SciELO - Scientific Electronic Library Online

vol.10 issue3Evolution of representation of the hands in plastic artsMechanical comparative analysis of three different types of sutures in different dimensions tendons: experimental study author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Acta Ortopédica Brasileira

Print version ISSN 1413-7852On-line version ISSN 1809-4406

Acta ortop. bras. vol.10 no.3 São Paulo July/Sept. 2002 

Anatomical study of lumbar vertebral pedicle and adjacent neural structures


Estudo anatômico do pedículo vertebral lombar e estruturas neurais adjacentes



Cláudia Maria MatuokaI; Roberto Basile JúniorII

IIDoctor in Orthopaedics

Address for correspondence




For the evaluation of the Lumbar pedicle morphometry and its relation to the neural structures, 14 male adult cadavers were dissected, and the size of the lumbar pedicle was assessed by measuring its sagittal and transversal diameter. It was found that the size of the pedicle increases from L2 to L5, both in the sagittal and transversal diameter, the first bigger. The relation of the lumbar pedicle to the neural structures was evaluated by measuring the distance between dura-mater and the pedicle medial area, the distance between the most distal area of the pedicle and the nerve root that appears under it, and , to obtain in an indirect way, the distance between the pedicle apex and the nerve root that appears over it. The acquired results showed that the distance between the most distal area of the pedicle and the nerve root that appears under it, and the distance between the pedicle medial area and dura-mater, do not increase from L2 to L5, and they are in average 1,98 and 3,02 respectively. The distance between the pedicle apex and the nerve root that appears over it, increases from L2 to L5, varying from 13,64 in L2 to 21,62 in L5. The location of the spinal ganglion in relation to the pedicle has also been found, and 87% of the spinal ganglions are located in the foraminal zone.

Key words: Anatomy; lumbar spine; nerve root; pedicular screw placement


Para avaliação da morfometria do pedículo vertebral lombar e sua relação com as estruturas neurais, foram dissecados 14 cadáveres do sexo masculino, adultos, onde foram avaliados o tamanho do pedículo vertebral lombar, através da mensuração do seu diâmetro longitudinal e transversal. Os dados obtidos mostraram aumento do pedículo de L2 a L5, tanto no diâmetro longitudinal quanto transversal, sendo o primeiro maior. A relação do pedículo lombar com as estruturas neurais, foram avaliadas através da mensuração da distância da dura-máter à borda medial do pedículo, entre a região mais distal do pedículo à raiz nervosa que emerge abaixo deste e, de maneira indireta, a distância entre o ápice do pedículo à raiz nervosa que emerge acima deste. Os resultados obtidos mostraram que a distância entre a parte mais distal do pedículo e a raiz nervosa que emerge abaixo deste, e a distância entre a borda medial do pedículo à dura-máter, não aumentam de L2 a L5. Quanto à distância entre o ápice do pedículo à raiz nervosa que emerge acima, aumenta de L2 a L5. Localizamos também o gânglio espinhal em relação ao pedículo, sendo que 87% dos gânglios espinhais localizam-se na zona foraminal.

Descritores: Cadáver; vértebras lombares; rede nervosa; dura-máter; gânglios/espinhais




The pedicular method of fixation of lumbar spine for different diseases is progressively being more used in the last decades. This method was created due to the high index of pseudarthrosis after arthrodesis without fixation of lumbar spine. Several authors(22,25) found 5 to 20% of pseudarthrosis after posterior-lateral arthrodesis. Thompson et al(24) evaluated several patients who underwent Hibbs' method arthrodesis and found an incidence of 55% of pseudarthrosis when performed from L4 to S1 and 12% from L5 to S1, that is, the higher the number of involved vertebrae, the higher the incidence of pseudarthrosis. Besides this, pedicular fixation grants great stability, avoiding the postoperative use of external immobilization.

According to Aebi et al(1) , treatment of any lumbar spine pathology should include: 1. effective decompression of spinal canal. 2. treatment of the spine without causing deformities, movement limitation, instability or pain, 3. early mobilization and simplified nurse care.

With arrival of pedicular fixation, spinal surgery got to reach the objectives mentioned by Aebi et alli. (1988) and Magerl et alli. (1984). This technique would be indicated in all cases with spine instability, that is, fractures, pseudarthrosis, degenerative changes, treatment of tumors, spondilolistesis from several causes.

The problem with pedicular fixation is the high index of complications due to the surgery, or the placement of the screw itself in the vertebral pedicle. Complications related to the surgery are, among others, infection, whether superficial or deep, pulmonary tromboembolism and several other complications linked to important surgeries. Those related to pedicular screw, reported by several authors(3,11,13,15,16,17,20,23,27,28), are the placement of the screw out of the pedicle, what may produce radicular injury, dura-mater injury, vascular injury(26), and pedicle fracture. Besides this, it can happen the implant to cleave. Esses et al(7) in a review of the complication found in analyzing several techniques applied by several authors an index of complications of 27.4%, being 9.6% of them intra-operative, and 17.8% postoperative. They did not find statistical differences among the different implants used. The incidence of complications is higher in patients with previous surgery.

Regarding the frequency, rupture of the screw is the most important, with an index ranging from 3 to 25%. This complication could be avoided with improvement of biomechanical knowledge. Regarding morbidity, radicular injury is the most important, ranging from 0 to 7%. This complication could be avoided with an improved knowledge of the morphometry of the vertebral pedicle, and its relation to neural structures.

Several authors(9,12,14,18,29) studied the morphometry of the vertebral pedicle both in cadavers and imagery examinations. These papers showed the lumbar vertebral pedicle to increase its diameter from L1 to L5 being the transversal axis smaller than the sagittal one, thus, the larger diameter of the screw should be determined by the first one. They also showed that the axis of the lumbar pedicle in the transverse plane ranges from 0-10° in L1, 10-20° from L2 to L4, and 20-30° in L5. At the pedicle axis, the distance between the lamina and the anterior part of the vertebral body increases 8 to 10 millimeters. Thus, the screw could be placed in a safest manner if placed in the axis of the pedicle. Krag et al(14) also found that the pedicle length in transverse axis is 40 to 46% of the distance from the lamina to the anterior part of the vertebral body. This measure would limit the minimum length of the screw.

The vertebral pedicle has an intimate relationship with neural structures, since it is part of the intervertebral foramen, as described by many authors(5,19). The intervertebral foramen has as its upper and inferior limit the pedicle of the vertebrae above and below, respectively. The floor is made by the posterior-inferior edge of the vertebral body above, the intervertebral disc, and the posterior-superior border of the vertebra below. The ceiling is formed by the yellow ligament. The pedicle is laterally placed in relation to the dural sac. The closest lumbar roots appears in an almost straight angle in relation to the dural sac, being shorter than the more distal ones. In its path inside the neural canal it is involved by epidural fat. Rauschning(19) reports that the involving structure of the lumbar roots are inferiorly anchored to the subjacent pedicle, limiting the medial and superior mobility of the root. Besides this, several papers(9,21) showed that the spinal ganglion has a preferentially intra-foraminal location, with exception of S1 that has a preferentially intra-spinal location. Thus, the spinal ganglion has an intimate relationship to the lumbar pedicle.

Through studies based on cadaver dissection, concluded(6) that the average distance between the pedicle and the inferior lumbar root is of 1.5 mm. Between dura-mater and the pedicle, it is also 1.5 mm. However between the pedicle and the nervous root above, it is in average 5.3 mm. Thus the medial and inferior placement of the pedicular screw could cause neural structure injuries. Also through dissection of cadavers studied the lateral zone of the lumbar pedicle, finding that this region has similar characteristics as the medial and inferior one(8).

The aim of this work was the anatomical study of the lumbar vertebral pedicle and its relationship to adjacent neural structures, objecting a better knowledge of this region.



Lumbosacral spines of 14 adult human fresh cadavers were dissected. All were male, died from a variety of causes, with no evidence of abnormalities of the lumbar spine. Dissection was performed under direct macroscopic vision.

Were evaluated 100 vertebral pedicles, being 28 from L2, 28 from L3, 28 from L4 and 16 from L5. The L5 pedicles were in less number because in 6 cadavers it was found a sacralization of the last lumbar vertebra. Even though these pedicles had their measures taken, they were excluded.

Dissections were performed at the Serviço de Verificação da Capital da Faculdade de Medicina da Universidade de São Paulo (SVOC - FMUSP), in 1998. (Board 1)



Age ranged from 28 to 75 years, average 52 years, and median 49 years. (Board 2).



Regarding race, eight were white, and 6 black. (Board 3).



The protocol of this work was approved by the Medical Ethics Committee from the Universidade de São Paulo.



Fourteen cadaver dissections were performed through posterior approach, being possible to analyze the lumbar canal, dura-mater, appearance of lumbar roots and the bases of the vertebral pedicles.

These dissections were performed with the corpses in ventral decubitus, and the incision performed from L1 to the sacrum. Soft tissues were dissected in a block, allowing to observe the posterior bony elements of the lumbar spine. Were removed the spinous processes and the yellow ligaments from L1 to S1, laminae, and were severed the pedicles in the same plane of the neural structures. The final anatomical aspect is shown in (Figure 1).



Measurements were performed for morphometric evaluation of the lumbar vertebral pedicle. This assesments were performed with a digital slide caliper, with a precision of 0.01 millimeter and a surgical magnifying lens with a 2.5 fold magnification. Were assessed (Figure 2):

1. LP ¾ (pedicle width), determined by the transversal axis of the pedicle.

2. CP ¾ (pedicle length), determined by the sagittal axis of the pedicle.

3. DDMP (distance from dura-mater to the pedicle), determined by the sagittal axis of the pedicle.

4. DNP (distance from the pedicle to the root below), is the distance from the most distal point of the pedicle to the root below.

5. CNE (length of the spinal nerve) is the measurement of the spinal nerve, from its appearance from dura-mater to the upper vertebral plateau of the vertebra below.

6. CDM (dura-mater length), is the length from the lateral edge of the dura-mater, extending since the appearance of the upper root to the upper vertebral plateau of the vertebra below.

7. DDMN (distance from dura-mater to the nerve) is the distance between the lateral edge of the dura-mater, at the upper vertebral plateau of the vertebra below to the medial edge of the spinal nerve, in the transversal axis.



From the works(4), it is known that the CNE, CDM and DDMN form a rectangle triangle called by these authors as "triangular safety zone", being CNE the hypotenuse (Figures 3 A and B). It was possible to calculate the angle of the root (angle ß) through the calculation of the tangent arch of a leg (CDM) divided by the other leg (DDMN). With this it was possible to calculate the angle between CNE and CDM (alpha angle), subtracting 90 from the appearance angle of the root. The need to know these parameters was for the calculation of the distance between the apex of the pedicle and the root appearing above it. As the distance between the dura-mater and the apex of the pedicle (DDMP) was known, this distance was subtracted from DDMN. With this, we constructed a new rectangle triangle having as legs the distance between the apex of the pedicle and the root above (DRAP) and the subtraction of DDMN and DDMP (Figure 4 A and B). As the angle between this last leg (DDMN-DDMP), that is the alpha angle, is known, it was possible to calculate the other leg (DRAP) by multiplying the known leg for the tangent to the alpha angle.




It was also observed the location of the spinal ganglion in relation to the vertebral pedicle (Figure 5). Zones were placed in order to a better standardization of location of the spinal ganglion. The limits of these zones were the lateral and medial edges of the pedicle, and a line dividing sagittally the pedicle into two parts, medial and lateral. Thus, we got four zones that were called:

1. GS1 ¾ intra-spinal zone, medially delimited by dura-mater, and laterally by the medial edge of the pedicle.

2. GS2 ¾ medial foraminal zone, medially delimited by the medial edge of the pedicle and laterally by the sagittal line dividing the pedicle.

3. GS3 ¾ lateral foraminal zone, medially delimited by the sagittal line dividing the pedicle, and laterally by the lateral edge of the pedicle.

4. GS4 ¾ extra-foraminal zone, medially limited by the lateral edge of the pedicle.



A descriptive statistics of the quantitative parameters was performed, calculating average, standard deviation, standard error of the mean, maximum and minimum value and number of cases in the sample.

A frequency distribution and proportion for description of the qualitative parameters was performed (absolute and relative frequencies).

In global comparison between the levels, it was used a variance analysis for parametric samples, and the Friedman's test for non-parametric samples.

For discrimination among the groups (levels) it was used the "t" paired test for parametric samples, and Wilcoxon test for non-parametric samples.

In all cases it was adopted a significance level of 5% (a = 0.005).



The results found in the evaluation of 100 lumbar vertebral pedicles, and their relationship to neural structures, is reported in the table below (for the Tables 1 to 6: L: level, average, DP: standard deviation, EP: standard error, Min: minimum value, Max: maximum value, N: number of cases):










Pedicular lumbar fixation is being increasingly used in treatment of lumbar spine pathologies generating instability, needing arthrodesis. The aim of instrumentation would be to reduce the incidence of pseudarthrosis following posterior or posterior-lateral arthrodesis.

For this study were selected only adult cadavers, with an average age of 52 years, ranging from 28 to 78 years old, since the majority of lumbar spine instability generating pathologies are from degenerative or development ones (spondilolistesis, tumors, fractures), being more frequent in skeletically mature individuals. Besides this, all works we evaluated for preparation of this dissertation dealt with adult population.

Were performed assessments for morphometric evaluations of the lumbar vertebral pedicle, that consisted in the transversal diameter (LP) and the sagittal one (CP), since lumbar vertebral pedicle has an elliptical form, with a sagittal diameter that is larger than the transversal. Once pedicle's size is known, it would be possible to evaluate if the pedicular screws would have diameters that could be considered as safe.

Were also performed measurements aiming to evaluate the relationship between the vertebral pedicle and neural structures (dura-mater, nerve roots). The distance between the dura-mater and the lumbar pedicle was assessed between the lateral edge of the dura-mater and the apex of the pedicle, at its transverse axis. This measurement was chosen for being the largest distance between the dura-mater and the vertebral pedicle, and for being the most trustful since just below the chosen place, at the lateral edge of the dura-mater, is the appearance of the root below, that is close together to the medial edge of the pedicle. Thus, there is no assessable measurement between the medial edge of the pedicle and the neural structure (root below).

Another evaluated measurement was the distance between the lumbar pedicle and the root below. This was chosen for being the shorter distance between the root and the lumbar pedicle. The distance between the pedicle and the root above is large as showed by the works on triangular safety zone. Thus, the root above would be protected from injury with introduction of a pedicular screw as reported(8). This measurement was not assessed due to its lack of reproducibility at the several levels. This because we were dependent on several anatomical parameters (apex of the pedicle, nerve root above, and a straight line linking these two points, and that would be parallel to the dura-mater), making this measurement too much subjective. Thus, this measurements were indirectly obtained as described above.

It was also evaluated the nerve root angle, due to the observation that the root has an angle that joins the axis of the pedicle. Thus, the nerve root above is close together to the lateral edge of the pedicle. This angle was indirectly measured, from the knowledge of the measures of the triangular safety zone. Ebraheim et al(6) report in their paper this indirectly performed measurement by means of a goniometer. It was not possible to directly assess this measure neither evaluation of the angle of the axis of the pedicle, neither directly nor indirectly.

We also evaluated the location of the spinal ganglion in relation to the vertebral pedicle. There are other works in literature that evaluated this parameter through magnetic resonance examinations Hasegawa(9), radiographs after injection of contrast in the nerve root Hasue et al(10), however only the work by Sato et al(21) as we evaluated directly in cadaver dissections. The difference was that Sato divided the location zones of the spinal ganglion only in extra-foraminal and intra-spinal. In our work we divided the foraminal zone into two halves, medial and lateral, where we located the center of the spinal ganglion. With this, we could better evaluate if there is a trend that the ganglion becomes progressively more medial in more distal roots, as suggested by the fact that spinal ganglion of S1 root in the works by Hasue, Sato, Hasegawa(9,10,21), has an intra-spinal location.

These assessments were performed by means of a digital slide caliper with a 0.01 mm precision, and a surgical magnifying glass with a 2.5 fold magnification. We believe this improved the assessments accuracy.

The obtained results show an increase both in the transversal as in the longitudinal diameter from L2 to L5 as shown in Tables 1 and 2, respectively. With use of the "t" paired test it was observed that there was an statistically significant increase from L2 to L3, from L3 to L4 and from L4 to L5. Thus, the more distal lumbar vertebral pedicles could receive more safely larger diameter pedicular screws. Besides this, by knowing the size of the lumbar vertebral pedicles it was possible to evaluate if the pedicular screws would have a suitable diameter for its safe introduction. As pedicular screws have their diameters ranging from 5.0 mm to 7.5 mm, and the average of the transversal diameter of L2 is 9.22 mm, we agree that the lumbar vertebral pedicle can receive the marketed pedicular screw.

The distance between the lumbar vertebral pedicle and the root emerging below it (DNP) showed to be small in all evaluated levels (Table 4), being larger at L5 root, followed by L2, L3 and last, L4. That means, it did not respect an increasing or decreasing order from L2 to L5. the difference between the roots from L2 to L3 was not statistically significant, and the same is true for L2-L5 and L3-L5. However, from L2-L4 and L3-L4 and L4-L5 the increase was statistically significant. Thus, the inferior region of the lumbar vertebral pedicle would be unsafe for introduction of a pedicular screw. This is in agreement with Ebraheim(6).

Aiming to evaluate region that is superior and medial to the lumbar vertebral pedicle, the following assessments were performed: 1. distance from dura-mater to the apex of the pedicle (DDMP), 2. distance from the apex of the pedicle to the root emerging above the pedicle (DRAP). This assessment was indirectly obtained. Regarding DDMP, its averages from L2 to L5 progressively increased. This increase was not significant between L2-L3, L2-L4, L2-L5, L3-L4, L3-L5, L4-L5. Thus, the region superior and medial to the pedicle would be safer for introduction of a pedicular screw, and this distance would improve at more distal levels.

We analyzed the angle of appearance of the nerve root at the described levels, since we observed that this angle joins the angle of the pedicle axis. We noticed that there is a progressive increase from L2 to L5, when evaluating the average of these assessments. Under statistical analysis there was no statistically significant increase between the levels L2-L3 and L4-L5, being however significant between L2-L4, L2-L5, L3-L4 and L3-L5. Thus, the results found were in agreement with the ones from Ebraheim(6) and Gu(8). Hasegawa(9) in an evaluation of the same parameter, by means of magnetic resonance analysis, showed a progressive reduction of the apperance angle of the nerve root from L2 to L5, in disagreement with our results.

In regard to the spinal ganglia location, we found that their majority was at the medial half of the foraminal region. Most of the remaining ganglia were at the lateral half of the foraminal region. A small part was at intra-spinal region, and only two ganglia were at the extra-foraminal region. Hasegawa(9), in a work made with magnetic resonance, found similar data. Thus, the majority of spinal ganglia were ate foraminal region, increasing the risk of neural injury with introduction of pedicular screws.



1. Lumbar vertebral pedicle increases in size from L2 to L5, both in sagittal and transversal diameters.

2. Transverse diameter of the pedicle is smaller than the sagittal.

3. Lumbar vertebral pedicle has an intimate relationship with the nerve root emerging below it, being its most distal part close together to the root, and its medial edge adjacent to the dura-mater.



1. AEBI, M.; ETTER, C. H. R.; KEHL, T. H.; THALGOTT, J. The internal skeletal fixation system. Clin. Orthop. 227: 30-43, 1988.        [ Links ]

2. AMERICAN ACADEMY OF ORTHOPAEDIC SURGEONS - AAOS. Glossary of spinal terminology, 1985. 71p.        [ Links ]

3. BLUMENTHAL, S.; GILL, K. Complications of the Wiltse pedicle screw fixation system. Spine 18, n. 13: 1867-1871, 1993.        [ Links ]

4. CHOI, P. S. Contribuição ao estudo da zona triangular de segurança Aplicado aos procedimentos percutâneos póstero-laterais lombar. São Paulo, 2000. 98p. Dissertação (Mestrado) ¾ Faculdade de Medicina, Universidade de São Paulo.        [ Links ]

5. CROCK, H. V. Normal and pathological anatomy of the lumbar spinal nerve root canals. J. Bone Joint Surg. [Br] 63, n. 4: 487-490, 1981.        [ Links ]

6. EBRAHEIM, N. A.; XU, R.; DARWICH, M.; YEASTING, R. A. Anatomic relations between the lumbar pedicle and the adjacent neural structure. Spine 22, n. 20: 2338-2341, 1997.        [ Links ]

7. ESSES, S. I.; SACHS, B. L.; DREYZIN, V. Complications associated with the techinique of pedicle crew fixation. Spine 18, n. 15: 2231-2239, 1993.        [ Links ]

8. GU, Y.; XU, R.; EBRAHEIM, N. A.; REZCALLAH, A. T.; YEASTING, R. A. The. quantitative study of tha lateral region to the lumbar pedicle. Surg Neurol. 52: 353-356, 1999.        [ Links ]

9. HASEGAWA, T.; MIKAWA, Y.; WATANABE, R.; AN, H. S. Morphometric analisis of the lumbosacral nerve roots and dorsal root ganglia by resonance imaging. Spine21, n. 9: 1005-1009, 1996.        [ Links ]

10. HASUE, M.; KUNOGI, J.; KONNO, S.; KIKUCHI, S. Classification by position of dorsal root ganglia in the lumbosacral region. Spine 14, n. 11: 1261-1264, 1989.        [ Links ]

11. HOROWITCH, A.; PEEK, R. D.; THOMAS, J. C.; WIDELL, E. H.; DIMARTINO, P. P.; SPENCER, C. W.; WEINSTEIN, J.; WILTSE, L. The Wiltse pedicle screw fixation system. Spine 14: 461 - 467, 1988.        [ Links ]

12. KIM, N. H.; LEE, H. M.; CHUNG, I. H.; KIM, H. J.; KIM, S. J. Morphometric study of the pedicles of thoracic and lumbar vertebrae in koreans. Spine 19, n. 12: 1390-1394, 1994.        [ Links ]

13. KINNARD, P.; GHIBELY, A.; GORDON, D.; TRIAS, A.; BASORA, J. Roy-Camille plates in unstable spinal conditions. A preliminary report. Spine 11, n. 2: 131-135, 1986.        [ Links ]

14. KRAG, M. H.; WEAVER, D. L.; BEYNNON, B. D.; HAUGH, L. D. Morphometric of the thoracolumbar and lumbar spine related to transpedicular screw placement for surgical spinal fixation. Spine 13, n. 1: 27-32, 1988.        [ Links ]

15. LONSTEIN, J. E.; DENIS, F.; PERRA, J. E.; PINTO, M. R.; SMITH, M. D.; WINTER, R. B. Complications associated with pedicle screw. J. Bone Joint Surg. 11: 1519-1528, 1999.        [ Links ]

16. LOUIS, R. Fusion of the lumbar and sacral spine by internal fixation whith screw plates. Clin. Orthop. 203: 18 ¾ 33, 1986.        [ Links ]

17. MATSUZAKI, H.; TOKUHASHI, Y.; MATSUMOTO, F.; HOSHINO, M.; KIUCHI, T.; TORIYAMA, S. Problems e solutions of pedicle screw plate fixation of lumbar spine. Spine 15, n. 11: 1159-1165, 1990.        [ Links ]

18. OLSEWSKI, J. M.; SIMMONS, E. H.; KALLEN, F. C.; MENDEL, F. C.; SEVERIN, C. M.; BERENS, D. L. Morphometric of the lumbar spine: Anatomical perspectives related to transpedicular fixation. J. Bone Joint Surg. [Am], n. 4: 541-549, 1990.        [ Links ]

19. RAUSCHNING, W. Normal and Pathologic Anatomy of the Lumbar Root Canals. Spine.12: 1008-1019, 1987.        [ Links ]

20. ROY-CAMILLE, R.; SAILLANT, G.; MAZEL, C. Internal Fixation of the lumbar spine with pedicle screw plating. Clin. Orthop. 203: 7-17, 1986        [ Links ]

21. SATO, K.; KIKUCHI, S. Na anatomic study of foraminal nerve root lesions in the lumbar spine. Spine 18, n. !5: 2246-2251, 1993.        [ Links ]

22. STAUFFER, R. N.; COVENTRY, M. B. Posterolateral lumbar-spine fusion. J. Bone Joint Surg.[Am] 54: 1195 ¾ 1204, 1972.24.        [ Links ]

23. STEFFEE, A. D.; BISCUP, R. S.; SITKOWSKY, D. J. Segmental Spine Plates Screw Fixation. Clin. Orthop. 203: 45-53, 1986.        [ Links ]

24. THOMPSON, W. A. L.; HALSTON, E. L. Pseudarthrosis following spine fusion. J. Bone Joint Surg. [Am] 31: 400-405, 1949.        [ Links ]

25. TRUCHLY, G; THOMPSON, W. A. L. Posterolateral fusion of the lumboscral spine. J. Bone Joint Surg.[Am] 40: 505 ¾ 512, 1962        [ Links ]

26. VANICHCHORN, J. S.; VACCARO, A. R.; COHEN, M. J.; COTLER, J. M. Potencial large vessel injury during thoracolumbar pedicle screw removal. Spine 22, n. 1: 110-113, 1997.        [ Links ]

27. WEST, J. L.; OLGIVIE, J. W.; BRADFORD, D. S. Complications of The variable screw plate pedicle screw fixation. Spine 16, n. 5: 576-579, 1991.        [ Links ]

28. WHITECLOUD, T. S.; BUTLER, J. C.; COHEN, J. L.; CANDELORA, P. D. Complications with the variable spinal platin system. Spine 14, n. 4: 472-476, 1989.        [ Links ]

29. ZINDRICK, M. R.; WILTSE, L. L.; DOORNIK, A.; WIDELL, E. H.; KNIGHT, G. W.; PATWARDHAN, M. A. G.; THOMAS, J. C.; ROTHMAN, S. L.; FIELDS, B. t. Analysis of the thoracic and the lumbar pedicles. Spine12, n. 2: 160-166, 1987.        [ Links ]



Address for correspondence
Rua Maracá, 266, apto 33, Vila Guarani
CEP 04313-190 São Paulo, SP

Trabalho recebido em 10/01/2001. Aprovado em 17/05/2002
Work performed at the Instituto de Ortopedia e Traumatologia do Hospital das Clínicas - FMUSP

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License