Facet Orientation and Tropism: Associations with Spondylolysis

Abstract

Study Design

Cross-sectional study.

Objective

To evaluate the association between lumbar spine facet joint orientation, facet joint tropism, and spondylolysis identified by multi-detector computed tomography (CT) in the community-based Framingham Heart Study.

Summary of Background Data

The association between lumbar spondylolysis and facet orientation and tropism remains unclear.

Methods

This study was an ancillary project to the Framingham Heart Study. 3529 participants of the Framingham Heart Study aged 40-80 underwent multi-detector CT imaging to assess aortic calcification. 191 subjects were included in this ancillary study. Facet joint features and spondylolysis were evaluated on CT scans. The final analyzed sample included 104 males with mean age 51.90±11.25 and 84 females with mean age 53.61±10.20. The association between spondylolysis and facet orientation and tropism was examined using univariate and multivariate analyses.

Results

Spondylolysis was prevalent in 11.5% of the total population. χ²-test demonstrated a significant sex difference in prevalence of spondylolysis (p=0.0154), with almost three times higher prevalence among males.

There was no statistically significant difference in facet orientation and continuous facet tropism between individuals with and without spondylolysis at the L5 level (p=0.49-0.91). After adjustment for age, sex and BMI, no significant association between the occurrence of spondylolysis and facet orientation and tropism was found.

In the studied sample the prevalence of facet joint osteoarthritis was significantly higher in individuals with spondylolysis than in those without spondylolysis at both sides of L4-L5 spinal level (p=0.044 at right side and p=0.003 at the left side) and at left side of L5-S1 level (p=0.038).

Conclusions

We did not find an association between facet orientation, facet tropism and spondylolysis. One of the possible explanations for this is that the high prevalence of facet joint osteoarthritis in individuals with spondylolysis in the studied sample may have led to diminished differences in facet orientation.

Introduction

Recent studies have suggested a relationship between facet orientation and the development of spondylolytic defects [1, 2]. Masharawi and colleagues [1], using a 3-dimentional digitizer, measured the facet orientation at the L1-L5 levels in skeletons of 115 males with bilateral spondylolysis at the L5 level and 120 age matched controls. They found a more coronal facet orientation (up to 13° at L4) in the isthmic spondylolysis group (right inferior facets). Three of the 4 articular facets of L5 (right and left inferior and right superior), corresponding to sites both above and below the pars defect, were significantly more coronally oriented in isthmic spondylolysis compared to the control group. A greater tendency towards facet tropism, an asymmetry in facet orientation was noticed in the isthmic spondylolysis group. Don and Robertson [2] measured the orientation of the facet joints on magnetic resonance image (MRI) at the L3-L4, L4-L5, and L5-S1 levels in 30 individuals with normal scans and 30 patients with isthmic spondylolisthesis at L5-S1. They found a more coronal angulation at the levels above the pars defect. This difference was highly statistically significant (P<0.001 at L3-4 and P<0.0001 at L4-5). Below the pars defect at L5-S1, orientations were the same in both groups.

There are notable limitations to the aforementioned studies. The Masharawi study is a cadaveric skeletal study, including specimens of young males born between 1825 and 1910. The Don study used MRI, which is inferior to computed tomography (CT) for the assessment of bony features such as spondylolysis and facet joint changes. In addition there are some discrepancies between the results of these studies. For example, in the Masharawi et al. [1] study, facets both above and below the L5 pars defect were significantly more coronally oriented in specimens with spondylolysis than in specimens without spondylolysis. In contrast, Don and Robertson [2] found that facets below the level of spondylolysis were slightly less coronally oriented.

Due to precise demonstration of osseous details [3, 4] CT is the preferred method for imaging lumbar facet joints [5] and allows precise measurement of facet orientation. CT also allows easy recognition of presence or absence of spondylolytic defect [6, 7].

The aim of the present community-based study was to evaluate the association between lumbar spine facet joint orientation, facet joint tropism, and spondylolysis identified by multi-detector CT in the community-based Framingham Heart Study.

Methods

Sample

This project was an ancillary project to the Framingham Heart Study. The Framingham Heart Study began in 1948 as a longitudinal population-based cohort study of the causes of heart disease. Initially, 5209 men and women between the ages of 30 and 60 years living in Framingham, Massachusetts were enrolled. Biennial examinations were conducted by trained research staff at the study clinic located in Framingham. In 1971, 5,124 offspring (and their spouses) of the original cohort were entered into the Offspring cohort. In 2002, 4095 men and women who were children of the Offspring cohort were enrolled in the Third Generation cohort. A description of the Offspring and Third Generation cohorts has been previously reported [8, 9]. 3529 participants of the Framingham study (participants in both the Offspring and Third Generation cohorts) aged 40-80 years underwent abdominal and chest multi-detector CT scanning to assess coronary and aortic calcification. The recruitment and conduct of CT scanning have been previously reported [10, 11]. During the later part of the CT study, 191 participants were consecutively enrolled in this ancillary study to assess the association between radiographic features of the lumbosacral spine and LBP. Three individuals were not analyzed because of insufficient CT data. We evaluated the imaging findings of spondylolysis, facet joint orientation, facet joint tropism, and facet joint osteoarthritis in the sample of 188 subjects

Imaging parameters

Study participants were imaged with an eight-slice multidetector CT scanner (Lightspeed Ultra, GE, Milwaukee, WI, USA). Each subject underwent unenhanced abdominal CT that was performed using a sequential scan protocol with a slice collimation of 8 mm × 2.5 mm (120 KVp, 320/400 mA for .220 lbs body weight, respectively) during a single end-inspiratory breath hold (typical duration 18 s). For the abdominal scan, sixty contiguous 2.5 mm thick slices of the abdomen were acquired covering 150 mm above the level of S1.

Evaluation of spondylolysis

CT scans were evaluated in blinded fashion with respect to clinical and personal data. The entire lumbar spine was reviewed for each case, using bone windows. Both axial views and sagittal multiplanar reconstruction were analyzed. Spondylolysis was marked as present or absent at right and/or left sides of the lumbar vertebrae. In the case where the image review was equivocal, it was evaluated again by a second reader, experienced research musculoskeletal radiologist (AG), and consensus between the two readers was obtained. CT evaluation of spondylolysis, especially using multiplanar reconstruction has previously been described as a reliable and accurate method [6, 7]. The overall image quality of the Framingham CT scans was excellent for visualization of the lumbar spine bony structures used to diagnose spondylolysis.

Evaluation of facet orientation and tropism

The facet joint angle relative to the coronal plane at L3-L4, L4–5 and L5-S1 was measured on the axial CT images. Continuous facet joint tropism was defined as the absolute difference of the angles between the right and the left facet. Overall image quality was excellent for the evaluation of facet joint changes.

Facet joint osteoarthritis (OA)

Four grades (0-normal, 1-mild, 2-moderae and 3-severe degeneration) of facet joint OA were defined using criteria that were described by Kalichman et al. [12]. This semi-quantitative score is derived from the earlier classification of Pathria [13] and accounts for such changes as joint space narrowing, osteophytes, hypertrophy of the articular process, subarticular sclerosis, subchondral cysts and vacuum phenomenon. Lumbar facet joints were graded on both sides at L2-L3, L3-L4, L4-L5, and L5-S1 levels. For the purposes of this study we dichotomized this index on the basis of the presence or absence of facet joint OA (≥grade 2) on any side at any level.

Reliability of CT readings

All readers were trained by an experienced research musculoskeletal radiologist (AG). The intra- and inter-rater reliability was calculated for two readers. All CT scans were then analyzed blinded to clinical status and to the hypothesis being tested here. To evaluate for reader-drift, intra-rater reliability was periodically reassessed by inserting one repeated “reliability” scan for every 10 new scans. Before analyzing each new set of CT scans, 5 previously analyzed CTs were reevaluated to “recalibrate” the readings to a standard. The intra-observer reliability for identification of spondylolysis was 1.00. The inter-observer reliability was 0.98, representing excellent reproducibility. The intra-observer reliability for grading facet joint OA indices varied between 0.64 and 0.91. The inter-observer reliability ranged from 0.59 to 0.94.

Body mass index (BMI)

BMI was computed as the ratio of weight (in kg) divided by height (in square meters).

Statistical analysis

We began by calculating descriptive statistics. We compared facet orientation and facet tropism (continuous) between males and females using the t-test for continuous variables. We compared the prevalence of spondylolysis (at any level) between males and females using the χ²-test.

To evaluate the association between spondylolysis and facet orientation and tropism we used two types of analyses. In the first analysis, the paired t-test was used to compare the facet joint orientation and tropism at the levels L3-L4, L4-L5 and L5-S1 between individuals with and without spondylolysis at L5 spinal level (the level where spondylolysis was most prevalent). In the second analysis we used a logistic regression model with spondylolysis at the L5 level as the dependent variable, and facet orientation and tropism as well as age, sex and BMI, as independent variables. The same univariate analyses were performed with spondylolysis at the L4 level as the dependent variable. Finally, to compare the prevalence of facet joint OA between individuals with and without spondylolysis we used χ²-test.

All statistical analyses were performed using SAS software, (SAS Institute Inc, Cary, North Carolina, release 9.1).

Results

The sample included 104 males with mean age 51.90±11.25 and 84 females with mean age 53.61±10.20. Individuals were slightly overweight on average: BMI=27.95±4.16 for males and 27.71±5.98 for females. Results of comparison between males and females showed no sex differences in facet orientation and tropism at any studied spinal level (p-values: 0.42-0.71).

Spondylolysis was prevalent in 16.5% of males (N=17) and 5% of females (N=4), 11.5% of the total population. Two subjects demonstrated unilateral spondylolytic defects while the remaining subjects were found to have bilateral defects. One man had spondylolysis at two spinal levels, L4-L5 (unilateral) and L5-S1 (bilateral). Spondylolysis was found at the L5 spinal level in 17 individuals, at the L4 level in 3 individuals, and at L3 level in 1 individual. χ²-test demonstrated a statistically significant sex difference in prevalence of spondylolysis (p=0.0154), with almost three times higher prevalence among males. The age of individuals with spondylolysis at the L5 level was 56.29±11.19 vs. 52.22±10.69 in individuals without spondylolysis. BMI was 27.82±3.51 and 27.90±5.20 respectively. The differences between groups in age and BMI were not statistically significant. There was a significantly lower prevalence of females among individuals with spondylolysis than among individuals without spondylolysis (17.7% vs. 49.4%, respectively; p=0.0229).

Comparisons of the mean facet orientation, mean facet tropism and prevalence of tropism at the level of the lytic defect between the individuals with and without spondylolysis are presented in Table 1. There was no significant difference in facet orientation and continuous facet tropism between individuals with and without spondylolysis at the L5 level.

Table 1.

Comparison of the facet angle and tropism between the individuals with and without spondylolysis at L5 level.

	Individuals with spondylolysis (N=17)	Individuals without spondylolysis (N=171)	Comparison (p-value) **
Age (years)	56.29±11.19	52.22±10.69	0.1381
BMI (kg/m2)	27.82±3.51	27.90±5.20	0.9509
Sex (females)(prevalence)	3 (17.7%)	77 (49.4%)	0.0229
Facet orientation L3-L4 Rt *	36.88±9.45	37.53±8.74	0.7740
Facet orientation L3-L4 Lt *	37.94±11.34	38.19±9.40	0.9191
Facet orientation L4-L5 Rt *	45.59±10.64	45.24±10.68	0.8985
Facet orientation L4-L5 Lt *	49.06±7.81	47.39±10.69	0.5307
Facet orientation L5-S1 Rt *	48.24±11.65	46.97±11.29	0.6614
Facet orientation L5-S1 Lt *	50.53±6.91	49.17+13.19	0.4932
Facet tropism (continuous) L3-L4	6.47±4.12	6.02±4.99	0.7214
Facet tropism (continuous) L4-L5	6.53±5.94	6.98±6.16	0.7755
Facet tropism (continuous) L5-S1	8.88±11.32	7.06±7.84	0.5256

^*Relative to sagittal plane.

^**Results of χ² test for dichotomous variables and of t-test for continuous variables. Statistically significant association at p<0.05 level marked bold.

Associations between spondylolysis and facet orientation and tropism while adjusted for age, sex and BMI are shown in Table 2. In the present study, after adjustment for age, sex and BMI, there was no significant association between the occurrence of spondylolysis and facet orientation and tropism.

Table 2.

Associations between spondylolysis and facet orientation while adjusted for age, sex and BMI. Results of multiple logistic regression analysis.

Dependent Variable	Independent Variable	OR	95%CI
Spondylolysis L5	Age	1.04	0.99	1.10
	Sex *	0.22	0.06	0.85
	BMI	1.00	0.88	1.13
	Facet orientation L3-L4 Rt	1.00	0.93	1.09
	Facet orientation L3-L4 Lt	0.99	0.92	1.07
	Facet orientation L4-L5 Rt	0.97	0.88	1.07
	Facet orientation L4-L5 Lt	1.04	0.95	1.15
	Facet orientation L5-S1 Rt	1.01	0.96	1.07
	Facet orientation L5-S1 Lt	1.00	0.94	1.05
	Facet tropism (continuous) L3-L4	1.05	0.94	1.18
	Facet tropism (continuous) L4-L5	0.93	0.82	1.06
	Facet tropism (continuous) L5-S1	1.03	0.96	1.10

^*Sex: Male as reference group; statistically significant association at p<0.05 level marked bold.

There were only three individuals with spondylolysis at L4 spinal level. The results of univariate analyses also showed no association between spondylolisthesis at L4 level and facet orientation and tropism at L3-L4, L4-L5 and L5-S1 levels.

Table 3 shows that in the studied sample the prevalence of facet joint OA was significantly higher in individuals with spondylolysis than in those without spondylolysis at both sides of L4-L5 spinal level (p=0.044 at right side and p=0.003 at the left side) and at left side of L5-S1 level (p=0.038).

Table 3.

Comparison of the facet joint OA prevalence between individuals with and without spondylolysis at L5 level.

	Individuals with spondylolysis	Individuals without spondylolysis	Comparison (p-value) **
Facet joint OA L3-L4 Rt *	9(52.9%)	69(40.8%)	0.3347
Facet joint OA L3-L4 Lt *	10(58.8%)	70(41.4%)	0.1671
Facet joint OA L4-L5 Rt *	14(82.4%)	96(57.1%)	0.0436
Facet joint OA L4-L5 Lt *	16(94.1%)	95(56.6%)	0.0026
Facet joint OA L5-S1 Rt *	14(82.4%)	97(59.2%)	0.0615
Facet joint OA L5-S1 Lt *	13(76.5%)	82(50.0%)	0.0375

^*Prevalence.

^**Results of χ² test. Statistically significant differences at p≤0.05 level are marked bold.

Discussion

We used a community-based sample of elderly patients (mean age 52.6±10.8) to evaluate the association between facet orientation and tropism and spondylolysis. In the present study we failed to find an association between facet orientation and tropism and spondylolysis.

These results contradict the previous findings of significantly more coronal orientation in certain facets in individuals with spondylolysis by Masharawi and colleagues [1] and Don and Robertson [2]. This is noteworthy since, as stated by Don, the relationship between spondylolysis and facet orientation has implications for surgical technique.

One possible explanation for these results could be that our study was underpowered due to a low number of cases (n=17). However, our power calculations using data from the work by Don and Robertson would have predicted 90% power to detect a statistically significant difference in joint orientation at the L4-L5 level, with a type I error probability of 0.05. This leads us to conclude that lack of power is an unlikely explanation for our finding of no significant association.

Another possible explanation for these conflicting results could be the high prevalence of facet joint OA in our sample (64.5%). Several previous studies found that facet joints with OA have significantly increased sagittal orientation [14, 15] as opposed to the increased coronal orientation found in prior studies of spondylolysis [1, 2]. Facet joint OA prevalence in the group of individuals with spondylolysis at the L5 level was significantly higher (at L4-L5 and L5-S1 levels) than in the group without spondylolysis (Table 3). In our study, the mean age of individuals with spondylolysis was 56.29±11.19 years, whereas in the Masharawi et al. [1] study the mean age of individuals with isthmic spondylolisthesis was 47.9±13.8, and in the Don and Robertson [2] study the mean age was 44.5. Albeit we do not have information about the prevalence of facet joint OA in the two previously mentioned studies, we do know that individuals in our sample are much older. Because OA is highly associated with age [16], we can infer that the prevalence of OA may have been higher as in our sample than in the prior studies. Remodeling of the facets due to an OA process in our sample may have led to a situation where facets that were previously more coronally oriented became oriented over time similarly to the facets of individuals without spondylolysis.

There are limitations of the present study that are worthy of mention. First, the cross-sectional design by its very nature limits any conclusions about causal associations. This work, however, serves to challenge the previously asserted association between facet orientation and spondylolysis, as well as to identify the novel concept of how remodeling from degenerative changes may obscure associations between facet orientation and spondylolysis. These concepts can be definitively examined in a future longitudinal study. Second, it should be noted that the CT images utilized for this study were originally obtained for visualization of the abdominal aorta, which is located immediately adjacent to the lumbar spine, and were not dedicated lumbar spine CTs per se. Nevertheless, these CTs provided high-quality imaging of spinal structures ideal for the bony pathology assessed in this study, and similar methods have been used in other published works in the spine literature [12, 17].

Conclusions

We did not find an association between facet orientation and spondylolysis. One of the possible explanations for this is that the high prevalence of facet joint OA in individuals with spondylolysis in the studied sample may have led to diminished differences in facet orientation.

Our finding that individuals with spondylolysis at the L5 spinal level have significantly higher prevalence of facet joint OA at L4-L5 and L5-S1 levels than individuals without spondylolysis, has not been previously reported. Additional studies are needed to confirm such an association and also to elucidate the possible causes for this phenomenon.