The Journal of International Medical Research 2006; 34: 316 – 330

Comparative Morphometry of the Lower Lumbar Vertebrae: Osteometry in Dry Bones and Computed Tomography Images of Patients With and Without Low Back Pain T VAROL1, C IYEM1, E CEZAYIRLI1, M ERTURK2, G KAYALIOGLU2 1Department

AND

C HAYRETDAG1

of Anatomy, Faculty of Medicine, Celal Bayar University, Manisa, Turkey; of Anatomy, Faculty of Medicine, Ege University, Izmir, Turkey

2Department

Various factors affect the development of the vertebral canal. The dimensions of the vertebral canal and the intervertebral foramen can be altered by these factors before or after birth. Sex differences in dimensions have also been reported. When there is a stenosis of the vertebral canal or the intervertebral foramen, neural structures confined within them can be affected easily, resulting in symptoms. Using computed tomography images, we compared vertebral canal dimensions in

100 patients with low back pain and/or radiculopathy with those in 40 healthy, non-symptomatic controls. We also measured the dimensions of 275 dry bones. We found significant correlations among the variables in the live subjects. We found significant differences between patients and controls in the variables that were measured. Stenoses were more prevalent in females. Dry bone measurements showed some sex differences, and stenosis mainly in vertebrae L4, L5 and S1.

KEY WORDS: LOW BACK PAIN; LUMBAR VERTEBRAE; ALLOMETRY; OSTEOMETRY; COMPUTED TOMOGRAPHY; SPINAL STENOSIS; SEX DIFFERENCES

Introduction Spinal stenosis, an important orthopaedic and neurological disease of the vertebral column, is defined as congenital or acquired pathological reduction of the osteoligamentous spinal canal area which leads to compression of the dural sac and/or spinal

nerve roots.1 – 4 Spinal stenosis can be classified as: (i) central, when it affects the spinal canal; (ii) foraminal, when it affects the intervertebral foramen; and (iii) lateral, when it affects the lateral recesses.5 – 7 The size of the spinal canal is determined by different factors. The phenotypic

316

T Varol, C Iyem, E Cezayirli et al. Bone morphometry in lower back pain expression of genetic factors can be altered by direct local injury (trauma, infection, etc.) and the effects of systemic disturbances (malnutrition, cardiovascular illness, etc.) on the spinal column during its development.8 – 11 Degenerative alterations are more frequent after the spinal development has ended, and therefore it is possible to differentiate these according to whether they are congenital or developmental alterations.12,13 The most significant and frequent causes of lumbar spinal stenosis are the degenerative diseases of the discs and articular processes (facets), and hypertrophy of the ligamentum flavum.1,14 To establish the diagnosis of spinal stenosis, the best method is computed tomography (CT) with myelography, but plain CT or magnetic resonance imaging can also be used with good results.5,15,16 Criteria described in the literature for the diagnosis of central lumbar spinal stenosis are: (i) symptoms and signs;3,4 and (ii) reduction in the anteroposterior diameter (sagittal diameter of the vertebral foramen; SAG) of the spinal canal at the level of stenosis and the corresponding reduction in the spinal canal area as estimated by CT, without taking into account the patient’s height and weight.5,6,16,17 The intervertebral foramen (IVF) has been considered to be the lateral portion or exit zone of the lateral canal;18 – 20 it has also been described as a distinct anatomical entity, a true canal with definite boundaries,21 differing from the lateral canal, which is an anatomical portion of the spinal canal that lacks definite anatomical boundaries. Cinotti et al.22 argued that the IVF should be considered an osteofibrous canal delimited by bone and soft tissues. Its superior and inferior boundaries are the pedicles of the adjacent vertebrae. Its anterior boundaries include the posteroinferior margin of the upper vertebral body, the intervertebral disc, and the postero-

superior margin of the lower vertebral body. The posterior boundaries are formed by the interarticular part, the superior facet joint, and the ligamenta flava lying adjacent to it.22 The IVF is a region where the nerve roots and mixed spinal nerves are particularly vulnerable to compression resulting from a variety of pathologies that can restrict foraminal width or height.23,24 The superior sagittal distance and the height of the IVF can be affected by disc-related problems (disc hernia, protrusion, bulging, etc.) and by posture and weight.22 The allometric relationship between two diameters can be calculated from ontogenetic data (measurements on individuals of different ages) or from static data (measurements on individuals of similar age but different size).25 This possibility prompted us to take a series of measurements in the lumbar region of adult patients suffering from low back pain and sciatica of mechanical origin, and of adult controls without a history of lumbar pain. This was a two-part study. In the first part, morphometric measurements were carried out on CT images of subjects with or without disc pathology and/or radiculopathy to reveal if, and how, anatomical differences play a role in the presentation of the symptoms. In the second part of the study we documented the biological variation in the measurements of dry L3, L4, L5 and S1 vertebrae. The information presented in this study about the normal biological variability of these measurements may create a basis for further studies.

Materials and methods RADIOGRAPHIC MEASUREMENTS After a review of clinical charts to exclude tumoral, inflammatory and traumatic low back pain, male and female patients suffering from mechanical low back pain and/or radiculopathy were selected. We also studied

317

T Varol, C Iyem, E Cezayirli et al. Bone morphometry in lower back pain a control group of males and females with no history of back pain requiring medical treatment. There was no disc pathology in the control group. The aims and method of the study were explained to the participants and informed consent was obtained from them. Celal Bayar University Ethics Board approved the study. The CT images were acquired with the subject in the supine position, and covered the area from the pedicle of L3 to the foraminal level of S1. With a slice thickness of 5 mm, images were obtained at a tube current of 90 mA and a tube voltage of 120 kV using a Hitachi-550 CT scanner (Hitachi Medical, Tokyo, Japan). The slices were acquired with the gantry inclined parallel to the intervertebral disc. All measurements were performed on the CT scanner console. L4, L5, the base of the sacrum (S1) and intervertebral disc spaces were used for measurements because more than 90% of cases of lumbar disc pathology are seen at levels L4 – L5 and L5 – S1.26 – 28 The value of SAG, the interpedicular distance of the vertebral foramen (IPD), the right and left lateral recesses (RLR and LLR, respectively) and the interfacet distance (IFD) were measured according to guidelines described elsewhere.29 – 31 Comparisons were made between patients and controls, with sexes pooled and separately for males and females. All radiological measurements were made by two of the authors blind to the clinical diagnosis.

OSTEOMETRY We made measurements on adult male and female dry bones without deformities. The dry bones were selected from the bone collection of the Department of Anatomy, Ege University, Izmir, Turkey. They were obtained from individuals whose background was known, i.e.

persons with no spinal abnormalities or diseases. Vertebrae were separated according to gender (male and female) or level (L3, L4, L5, sacrum). Measurements were carried out using a calliper with a precision of 0.01 mm. SAG, IPD, RLR and LLR, the distance between the superior articular processes (SAD), the distance between the inferior articular processes (IAD), and the sagittal distances of the superior and inferior vertebral notches (SVN and IVN, respectively) were measured. Figure 1 illustrates how these measurements were made. The SAG was measured between the superior border of the lamina and the posterior cortex of the vertebral body on the midline. A SAG value between 11 and 12 mm was regarded as relative stenosis and a value less than 11 mm as absolute stenosis32 – 34 (Fig. 1A and 1C). The IPD was measured as the horizontal distance between the inner cortices of the vertebral pedicles, from the superior endplate. IPD less than 16 mm was regarded as stenosis17 (Fig. 1A and 1C). The lateral recess diameter (LRD) was the distance between the posterior aspect of the vertebral body and the anteromedial point of the facet joint. LRD between 3 and 5 mm and LRD less than 3 mm were taken as relative and absolute stenosis, respectively32,33,35,36 (Fig. 1A and 1C). The IFD was the distance between facet joints at the level of the mid-intervertebral disc. IFD was measured at levels L4 – L5 and L5 – S1. IFD less than 15 mm was taken as stenosis37 – 40 (Fig. 1B). The SAD was measured as the horizontal distance between the inner edges of the superior articular processes at the mid-point of the superior articular facet (Fig. 1C). The IAD was the horizontal distance between the inner edges of the inferior articular processes at the mid-point of the inferior articular facet.

318

T Varol, C Iyem, E Cezayirli et al. Bone morphometry in lower back pain

FIGURE 1: Illustrations demonstrating how the measurements were performed. (A) LRD, lateral recess diameter; SAG, sagittal diameter of the vertebral foramen; IPD, interpedicular distance of the vertebral foramen. (B) IFD, interfacet distance. (C) SAD, distance between the superior articular processes. (D) SVN, sagittal distance of the superior vertebral notch; IVN, sagittal distance of the inferior vertebral notch

The SVN was measured as the distance between the posterior cortex of the vertebral body and the origin of the superior articular process from the pedicle (pedicle length) (Fig. 1D). The IVN was the distance between the posterior cortex of the vertebral body and the origin of the inferior articular process from the pedicle (Fig. 1D).

STATISTICAL ANALYSIS The allometric relationships between different measurements of the vertebral canal and dry bones were explored using Pearson’s correlation. Differences between radiological and osteometric measurements with respect to sex, laterality and the vertebrae were tested using the independent samples t-test and one-way analysis of variance. SPSS for Windows version 11.0 (SPSS, Chicago, IL, USA) was used for

statistical analysis and a P-value < 0.05 was considered significant.

Results Data were obtained on a total of 100 patients (50 males and 50 females) suffering from mechanical low back pain and/or radiculopathy. Mean age and range were 42.3 and 25 – 72 years, respectively for females and 39.8 and 25 – 67 years, respectively for males. Sixty-nine of the patients suffered from disc hernia, 34 from bulging and one from spondylolisthesis (with or without disc hernia). The control group comprised 40 subjects (20 males and 20 females), none of whom had any history of back pain requiring medical treatment. There was no disc pathology in the control group. A total of 275 adult dry bones (206 vertebrae and 69 sacrums) without deformities were studied. We had a total of 36 L3, 63 L4 and

319

T Varol, C Iyem, E Cezayirli et al. Bone morphometry in lower back pain 26 L5 vertebrae and 37 sacrums from males and 32 L3, 25 L4 and 24 L5 vertebrae and 32 sacrums from females. Our first study was carried out to assess repeatability (intra-observer variability) and reproducibility (inter-observer variability) of the measurement. To assess inter-observer variability, all variables were measured by two authors on 10 randomly selected dry vertebrae and 10 CT images. As an index of reliability, we used the intra-class correlation (ICC), which combines information from order and between-observation distance and from the discrepancy between raters.41,42 The ICCs between the two authors who performed the measurements and the coefficients of variation between the measurements obtained by the two authors are given in Table 1. To assess intra-observer variability, measurements on the 10 vertebrae and 10 CT images were repeated a week later. The

coefficients of variation between the measurements and the ICC coefficients are presented in Table 1. The ICC coefficients for repeatability and reproducibility were generally good (> 0.800), showing that there was good agreement between the two authors’ measurements as well as consistency within each author. The measurements on dry bones and CT images, their means, maximum and minimum values, and the SDs are tabulated separately for males and females in Tables 2 and 3, respectively. Pearson’s product moment correlation was used to examine the relationship between individual parameters and test the significance of each correlation. Measurements on CT images revealed that SAG, IPD, RLR and LLR measurements at the level of L4 correlated positively with each other (range 0.475 – 0.979; P < 0.01), as did those at

TABLE 1: Coefficients of variation and intra-class correlation (ICC) coefficients between the two authors’ measurements (inter-observer reliability) and the intra-observer reliability, measured using an ICC coefficient derived from repeated measurements under blind conditions and the coefficients of variation Inter-observer

Intra-observer

Coefficient of variation

ICC

Coefficient of variation

ICC

SAG

5.3%

0.956

6.9%

0.967

IPD

3.9%

0.990

4.9%

0.950

LRD

9.6%

0.890

13.4%

0.848

IFD

6.2%

0.961

6.7%

0.916

SAD

4.9%

0.950

5.4%

0.980

IAD

6.7%

0.925

7.8%

0.920

SVN

10.2%

0.842

13.8%

0.834

IVN

7.0%

0.910

7.7%

0.941

Measurement

SAG, sagittal diameter of the vertebral foramen; IPD, interpedicular distance of the vertebral foramen; LRD, lateral recess diameter; IFD, interfacet distance; SAD, distance between the superior articular processes; IAD, distance between the inferior articular processes; SVN, sagittal distance of the superior vertebral notch; IVN, sagittal distance of the inferior vertebral notch.

320

321

S1 (n = 37)

L5 (n = 26)

L4 (n = 63)

L3 (n = 36)

Males

Vertebra

15.92

Mean

15.60

Mean

16.46

Mean

15.28

Mean

2.45

21.40

Max

SD

10.32

Min

2.36

21.76

Max

SD

12.56

Min

2.03

20.60

Max

SD

11.38

Min

2.01

18.80

Max

SD

9.62

Min

SAG

2.91

31.02

38.52

24.38

3.09

26.14

32.56

21.46

1.91

23.22

29.26

18.96

1.30

23.00

25.82

19.76

IPD

1.06

6.35

8.42

4.04

0.95

5.76

7.84

4.28

1.46

6.90

10.10

3.64

1.46

8.81

11.92

6.08

RLR

1.22

5.95

8.26

3.96

1.09

5.39

7.44

3.42

1.47

6.41

9.80

3.50

1.60

8.50

11.28

5.10

LLR

TABLE 2: Descriptive statistics of male and female dry vertebral bones (mm)

4.36

27.73

36.78

18.48

3.82

26.30

33.88

18.10

3.35

23.29

34.00

14.30

3.70

23.16

32.76

18.38

SAD

4.94

33.17

40.36

23.84

4.81

22.17

38.82

13.66

4.49

19.94

31.40

12.96

IAD

0.76

4.47

6.36

3.14

0.83

4.15

6.46

3.00

0.95

4.66

6.60

2.76

1.08

5.97

8.54

4.24

RSVN

0.61

4.23

5.52

3.18

0.71

3.83

5.30

2.54

1.12

4.46

8.30

2.46

1.01

5.79

8.30

3.96

LSVN

1.42

8.07

10.80

4.90

1.36

8.46

11.98

6.18

1.69

9.73

13.74

6.60

RIVN

1.42

7.63

10.10

4.20

1.21

8.63

12.50

6.00

1.50

9.87

13.50

7.08

LIVN

T Varol, C Iyem, E Cezayirli et al. Bone morphometry in lower back pain

322

15.79

Mean

14.75

Mean

15.30

Mean

13.97

Mean

2.48

19.88

Max

SD

8.78

Min

1.73

17.78

Max

SD

11.58

Min

1.68

17.56

Max

SD

11.32

Min

1.86

20.80

Max

SD

13.18

Min

SAG

3.09

30.51

36.24

22.52

2.51

24.11

29.18

20.46

1.65

21.76

25.08

18.74

1.73

22.44

25.00

17.90

IPD

1.17

5.93

8.08

3.94

1.10

5.72

8.00

4.00

0.98

6.58

8.54

5.00

1.19

8.28

11.02

6.40

RLR

1.21

5.76

8.72

3.40

1.03

5.20

7.84

3.50

1.10

6.25

8.14

4.12

1.20

8.05

10.34

6.20

LLR

3.92

27.29

34.00

20.46

2.85

24.17

29.16

17.84

2.57

22.42

28.24

16.36

3.12

21.97

27.02

12.84

SAD

3.83

31.17

37.16

20.10

3.21

21.74

30.08

15.98

3.46

19.06

26.78

12.74

IAD

0.81

4.26

5.82

2.72

0.68

4.04

5.50

2.84

0.74

4.09

5.50

2.96

1.07

5.48

7.48

3.18

RSVN

0.75

4.36

6.60

2.94

0.61

3.72

4.88

2.52

0.84

4.14

5.98

2.76

1.08

5.43

7.88

3.90

LSVN

1.13

8.01

10.74

5.72

1.28

8.08

10.12

5.58

1.72

9.20

13.42

6.56

RIVN

1.24

7.82

9.12

4.10

1.04

8.06

10.24

6.32

1.68

9.07

13.10

6.08

LIVN

SAG, sagittal diameter of the vertebral foramen; IPD, interpedicular distance of the vertebral foramen; RLR, right lateral recess; LLR, left lateral recess; SAD, distance between the superior articular processes; IAD, distance between the inferior articular processes; RSVN, sagittal distance of the right superior vertebral notch; LSVN, sagittal distance of the left superior vertebral notch; RIVN, sagittal distance of the right inferior vertebral notch; LIVN, sagittal distance of the left inferior vertebral notch.

S1 (n = 32)

L5 (n = 24)

L4 (n = 25)

L3 (n = 32)

Females

Vertebra

TABLE 2 (continued): Descriptive statistics of male and female dry vertebral bones (mm)

T Varol, C Iyem, E Cezayirli et al. Bone morphometry in lower back pain

323

Min Max Mean SD Min Max Mean SD Min Max Mean SD

Min Max Mean SD Min Max Mean SD Min Max Mean SD

13.4 20.6 17.16 1.82 16.3 24.5 18.68 2.13

13.5 22.2 17.15 2.23 14.2 22.2 18.61 2.05

SAG

22.0 30.2 26.17 2.62 23.7 33.5 28.40 2.12

20.4 29.1 25.86 2.47 24.1 33.3 28.24 2.39

IPD

4.5 7.8 6.11 0.72 4.1 7.4 6.32 0.81

4.6 7.8 6.23 1.06 4.4 9.2 6.53 1.19

RLR

4.6 7.8 6.17 0.71 4.2 7.5 6.50 0.79

4.6 7.9 6.29 0.87 4.5 9.3 6.59 1.17

LLR

19.8 27.3 22.92 2.33

20.1 2.7 23.70 1.76

21.3 32.3 26.55 2.64

24.2 31.5 27.67 2.42

L4 – L5 L5 – S1

9.3 18.5 13.33 2.40 11.1 22.2 15.96 2.52

11.1 18.5 13.94 1.88 11.1 22.2 16.31 2.17

SAG

14.7 25.7 19.02 3.06 15.4 28.7 22.45 3.38

13.2 27.6 20.96 3.44 16.7 31.45 24.95 2.91

IPD

3.7 9.3 5.15 1.28 3.7 7.4 5.75 1.03

3.7 9.3 5.35 1.13 3.7 9.25 5.97 1.13

RLR

3.7 9.3 5.18 1.29 3.7 7.4 5.78 1.03

3.7 7.4 5.3 0.99 3.7 7.4 5.95 1.04

LLR

Patients

13.0 24.1 18.26 3.63

14.8 25.9 20.15 2.51

11.1 29.5 23.92 3.94

16.6 31.45 25.65 3.10

L4 – L5 L5 – S1

SAG, sagittal diameter of the vertebral foramen; IPD, interpedicular distance of the vertebral foramen; RLR, right lateral recess; LLR, left lateral recess; IFD interfacet distance.

IFD

L5

Females L4

IFD

L5

Males L4

Vertebra

Controls

TABLE 3: Descriptive statistics from computed tomography (CT) images of vertebral bones of male and female patients and controls (mm)

T Varol, C Iyem, E Cezayirli et al. Bone morphometry in lower back pain

T Varol, C Iyem, E Cezayirli et al. Bone morphometry in lower back pain the level of L5 (range 0.347 – 0.989; P < 0.01). IFD measurements at the level of L4 – L5 correlated positively with those at the level of L5 – S1 in all groups (range 0.646 – 0.828, P < 0.01). Radiological measurements of the lumbar vertebrae were not significantly correlated with either subject age or subject height. The SAG, IPD, right SVN (RSVN), left SVN (LSVN), right IVN (RIVN) and left IVN (LIVN) measurements were made and except for IPD, for which there was a weak negative correlation with RSVN and RLR (P < 0.05), other measurements showed positive correlations with each other (range 0.239 – 0.898; P < 0.01). There were high correlations (P < 0.01) between RLR and LLR (range 0.492 – 0.717) and between SVN and IVN (range 0.520 – 0.661). Measurements of the lateral canal components (RLR, LLR, RSVN, LSVN, RIVN and LIVN) showed positive correlations in both male and female bones (range 0.460 – 0.875 and 0.464 – 0.802, respectively; P < 0.01).

DIFFERENCES BETWEEN PATIENT AND CONTROL GROUPS On CT images, when the patient and control groups were compared, separately for males and females and for males and females together, we found significant differences between patient and control groups in SAG, IPD, RLR and LLR (Tables 4 and 5). When the measurements of male and female patients were compared, we found significant differences in IPD at L4 (P < 0.05) and L5 levels (P < 0.001) and a significant difference in IFD at L4 – L5 and L5 – S1 levels (P < 0.05) (Table 6). In control subjects there were no significant differences between sexes with respect to SAG, IPD, RLR or LLR at levels L4 and L5, or to IFD at levels L4 – L5 or L5 – S1 (Table 7). Measurements on dry bones (Table 8) did not reveal sex differences at the L3 level in any

of the parameters except for LIVN (P < 0.05). At the L4 level, on the other hand, differences between males and females were significant for SAG (P < 0.001), IPD (P < 0.001), RLR (P = 0.001), LLR (P = 0.001), RSVN (P < 0.01) and LIVN (P < 0.05). At the L5 level SAG (P < 0.001), IPD (P < 0.001), RLR (P < 0.001), LLR (P < 0.001) and SAD (P < 0.05) significantly differed between males and females, whereas at the S1 level only SAG (P < 0.05) showed a significant sex difference.

STENOTIC FINDINGS OF THE RADIOLOGICAL AND DRY BONE MEASUREMENTS On the basis of SAG measurements on CT images, four male patients (8%) had relative, five female patients (10%) had absolute and 10 female patients (20%) had relative stenosis. No stenosis was observed in the male and female controls. According to lateral recess measurements, 11 male patients had bilateral relative stenosis (22%) and 20 female patients (40%) had bilateral relative stenosis. Among the controls, only one case in each sex group (5%) had bilateral relative stenosis. The IPD measurements revealed stenosis in two male patients (4%) at the L4 level, one of which was relative stenosis of the lateral recess, whereas there was no stenosis at the L5 level. In female patients, eight cases (16%) had stenosis at the L4 level; five of these stenoses were relative. No stenosis was observed in the control subjects. The IFD measurements revealed stenosis in one male patient (2%) at the L4 – L5 level, whereas among female patients 12 cases (24%) had stenosis at the L4 – L5 level and one case (2%) at the L5 – S1 level. No stenosis was found in control groups with respect to IFD. In the study of dry bone measurements we identified two cases of absolute stenosis (1.23%, one L3 and one S1) and six cases of

324

T Varol, C Iyem, E Cezayirli et al. Bone morphometry in lower back pain

TABLE 4: Comparison of measurements of vertebral bones in patients and controls (males and females separately) on computed tomography (CT) images (mm) Males Measurement

Patients n = 50

Controls n = 20

Females P-value

Patients n = 50

Controls n = 20

P-value

L4 SAG IPD RLR LLR

13.94 20.96 5.35 5.30

± ± ± ±

1.88 3.44 1.13 0.99

17.15 25.86 6.23 6.29

± ± ± ±

2.23 2.47 1.06 0.87

< 0.001 < 0.001 0.002 < 0.001

13.33 19.02 5.15 5.18

± ± ± ±

2.40 3.06 1.28 1.29

17.16 26.17 6.11 6.17

± ± ± ±

1.82 2.62 0.72 0.71

< < <