Sidebending versus flexion-extension radiographs in lumbar spinal instability

ClinicalRadiology(1994) 49, 109-114

Sidebending Versus Flexion-Extension Radiographs in Lumbar Spinal Instability M. P I T K A N E N and H. M A N N I N E N

Department of Diagnostic Radiology, Kuopio University Hospital, Finland Flexion-extension radiography is accepted as an effective method for the diagnosis of lumbar spinal instability, but the usefulness of sidebending films is less well known. Flexion-extension and sidebending radiographs of 300 patients with clinically suspected lumbar spinal instability were analysed retrospectively. Generally used criteria for lumbar spine instability were applied in the film analysis. Although flexion-extension and sidebending films were statistically significantly interrelated in the diagnosis of instability, intertech,lique agreement remained poor. Flexion-extension films more frequently revealed signs of instability than sidebending films; 84 vs 50 patients. Signs of instability on sidebending films showed the best correlation with the findings of angular motion and posterior sliding instability on flexion-extension films. Sidebending films are complementary to flexion-extension films but are unlikely to be helpful on a routine basis. Pitk/inen, M. & Manninen, H. (1994). Clinical Radiology 49, 109-114. Sidebending Versus Flexion-Extension Radiographs in Lumbar Spinal Instability

Accepted for Publication 24 August 1993

Degenerative changes of the lumbar spine have been divided into three stages: (1) temporary dysfunction; (2) unstable phase; and (3) stabilization [1]. The duration of each stage varies greatly, and there is no clear-cut distinction by clinical signs and symptoms [1]. In the unstable phase, signs of instability may be detected between two consecutive vertebrae. Instability can also occur in the absence of any noticeable degenerative changes on plain films [2]. Although instability cannot be diagnosed by plain films alone, the following findings have been associated with instability: (1) retrolisthesis [2]; (2) traction spur [3]; (3) spondylolisthesis [4]; (4) previous total laminectomy or fusion operation below the motion segment [5]; (5) the presence of gas in the disc [1]; (6) disc space narrowing [6]; (7) facet degeneration [6]; (8) malalignment of the spinous processes at the affected level [6]; and (9) a rotational deformity of the pedicles [6]. Several techniques have been proposed for the radiologic diagnosis of lumbar spinal instability [7-11]. The most thoroughly studied and the most widely used is radiography during flexion and extension of the lumbar spine. For precise measurement of the motion, a biplanar imaging technique with orthogonal projections is required (3-D computer analysis). However, the biplanar technique requires special equipment and more radiographs with a considerable radiation dose [7]. Radiography in the antero-posterior projection during maximal sidebending is an uncommon technique for the diagnosis of instability. Flexion-extension and sidebending radiography can be performed in the upright or supine positions [7,10,12 14]. In flexion-extension radiography, various types of instability have been defined [1,7,14,15] and various quantitative criteria of abnormal motion have been developed [4,6,7,16-18]. However, only a qualitative analysis of the signs of instability has been performed for sidebending views, and there are no data available concerning a comparison of the two techniques. Correspondence to: Dr Marja Pitk/inen,Department of Diagnostic Radiology,Kuopio UniversityHospital, SF-70210, Kuopio, Finland.

The main purpose of this study was to correlate the findings of instability of both flexion-extension and sidebending radiographs. In addition, the association of plain film signs was correlated with instability.

PATIENTS AND M E T H O D S Radiographic Techniques

The study population consisted of 300 consecutive patients admitted for functional lumbar spine radiography for clinically suspected instability based mainly on the following symptoms: (1) central low back pain on a prolonged static weightbearing posture; (2) visible and palpable step in the low lumbar spine; (3) so called instability catch [1,4,5]. The mean age of patients was 42 years (range 13-78 years), 125 (42%) men and 175 (58%) women. Flexion-extension and sidebending radiographs were taken of each patient in the upright position. For flexion-extension examinations, a special apparatus was used to fix the pelvis. Flexion-extension and sidebending radiographs were taken with a large-field image intensifier on 100 x 100 mm photospot film (photo-fluorography), Optilux 57 with a Sircam 100L 100 mm magazine camera (Siemens; Germany) and Agfa Gevaert Scopix RPIS film. This device has a 12:1 stationary grid and an automatic exposure timing control, A fixed-kV technique of 85 kVp and 140 cm focus-to-image-intensifier distance gave a radiation exposure to the patient's skin about half that of full-size radiography [19]. All radiographs were read by a team of two radiologists (MP and HM). The flexion-extension and sidebending photospot films were read with a magnifying glass film viewer, but the quantitative angle and sliding measurements were made on the surface of a spot-film viewer at 3.5 x magnification. Four types of instability findings were recorded for the flexion-extension films: (1) translational forward dis-

110

CLINICAL RADIOLOGY

(a)

(e)

(b)

(e)

Fig. 1 - A 54-year-old female with 2 years low back pain. Flexion-extension films (a, b) reveal a 4 m m anterior displacement and an angular displacement of - 13~ in flexion at the L4-5 level (arrow). In contrast, the sidebending films (c, d) are normal.

SIDEBENDING RADIOGRAPHS IN LUMBAR SPINAL INSTABILITY

111

(a)

(b)

(O

(a)

Fig. 2 - A 34-year-old welder with a history of two lumbosacral disc operations presented with acute back pain. Plain films showed an overcrossing of S 1 and a left-sided hemilaminectomy at L5. Flexion-extension films (a, b) are normal but sidebending films (c, d) show a slight limitation of bending to the left, especially at the L4-5 level (arrow). During sidebending to the right the spinous process of L4 remains at the convex side denoting pathological axial rotation.

112

CLINICAL RADIOLOGY

placement of one vertebra on another (anterior sliding instability) (Fig. 1); (2) backward displacement of one vertebra on another (posterior sliding instability) [1]; (3) excessive angular movement of a motion segment (angular instability) (Fig. 1); (4) pathologic axial rotation in which the back margin of the vertebral body has a local double contour during bending [7], excluding the thoracolumbar region where this phenomenon can occur normally as a result of the projection. A constant double contour of the entire lumbar spine is a consequence of oblique imaging projection. In sidebending films, signs of pathologic axial rotation and asynchronously moving segments were interpreted as signs of instability. During sidebending, normal axial rotation may cause the spinous processes to move from the central line towards the concavity of the curve (i.e. towards the direction of bending) due to a coupling movement. Pathologic axial rotation can be detected if the spinous processes move to the convex side producing an asynchronous spinous process line (Fig. 2). Pathologic rotation can also manifest as a lateral translation of one vertebra on another during sidebending [7]. Additional signs of abnormal movement include asymmetric closure of the disc space on bending, and paradoxical opening of the disc space on the bending side [!]. Translational and angular motion were measured by the method described by Dupuis et al. [7]. The criteria for pathologic movement in flexion-extension films included: (1) translational movement ~>4 mm; (2) axial rotation of a motion segment; (3) angular displacement of more than - 9 ~ in flexion; or (4) flexion-extension angular displacement at least 10~ greater than that in the adjacent segment. In the sidebending films, abnormal axial rotation and asynchronous bending of the motion segments were interpreted qualitatively. All the films were examined for: (1) traction spurs [3]; (2) loss of disc height; (3) spondylarthrosis; (4) retrolisthesis; (5) degenerative spondylolisthesis; and (6) spondylolisthesis with spondylolysis. The McNemar ~2 test (exact Fisher's test if expected frequency < 5) was used to correlate instability findings from the two imaging methods. A P-value of 0.05 or less was considered to be significant. In addition, the statistic kappa (~) was calculated to determine the intertechnique agreement and the intraobserver variation. Kappa is an interclass correlation coefficient that corrects for chance agreements. According to Fleiss [20], a ~ value of less than 0.40 denotes poor agreement; 0.40 0.75, fair to good agreement; and greater than 0.75, excellent agreement. RESULTS Plain Film Findings and Instability

Pathologic plain film findings were present in 180 (60%) cases. The prevalences of spondylarthrosis, retrolisthesis, degenerative spondylolisthesis and spondylolisthesis with spondylolysis were 13% (40/300), 21% (63/ 300), 8% (25/300) and 4% (13/300) respectively. The difference in the prevalences of these findings on bending films in the normal and instability groups was not statistically significant. The presence of disc degeneration and traction spurs (prevalences 36% (108/300) and 16% (47/300)) showed a statistically significant correlation with instability, but individual motion segments differed greatly from each other in this respect. Statistically

Table 1 - Frequency of instability findings in each motion segment Type

Flexion-extensionfilms posterior sliding anterior sliding axial rotation angular motion Sidebending films

Number of positive findings, by level L1-2

L2-3

L3-4

L4-5

L5-S1

All

1 0 0 1 1

3 0 3 3 15

15 5 11 1 23

19 2l 5 12 19

2 2 I 0 2

40 28 20 17 60

significant correlations were found on levels L2-L5 for traction spurs (P ~<0.01) and on levels L2-L4 (P ~<0.001) for disc degeneration. Instability on Flexion-Extension

and Sidebending Films

Signs of instability were present on the bending radiographs of 36% (107/300) of the patients. In flexionextension films, instability was diagnosed in 93 motion segments of 84 patients (Fig. 1), and in sidebending films, in 60 motion segments of 50 patients (Fig. 2). Table 1 shows the frequencies of various types of instability in each motion segment for both imaging techniques. Table 2 shows the intertechnique agreement of flexionextension and sidebending films in revealing signs of instability when all types of instability are considered. Although the techniques show a highly significant statistical correlation, statistic kappa for intertechnique agreement remains poor. For individual motion segments, the imaging techniques showed a statistically significant correlation only at the L2-3 and L3-4 levels (P < 0.001). By statistic kappa, intertechnique agreement was best for L2 3 level (kappa=0.71) followed by L3 4 (kappa =0.26); other levels showed very poor values of kappa ( ~<0.10). When types of instability of flexion-extension were considered individually statistically significant intertechnique correlation was found for angular motion and posterior sliding instability (P < 0.001). Values of statistic kappa for intertechnique agreement were, however, only 0.14 and 0.17 for angular motion and posterior sliding instability. For anterior sliding instability and pathological axial rotation values of kappa were even poorer ( - 0 . 0 2 and 0.12, correspondingly). To determine the researcher's intraobserver variation for film interpretation, radiographs of 15 randomized patients were analysed twice with 3 month intervals between film readings. The intraobserver variation between the first and second film interpretation was good for flexion-extension films (~: = 1) but poor for sidebending films (~ = 0.28). Table 2 - Intertechniqnc agreement of flexion-extension and sidebending films in the diagnosis of instability of the 300 patients Sidebending instability

Flexion-extension instability

+ Total

+

-

Total

27 23 50

57 193 250

84 216 300

P < 0.001, McNemar g2 test, statistic kappa = 0.25.

S I D E B E N D I N G R A D I O G R A P H S IN L U M B A R S P I N A L I N S T A B I L I T Y

DISCUSSION

Diagnosis of Segmental Instability Frymoyer [6] has defined segmental instability as a loss of motion segment stiffness such that force applied to that motion segment will produce greater displacement than would occur in a normal structure. Farfan [21] has verified that clinical instability is a symptomatic condition in which, in the absence of new injury, a physiological load induces abnormally large deformations at the intervertebral joint. A lumbar motion segment is considered to be unstable when it exhibits movement which is either qualitatively or quantitatively abnormal. Instability can be symptomatic or asymptomatic. Roentgenographic evaluation of abnormal spinal movement is difficult. At present we cannot correlate specific symptoms with a specific type of instability [7], and we cannot make a diagnosis of instability solely on the basis of abnormalities on bending films, but must support these findings with a clinical diagnosis. Although some investigators have suggested that the flexion-extension bending films should be questioned as the primary determinant of lumbar segmental instability [15], the criteria used in our study for the radiologic diagnosis of instability are widely accepted but not agreed upon [4,6,7,17,18]. Sidebending films do not tell us about the whole performance of motion. Hence, it is difficult to estimate asynchronous bending only on two films. Two major problems complicate the clinical diagnosis of lumbar segmental instability. The first is the concept of 'three joint complex'. When the posterior joints are involved, the disc is also affected, and vice versa: for example, when the main lesion is in the posterior joints, the disc also may be abnormal and may be a source of pain. Secondly, several quite distinct lesions commonly yield much the same symptom complex [9]. In our study, these difficulties in clinical diagnosis manifested as a relatively low prevalence of radiologic instability. Although all radiographs were taken to substantiate a clinical suspicion of instability, only 28% of the flexionextension films and 17 % of the sidebending films showed radiologic signs of instability (Table 2). In our study, the referring physician was either a physiotherapist, orthopaedic surgeon or neurosurgeon. Studied in retrospect it was not possible to standardize clinical examination.

Plain Film Findings and Instability Knutsson [2] has verified that antero-posterior sliding could be demonstrated long before there was any other radiologic evidence of disc degeneration, such as disc space narrowing. MacNab [3] suggested that the traction spur was significant because it denotes segmental instability, which may or may not cause symptoms. On the other hand, Pate et al. [22] and Johnson [23] have both discussed the nature of the traction spur and suggest that it may be a normal part of the ossification process without any relation to instability and merely a phenomenon caused by physiologic movement. Our study found a statistically significant correlation between traction spurs and instability and between disc degeneration and instability. None of the other plain film findings that we registered showed any significant correlation with instability. A previous

113

history of total laminectomy or fusion has been reported to cause instability [5] but in our study the numbers were too small to draw any firm conclusions.

Instability in Flexion-Extension and Sidebending Films Lumbar instability has been reported to be most common at the L4-5 level and to be rare at L5-S1, because the facets between L5 S1 are normally coronally aligned and thus resist anterior sliding [14]. In our study, instability on flexion-extension films was most usually found at level L4-5 followed by L3-4, but on sidebending films signs of instability occurred most frequently at the L3-4 level, followed by L4-5 and L2-3 (Table 1). This highlights the different nature of instability in the two movement axes. In our study, the relationship of instability on flexionextension films to abnormalities on sidebending films was statistically significant; however, flexion-extension instability was more frequent (Table 2). Altogether, over two thirds (79%) of the instability cases were revealed by flexion-extension films, but only half (47%) by sidebending films. In individual motion segments the correlation between the two imaging techniques varied considerably. Significant correlations were found only at the L2 3 and L3-4 levels. The small number of positive cases at the L1 2 and L5-S1 levels does not permit reliable conclusions to be made at these levels. Only level L2 3 showed excellent intertechnique agreement, while only poor agreement was seen at all other levels. The correlation of signs of instability on flexionextension films with abnormalities on sidebending films was also dependent on the type of instability under consideration. Sidebending instability was found to be most closely related to angular motion and posterior sliding instability, while anterior sliding instability showed practically no association with sidebending instability. The reasons for these differences are not evident to US,

Finally, the radiation dose associated with bending radiography is significant and must be considered when flexion-extension and especially sidebending films are proposed. Sidebending films increase the radiation dose to the genitals as much as 2.8 times and to the bone marrow 1.6 times [24,25].

CONCLUSION We conclude that sidebending films should not be routinely combined with flexion-extension films in the radiologic diagnosis of segmental lumbar instability. Sidebending films complement flexion-extension films and should be taken if sidebending instability is clinically suspected, especially when flexion-extension films are normal. Sidebending radiography could also be used for planning treatment when posterior sliding or angular instability is diagnosed at the levels L2-5 on flexionextension films. On the other hand, when anterior sliding instability is seen on flexion-extension films, sidebending films give no additional information.

Acknowledgement.This study has been financiallysupported by The Northern Savo Regional Fund of the Finnish Cultural Foundation.

114

CLINICAL RADIOLOGY

REFERENCES 1 Kirkaldy-Willis WH, Farfan HF. Instability of the lumbar spine.

2 3 4 5 6 7 8 9 10 11 12 13

14

Clinical Orthopaedics and Related Research 1982;165:110-123. Knutsson F. The instability associated with disc herniation in the lumbar spine. Acta Radiologica 1944;25:593-609. MacNab I. The traction spur. An indicator ofsegrnental instability. Journal of Bone and Joint Surgery 1971;(Am)53:663 670. Nachemson A. The role of spine fusion. Spine 1981;6:306 307. Nachemson A. Lumbar spine instability: a critical update and symposium summary. Spine 1985;10:290-291. Frymoyer JW, Selby DK. Segmental instability: rationale for treatment. Spine 1985;10:280 286. Dupuis PR, Yong-hing K, Cassidy JD, Kirkaldy-Willis WH. Radiologic diagnosis of degenerative lumbar spinal instability. Spine 1985;10:262-276. Friberg O. Lumbar instability: a dynamic approach by tractioncompression radiography. Spine 1987; 12:119-129. Kirkaldy-Willis WH, Hill RJ. A more precise diagnosis for low-back pain. Spine 1979;4:102 109. Pearcy MJ, Tibrewal SB. Axial rotation and lateral bending in the normal lumbar spine measured by three-dimensional radiography. Spine 1984;9:582-587. Putto E, Tallroth K. Extension-flexion radiographs for motion studies of the lumbar spine. Spine 1990;15:107-110. Biering-Sorensen F, Hansen FR, Schroll M, Runeberg O. The relation of spinal X-ray to low-back pain and physical activity among 60-year-old men and women. Spine 1985; 10:445-451Y. Keessen W, Becker THW, Goudfrooij H, Crowe A. Recordings of the movement at the intervertebral segment L5 SI: a technique for the determination of the movement in the L5 S1 spinal-segment by using three specified postural positions. Spine 1984;9:83-90. Morgan FP, King T. Primary instability of lumbar vertebrae as a

15 16 I7 18 19

20 21 22 23 24 25

common cause of low back pain. Journal of Bone and Joint Surgery 1957;(Br)39:6-22. Hayes MA, Howard TC, Gruel CR, Kopta JA. Roentgenographic evaluation of lumbar spine flexion-extension in asymptomatic individuals. Spine 1989;14:327-331. Boden SD, Wiesel SW. Lumbosacral segmental motion in normal individuals: have we been measuring instability properly? Spine 1990;15:571-575. Penning L, Blickman JR. Instability in lumbar spondylolisthesis: a radiologic study of several concepts. American Journal of Roentgenology 1980; 134:293-301. Posner I, White AA, Edwards WT, Hayes WC. A biomechanical analysis of the clinical stability of the lumbar and lumbosacral spine. Spine 1982;7:374 389. Manninen H, Kiekara O, Soimakallio S, Vainio J. Reduction of radiation dose and imaging costs in scoliosis radiography: application of large-screen image intensifier photofluorography. Spine 1988; 13:409-412. Fleiss JL. Statisticalmethodsfor rates and proportions, 2nd ed. New York: Wiley, 1981:212-236. Farfan HF, Gracovetsky S. The nature of instability. Spine 1984; 9:714 719. Pate D, Goobar J, Resnick D, Haghighi P, Sartoris D J, Pathria MN. Traction osteophytes of the lumbar spine: radiographic pathologic correlation. Radiology 1988;166:843-846. Johnson RG. MacNab's traction spur revisited. Neuro-Orthopedics 1989;7:92-93. Johns HE, Cunningham JR. The physics of radiology, 4th ed. New York: Charles C. Thomas, I983:648-653. Kereiakes JG, Rosenstein M. Handbook of radiation doses in nuclear medicine and diagnostic X-ray. Florida: CRC Press Inc, 1980: 185 197.