling the four corners of each vertebra on the flexion and extension views then drawing lines connecting the respective points on the flexion and extension views. Extending perpendicular lines from each of the four lines intersect in the center of rotation for each vertebra as illustrated in Figure 3. The authors determined that in the normal spine there is a fixed pattern of movement about an axis of rotation, while in cases of instability there is a change in the axis of rotation. They found that hypermobility was a normal finding in the presence of spondylolisthesis and argued that the traditional concept of vertebral translation as representing instability was not valid. Instead, they believed that there was a larger than normal spread of the axes of rotation present with spondylolisthesis related to underlying disc degeneration.    

The majority of patients undergoing decompression for lumbar stenosis have a preoperative MRI unless there is a contraindication. The common use of advanced imaging studies has allowed investigators to assess the predictive value of these studies for instability. Rihn et al. performed a retrospective MRI study to assess the relationship of facet joint fluid with instability in patients with degenerative lumbar disease.³² Fifty-one patients were included in the study that had undergone either lumbar laminectomy alone or laminectomy with fusion at L4-L5. Preoperative dynamic radiographs were used to measure the percentage slip of L4 on L5. Preoperative MRIs were used to measure the facet fluid index (ratio of the sum of the width of fluid in each facet to the sum of the width of both facets). A positive linear association was identified between the facet fluid index and the percent anterior slip (p<0.001). The positive predictive value of facet fluid on MRI at the L4-L5 level as an indicator of instability on dynamic radiographs was 82%. The authors continued to recommend the use of both dynamic radiographs and MRI as they found 15% of patients with radiographic evidence of instability had no facet fluid on MRI, and 19% of patients had facet fluid on MRI without corresponding radiographic instability.  

Chaput et al. performed a retrospective imaging review of 193 patients with and without degenerative spondylolisthesis.³³ All patients had dynamic lumbar radiographs and MRI. Measurements included grade of L4-L5 facet osteoarthritis based on the Weishaupt system,34 amount of spondylolisthesis on radiographs and MRI based on the Taillard system,³⁵ facet effusion, presence of interspinous ligament intensity, and presence of synovial cysts. Subjects with degenerative spondylolisthesis were associated with increased age (p<0.0001), female gender (p=0.0042), presence of synovial facet cysts (p<0.0001), higher osteoarthritis grade (p<0.0001), and larger facet effusion (p<0.0001). A facet effusion greater than 1.5mm was highly predictive of degenerative spondylolisthesis on x-ray without spondylolisthesis on supine MRI. The authors recommended the use of dynamic radiographs to assess for spondylolisthesis in all cases with greater than 1.5 mm of facet joint effusion to further evaluate for the presence of instability.

Lattig et al. performed a retrospective imaging study on 160 patients who had undergone either lumbar decompression or decompression with instrumented fusion for degenerative spondylolisthesis and stenosis.³⁶ Preoperative dynamic radiographs and MRI scans were utilized to measure percentage of slip, value of facet joint effusion, facet angle, degree of facet joint degeneration³⁴, degree of central spinal canal narrowing, disc height, presence of facet cysts, and presence of rotational translation on AP radiograph. The amount of facet joint effusion was significantly correlated with the amount of slip between standing and supine images (p<0.001). The amount of variation in facet joint effusion from side to side was correlated with the presence of rotational translation (p<0.0001). This study added to the recommended use of facet effusion but was the only study that also showed that asymmetric facet effusions signified the presence of rotational instability.

Blumenthal et al. performed a radiographic analysis of 40 patients with degenerative grade I spondylolisthesis that were prospectively enrolled for lumbar laminectomy without fusion.³⁷ Patients were eliminated from the study if they had mechanical back pain or dynamic instability > 3 mm on preoperative dynamic radiographs. Of the 40 patients, 15 (37.5%) went on to an additional fusion surgery for pain attributed to instability at the index level. Risk factors for subsequent fusion surgery included facet angle >50 degrees (39%), disc height >6.5 mm (45%) and spondylolisthesis motion >1.25 mm (54%). The presence of all three risk factors carried a 75% risk of revision surgery versus 0% with none of the risk factors (p=0.14). The limitations of this study include a limited patient and no standardization based on level of surgery. As discussed below, there can be significant variability in the mechanics of the varying lumbar levels. Additionally, a 37.5% risk for iatrogenic instability going on to subsequent fusion is significantly greater than typically reported in the literature.  

Lao et al. performed a retrospective imaging analysis of 162 patients with symptomatic back pain using dynamic MRI.38 For grades I and II degeneration, the L2-L3 and L3-L4 levels produced the greatest angular mobility, while these levels decreased significantly in grade V. In agreement with Kirkaldy-Willis the authors found a progression: in the earlier stages of disc degeneration there was an increase in segmental motion coupled with reduced stability, followed by an ankylosed phase characterized by decreased motion and greater stability.^39,40

Hipp et al. performed a validation study to test the accuracy of the Quantitative Stability Index (QSI) as an objective radiographic parameter for identifying patients with sagittal plane lumbar instability.⁴¹ The QSI is calculated as the ratio of the amount of translation over the degree of rotation in the sagittal plane. Previous studies have reported in the normal disc that the relationship between translation and rotation are nearly linear once the rotation is beyond the neutral zone, or a minimum of 3 degrees.^7,42,43 The presence of a facet fluid sign was used as the “gold standard” indicator for instability.^{32,33,36,44-46} Images from 66 patients were studied by an independent musculoskeletal radiologist. The mean QSI was significantly greater at levels with a definite fluid sign in the facet joints on MRI (2.3 versus 0.60, p=0.047). The authors concluded that the QSI was a reliable test for sagittal plane lumbar instability. However, the authors also discussed that the amount of motion generated by patients during dynamic radiographs can be restricted by pain, prior fusion, and fear of further injury as discussed in previous reports.⁴⁷ This potential for self-limited motion by patients could have a significant effect on measurements of instability leading to false negative results. Previous studies have shown that the degree of flexion and extension at each level needs to be great enough to stress the restraints to motion.^48,49 This could account for the lack of predictability of dynamic radiographs in the studies by Donnarumma and Penning.^19,31 The analogy the authors used was that testing for disruption of the anterior cruciate ligament requires sufficient applied stress to activate the normal restraints of the ligament. The ratio of translation to rotation used in the QSI has been previously recommended as a potential measurement tool to correct for the effects of patient effort in dynamic motion.^43,50,51

Dombrowski et al. performed a clinical radiographic study on seven patients with degenerative spondylolisthesis matched to a control group of asymptomatic subjects.⁵² All subjects underwent dynamic biplanar flexion-extension radiographs in an upright position without knee bending to as far as comfortably possible. Images were obtained at 20 frames per seconds for a 4-8 second trial. Surface markers were tracked with a motion analysis system simultaneously. CT scans were obtained to create three-dimensional bone models. In vivo flexion-extension and translation motions were calculated and compared to values from the standard dynamic flexion-extension radiographs. The calculations from the standard dynamic radiographs were found to underestimate the amount of translation as compared to the dynamic in vivo imaging system (1.0 mm versus 3.1 mm, p=0.03). Additionally, 3 of 7 (42%) of patients had aberrant mid-range motion identified on the continuous dynamic in vivo motion. The authors suggested that there is aberrant motion with kinematic heterogeneity that is not evident on traditional dynamic imaging evaluations. This topic is more thoroughly investigated in the biomechanics section below. Limitations of this study include the small sample and potential for effort-based limitations in motion of patients due to pain.

Kitanaka et al. performed a retrospective imaging study on 50 patients with L4-L5 stenosis with or without degenerative spondylolisthesis.⁵³ All patients had dynamic CT measurements of disc height, foraminal height, distance between craniocaudal edges of the facet joint, and interspinous distance. MRI was used to grade disc degeneration and facet joint osteoarthritis. No significant differences were identified for any degenerative findings of the disc between those with and without spondylolisthesis. Increased degrees of facet joint osteoarthritis were found in patients with spondylolisthesis. Additionally, there were significantly increased distances between the craniocaudal edges of the facet joint in those with spondylolisthesis. The authors believed that the ligamentous laxity associated with facet joint osteoarthritis significantly affected the amount of spinal motion in vivo.

Sabnis et al. performed a retrospective study of 72 consecutive patients with chronic back pain to determine the effects of disc and facet degeneration on segmental kinematics at each level.⁵⁴ Degree of disc degeneration and the presence of facet joint arthritis were obtained from MRI images. Upright neutral and dynamic flexion-extension radiographs were obtained, and segmental kinematics measured from L3-S1. At the L5-S1 level the presence of disc degeneration and facet arthritis occurred independently from each other (p=0.188). In contrast, at L3-L4 (p<0.05) and L4-L5 (p<0.001) there was an association identified. Without facet degeneration at L5-S1 there was no change in kinematics associated with increased degrees of disc degeneration, but with facet degeneration at L5-S1 increased levels of disc degeneration led to increased stiffness at L5-S1. The author felt the changes found at L5-S1 as compared to L3-L5 could be related to the presence of the iliolumbar and lumbosacral ligaments, coronal orientation of the facet joints and the wedge shape of the disc. This study highlights the differences in mechanical properties of the L5-S1 functional spinal unit and suggests that previous reports on spinal kinematics that do not separate results based on specific level could have limitations in their findings. This is also true when evaluating the results of biomechanical studies using animal models with different numbers of vertebrae, different loading patterns, and sagittal plane alignment.

The body of work regarding radiographic predictors of clinical instability points to several important signs that surgeons should be aware of: excess facet fluid on MRI, facet arthritis, posterior disc space wedging/gapping with forward flexion, and the listhesis/disc space angle index. Importantly, there are significant limitations to the utility of flexion/extension x-rays in the standing position including: pain guarding, body habitus, patient effort, core muscle strength, and pelvic mechanics. Several recent studies have indicated that lateral flexion x-rays in the seated position may in fact be superior stress views of the lumbar spine compared with standing flexion, as sitting obligates the pelvis to retrovert and place the lumbar spine into relative kyphosis.^55,56 Surgeons should be aware that while dynamic radiographs are the mainstay of diagnosis major instability, there are several other indicators of instability (facet fluid, facet arthrosis, disc wedging) that portend subtle, yet clinically meaningful, instability. The articles reviewed in this section can be found in Table 4.

Biomechanical Studies

In addition to the many clinical and radiologic studies that have investigated lumbar instability, several biomechanical studies have also sought to clarify the concept. Many of these studies involve testing in vitro spine specimens from either cadaveric lumbar spines or from various animal models. The strength of biomechanical testing is that investigators can simulate various injuries, levels of degeneration, and surgical procedures in a controlled setting and analyze their effect on the stability of the spine in vitro. Such biomechanical data may add greatly to our understanding of spinal stability; however, a significant limitation is that the exact responses in the in vivo environment cannot be precisely reproduced in a mechanical in vitro experiment. Additionally, cadaveric studies cannot reproduce the in-vivo effects that active muscle stabilization, proprioception, and pain may have on lumbar stability.

Seligman et al. studied 47 cadaveric spine specimens to determine the effect of disc degeneration on segmental motion.⁵⁷ The authors studied the location of the centroid of the instantaneous axis of rotation as a function of degree of disc degeneration. They found significantly increased length of the centroid with early stages of disc degeneration and an inferior migration of the centroid with more advanced degeneration. The addition of an axial load during testing did not significantly alter their results. They report that changes of the centroid could identify 94% of abnormal discs versus only 25% being detected by increased range of motion on dynamic radiographs. This in-vitro study seems to agree with the clinical and radiographic studies that call into question the accuracy of measuring translation on dynamic radiographs to identify instability.^19,31,41

McNally et al. performed an in vivo analysis of 15 patients, each with three or more levels of confirmed lumbar disc degeneration (31 levels total), to determine the relationship between internal disc mechanics and discogenic low back pain.⁵⁸ The distribution of stress within the intervertebral discs of these subjects was measured by inserting a miniature stress transducer through the disc and creating a graph of the internal stress versus position in the disc. Subjects then underwent provocative discography to determine the presence and severity of pain at each disc. Stress profiles were determined in the nucleus and in the posterolateral annulus for each level. There were stress concentrations found in each posterolateral annulus, but there was significant variation. The authors found the posterolateral annulus was broadened or contained multiple stress concentrations in 19 of 31 levels tested. Pain could be predicted by these factors with a sensitivity of 0.94 and specificity of 0.85.  The nucleus was depressurized in 11 of 31 levels and predicted pain with a sensitivity of 0.91 and specificity of 1.00.  The posterolateral annulus was 38% wider in the painful disc as compared to those that were painless (p<0.023).  The mean stress of the functional nucleus was 63% (p<0.017) and 64% (p<0.002) less in the painful discs in the vertical and horizontal planes, respectively.  The best predictor for the presence of pain was abnormal loading in the posterolateral annulus.  This was explained by the abnormal loading of the pain-sensitive nerve fibers of the annulus and endplate known to be in that location.⁵⁹ This study was unique in that it was able to correlate changes in the mechanical function of the disc in vivo with clinical symptoms of pain.  While the patterns of stress distribution were similar to reports of cadaveric studies, other findings such as the depressurized nucleus, stress concentrations in the nucleus, and multiple stress concentrations in the annulus were not commonly reported in cadaveric specimens. The findings of this study highlight the limitations of cadaveric studies in predicting actual clinical symptoms.  The authors also noted there was significant variability in stress distribution found in the discs even if they had similar radiographic or MR characteristics.  This confirmed that the mere presence of abnormal imaging findings may not accurately predict the presence or severity of clinical symptoms, which agrees with previous studies.^60,61

One common criticism of many biomechanical studies is failure to accurately reproduce the normal in vivo environment including the soft tissue envelope of muscle attachments to the spine.  In an attempt to correct for this, Quint et al. performed a biomechanical study on six fresh frozen human lumbar spine specimens (L2-S1).⁶² A kinematic analysis was performed on the intact specimens, followed by a decompression surgery and then application of pedicle screw fixation. The decompression consisted of laminectomy, and removal of the supraspinous and interspinous ligaments, ligamentum flavum and bilateral facet joints. Then the stabilizing effect of simulated deep intersegmental muscles was determined for each case. There were significant decreases in range of motion in both the intact specimens and following decompression during lateral bending and axial rotation after the simulated muscle stabilizing forces were applied. The authors concluded that the impairment caused by wide laminectomy was corrected by application of the muscle forces. The authors reasoned that during the degenerative process there is narrowing of the intervertebral disc and weakening of the normal pre-tensioning of the facet capsule and anterior and posterior longitudinal ligaments that allows for subluxation of the facet joints and subsequent spondylolisthesis. A wide decompression of the spinal canal leads to further laxity and potential instability. Application of the agonist and antagonist muscle forces helps to restore the stability to the functional spinal unit. A similar situation occurs during a transforaminal interbody fusion where a portion of the disc and facet joint are removed, causing a loss of the tensioning mechanism created by the longitudinal ligaments and annulus. Once the interbody device is placed, the ligaments and annulus are again under tension and restore much of the lost stability of the segment. This is analogous to the case of total knee arthroplasty. During that procedure, after the femoral and tibial cuts and bony resections are made, there is a loss of the normal tensioning mechanism of the knee joint. Once the arthroplasty device is inserted the normal tension and stability from the soft tissues is restored.  

Brown et al. performed a biomechanical study using 50 motion segments from 12 cadavers to determine the degree of motion segment stiffness as a function of the degree of disc degeneration.63 The stiffness, maximum displacement, and hysteresis of the motion segment stiffness curves were determined, and the authors found there was an inverse, nonlinear association between the stiffness and maximum displacement. Motion segment stiffness was found to decrease with initial stages of disc degeneration and then increase with more severe levels of degeneration. This finding was found to quantitatively substantiate the three phases of disc degeneration proposed by Kirkaldy-Willis.^39,40 Limitations of this study included the lack of simulated muscle forces and the use of single motion segment specimens from various levels. As shown in the previous study by Quint et al. mechanical testing results were significantly different after accounting for the muscle forces.62 It is also likely that different motion segments of the lumbar spine could behave differently due to the differences in normal anatomic features including facet orientation and amount of disc height and lordosis.⁵⁴

Gay et al. performed a biomechanical study using 15 human lumbar motion segments to determine the effects of loading rate and disc degeneration on dynamic motion parameters.⁶⁴ They found that both the rate of loading as well as the degree of disc degeneration affected motion parameters. Specifically, as loading rate was increased, the range of motion, area and width of the hysteresis loop increased while the width of the transition zone decreased. They also found lower levels of stiffness as disc degeneration increased from grade 1 to 3 and increased stiffness with grade 4 degeneration. The slope of the transition zone was found to be a useful measure of neutral zone stiffness during dynamic motion testing and was best able to differentiate between normal and degenerative discs. The authors discussed that many different parameters have been used to characterize clinical instability, but there remains a disconnect between laboratory measurements of segmental motion and clinical application in vivo. The neutral zone is a commonly used parameter to determine the amount of laxity around the neutral position of spinal segments where spinal motion is produced with minimal internal resistence.⁹ Continuous loading methods provide a better representation of physiologic motion and have been described as a dynamic equivalent of the neutral zone. The width of the hysteresis loop was reported to be a measure of neutral laxity by Wilke et al.⁶⁵  A comparison of the range of motion and hysteresis loop between the quasi-static versus continuous loading performed by Goertzen et al. determined both the range of motion and the width of the hysteresis loop to be less in the continuous motion group.⁶⁶  The lax zone has been defined as the zone that extends beyond the neutral zone where there is minimal ligamentous resistance to motion and was found to be distinct from the neutral zone and less variable.⁶⁷  In an earlier study by Gay et al. the slope of the transition zone between the neutral zone and the elastic zone had the strongest correlation with the neutral zone and best identified variations in the amount of disc degeneration.⁶⁴  

Tai et al. performed a biomechanical study using 8 porcine lumbar spine specimens to compare spinal stability between a laminectomy versus bilateral laminotomy.⁶⁸ The bilateral laminotomy entailed removing the inferior margin of L4 and superior margin of L5 lamina, and the ligamentum flavum was undercut. In the laminectomy model the complete lamina were removed with the spinous processes, ligamentum flavum and supraspinous ligaments. The facet joints were maintained in all testing cases. No significant differences were found in any models during extension testing. During flexion, there were no differences between the intact and the bilateral laminotomy models, but there was a significantly greater motion found in the laminectomy model compared to both the intact and bilateral laminotomy models. The authors concluded that posterior ligamentous complex integrity was a significant factor in maintaining lumbar spinal stability during flexion in agreement with previous clinical studies.¹⁶

Kettler et al. utilized a large database of 203 previously tested lumbar spine specimens to study the effects of disc degeneration on lumbar spine stability.⁶⁹ The radiographic level of disc degeneration grade from 0 (no degeneration) to 3 (severe degeneration), and its influence on range of motion and neutral zone was determined. They found a decrease in range of motion during flexion-extension and lateral bending when going from grade 0 to grade 3 degeneration (p<0.05). During axial rotation there was a trend for range of motion to increase from grade 0 to grade 3 (p>0.05). Results for neutral zone testing had a similar trend but did not reach statistical significance. These results dispute the proposals of Kirkaldy-Willis⁴⁰, and instead of increased motion during initial stages of degeneration, these authors found a continuous decrease in motion. The authors discuss the disparity in results among different authors on this topic and argue that results could be affected by the pooling of all lumbar segments together, and the use of different grading systems being used for determining the degree of disc degeneration.

    Bisschop et al. performed a biomechanical study on 10 thoracolumbar (T12 – L5) human cadaveric specimens to determine factors associated with instability.⁷⁰ All specimens had radiographs and MRI to grade levels of degeneration of the disc and facet joints.  Dual x-ray absorptiometry scans determined bone mineral content and density for each specimen. Biomechanical testing was then performed on intact specimens and compared to specimens following facet joint-sparing laminectomy. Specimens were tested with an applied axial load and a shear force. Measurements were obtained for shear stiffness, shear yield force and shear force to failure. Significant decreases in strength and shear stiffness were predicted by bone mineral content, bone mineral density, degeneration, and disc geometry. The authors concluded that lower levels of bone mineral density and content, smaller discs and a lack of osteophytes could predict instability following laminectomy in agreement with the clinical study by Hirasuka et al.³⁰  In a later study, the same group found that intervertebral width, frontal area, and facet joint tropism were strong predictors of early and late torsional stiffness and that torsion moment to failure following laminectomy could be used to predict postoperative rotational instability.⁷⁰  

Hasegawa et al. performed a biomechanical study using 30 porcine functional spinal units to determine the effects of different decompressive procedures on spinal stability.⁴⁶ Specimens were divided into groups of six and tested intact and following interlaminar decompression, bilateral medial facetectomy, left unilateral total facetectomy, and bilateral total facetectomy. Results showed maintained stability following the interlaminar decompression and bilateral medial facetectomy as compared to either left unilateral total facetectomy or bilateral total facetectomy. Limitations of this study included a small sample in each test group, use of a single functional spinal unit for testing, and use of an animal model without a comparable alignment and loading to the human lumbar spine.

Sengupta and Fan performed a biomechanical study using 39 human cadaveric lumbar motion segments to determine the effects of disc degeneration grade on mechanical instability.⁸  Prior to testing the specimens underwent radiographic, MRI and bone density studies to grade degeneration based on the Pfirrmann classification and to rule out significant osteoporosis.⁷¹ There were small trends toward increased range of motion with increased grades of degeneration, but they did not reach statistical significance. There was also a trend towards increased neutral zone motion with disc degeneration increasing from grade I to grade III, and a decreasing trend from grade III to grade IV in all motions. The neutral zone changes were only significant during flexion-extension between grade I and grade III and between grade I and grade IV. Intradiscal pressures were measured during testing, and they found changes with degeneration. The pressure profile became more irregular and had lower pressure in the nucleus and greater pressure spikes in the annulus as the grade of disc degeneration increased similar to findings of McNally et al.⁵⁸ Evaluation of the center of rotation and the instantaneous axis of rotation showed a smooth transition during flexion-extension with grade I specimens, but an increasingly irregular translation with increased grades of degeneration. The center of rotation dispersion was significantly increased in the vertical direction in degenerative discs as compared to normal discs. While this study showed a trend toward increasing range of motion with degeneration and then a decreased range of motion with grade IV degeneration, there were more significant changes in the neutral zone as a function of degeneration. The authors did discuss that while mechanical instability could be identified in their study it might not suggest clinical instability in all the specimens. Additionally, it was noted that mechanical instability did not always correlate with an overall increase in range of motion. Instead, they argued that instead of using simple range of motion measurements to define instability, the quality of motion was better defined by the center of rotation as previous reported by Seligman et al.⁵⁷  While this study does effectively identify that mechanical instability can be identified using variation in the center of rotation even in the absence of changes in overall range of motion, it does not definitively link mechanical instability to clinical instability or clinical symptoms of back pain.

Fry et al. performed a biomechanical study using eight human cadaveric lumbar spine specimens (L1-sacrum) to determine the effects of a compressive follower preload on a destabilized lumbar spine.⁷² Specimens were evaluated with radiographs and MRI to determine the grade of disc degeneration prior to testing. Specimens were tested intact and following an L4-L5 nucleotomy, transection of the supraspinous and interspinous ligaments at L4-L5, and after bilateral laminotomy with foraminotomy and partial medial facetectomy at L4-L5. Preloads were tested at 0 N, 200 N and 400 N as previously described by Patwardhan et al.⁷³ They found a progressive increase in the range of motion and high flexibility zone with increasing levels of destabilization. The addition of the compressive preload did not have a significant change on the range of motion; however, it did significantly increase the segmental stiffness in the high flexibility zone. Even with the greatest level of applied preload to the most destabilized specimen, the segmental stiffness did not return to the level of the intact spine (p=0.01), suggesting the increasing core strength can help restore stability, but alone it might not be sufficient to restore segmental stability to normal levels after wide decompression surgery. The authors showed that the application of the preload which simulates the activation of the paraspinous musculature did not significantly change the overall range of motion but did have a significant effect on the segmental stiffness of the high flexibility zone where many activities of daily living occur. It is believed that the compression from the preload (and thus the paraspinous musculature) increases the stiffness in the disc and engagement of the facet joints. Increasing the pressure in the disc leads to increased tension in the annulus and improved segmental stiffness and stability in agreement with Quint et al.⁶²  

Chamoli et al. performed an in vitro biomechanical study to investigate the global and segmental kinematic changes in the lumbar spine following transection of the interspinous and supraspinous ligament and following total bilateral facetectomy at L4-L5.74 Six fresh frozen kangaroo specimens were utilized. Transection of the supraspinous and interspinous ligaments did not lead to any significant change in global range of motion in any plane. This finding disputes the important role of the posterior ligamentous complex in maintaining stability reported by previous investigators.^16,68  The total bilateral facetectomy resulted in significant torsional instability at both the global and segmental levels with associated decreased motion at the caudal level. Limitations of this study included the unconventional animal model and the use of pure moment testing. A follower load was not utilized to represent compressive and shear loading found in the in vivo case.  

Melnyk et al. conducted a biomechanical study using 30 human cadaveric functional spinal units (L3-L4 and L4-L5) to determine the effects of disc degeneration under shear loading.⁷⁵ Each of the specimens underwent pretesting MRI to grade the level of disc degeneration according to the Pfirrmann classification.⁷¹ The testing conditions included intact, facet destabilization and disc destabilization models. The facet joints were destabilized using a burr to create a 4 mm gap in the facet joint. The disc was destabilized by creating a window in the posterior annulus and performing a nucleotomy. During testing an axial force was applied to simulate in vivo loading. Shear forces were applied to simulate the upper limits of expected in vivo shear forces. The results showed no significant effect of disc degeneration on anterior shear translation during any of the testing conditions. Additionally, the authors found some of their specimens had increased motion following disc destabilization, but others had decreased motion. In the specimens that had increased motion the disc height was maintained after axial loading was applied, while in those with decreased motion the disc space collapsed with axial loading. The authors felt that in the cases of increased motion, the annulus maintained sufficient integrity and thickness to resist the axial forces. As discussed above, there is variability in results regarding the effects of degeneration on spinal motion and associated degrees of suspected instability. The authors believed this variability is related to a combination of issues including variability in grading schemes for degeneration, limited number and variability in tested specimens, lack of direct relationship between change in cadaveric motion associated with degeneration to clinical instability associated with degeneration, and inability to predict clinically significant differences in motion among varying grades of degeneration.

Although limited, the biomechanical data on lumbar instability provides general insights into spine instability and the lumbar degeneration process. A summary of the biomechanical literature reviewed in this section can be found in Table 5.

Finite Element Studies

The final category of studies included in this review utilize a mathematical modeling technique called finite element analysis (Table 6).  In this technique data from either CT or MRI scans are used to create a mathematical model of the spine. The more advanced models take into consideration the actual mechanical properties and anatomic configuration of each of the components of the spine including the annulus, nucleus, ligament, facet joints, cartilage, and bone. A major benefit of this technique is that multiple changes can be made to the components of the model to represent changes in degeneration, surgical procedures, loading patterns, etc.  By allowing for each of the changes to be made to the model, it can provide more consistent results than possible when using cadaveric or animal specimens where there can be great variability among tested specimens. However, as discussed above, McNally et al. found limitations in the ability of MRI to predict significant variability in stress distribution found in the discs.⁵⁸ Limitations in the data used to create the finite element models might limit their accuracy in predicting actual clinical results.

Zander et al. constructed a nonlinear, three-dimensional finite element model of the lumbar spine (L2-S1) to determine the effect of sequential degrees of posterior decompression on mechanical properties of the spine.⁷⁶ They found no significant effect of either facetectomy or laminectomy during extension or lateral bending. Significant increases in flexion occurred after bilateral laminectomy and facetectomy. Significant increases in axial rotation occurred following hemifacetectomy in addition to increased stresses in the annulus and increased intradiscal pressures. No further significant effects were identified in axial rotation following further decompression beyond hemifacetectomy. Decompression with bilateral facetectomy even at two levels had no significant effect on the motion, stresses, strains, intradiscal pressures or facet forces at the adjacent levels. Also, disc degeneration resulted in decreased motions and increased maximum von Mises stresses suggesting that the lack of the normal compressive strength of the nucleus pulposus led to increased stresses in the annulus fibrosis in agreement with Quint et al.⁶² The authors concluded that patients undergoing laminectomy and facetectomy should be cautioned to avoid excessive axial rotation motions.

Lee et al. utilized a three-dimensional finite element model of the L2-L3 functional spinal unit to study the effect of graded facetectomy on stability during sagittal plane loading.⁷⁷ The model simulated unilateral and bilateral facetectomies performed with grades of 25%, 50%, 75% and 100%. A compressive preload of 400 N was applied to the model. During flexion, there were no significant changes found with unilateral or bilateral facetectomy as long as the posterior ligaments remained intact, in agreement with the biomechanical study by Pintar et al.⁷⁸ When the posterior ligaments were removed, the current model predicted a 43% increase in flexibility compared to the intact case, but the addition of facetectomy produced no further significant change in stiffness during flexion. This suggested that the role of the posterior ligamentous complex actually provides a much more crucial role for segmental flexion stability than the facet joints. Total bilateral facetectomy resulted in 30% increased rotational flexibility compared to the intact case. During extension, unilateral complete facetectomy and contralateral resection of the facet resulted in marked changes in the rotational flexibility and coupled motions. These results were similar to those of Natarajan et al. who reported resection of greater than 75% of the facet joint led to substantial change in rotational motion.⁷⁹  Limitations of this study include the use of a single spinal unit and failure to account for the stabilizing role of muscles.

Homminga J et al. constructed a three-dimensional, nonlinear finite element model of the L2-L3 functional spinal unit.⁸⁰ The model simulated varying grades of disc degeneration and were loaded in pure torsion with and without an applied shear stress in the anterior and posterior direction. The models predicted that with disc degeneration the annular fibers and the spinal ligament became more slack due to a loss of disc height. This led to a decreased torsional stiffness. With further levels of simulated disc degeneration that modeled loss of disc height with fibrosis and disorganization of the annular fibers, the torsional stiffness was restored. The authors argued that the initial loss of torsional stiffness seen with early disc degeneration is related to the slack in the annulus and ligament as opposed to either dehydration or fibrosis of the nucleus. Limitations of the study include using a single motion segment, simplified modeling of the facet joint as two parallel flat surfaces and lack of applied axial loading.

Du et al. created a three-dimensional, nonlinear finite element model of the lumbar spine (L1-S1) to investigate the response of the facet joints to different values of follower preload during testing.⁸¹ A major benefit of this study was the utilization of a full lumbar spine model (L1-S1), the ability to account for follower preload to better simulate the in vivo active muscle forces, and improved facet joint modeling. This model created complex facet joint geometry to account for the normal curved surface and inhomogeneous thickness.⁸² A graded follower preload from 0 N to 1200 N was applied to determine the effects on facet joint contact force, contact area, mean contact pressure, and pressure distribution. The applied preload amplified the facet forces, contact area, and contact pressure during flexion and extension; however, the effects weakened with increasing levels of preload.

Zeng et al. constructed a three-dimensional, nonlinear finite element model of L3-L5 to investigate the effects of different grades of facetectomy on intervertebral rotation, intradiscal pressure, facet joint forces, and maximum von Mises equivalent stresses in the annulus.⁸³ Total unilateral and bilateral facetectomy had minimal effect on intervertebral rotation, facet forces or intradiscal pressure during flexion and lateral bending. Unilateral and bilateral facetectomy increased extension motion by 11.7% and 40.5%, respectively. Following complete unilateral facetectomy, the contralateral facet joint force was increased by 108.1% during extension while also increasing the intradiscal pressure and maximum von Mises stresses in the annuli. Axial rotation motion was also significantly affected with an increase of 354.3% and 265.3% in right and left axial rotation following bilateral facetectomy, respectively. Limitations of this study included not accounting for muscle forces or preload and modeling a limited portion of the lumbar spine (L3-L5). The authors concluded that following complete unilateral or bilateral facetectomy, axial rotation and extension movement could be unstable and in need of further fixation. There were, however, no significant changes with less than 50% facet removal compared to the intact case. Thus, removal of up to 50% of each facet joint was felt to be safe without the risk of developing instability.    

Conclusions

There have been many important advances in our understanding of lumbar spine instability since the 1940’s resulting from advancements in imaging modalities, mechanical testing equipment, and computer simulation technology; however, there is still much to learn about lumbar instability and how it evolves. This review is intended to present key literature in the clinical, radiographic, and biomechanical fields, and to specifically focus on what is known about how degenerative changes and surgical interventions may affect lumbar instability.

In terms of the effects of degenerative changes, the majority of studies agree that the presence of a facet effusion is a good predictor of increased motion on dynamic radiographs as well as lumbar instability. However, there is no universal agreement on the effects of other measures of facet degeneration on lumbar stability. There is agreement in most studies that the orientation of the facet joints and facet angles do play a role in lumbar range of motion. This is important in that the different lumbar levels are known to have different anatomical orientations that can affect range of motion in both normal and degenerative cases. One common limitation in many of the reported studies is the grouping of multiple different lumbar levels together in studies. The normal variants in the anatomy are sufficient to influence study results. Future studies should account for these differences.

The effect of disc degeneration also varies among studies. There is some agreement, but not universal, that with increasing amounts of disc degeneration there is an associated initial increase in lumbar range of motion and instability, followed by a decrease in motion and improved stability with more advanced stages of disc degeneration. The increase in motion is believed to be the result of loss of disc height, and less tension in the ligaments and annular fibers and thus less restraint to motion.  More advanced disc degeneration leads to osteophyte formation, potential ankylosis, and decreased motion.

In terms of iatrogenic postsurgical changes, there is reasonable agreement that lumbar laminectomy and at least partial facetectomy can be safely performed without increased risk for instability. In the majority of studies, the incidence of patients going on for subsequent fusion was less than 10%. Importantly, in many of these studies it was not specified if the subsequent fusion was necessary due to instability or for other pain complaints related to degeneration. The presence of preoperative grade I to II spondylolisthesis does not appear to be a strong predictor of post laminectomy instability, while an intact posterior ligamentous complex does appear to be more important for postsurgical instability. Additionally, the effects of the muscle forces on the lumbar spine have been shown in both clinical and biomechanical studies to be an important factor in stability. Advancements in less invasive surgical techniques that spare the posterior ligaments and minimize muscle destruction may thus be beneficial in maintaining stability, even when combined with partial or complete facetectomy.

Table 7 summarizes some of the key concepts discussed in this review both in terms of predictors of instability and limitations in prior research studies, respectively. The concept of instability is vitally important when caring for patients with spinal disorders. The ability to accurately predict which preoperative imaging findings and which surgical procedures place the patient at risk for instability is necessary to properly plan for the most successful and least aggressive or invasive surgical treatments.  It is also crucial to understand that the presence of instability based on imaging studies does not directly correlate with clinical manifestations.  Despite 75 years of research on this topic, there are still many disagreements and unanswered questions. It is hoped that identification of the shortcomings of prior studies and further improvements in technology will allow for many of these questions to be more definitively answered.

Figure 8 | Figure 9 | Figure 10

Acknowledgments

This study did not receive funding from any institution or grant.

References

1. Knutsson F. The instability associated with disk degeneration in the lumbar spine. Acta Radiologica. 1944;25(5-6):593-609.
2. Meschan IR. Spondylolisthesis-a commentary on etiology, and an improved method of roentgenographic mensuration and detection of instability. Am J Roentgenol. 1945;53(3):230-243.
3. Meschan I. A radiographic study of spondylolisthesis with special reference to stability determination. Radiology. 1946;47:249-262.
4. Barr J. Editorial back pain. J Bone Joint Surg Br. 1950;32:461-569.
5. Harris RI, Macnab I. Structural changes in the lumbar intervertebral discs; their relationship to low back pain and sciatica. J Bone Joint Surg Br. 1954;36-b(2):304-322.
6. White A. The problems of clinical instability in the human spine: a systematic approach. Clin Biomech Spine. 1990;278.
7. Panjabi MM. The stabilizing system of the spine. Part II. Neutral zone and instability hypothesis. J Spinal Disord. 1992;5(4):390-396; discussion 397.
8. Sengupta DK, Fan H. The basis of mechanical instability in degenerative disc disease: a cadaveric study of abnormal motion versus load distribution. Spine (Phila Pa 1976). 2014;39(13):1032-1043.
9. Panjabi MM. Biomechanical evaluation of spinal fixation devices: I. A conceptual framework. Spine (Phila Pa 1976). 1988;13(10):1129-1134.
10. Resnick DK, Choudhri TF, Dailey AT, et al. Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine. Part 10: fusion following decompression in patients with stenosis without spondylolisthesis. J Neurosurg Spine. 2005;2(6):686-691.
11. Lee CK. Lumbar spinal instability (olisthesis) after extensive posterior spinal decompression. Spine (Phila Pa 1976). 1983;8(4):429-433.
12. Garrido E, Connaughton PN. Unilateral facetectomy approach for lateral lumbar disc herniation. J Neurosurg. 1991;74(5):754-756.
13. Epstein NE. Evaluation of varied surgical approaches used in the management of 170 far-lateral lumbar disc herniations: indications and results. J Neurosurg. 1995;83(4):648-656.
14. Ofluoglu AE, Karasu A, Ekinci B, Toplamaoglu H. The effect of laminectomy on instability in the management of degenerative lumbar stenosis surgery: a retrospective radiographic assessment. Turk Neurosurg. 2007;17(3):178-182.
15. Sasai K, Umeda M, Maruyama T, Wakabayashi E, Iida H. Microsurgical bilateral decompression via a unilateral approach for lumbar spinal canal stenosis including degenerative spondylolisthesis. J Neurosurg Spine. 2008;9(6):554-559.
16. Alimi M, Hofstetter CP, Pyo SY, Paulo D, Hartl R. Minimally invasive laminectomy for lumbar spinal stenosis in patients with and without preoperative spondylolisthesis: clinical outcome and reoperation rates. J Neurosurg Spine. 2015;22(4):339-352.
17. Guha D, Heary RF, Shamji MF. Iatrogenic spondylolisthesis following laminectomy for degenerative lumbar stenosis: systematic review and current concepts. Neurosurg Focus. 2015;39(4):E9.
18. Sato S, Yagi M, Machida M, et al. Reoperation rate and risk factors of elective spinal surgery for degenerative spondylolisthesis: minimum 5-year follow-up. Spine J. 2015;15(7):1536-1544.
19. Donnarumma P, Tarantino R, Nigro L, et al. Decompression versus decompression and fusion for degenerative lumbar stenosis: analysis of the factors influencing the outcome of back pain and disability. J Spine Surg. 2016;2(1):52-58.
20. Dupuis PR, Yong-Hing K, Cassidy JD, Kirkaldy-Willis WH. Radiologic diagnosis of degenerative lumbar spinal instability. Spine (Phila Pa 1976). 1985;10(3):262-276.
21. Ghogawala Z, Dziura J, Butler WE, et al. Laminectomy plus fusion versus laminectomy alone for lumbar spondylolisthesis. N Engl J Med. 2016;374(15):1424-1434.
22. Kuo CC, Merchant M, Kardile MP, Yacob A, Majid K, Bains RS. In degenerative spondylolisthesis, unilateral laminotomy for bilateral decompression leads to less reoperations at 5 years when compared to posterior decompression with instrumented fusion: a propensity-matched retrospective analysis. Spine (Phila Pa 1976). 2019;44(21):1530-1537.
23. Forsth P, Olafsson G, Carlsson T, et al. A randomized, controlled trial of fusion surgery for lumbar spinal stenosis. N Engl J Med. 2016;374(15):1413-1423.
24. Ahmad S, Hamad A, Bhalla A, Turner S, Balain B, Jaffray D. The outcome of decompression alone for lumbar spinal stenosis with degenerative spondylolisthesis. Eur Spine J. 2017;26(2):414-419.
25. Pichelmann MA, Atkinson JLD, Fode-Thomas NC, Yaszemski MJ. Total lumbar facetectomy without fusion: short and long term follow-up in a single surgeon series. Br J Neurosurg. 2017;31(5):531-537.
26. Tye EY, Anderson JT, Haas AR, et al. Decompression versus decompression and fusion for degenerative lumbar stenosis in a workers’ compensation setting. Spine (Phila Pa 1976). 2017;42(13):1017-1023.
27. Wang M, Luo XJ, Ye YJ, Zhang Z. Does concomitant degenerative spondylolisthesis influence the outcome of decompression alone in degenerative lumbar spinal stenosis? A meta-analysis of comparative studies. World Neurosurg. 2019;123:226-238.
28. Patwardhan AG, Khayatzadeh S, Faundez AA, et al. Effect of L4-Sacrum fusion alignment on biomechanics of the proximal lumbar segments in sitting postures. Spine J. 2017;17(10):S118-S119.
29. Matsumoto T, Okuda S, Nagamoto Y, Sugiura T, Takahashi Y, Iwasaki M. Effects of concomitant decompression adjacent to a posterior lumbar interbody fusion segment on clinical and radiologic outcomes: comparative analysis 5 years after surgery. Global Spine J. 2019;9(5):505-511.
30. Hiratsuka S, Takahata M, Hojo Y, et al. Increased risk of symptomatic progression of instability following decompression for lumbar canal stenosis in patients receiving chronic glucocorticoids therapy. J Orthop Sci. 2019;24(1):14-18.
31. Penning L, Blickman J. Instability in lumbar spondylolisthesis: a radiologic study of several concepts. Am J Roentgenol. 1980;134(2):293-301.
32. Rihn JA, Lee JY, Khan M, et al. Does lumbar facet fluid detected on magnetic resonance imaging correlate with radiographic instability in patients with degenerative lumbar disease? Spine (Phila Pa 1976). 2007;32(14):1555-1560.
33. Chaput C, Padon D, Rush J, Lenehan E, Rahm M. The significance of increased fluid signal on magnetic resonance imaging in lumbar facets in relationship to degenerative spondylolisthesis. Spine (Phila Pa 1976). 2007;32(17):1883-1887.
34. Weishaupt D, Zanetti M, Hodler J, Boos N. MR imaging of the lumbar spine: prevalence of intervertebral disk extrusion and sequestration, nerve root compression, end plate abnormalities, and osteoarthritis of the facet joints in asymptomatic volunteers. Radiology. 1998;209(3):661-666.
35. Dubousset J. Treatment of spondylolysis and spondylolisthesis in children and adolescents. Clin Orthop. 1997;337:77-85.
36. Lattig F, Fekete TF, Grob D, Kleinstuck FS, Jeszenszky D, Mannion AF. Lumbar facet joint effusion in MRI: a sign of instability in degenerative spondylolisthesis? Eur Spine J. 2012;21(2):276-281.
37. Blumenthal C, Curran J, Benzel EC, et al. Radiographic predictors of delayed instability following decompression without fusion for degenerative grade I lumbar spondylolisthesis. J Neurosurg Spine. 2013;18(4):340-346.
38. Lao L, Daubs MD, Scott TP, et al. Effect of disc degeneration on lumbar segmental mobility analyzed by kinetic magnetic resonance imaging. Spine (Phila Pa 1976). 2015;40(5):316-322.
39. Kirkaldy-Willis W. Presidential symposium on instability of the lumbar spine: introduction. Spine. 1985;10(3):254.
40. Kirkaldy-Willis W, Farfan H. Instability of the lumbar spine. Clin Orthop. 1982;165:110-123.
41. Hipp JA, Guyer RD, Zigler JE, Ohnmeiss DD, Wharton ND. Development of a novel radiographic measure of lumbar instability and validation using the facet fluid sign. Int J Spine Surg. 2015;9:37.
42. Oxland TR, Panjabi MM. The onset and progression of spinal injury: a demonstration of neutral zone sensitivity. J Biomech. 1992;25(10):1165-1172.
43. Weiler PJ, King GJ, Gertzbein SD. Analysis of sagittal plane instability of the lumbar spine in vivo. Spine (Phila Pa 1976). 1990;15(12):1300-1306.
44. Caterini R, Mancini F, Bisicchia S, Maglione P, Farsetti P. The correlation between exaggerated fluid in lumbar facet joints and degenerative spondylolisthesis: prospective study of 52 patients. J Orthop Traumatol. 2011;12(2):87-91.
45. Oishi Y, Murase M, Hayashi Y, Ogawa T, Hamawaki J. Smaller facet effusion in association with restabilization at the time of operation in Japanese patients with lumbar degenerative spondylolisthesis. J Neurosurg Spine. 2010;12(1):88-95.
46. Hasegawa K, Kitahara K, Shimoda H, Hara T. Facet joint opening in lumbar degenerative diseases indicating segmental instability. J Neurosurg Spine. 2010;12(6):687-693.
47. Pope MH, Rosen JC, Wilder DG, Frymoyer JW. The relation between biomechanical and psychological factors in patients with low-back pain. Spine (Phila Pa 1976). 1980;5(2):173-178.
48. Panjabi M. Physical properties and functional biomechanics of the spine. Clin Biomech Spine. 1990:1-84.
49. Panjabi MM, Goel VK, Takata K. Physiologic strains in the lumbar spinal ligaments. An in vitro biomechanical study 1981 Volvo Award in biomechanics. Spine (Phila Pa 1976). 1982;7(3):192-203.
50. Stokes I, Frymoyer JW. Segmental motion and instability. Spine. 1987;12(7):688-691.
51. Frobin W, Brinckmann P, Leivseth G, Biggemann M, Reikeras O. Precision measurement of segmental motion from flexion-extension radiographs of the lumbar spine. Clin Biomech (Bristol, Avon). 1996;11(8):457-465.
52. Dombrowski ME, Rynearson B, LeVasseur C, et al. ISSLS Prize in bioengineering science 2018: dynamic imaging of degenerative spondylolisthesis reveals mid-range dynamic lumbar instability not evident on static clinical radiographs. Eur Spine J. 2018;27(4):752-762.
53. Kitanaka S, Takatori R, Arai Y, et al. Facet joint osteoarthritis affects spinal segmental motion in degenerative spondylolisthesis. Clin Spine Surg. 2018;31(8):E386-E390.
54. Sabnis AB, Chamoli U, Diwan AD. Is L5-S1 motion segment different from the rest? A radiographic kinematic assessment of 72 patients with chronic low back pain. Eur Spine J. 2018;27(5):1127-1135.
55. Hey HW, Lau ET, Lim JL, et al. Slump sitting X-ray of the lumbar spine is superior to the conventional flexion view in assessing lumbar spine instability. Spine J. 2017;17(3):360-368.
56. Hey HW, Wong CG, Lau ET, et al. Differences in erect sitting and natural sitting spinal alignment-insights into a new paradigm and implications in deformity correction. Spine J. 2017;17(2):183-189.
57. Seligman JV, Gertzbein SD, Tile M, Kapasouri A. Computer analysis of spinal segment motion in degenerative disc disease with and without axial loading. Spine (Phila Pa 1976). 1984;9(6):566-573.
58. McNally DS, Shackleford IM, Goodship AE, Mulholland RC. In vivo stress measurement can predict pain on discography. Spine. 1996;21(22):2580-2587.
59. Yoshizawa H, O’Brien JP, Smith WT, Trumper M. The neuropathology of intervertebral discs removed for low-back pain. J Pathol. 1980;132(2):95-104.
60. Boden SD, Wiesel SW. Lumbosacral segmental motion in normal individuals. Have we been measuring instability properly? Spine (Phila Pa 1976). 1990;15(6):571-576.
61. Borenstein DG, O’Mara JW, Jr., Boden SD, et al. The value of magnetic resonance imaging of the lumbar spine to predict low-back pain in asymptomatic subjects: a seven-year follow-up study. J Bone Joint Surg Am. 2001;83(9):1306-1311.
62. Quint U, Wilke HJ, Loer F, Claes L. Laminectomy and functional impairment of the lumbar spine: the importance of muscle forces in flexible and rigid instrumented stabilization–a biomechanical study in vitro. Eur Spine J. 1998;7(3):229-238.
63. Brown MD, Holmes DC, Heiner AD. Measurement of cadaver lumbar spine motion segment stiffness. Spine. 2002;27(9):918-922.
64. Gay RE, Ilharreborde B, Zhao K, Zhao C, An KN. Sagittal plane motion in the human lumbar spine: comparison of the in vitro quasistatic neutral zone and dynamic motion parameters. Clin Biomech (Bristol, Avon). 2006;21(9):914-919.
65. Wilke HJ, Wenger K, Claes L. Testing criteria for spinal implants: recommendations for the standardization of in vitro stability testing of spinal implants. Eur Spine J. 1998;7(2):148-154.
66. Goertzen DJ, Lane C, Oxland TR. Neutral zone and range of motion in the spine are greater with stepwise loading than with a continuous loading protocol. An in vitro porcine investigation. J Biomech. 2004;37(2):257-261.
67. Crawford NR, Peles JD, Dickman CA. The spinal lax zone and neutral zone: measurement techniques and parameter comparisons. J Spinal Disord. 1998;11(5):416-429.
68. Tai CL, Hsieh PH, Chen WP, Chen LH, Chen WJ, Lai PL. Biomechanical comparison of lumbar spine instability between laminectomy and bilateral laminotomy for spinal stenosis syndrome – an experimental study in porcine model. BMC Musculoskelet Disord. 2008; 9:84.
69. Kettler A, Rohlmann F, Ring C, Mack C, Wilke HJ. Do early stages of lumbar intervertebral disc degeneration really cause instability? Evaluation of an in vitro database. Eur Spine J. 2011;20(4):578-584.
70. Bisschop A, Kingma I, Bleys RL, et al. Which factors prognosticate rotational instability following lumbar laminectomy? Eur Spine J. 2013;22(12):2897-2903.
71. Pfirrmann CW, Metzdorf A, Zanetti M, Hodler J, Boos N. Magnetic resonance classification of lumbar intervertebral disc degeneration. Spine (Phila Pa 1976). 2001;26(17):1873-1878.
72. Fry RW, Alamin TF, Voronov LI, et al. Compressive preload reduces segmental flexion instability after progressive destabilization of the lumbar spine. Spine (Phila Pa 1976). 2014;39(2): E74-81.
73. Patwardhan AG, Havey RM, Meade KP, Lee B, Dunlap B. A follower load increases the load-carrying capacity of the lumbar spine in compression. Spine (Phila Pa 1976). 1999;24(10):1003-1009.
74. Chamoli U, Korkusuz MH, Sabnis AB, Manolescu AR, Tsafnat N, Diwan AD. Global and segmental kinematic changes following sequential resection of posterior osteoligamentous structures in the lumbar spine: An in vitro biomechanical investigation using pure moment testing protocols. Proc Inst Mech Eng H. 2015;229(11):812-821.
75. Melnyk AD, Kelly A, Chak JD, et al. The effect of disc degeneration on anterior shear translation in the lumbar spine. J Orthop Res. 2015;33(4):450-457.
76. Zander T, Rohlmann A, Klockner C, Bergmann G. Influence of graded facetectomy and laminectomy on spinal biomechanics. Eur Spine J. 2003;12(4):427-434.
77. Lee KK, Teo EC, Qiu TX, Yang K. Effect of facetectomy on lumbar spinal stability under sagittal plane loadings. Spine (Phila Pa 1976). 2004;29(15):1624-1631.
78. Pintar FA, Cusick JF, Yoganandan N, Reinartz J, Mahesh M. The biomechanics of lumbar facetectomy under compression-flexion. Spine (Phila Pa 1976). 1992;17(7):804-810.
79. Natarajan RN, Andersson GB, Patwardhan AG, Andriacchi TP. Study on effect of graded facetectomy on change in lumbar motion segment torsional flexibility using three-dimensional continuum contact representation for facet joints. J Biomech Eng. 1999;121(2):215-221.
80. Homminga J, Lehr AM, Meijer GJ, et al. Posteriorly directed shear loads and disc degeneration affect the torsional stiffness of spinal motion segments: a biomechanical modeling study. Spine (Phila Pa 1976). 2013;38(21):E1313-1319.
81. Du CF, Yang N, Guo JC, Huang YP, Zhang C. Biomechanical response of lumbar facet joints under follower preload: a finite element study. BMC Musculoskelet Disord. 2016;17:126.
82. Simon P, Espinoza Orias AA, Andersson GB, An HS, Inoue N. In vivo topographic analysis of lumbar facet joint space width distribution in healthy and symptomatic subjects. Spine (Phila Pa 1976). 2012;37(12):1058-1064.
83. Zeng ZL, Zhu R, Wu YC, et al. Effect of graded facetectomy on lumbar biomechanics. J Healthc Eng. 2017;2017:7981513.