Automated radiographic metrics for diagnosing lumbar spine instability: a cross-sectional observational study

Introduction

Disc degeneration is common in the lumbar spine and contributes to the overall burden of lumbar degenerative disease (1,2). A phase of instability may occur as lumbar intervertebral disc degeneration progresses, and instability may partially explain symptoms in some patients (3-5). Instability is among the factors included in the differential diagnosis of a symptomatic patient with stenosis and spondylolisthesis (6,7). Instability is a common indication for fusion surgery in these patients (8-10). Despite the frequent use of flexion and extension radiographs to determine clinical instability, no high-quality evidence supports a routine and reliable diagnostic test for spinal instability (3,4,11-18). The lack of a validated clinical test for lumbar spinal instability contributes to the considerable variation in indications for adding a fusion among surgeons when performing lumbar decompression (3,11-17).

Many variables, including neuromuscular factors, likely contribute to the overall assessment of abnormal motion in the sagittal, coronal, and axial planes (3,19-24). However, this study concentrates on the automated, quantitative, imaging-based diagnosis of abnormal segmental intervertebral motion between flexion and extension in the sagittal plane, which is attributed to inadequate passive motion constraints, including the vertebrae, intervertebral discs, ligaments, and facet joints (25).

Currently, many spinal surgeons use >2–4 mm translation of one vertebral body over the other to indicate instability (26). This simple measurement may not distinguish physiological from pathological movement (18,27,28). In this manuscript, we extend the use of basic intervertebral translation measurements by adding two new measurements made from flexion-extension radiographs. The first, rotation-dependent translation (RDT), measures sagittal plane translation adjusted for rotation. Sagittal plane translation can decrease the space available for neural elements in the central spinal canal, the foramen, and the lateral recess. Abnormally high levels of translation may also result in mechanical irritation of nerve roots and other tissues, which is why quantifying it is so clinically relevant (29-31). Translation per degree of sagittal rotation has been discussed to a limited extent in prior studies (28,32-34). Data have been published to help define normal RDT and differentiate between normal versus abnormal RDT (28). RDT is level-dependent, requiring level-specific reference tables to interpret measurements, and can be reported as the statistical deviation from the average (32,35). The second test aims to quantify sagittal rotation-dependent disc widening (RDDW), calculated from anterior and posterior disc widening between flexion and extension, adjusted for sagittal rotation. This second measurement is intended to quantify vertical instability as a type of instability distinct from RDT instability (36,37). Vertical instability can be quantified from flexion-extension radiographs as abnormally high intervertebral motion in a direction perpendicular to the endplates.

This paper explores the potential of RDT and RDDW to diagnose sagittal plane, translational, and vertical lumbar spinal instabilities by retrospectively analyzing the prevalence of abnormalities in a range of patient populations. We present this article in accordance with the STROBE reporting checklist (38) (available at https://qims.amegroups.com/article/view/10.21037/qims-24-1931/rc).

Methods

Rotation, as used in this paper, refers to the angular change between vertebrae in the sagittal plane between flexion and extension (Figure 1).

Figure 1 Details of intervertebral rotation and translation measurements. Translation is the sagittal plane displacement, between extension and flexion, of the posterior-inferior corner of the superior vertebra in the direction defined by the superior endplate of the inferior vertebra. Translation is reported as percent EPW to eliminate the need to correct for radiographic magnification and to control for EPW variability between individuals. Rotation is the absolute difference in disc angles (da_E and da_F). EPW, end plate width.

Automated measurement of intervertebral motion from flexion-extension radiographs

A fully automated iteration of the previously validated (39-41) quantitative motion analysis (QMA) method was used to measure intervertebral motion from flexion-extension radiographs. The fully automated and FDA-cleared device (SpineCAMP^TM, Medical Metrics, Inc., Houston, TX, USA) uses a pipeline of neural networks and coded logic to produce four anatomic landmarks for each vertebra (42), determine transformation matrices to move landmarks from the flexion to the extension image, and then perform intervertebral motion measurements. The landmarks are intended to be consistent with those used in prior research studies (43-47). Multiple machine-learning/artificial intelligence approaches have been used to produce vertebral landmarks in previous studies (48-53); however, none have been widely adopted in clinical use to quantify instability. All intervertebral motion measurements were calculated from the coordinates of landmarks automatically placed on the four corners of every vertebral body in flexion and extension radiographs from L1 to S1. Intervertebral rotation and translation were measured, as illustrated in Figure 1. Note that translation is calculated as percent endplate width to correct for variability in vertebral sizes and to avoid the need to correct for variable radiographic magnification (54).

Relationships between intervertebral motion measurements

To help develop standardized metrics for reporting RDT and RDDW measurements and to help understand the relationships between RDT, RDDW, and other variables, a re-analysis of previously reported, IRB-approved, flexion-extension radiographs for 162 asymptomatic volunteers (78 males, 84 females) was completed. The mean age of participants was 42.2 years (47% 18 to <40 years old, 39% 40 to <60 years old, 14% 60 to 82 years old). Complete details of the study protocol have been previously described (28). No imaging other than the sagittal plane flexion-extension radiographs was available for analysis. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study (protocol No. H-12858) was approved by the Institutional Review Board of Baylor College of Medicine (BCM), and all volunteers provided informed consent. Some findings need to be reported at this point since they are required to explain the methods further. Figure 2 documents the relationship between translation and rotation in radiographically normal levels of asymptomatic volunteers. These data allow the determination of the average translation for any amount of rotation in radiographically normal spines, within the range of rotations assessed. The gray shaded area defines the 95% confidence interval. To facilitate the clinical utility of RDT measurements, they are reported as an index relative to the average and 95% confidence interval for an asymptomatic population. We refer to this as the translational instability index (TI-Index). The TI-Index for a specific level in a patient is calculated as the measured translation minus the average translation (adjusted for rotation) in healthy discs, divided by the standard error of the forecast (adjusted for rotation). The standard error of the forecast is used instead of a standard deviation, as the standard error of the forecast is a point estimate of the variability at a specific level of rotation. If the TI-Index is between −2 and +2, it is considered to be within normal limits. When the TI-Index is zero, translation is average relative to the asymptomatic population at the assessed level. When the TI-Index is >2, RDT is above the upper limit of the 95% confidence interval at radiographically normal levels in an asymptomatic population.

Figure 2 Relationship between intervertebral translation and rotation at the L4–5 level at radiographically normal levels in asymptomatic volunteers (27). CI, confidence interval.

RDDW is calculated as the difference in disc heights between flexion and extension and is adjusted for rotation (Figure 3). Disc heights are normalized by expressing them as a percentage of the width of the superior endplate of the inferior vertebra, thereby accounting for size variations among vertebrae.

Figure 3 Measurement details for the change in disc heights between flexion and extension. Disc heights are measured at the anterior-most aspect of the disc space in flexion and extension (dh_AE and dh_AF) and the posterior-most aspects of the disc space (dh_PE and dh_PF). Disc heights are reported as percent EPW. Rotation is the absolute difference in disc angles (da_E and da_F). EPW, end plate width.

An anterior vertical instability index (AVI-Index) and a posterior vertical instability index (PVI-Index) were calculated by first predicting the average disc widening for the measured rotation at radiographically normal levels (Figures 4,5). The gray shaded area in Figures 4,5 defines the 95% confidence interval for radiographically normal discs. The standard error of the forecast is used to determine the 95% confidence interval for the measured rotation. The AVI-Index or PVI-Index for a level in a patient is calculated as the measured change in disc height minus the average change in disc height in healthy discs (adjusted for rotation), divided by the standard error of the forecast (adjusted for rotation). If the AVI-Index or PVI-Index is between −2 and +2, it is considered to be within normal limits.

Figure 4 Across a population of asymptomatic volunteers, the change in anterior disc height is strongly related to the amount of intervertebral rotation. CI, confidence interval.

Figure 5 Across a population of asymptomatic volunteers, the change in posterior disc height is strongly related to the amount of intervertebral rotation. CI, confidence interval.

Statistical analysis

Prior research has documented the importance of level when predicting translations (28). To determine whether level should be included alongside rotation when predicting the change in disc height from flexion to extension, we conducted a multivariate analysis of variance (ANOVA) using Stata ver 15. In this ANOVA model, change in disc height was the outcome variable, and rotation and level were factor variables. The F statistic from the ANOVA assessed each factor’s relative contribution to explaining variability, while the R² indicated the proportion of variation explained by the model. A higher R² signifies greater explanatory power. We compared the R² from the two-factor model to that from a rotation-only model; if the increase in R² was minimal, “level” was deemed unnecessary for predicting disc height change. Finally, scatter plots and linear regression analyses were performed to illustrate relationships between variables.

RDT and RDDW abnormalities in different patient populations

A retrospective analysis of previously collected pre-treatment lumbar spine flexion-extension radiographs was completed to understand the prevalence of TI-Index, AVI-Index, and PVI-Index abnormalities in different patient populations. The flexion-extension exams were sourced from a quality-control database maintained by an imaging core laboratory (Medical Metrics, Inc., Houston, TX, USA). This database contains only de-identified raw images with no image meta-data, patient details, or inclusion/exclusion criteria. Only generic study descriptions were available, such as treatment for lumbar stenosis, selection for fusion, dynamic stabilization, disc arthroplasty, and biologics for disc treatment. Specific details of the treatments were not available in the database; however, the level(s) that were treated were known. Although this database is very limited by this lack of details about the patients and treatments, it was analyzed to provide rough estimates of expected prevalences in different patient populations. Pearl IRB determined the retrospective analysis of anonymized images using fully automated methods to be exempt research according to 45 CFR 46.104(d)(4) Secondary Research Uses of Data or Specimens.

Using SpineCAMP^TM, the TI-Index, AVI-Index, and PVI-Index were computed for all levels (between L1–2 and L5–S1) that could be analyzed in 7,621 pre-treatment flexion/extension radiographs from the various studies. Some levels were not analyzable as they were not included in the field-of-view in the flexion or extension radiographs or were obscured by labels or artifacts. The TI-, AVI-, and PVI-Index calculations were only made when intervertebral rotation exceeded 5 degrees, assuming that 5 degrees of rotation ensures sufficient stress on intervertebral motion restraints to detect abnormal motion. This 5 deg criterion was based on a review of existing neutral/lax zone research data (55-61). This is an interim criterion; formal validation is needed to establish standards for determining whether a spine has been adequately stressed.

Descriptive statistics

We calculated the prevalence of treatment and adjacent levels where the TI-Index, AVI-Index, or PVI-Index exceeded thresholds of 2 and 3. A threshold of 2 included levels slightly beyond the 95% confidence interval, while a threshold of 3 indicated more significant abnormalities.

Results

Relationships between intervertebral motion measurements at radiographically normal levels

In the radiographically normal levels of the asymptomatic volunteers, the relationship between translation and rotation was approximately linear for all levels (L1–2 to L5–S1) with an R² of between 0.29 and 0.58 (Table 1 and Figure 2). Including all levels in the asymptomatic population (not just the radiographically normal levels that were used to define normal motion), the data document that a small proportion (<4%) of levels in the asymptomatic population had a TI-Index >2 and approximately 1% had a TI-Index >3. In the asymptomatic population, disc degeneration had been graded using the Kellgren-Lawrence (KL) grading system (62). Figure 6 shows the average TI-Index for the different KL grades of disc degeneration. Although the association between disc degeneration and TI-Index was significant based on a one-way ANOVA (P<0.0001), the difference between KL grade 4 and the other grades was insignificant based on Bonferroni post hoc analysis using the available data. There were only 15 KL grade 4 discs. There was no significant association between the AVI-or PVI-Index and the KL grade of disc degeneration (P>0.1).

Figure 6 The TI-Index is significantly related to the grade of disc degeneration. P<0.0001, one-way ANOVA. Errors bars show the SEM. ANOVA, analysis of variance; KL, Kellgren-Lawrence; SEM, standard error of the mean; TI, translational instability.

The change in anterior and posterior disc heights was also linearly related to intervertebral rotation at the radiographically normal levels of the asymptomatic volunteers (Figures 4,5). Based on a multi-variable ANOVA, both rotation (P<0.0001) and intervertebral level were significant for the change in anterior disc height (P=0.0001). However, the R² changed an exceedingly small amount (0.9389 to 0.9379) when level was removed, supporting that data for all levels can be pooled.

Prevalence of abnormal rotation-dependent translation

TI, AVI and PVI measurements were calculated for 7,802 treatment levels and 27,707 adjacent levels. Requiring 5 degrees of rotation (to avoid analysis of inadequately stressed levels) reduced the number of analyzable treatment levels to 3,714 and analyzable adjacent levels to 15,189. A higher prevalence of the TI-Index abnormalities was seen in patients being treated for lumbar stenosis and patients treated using lumbar fusion or dynamic stabilization compared to patients enrolled in disc arthroplasty studies or studies investigating biologics for the treatment of disc disease (Table 2). Nevertheless, the prevalence of an abnormally high TI-Index was relatively small, particularly that of more pronounced abnormalities where the TI-Index was >3. The data also documents that a proportion of adjacent levels had an abnormal TI-Index.

Additional observations: (I) in the radiographically normal discs that were analyzed, where rotation exceeded 5 deg (N=579), the R² was 0.73 (P<0.0001) between the TI-Index and the cranial-caudal coordinate of the center of rotation (COR). (II) In the radiographically normal discs, the TI-Index and quantitative stability index (QSI) [aka sagittal plane shear index (SPSI)] were highly correlated (R²=0.90).

Examples of normal and abnormal TI-Index

Video S1 provides a stabilized animation showing intervertebral motion, with the TI-Index well within radiographically normal limits at all levels. “Stabilized” means that one vertebral body (L5 in this example) is held in a constant position on the display as the flexion and extension images are alternately displayed. In Video S1, rotation is >14 degrees at L4–5 and L5–S1.

In contrast, Video S2 provides a stabilized animation of a spine with a TI-Index >4 at L3–4 and L4–5. The TI-Index data document that the amount of translation (adjusted for rotation) is abnormal and would not be found in a healthy motion segment. Note that the intervertebral translations are visually higher than in Video S1. Video S3 provides another example of a stabilized animation with an abnormal TI-Index at multiple levels (5.7 at L3–4, 4.6 at L4–5, and 2.5 at L5–S1).

Prevalence of abnormal RDDW

A higher prevalence of AVI-Index abnormalities was seen in patients treated for lumbar stenosis and patients selected for fusion or dynamic stabilization compared to patients enrolled in disc arthroplasty studies or studies investigating biologics for disc disease treatment (Table 3). The data also document abnormal AVI-Index at a proportion of adjacent levels. The prevalence of AVI-Index abnormalities was much higher than the prevalence of TI-Index abnormalities in some treatment groups. Table 4 provides the prevalence of an abnormal PVI-Index and shows that these abnormalities are uncommon, though somewhat more common than TI-Index abnormalities in patients enrolled in disc arthroplasty and biologic disc treatment studies.

Video S4 provides an example of an abnormally high AVI-Index. The AVI-Index is 7.6 at L4–5 and is within radiographically normal limits at the adjacent levels. The TI-Index is near zero at all levels from L3–4 to L5–S1, though rotation may be too low to assess the TI-Index at some levels reliably. Note the vacuum disc sign in the extension. This was seen to be common with high AVI-Index and low disc space. Though motion at L4–5 is very abnormal based on RDDW, this level would not be diagnosed as abnormal based on rotation, translation, or the TI-Index.

Discussion

Instability is a common indication for spinal surgery; however, multiple reviews of the criteria used to diagnose spinal instability conclude that there is no well-validated test for instability and no uniform definition of stability in the degenerative spine (3,11-17). The current study addresses two specific tests for abnormal intervertebral motion: the TI-Index to diagnose abnormal rotation-dependent sagittal plane translations and the AVI-Index or the PVI-Index to diagnose abnormal RDDW. The TI-Index is an evolution of the QSI and SPSI metrics previously used in earlier research studies (32,35,63-65). QSI and SPSI are the same metrics but with different acronyms. These metrics were based on translation per degree of rotation (TPDR). These metrics became unstable at very low rotations since rotation was the denominator. Rather than using TPDR, the TI-Index uses an estimate of what translation should be for any degree of rotation (e.g., Figure 2). The TI-Index is, therefore, more robust and reliable in quantifying RDT, particularly with low magnitudes of rotation. The results document that the prevalence of levels with abnormal TI-Index, AVI-Index, and PVI-Index was higher in patients enrolled in studies investigating treatments for lumbar stenosis and patients selected for lumbar fusion or dynamic stabilization compared to patients enrolled in disc arthroplasty studies or studies investigating biologics for the treatment of disc disease.

Since this is the first report on the TI-Index, there is currently no published evidence of the clinical relevance of an abnormally high TI-Index. However, with support from prior research (28,33,66), and a review of Videos S2,S3, it is reasonable to hypothesize that an abnormally high TI-Index would be associated with variable neural compression and clinical symptoms (67,68). Conversely, diagnosing an abnormally low TI-Index may also be of clinical utility. Patients can also be symptomatic if the spine is hypomobile; thus, recognizing that a spine is stiff can also be important in surgical decision-making (69,70).

Consistent with the Kirkaldy-Willis theory, which suggests instability increasing with degeneration up to a point, followed by re-stabilization, the TI-Index increased until KL grade 3, then decreased with KL grade 4 degeneration (Figure 6) (3). This apparent decrease in the TI-Index in grade 4 discs was insignificant with the available data, though it was consistent with some prior studies (71-73). With more research, the TI-Index may help surgeons decide between a minimally invasive decompression (low or normal TI-Index) and a more significant fusion with instrumentation (high TI-Index) (35).

The potential to use the amount of sagittal plane translation between vertebrae, corrected for rotation, as the basis for an imaging-based test for lumbar instability has previously been described; however, this has not been done using fully automated methods (33,35). It is also noteworthy that, in the radiographically normal discs that were analyzed (N=579), the R² was 0.73 (P<0.0001) between the TI-Index and the cranial-caudal coordinate of the COR. In the healthy discs, the TI-Index and QSI (aka SPSI) are highly correlated (R²=0.90), which helps to relate the TI-Index data to previously reported QSI and SPSI data (32,35,63-65). A test for abnormal RDT can potentially detect the larger sagittal plane translations that can occur with this type of instability (28,32,33). Data documenting that abnormal RDT is associated with a wide facet gap and the facet fluid sign support this (32).

The intervertebral motion that can be measured from clinical flexion-extension radiographs depends on patient effort and may be less than what can be achieved in studies of intervertebral motion in healthy, asymptomatic volunteers (74). A patient may have a very unstable level, but because they minimally flexed and extended, the measured motion did not exceed the thresholds established based on studies of healthy spines. Using the proposed RDT and RDDW measurements, along with good-quality flexion-extension exams, may help detect instabilities that could otherwise be missed.

The concept of vertical instability has been proposed to be caused by “axial laxity” in the intervertebral disc (36). Disc height changes between flexion and extension have been previously shown to depend on the level and the amount of rotation (75). In the asymptomatic population, based on multivariate ANOVA to understand what effects disc height changes, the F-statistic for the level was 5.9 compared to 15,813 for rotation. Dropping the level in ANOVA reduced the R² from 0.9374 to 0.9364. Based on that observation, a decision was made not to include level when calculating the AVI- and PVI-indices. Like the rotation-dependent translation test for anterior-posterior (shear) instability, vertical instability might be diagnosed from flexion-extension radiographs as an abnormal amount of disc widening for the amount of rotation measured. Comparing prevalence data suggests the abnormal disc widening (Tables 3,4) may be more common than abnormal translation (Table 2), suggesting that it is possible to have vertical instability but not translational instability. It may be that a specific treatment could prove effective in the presence of an abnormal TI-Index but not an abnormal AVI-Index or PVI-Index or vice-versa, or simply that these all are varying forms of instability that might benefit from a specific stabilization procedure. The customary method for evaluating instability to date is mainly based on sagittal translation and, to some extent, angulation (26). Assessing for multiple forms of instability that control for variability in the range of motion and patient anatomy may lead to previously under-appreciated instabilities. If this becomes adopted, it will be interesting to look at patterns of motion that appear between the sagittal and vertical planes of translation. The TI-, AVI-, and PVI-indices are all calculated as the difference between an actual measurement and what the measurement is predicted to be in a normal spine, divided by an estimate of the error in the measurement. The gray shaded area in Figure 2 appears wider than the shaded areas in Figures 3,4. We hypothesize that a narrower confidence interval results in a relatively smaller denominator used to calculate the AVI- and PVI-Indices and that it thereby takes less of a deviation from normal to result in a high AVI- or PVI-index versus a high TI-Index. This may explain some of the higher prevalence of AVI-Index abnormalities compared to TI-Index abnormalities in some patient populations.

The intervertebral disc may become more flexible in the initial stages of disc degeneration (71,72,76-80). Hypothetically, abnormally high cranial-caudal or “vertical” translations between vertebrae might be diagnostic of loss of pressure in the nucleus, softening of the annulus, incompetence of the longitudinal ligaments, annular avulsions, or other causes (36,37,81-83). Such excessive vertical translations may be associated with abnormal intervertebral loading patterns (84-86). Disc height changes with loading have been reported in several studies (87,88). Since disc height changes occur with loading in healthy discs (89) and in patients (90,91), reference data to differentiate between normal and abnormal disc compressibility are required. That is the goal of the AVI-Index and PVI-Index metrics, although additional research is essential.

In the late stages of disc degeneration, where there is minimal remaining disc height, yet angular motion still occurs between flexion and extension, there is little restraint to vertical separation between vertebrae (37). In this condition, where there is near complete loss of disc height, the motion segment may also exhibit vertical instability characterized by abnormally high disc height changes for the amount of intervertebral rotation that occurred. It is also possible that any vertical instability may resolve in the restabilization stage of disc degeneration. Further research is needed to understand changes in vertical instability over time. Vertical instability may also be associated with a vacuum sign in the disk, which can be associated with symptoms and help predict outcomes (37). A vacuum sign is evident in X-rays or CT exams when gas pockets develop within the intervertebral disc, and this is believed to be associated with a stage of instability (92).

Notably, a small percentage of asymptomatic subjects demonstrated abnormal RDT, including (<4%) of levels with a TI-Index >2 and approximately 1% with a TI-Index >3. Adjacent level instability was also found in a small percentage of symptomatic patients. In the same way that disc or facet disease can be seen in asymptomatic patients (93,94), the TI-Index and disc widening metrics must be interpreted cautiously and always in the context of the clinical picture. Future longitudinal studies are required to determine if these metrics are irrelevant outliers or relevant predictors of eventual symptomatic and treatable pathology.

This study has several limitations. First, we did not have paired magnetic resonance imaging (MRI) exams along with the flexion-extension exams of the asymptomatic volunteers used to define “normal” intervertebral motion. We can, therefore, not rule out the possibility of early-stage degeneration at the levels used to define “normal” motion. In addition, it is possible that with a much more extensive collection of flexion-extension exams from a very diverse population of asymptomatic volunteers with no evidence of degeneration on MRI and X-rays, it would be discovered that the data used to define “normal” motion needs to be personalized for each patient. It has also yet to be determined how “normal” discs must be to justify inclusion in the definition of “normal” for purposes of a diagnostic test. Second, it would have been preferable to have details about the patients and image acquisition protocols used to obtain the flexion-extension studies that were analyzed to determine the prevalence of motion abnormalities. In particular, clinical histories, validated patient-reported outcomes, and MRI exams would have helped understand confounding issues that might be commonly assessed in clinical practice. These details were unavailable, and the lack of these desirable details limits the external validity of the results. Note that the difference in prevalence data between the treatment groups was deliberately not statistically analyzed due to the limitations in the definition of the treatment groups. Statements such as “There is a higher prevalence of abnormalities in treatment group A versus treatment group B” would not be justified given the limitations in group definitions in the pooled imaging database. The data should be used only for preliminary estimates of proportions.

Third, there is currently no accepted and well-validated test for instability that can be used as the essential “gold standard” to assess the sensitivity and specificity of the translational and vertical instability metrics.

Although not necessarily a limitation, it should be noted that endplate width was used to standardize the intervertebral translation and disc widening measurements instead of vertebral body height, based on concerns that the disc widening measurements could be compromised by the collapse of vertebral body height due to fractures. This concern was supported by the previously reported observation, based on analysis of 7,364 lumbar spines, that the superior/inferior endplate width ratio had much lower standard deviations than the anterior/posterior vertebral body height ratio or the ratio of vertebral body height to width (42). Alternative approaches to normalizing disc heights were not explored (95). In addition, disc height has been shown to be correlated with vertebral body height (96). That lack of independence might complicate the interpretation of aggregated disc heights normalized to vertebral body heights.

While the translational and vertical instability tests described here are promising, three requirements/details need to be acknowledged: (I) a good quality flexion-extension exam is required where the patient flexed and extended enough to provoke measurable abnormal motion if it can occur. In addition to sufficient patient effort, the proposed instability metrics are intended for lateral spine radiographs, not oblique or AP radiographs. The SpineCAMP^TM technology attempts to identify the mid-sagittal plane of each vertebral body (42). This approach has been shown to provide accurate intervertebral motion measurements even when the X-ray beam is not perfectly perpendicular to the mid-sagittal plane of the vertebral body (39). However, it has not yet been determined how far the X-rays can be from perpendicular before introducing an unacceptable error. The protocol used to obtain flexion-extension radiographs can substantially influence the amount of intervertebral motion (9-13). In our analysis of pooled lumbar pre-treatment flexion-extension exams, 52.3% of levels were not analyzed due to rotation <5 deg. A flexion-extension protocol designed to stress the spine adequately can increase the proportion of levels with rotation > 5 degrees to over 80% (35,97). The actual protocol used to obtain the pooled flexion-extension exams is unknown: (II) in this study, linear regressions and confidence intervals developed using data for asymptomatic volunteers were used to determine abnormal motion. The level-dependent regressions relating translation to rotation were not as strong (R²=0.3 to 0.59) as they were for the regression relating disc widening to rotation (R²=0.94). This is likely due in part to the fact that the average magnitude of disc widening (10.6%±4.0% endplate width) is higher than the average magnitude of translation (5.8%±3.0% endplate width). It is, therefore, somewhat easier to measure widening accurately: (III) the TI-Index, AVI-Index, and PVI-Index only measure sagittal plane motion. Coronal and axial plane intervertebral motion abnormalities and abnormal coupling between motion planes are also important. It is also possible that any clinical significance of abnormalities in the TI-Index and AVI-Index or PVI-Index depends on covariates, such as spondylolisthesis, scoliosis, and disc space narrowing.

Our goal is to develop objective, imaging-based diagnostic tests for spinal instability. Using flexion/extension X-rays, we can measure two linear degrees of freedom (anterior-posterior and superior-inferior) and one angular degree of freedom (sagittal plane rotation). The two instability metrics we present are designed to quantify linear motion while controlling for angular motion and anatomical differences between patients. With these metrics, we aim to characterize the dynamic nature of spinal instability more accurately. If validated, such tests could supplement current clinical definitions of instability—e.g., “the loss of the spine’s ability to maintain its patterns of displacement under physiologic loads without neurologic deficit, major deformity, or incapacitating pain” (98,99), or “Instability can be defined as the clinical status of the patient with back problems who with the least provocation steps from the mildly symptomatic to the severe episode.” (3). Objective imaging-based assessments could also be combined with physical tests for a more comprehensive evaluation of each patient (100).

Conclusions

Despite hundreds of research studies, no well-validated clinical tests for instability defined as abnormal intervertebral motion exist. Two instability indexes have been identified and investigated using retrospective automated analysis to assess the prevalence of abnormal sagittal plane intervertebral translation and abnormal disc widening across a range of patient populations. The prevalence estimates may help decide whether to explore the instability indexes in future studies and also help with sample size estimation. Further testing in populations with known clinical instability would help to assess whether and how the abnormal sagittal plane and abnormal disc space widening might be effective in diagnosis and treatment algorithms. With advancements in imaging technologies, software analytics, and data processing capabilities, significant progress in this area of research is anticipated. It is possible that these metrics may help to diagnose instability more reliably and may eventually help guide appropriate interventions that result in improved outcomes.

Acknowledgments

Medical Metrics, Inc. was started by Baylor College of Medicine (BCM) and through a close working relationship, collaborated in several research studies. Medical Metrics, Inc. funded the original study at BCM that collected flexion-extension X-rays of asymptomatic volunteers and has ownership of the collected images and data. This article includes a retrospective analysis of those data that is aligned with the aims of the original study.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-24-1931/rc

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-24-1931/coif). J.A.H. is a part-time salaried employee of Medical Metrics, Inc. and owns stock in that company, and is a coinventor on a US patent peripherally related to the work described in the manuscript. C.A.R. gets consulting fees from Medical Metrics, Inc. Z.B. has had multiple past research grants paid directly to an institution; has received consulting fees from several organizations; has received non-financial support for travel to professional society meetings; is a coinventor on an issued US patent, serves in leadership roles at several professional societies; and is an adjunct faculty at a US academic institute. C.D.C. has grants from several companies, all paid to his institution, and receives royalties from Globus. Z.G. owns stock and has intellectual property in NidusAI, Inc., is past-president of the North American Spine Society, and is treasurer of the Cervical Spine Research Society. T.F.G. is a full-time salaried employee of Medical Metrics, Inc. (MMI) and has stock options in Medical Metrics Diagnostics, Inc., a subsidiary of MMI. The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by Institutional Review Board of Baylor College of Medicine (No. H-12858) and informed consent was obtained from all the volunteers. J.A.H. conducted the study while serving as Director of the Spine Research Laboratory at BCM, which founded Medical Metrics, Inc.—the medical imaging core research laboratory where J.A.H. is now employed. Medical Metrics, Inc. funded the original study at BCM that collected flexion-extension X-rays of asymptomatic volunteers. The retrospective analysis of these flexion-extension X-rays aligns with the original IRB-approved goal of providing data on normal ranges of intervertebral motion to aid in interpreting motion measurements in patients with lumbar spine disorders. Pearl IRB determined the investigation of intervertebral motion abnormalities in different patient populations, which was a retrospective analysis of anonymized images using fully automated methods, to be exempt research according to 45 CFR 46.104(d)(4) Secondary Research Uses of Data or Specimens.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Teraguchi M, Yoshimura N, Hashizume H, Muraki S, Yamada H, Minamide A, Oka H, Ishimoto Y, Nagata K, Kagotani R, Takiguchi N, Akune T, Kawaguchi H, Nakamura K, Yoshida M. Prevalence and distribution of intervertebral disc degeneration over the entire spine in a population-based cohort: the Wakayama Spine Study. Osteoarthritis Cartilage 2014;22:104-10. [PubMed]
2. Ravindra VM, Senglaub SS, Rattani A, Dewan MC, Härtl R, Bisson E, Park KB, Shrime MG. Degenerative Lumbar Spine Disease: Estimating Global Incidence and Worldwide Volume. Global Spine J 2018;8:784-94. [PubMed]
3. Kirkaldy-Willis WH, Farfan HF. Instability of the lumbar spine. Clin Orthop Relat Res 1982;110-23. [PubMed]
4. Wang Y, Huang K. Research progress of diagnosing methodology for lumbar segmental instability: A narrative review. Medicine (Baltimore) 2022;101:e28534. [PubMed]
5. Fujiwara A, Tamai K, An HS, Kurihashi T, Lim TH, Yoshida H, Saotome K. The relationship between disc degeneration, facet joint osteoarthritis, and stability of the degenerative lumbar spine. J Spinal Disord 2000;13:444-50. [PubMed]
6. Anderson DB, Beard DJ, Rannou F, Hunter DJ, Suri P, Chen L, Van Gelder JM. Clinical assessment and management of lumbar spinal stenosis: clinical dilemmas and considerations for surgical referral. The Lancet Rheumatology 2024;6:e727-e32. [PubMed]
7. Dorsi MJ, Buchanan P, Vu C, Bhandal HS, Lee DW, Sheth S, Shumsky PM, Brown NJ, Himstead A, Mattie R. Pacific Spine and Pain Society (PSPS) evidence review of surgical treatments for lumbar degenerative spinal disease: a narrative review. Pain and Therapy 2024;13:349-90. [PubMed]
8. Jenis LG, An HS. Spine update. Lumbar foraminal stenosis. Spine (Phila Pa 1976) 2000;25:389-94. [PubMed]
9. Katz JN, Zimmerman ZE, Mass H, Makhni MC. Diagnosis and Management of Lumbar Spinal Stenosis: A Review. JAMA 2022;327:1688-99. [PubMed]
10. Schroeder GD, Kepler CK, Kurd MF, Vaccaro AR, Hsu WK, Patel AA, Savage JW. Rationale for the Surgical Treatment of Lumbar Degenerative Spondylolisthesis. Spine (Phila Pa 1976) 2015;40:E1161-6. [PubMed]
11. Muggleton JM, Kondracki M, Allen R. Spinal fusion for lumbar instability: does it have a scientific basis? J Spinal Disord 2000;13:200-4. [Crossref] [PubMed]
12. Nizard RS, Wybier M, Laredo JD. Radiologic assessment of lumbar intervertebral instability and degenerative spondylolisthesis. Radiol Clin North Am 2001;39:55-71. v-vi. [Crossref] [PubMed]
13. Chin KR. Radiology of Lumbar Instability. Semin Spine Surg 2005;17:243-50. [Crossref]
14. Leone A, Cassar-Pullicino VN, Guglielmi G, Bonomo L. Degenerative lumbar intervertebral instability: what is it and how does imaging contribute? Skeletal Radiol 2009;38:529-33. [Crossref] [PubMed]
15. Leone A, Guglielmi G, Cassar-Pullicino VN, Bonomo L. Lumbar intervertebral instability: a review. Radiology 2007;245:62-77. [Crossref] [PubMed]
16. Simmonds AM, Rampersaud YR, Dvorak MF, Dea N, Melnyk AD, Fisher CG. Defining the inherent stability of degenerative spondylolisthesis: a systematic review. J Neurosurg Spine 2015;23:178-89. [Crossref] [PubMed]
17. Elmose SF, Andersen GO, Carreon LY, Sigmundsson FG, Andersen MO. Radiological Definitions of Sagittal Plane Segmental Instability in the Degenerative Lumbar Spine - A Systematic Review. Global Spine J 2023;13:523-33. [Crossref] [PubMed]
18. HippJAReitmanCAChaputCDBuserZGriecoTF. The challenge of diagnosing lumbar segmental instability.Research Square 2023. doi: .10.21203/rs.3.rs-3659900/v1
19. Pitkänen M, Manninen H. Sidebending versus flexion-extension radiographs in lumbar spinal instability. Clin Radiol 1994;49:109-14. [Crossref] [PubMed]
20. Ochia RS, Inoue N, Takatori R, Andersson GB, An HS. In vivo measurements of lumbar segmental motion during axial rotation in asymptomatic and chronic low back pain male subjects. Spine (Phila Pa 1976) 2007;32:1394-9. [Crossref] [PubMed]
21. McFadden KD, Taylor JR. Axial rotation in the lumbar spine and gaping of the zygapophyseal joints. Spine (Phila Pa 1976) 1990;15:295-9. [Crossref] [PubMed]
22. Bogduk N. Clinical and Radiological Anatomy of the Lumbar Spine. Chapt 16. Elsevier Health Sciences; 2012.
23. Mellor FE, Muggleton JM, Bagust J, Mason W, Thomas PW, Breen AC. Midlumbar lateral flexion stability measured in healthy volunteers by in vivo fluoroscopy. Spine (Phila Pa 1976) 2009;34:E811-7. [Crossref] [PubMed]
24. Pearcy MJ, Tibrewal SB. Axial rotation and lateral bending in the normal lumbar spine measured by three-dimensional radiography. Spine (Phila Pa 1976) 1984;9:582-7. [Crossref] [PubMed]
25. Panjabi MM. The stabilizing system of the spine. Part I. Function, dysfunction, adaptation, and enhancement. J Spinal Disord 1992;5:383-9; discussion 397. [Crossref] [PubMed]
26. Spina N, Schoutens C, Martin BI, Brodke DS, Lawrence B, Spiker WR. Defining Instability in Degenerative Spondylolisthesis: Surgeon Views. Clin Spine Surg 2019;32:E434-9. [Crossref] [PubMed]
27. Foster WS, Hipp JA. Should the AMA change current structural integrity guidelines for rating spinal impairment? Disability Medicine 2005;5:3.
28. Staub BN, Holman PJ, Reitman CA, Hipp J. Sagittal plane lumbar intervertebral motion during seated flexion-extension radiographs of 658 asymptomatic nondegenerated levels. J Neurosurg Spine 2015;23:731-8. [Crossref] [PubMed]
29. Mayoux-Benhamou MA, Revel M, Aaron C, Chomette G, Amor B. A morphometric study of the lumbar foramen. Influence of flexion-extension movements and of isolated disc collapse. Surg Radiol Anat 1989;11:97-102. [Crossref] [PubMed]
30. Inufusa A, An HS, Lim TH, Hasegawa T, Haughton VM, Nowicki BH. Anatomic changes of the spinal canal and intervertebral foramen associated with flexion-extension movement. Spine (Phila Pa 1976) 1996;21:2412-20. [Crossref] [PubMed]
31. Fujiwara A, An HS, Lim TH, Haughton VM. Morphologic changes in the lumbar intervertebral foramen due to flexion-extension, lateral bending, and axial rotation: an in vitro anatomic and biomechanical study. Spine (Phila Pa 1976) 2001;26:876-82. [Crossref] [PubMed]
32. 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. [PubMed]
33. Weiler PJ, King GJ, Gertzbein SD. Analysis of sagittal plane instability of the lumbar spine in vivo. Spine (Phila Pa 1976) 1990;15:1300-6. [PubMed]
34. Kristjansson E, Leivseth G, Brinckmann P, Frobin W. Increased sagittal plane segmental motion in the lower cervical spine in women with chronic whiplash-associated disorders, grades I-II: a case-control study using a new measurement protocol. Spine (Phila Pa 1976) 2003;28:2215-21. [Crossref] [PubMed]
35. Reijmer JFH, de Jong LD, Kempen DHR, Arts MP, van Susante JLC. Clinical Utility of an Intervertebral Motion Metric for Deciding on the Addition of Instrumented Fusion in Degenerative Spondylolisthesis. Spine (Phila Pa 1976) 2024;49:E355-60. [Crossref] [PubMed]
36. Luk KD, Chow DH, Holmes A. Vertical instability in spondylolisthesis: a traction radiographic assessment technique and the principle of management. Spine (Phila Pa 1976) 2003;28:819-27. [Crossref] [PubMed]
37. Lewandrowski KU, Zhang X, Ramírez León JF, de Carvalho PST, Hellinger S, Yeung A. Lumbar vacuum disc, vertical instability, standalone endoscopic interbody fusion, and other treatments: an opinion based survey among minimally invasive spinal surgeons. J Spine Surg 2020;6:S165-78. [Crossref] [PubMed]
38. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007;370:1453-7. [Crossref] [PubMed]
39. Zhao KD, Ben-Abraham EI, Magnuson DJ, Camp JJ, Berglund LJ, An KN, Bronfort G, Gay RE. Effect of Off-Axis Fluoroscopy Imaging on Two-Dimensional Kinematics in the Lumbar Spine: A Dynamic In Vitro Validation Study. J Biomech Eng 2016;138:054502. [Crossref] [PubMed]
40. Zhao K, Yang C, Zhao C, An KN. Assessment of non-invasive intervertebral motion measurements in the lumbar spine. J Biomech 2005;38:1943-6. [Crossref] [PubMed]
41. Pearson AM, Spratt KF, Genuario J, McGough W, Kosman K, Lurie J, Sengupta DK. Precision of lumbar intervertebral measurements: does a computer-assisted technique improve reliability? Spine (Phila Pa 1976) 2011;36:572-80. [Crossref] [PubMed]
42. Hipp JA, Grieco TF, Newman P, Reitman CA. Definition of Normal Vertebral Morphometry Using NHANES-II Radiographs. JBMR Plus 2022;6:e10677. [Crossref] [PubMed]
43. Hurxthal LM. Measurement of anterior vertebral compressions and biconcave vertebrae. Am J Roentgenol Radium Ther Nucl Med 1968;103:635-44. [Crossref] [PubMed]
44. Diacinti D, Guglielmi G. Vertebral morphometry. Radiol Clin North Am 2010;48:561-75. [Crossref] [PubMed]
45. Frobin W, Brinckmann P, Biggemann M, Tillotson M, Burton K. Precision measurement of disc height, vertebral height and sagittal plane displacement from lateral radiographic views of the lumbar spine. Clin Biomech (Bristol) 1997;12:S1-S63. [Crossref] [PubMed]
46. Keynan O, Fisher CG, Vaccaro A, Fehlings MG, Oner FC, Dietz J, Kwon B, Rampersaud R, Bono C, France J, Dvorak M. Radiographic measurement parameters in thoracolumbar fractures: a systematic review and consensus statement of the spine trauma study group. Spine (Phila Pa 1976) 2006;31:E156-65. [Crossref] [PubMed]
47. Quint DJ, Tuite GF, Stern JD, Doran SE, Papadopoulos SM, McGillicuddy JE, Lundquist CA. Computer-assisted measurement of lumbar spine radiographs. Acad Radiol 1997;4:742-52. [Crossref] [PubMed]
48. Zhang T, Li Y, Cheung JPY, Dokos S, Wong K-YK. Learning-based coronal spine alignment prediction using smartphone-acquired scoliosis radiograph images. IEEE Access 2021;9:38287-95.
49. Galbusera F, Bassani T, Costa F, Brayda-Bruno M, Zerbi A, Wilke HJ. Artificial neural networks for the recognition of vertebral landmarks in the lumbar spine. Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization 2018;6:447-52. [Crossref]
50. Yeh YC, Weng CH, Huang YJ, Fu CJ, Tsai TT, Yeh CY. Deep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs. Sci Rep 2021;11:7618. [Crossref] [PubMed]
51. Zhou S, Yao H, Ma C, Chen X, Wang W, Ji H, He L, Luo M, Guo Y. Artificial intelligence X-ray measurement technology of anatomical parameters related to lumbosacral stability. Eur J Radiol 2022;146:110071. [Crossref] [PubMed]
52. Cina A, Bassani T, Panico M, Luca A, Masharawi Y, Brayda-Bruno M, Galbusera F. 2-step deep learning model for landmarks localization in spine radiographs. Sci Rep 2021;11:9482. [Crossref] [PubMed]
53. Al Arif SMMR, Gundry M, Knapp K, Slabaugh G. Improving an Active Shape Model with Random Classification Forest for Segmentation of Cervical Vertebrae. In: Yao J, Vrtovec T, Zheng G, Frangi A, Glocker B, Li S. editors. Computational Methods and Clinical Applications for Spine Imaging. CSI 2016. Lecture Notes in Computer Science, vol 10182. Springer, Cham. doi: 10.1007/978-3-319-55050-3_1.10.1007/978-3-319-55050-3_1
54. Ravi B, Rampersaud R. Clinical magnification error in lateral spinal digital radiographs. Spine (Phila Pa 1976) 2008;33:E311-6. [Crossref] [PubMed]
55. Mimura M, Panjabi MM, Oxland TR, Crisco JJ, Yamamoto I, Vasavada A. Disc degeneration affects the multidirectional flexibility of the lumbar spine. Spine (Phila Pa 1976) 1994;19:1371-80. [Crossref] [PubMed]
56. Crawford NR, Peles JD, Dickman CA. The spinal lax zone and neutral zone: measurement techniques and parameter comparisons. J Spinal Disord 1998;11:416-29. [Crossref] [PubMed]
57. Busscher I, van Dieën JH, Kingma I, van der Veen AJ, Verkerke GJ, Veldhuizen AG. Biomechanical characteristics of different regions of the human spine: an in vitro study on multilevel spinal segments. Spine (Phila Pa 1976) 2009;34:2858-64. [Crossref] [PubMed]
58. 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;20:578-84. [PubMed]
59. Smit TH, van Tunen MS, van der Veen AJ, Kingma I, van Dieën JH. Quantifying intervertebral disc mechanics: a new definition of the neutral zone. BMC Musculoskelet Disord 2011;12:38. [PubMed]
60. Käfer W, Cakir B, Midderhoff S, Reichel H, Wilke HJ. Circumferential dynamic stabilization of the lumbar spine: a biomechanical analysis. Eur Spine J 2014;23:2330-9. [Crossref] [PubMed]
61. Muriuki MG, Havey RM, Voronov LI, Carandang G, Zindrick MR, Lorenz MA, Lomasney L, Patwardhan AG. Effects of motion segment level, Pfirrmann intervertebral disc degeneration grade and gender on lumbar spine kinematics. J Orthop Res 2016;34:1389-98. [PubMed]
62. Kellgren JH, Jeffrey MR, Ball J. Atlas of standard radiographs of arthritis. The epidemiology of chronic rheumatism. Oxford: Blackwell Scientific Publications; 1963:46-72.
63. Braunstein J, Hipp JA, Browning R, Grieco TF, Reitman CA. Analysis of translation and angular motion in loaded and unloaded positions in the lumbar spine. N Am Spine Soc J 2020;4:100038. [PubMed]
64. Grieco TF, McKnight B, Wang JC, Buser Z. Evaluating the Prevalence of Motion Abnormalities at Treatment Levels and Nontreatment Levels in Lumbar Stenosis and Spondylolisthesis Patients. Spine (Phila Pa 1976) 2023;48:1405-7. [PubMed]
65. Hedman T, Yu J, Singh H, Deer T. Early Clinical Results of Intervertebral Joint Stabilization by Injectable Load-Sharing Polymers. J Pain Res 2023;16:2777-89. [PubMed]
66. Kong MH, Hymanson HJ, Song KY, Chin DK, Cho YE, Yoon DH, Wang JC. Kinetic magnetic resonance imaging analysis of abnormal segmental motion of the functional spine unit. J Neurosurg Spine 2009;10:357-65. [PubMed]
67. Morishita Y, Hida S, Naito M, Arimizu J, Matsushima U, Nakamura A. Measurement of the local pressure of the intervertebral foramen and the electrophysiologic values of the spinal nerve roots in the vertebral foramen. Spine (Phila Pa 1976) 2006;31:3076-80. [Crossref] [PubMed]
68. Morishita Y, Hida S, Naito M, Arimizu J, Takamori Y. Neurogenic intermittent claudication in lumbar spinal canal stenosis: the clinical relationship between the local pressure of the intervertebral foramen and the clinical findings in lumbar spinal canal stenosis. J Spinal Disord Tech 2009;22:130-4. [Crossref] [PubMed]
69. Dvorák J, Panjabi MM, Novotny JE, Chang DG, Grob D. Clinical validation of functional flexion-extension roentgenograms of the lumbar spine. Spine (Phila Pa 1976) 1991;16:943-50. [Crossref] [PubMed]
70. Mayer TG, Robinson R, Pegues P, Kohles S, Gatchel RJ. Lumbar segmental rigidity: can its identification with facet injections and stretching exercises be useful? Arch Phys Med Rehabil 2000;81:1143-50. [Crossref] [PubMed]
71. Fujiwara A, Lim TH, An HS, Tanaka N, Jeon CH, Andersson GB, Haughton VM. The effect of disc degeneration and facet joint osteoarthritis on the segmental flexibility of the lumbar spine. Spine (Phila Pa 1976) 2000;25:3036-44. [Crossref] [PubMed]
72. Axelsson P, Karlsson BS. Intervertebral mobility in the progressive degenerative process. A radiostereometric analysis. Eur Spine J 2004;13:567-72. [Crossref] [PubMed]
73. 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;39:1032-43. [Crossref] [PubMed]
74. Matheve T, De Baets L, Bogaerts K, Timmermans A. Lumbar range of motion in chronic low back pain is predicted by task-specific, but not by general measures of pain-related fear. Eur J Pain 2019;23:1171-84. [Crossref] [PubMed]
75. Fu M, Ye Q, Jiang C, Qian L, Xu D, Wang Y, Sun P, Ouyang J. The segment-dependent changes in lumbar intervertebral space height during flexion-extension motion. Bone Joint Res 2017;6:245-52. [Crossref] [PubMed]
76. Tanaka N, An HS, Lim TH, Fujiwara A, Jeon CH, Haughton VM. The relationship between disc degeneration and flexibility of the lumbar spine. Spine J 2001;1:47-56. [Crossref] [PubMed]
77. Brown MD, Holmes DC, Heiner AD. Measurement of cadaver lumbar spine motion segment stiffness. Spine (Phila Pa 1976) 2002;27:918-22. [Crossref] [PubMed]
78. Ebara S, Harada T, Hosono N, Inoue M, Tanaka M, Morimoto Y, Ono K. Intraoperative measurement of lumbar spinal instability. Spine (Phila Pa 1976) 1992;17:S44-50. [Crossref] [PubMed]
79. Rohlmann A, Zander T, Schmidt H, Wilke HJ, Bergmann G. Analysis of the influence of disc degeneration on the mechanical behaviour of a lumbar motion segment using the finite element method. J Biomech 2006;39:2484-90. [Crossref] [PubMed]
80. Knutsson F. The instability associated with disc degeneration in the lumbar spine. Acta Radiol 1944;25:593-8. [Crossref]
81. Friberg O. Lumbar instability: a dynamic approach by traction-compression radiography. Spine (Phila Pa 1976) 1987;12:119-29. [Crossref] [PubMed]
82. Son S, Lee SG, Kim WK, Ahn Y, Jung JM. Disc height discrepancy between supine and standing positions as a screening metric for discogenic back pain in patients with disc degeneration. Spine J 2021;21:71-9. [Crossref] [PubMed]
83. Abdollah V, Parent EC, Su A, Wachowicz K, Battié MC. The effects of axial loading on the morphometric and T(2) characteristics of lumbar discs in relation to disc degeneration. Clin Biomech (Bristol) 2021;83:105291. [Crossref] [PubMed]
84. Horst M, Brinckmann P. 1980 Volvo award in biomechanics. Measurement of the distribution of axial stress on the end-plate of the vertebral body. Spine (Phila Pa 1976) 1981;6:217-32. [Crossref] [PubMed]
85. McNally DS, Adams MA. Internal intervertebral disc mechanics as revealed by stress profilometry. Spine (Phila Pa 1976) 1992;17:66-73. [Crossref] [PubMed]
86. McNally DS, Shackleford IM, Goodship AE, Mulholland RC. In vivo stress measurement can predict pain on discography. Spine (Phila Pa 1976) 1996;21:2580-7. [PubMed]
87. Shymon S, Hargens AR, Minkoff LA, Chang DG. Body posture and backpack loading: an upright magnetic resonance imaging study of the adult lumbar spine. Eur Spine J 2014;23:1407-13. [PubMed]
88. Tarantino U, Fanucci E, Iundusi R, Celi M, Altobelli S, Gasbarra E, Simonetti G, Manenti G. Lumbar spine MRI in upright position for diagnosing acute and chronic low back pain: statistical analysis of morphological changes. J Orthop Traumatol 2013;14:15-22. [PubMed]
89. Alamin TF, Agarwal V, Zagel A, Qeli A. The effect of standing vs. variants of the seated position on lumbar intersegmental angulation and spacing: a radiographic study of 20 asymptomatic subjects. J Spine Surg 2018;4:509-15. [PubMed]
90. Axelsson P, Karlsson BS. Standardized provocation of lumbar spine mobility: three methods compared by radiostereometric analysis. Spine (Phila Pa 1976) 2005;30:792-7. [PubMed]
91. Axelsson P, Johnsson R, Strömqvist B. Adjacent segment hypermobility after lumbar spine fusion: no association with progressive degeneration of the segment 5 years after surgery. Acta Orthop 2007;78:834-9. [PubMed]
92. Kelm NE, Aftab MA. Vacuum phenomenon in the lumbar spine: a useful tool for neuroradiologists and spine surgeons? Neurographics 2021;11:59-64.
93. Jensen MC, Brant-Zawadzki MN, Obuchowski N, Modic MT, Malkasian D, Ross JS. Magnetic resonance imaging of the lumbar spine in people without back pain. N Engl J Med 1994;331:69-73. [Crossref] [PubMed]
94. Boden SD, Davis DO, Dina TS, Patronas NJ, Wiesel SW. Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg Am 1990;72:403-8. [PubMed]
95. Chen X, Sima S, Sandhu HS, Kuan J, Diwan AD. Radiographic evaluation of lumbar intervertebral disc height index: An intra and inter-rater agreement and reliability study. J Clin Neurosci 2022;103:153-62. [Crossref] [PubMed]
96. Kunkel ME, Herkommer A, Reinehr M, Böckers TM, Wilke HJ. Morphometric analysis of the relationships between intervertebral disc and vertebral body heights: an anatomical and radiographic study of the human thoracic spine. J Anat 2011;219:375-87. [Crossref] [PubMed]
97. Gornet MF, Hipp JA, Schranck FW, Copay AG, Schranck CR, Wymore K, editors. A flexion-extension stress-test for the lumbar spine. International Society for the Advancement of Spine Surgery; 2016.
98. White AA, III, Panjabi MM. Clinical Biomechanics of the spine. 2nd Ed. Philadelphia: J.B. Lippincott; 1990.
99. Panjabi MM. Clinical spinal instability and low back pain. J Electromyogr Kinesiol 2003;13:371-9. [PubMed]
100. Jiang J, Hu J, Cai HP, Niu L, Zheng ML, Chen X, Zhang WZ. Radiographic analysis of dynamic lumbar motion during the five-repetition sit-to-stand test in degenerative lumbar spondylolisthesis. BMC Musculoskelet Disord 2022;23:800. [PubMed]