Morphometric analysis of the thoracic pedicle: An anatomico-radiological study
Background: Knowledge of the morphometric anatomy of the pedicles in thoracic spine is essential for the surgeon attempting thoracic pedicle screw placement. Aim: To study the morphometry of the pedicles in the thoracic spine in Indian subjects and note the anatomical landmarks required for safe and accurate pedicle screw placement. Settings and Design: An anatomico-radiological study with cadaveric confirmation was conducted in a medical college and tertiary care center. Materials and Methods: Morphometric anatomy of 720 pedicles from T1-T12 was studied. The study consisted of 15 separated thoracic vertebral cadavers and 15 computed tomographic scans of undiseased thoracic spine. Transverse and sagittal pedicle diameters, transverse pedicle angle, chord length and the pedicle entry point landmarks were studied. Using the results, accuracy of the pedicle screw placement was confirmed in 120 pedicles from five unseparated cadavers. Results: Transverse pedicle diameter was narrowest at T4 and gradually increased craniocaudally. Twenty-two percent of the pedicles from T4 to T6 were less than 4.5 mm, but none were less than 3.5 mm. The transverse pedicle angle was widest at T1 (26.50) and decreased caudally to T12 (8.10). The chord length gradually increased caudally from a minimum of 30.4 mm at T1 to a maximum of 43.3 mm at T12. Landmarks (transverse and vertical distance) for the pedicle entry point were measured from the reference point taken as the center of the junction of lamina and the transverse process. The pedicle entry point was always superolateral to this reference point. Conclusion: Knowledge of the pedicle diameter and chord length is essential for choosing the appropriate pedicle screw, whereas the pedicle angle and the entry point are important for accurate screw placement.
Keywords: Chord length, pai chart, pedicle, pedicle angle, pedicle diameter, pedicle screw entry point, thoracic spine
The advent and general acceptance of pedicle screw fixation of the thoracic spine has made the morphometric analysis of the thoracic pedicle a clinical necessity for all spinal surgeons practicing this procedure.  Pedicle screw fixation is considered biomechanically superior to the hook rod fixation. Various authors have studied the thoracic pedicle morphometry and have arrived at differing conclusions regarding the safety of thoracic pedicle screw placement. ,,, Datir et al. have conducted a morphometric study of the thoracic pedicle in an Indian population.  The authors feel that a thorough knowledge of the morphometric anatomy of the thoracic pedicle at various levels is essential for the spinal surgeon attempting thoracic pedicle screw placement.
The authors have studied the morphometric anatomy of 720 thoracic pedicles in two groups, cadaveric group and radiological group.
(a) Cadaveric group: This group consisted of 15 cadavers, age ranging from 51 to 74 years at the time of death. The specimens were free of any diseases of the spine and the vertebrae were separated and meticulously cleaned off all soft tissue before making the measurements. The following measurements were made in this group on either side:
(b) Radiological group: This group consisted of a morphometric anatomical study of the thoracic pedicles in a group of 15 patients who underwent computed tomography (CT) scan of the thoracic region for reasons other than any spinal pathology. The following measurements were made in this group on either side:
These measurements were made on the console of the CT scan using the scale provided in the software [Figure 1].
(c) Cadaveric confirmation: Using the pedicle entry point information derived from the cadaveric group and the pedicle angles from the cadaveric and radiological study, thoracic pedicle screw placement was performed from the T1 to T12 levels on both sides in five unseparated cadaveric specimens. The screws were then removed and the vertebrae separated and the screw tract studied for any pedicle fractures, cord or root injury and anterior vertebral cortex perforation.
Data were first analyzed to determine the variation of dimensions from left to right, but were not found to be statistically significant and hence were pooled. The data between the cadaveric and the radiological groups were compared and also pooled. There was no correlation between pedicle morphometrics and gender or side in this series.
Transverse pedicle diameter at the isthmus (PD)
The PD of the various levels is tabulated and shown in [Table 1]. There was no statistically significant difference in the values between the cadaveric and radiological groups. The PD started at a mean value of 7.3 mm at D1 diminishing caudally, reaching a minimum value of 4.5 mm at the D4 level and increasing again caudally to a maximum of 7.8 mm at D12.
Sagittal pedicle diameter at the isthmus (SD)
The SD of the various levels is tabulated and shown in [Table 2]. The SD gradually increased caudally from a minimum of 8.8 mm at D1 to a maximum of 15.8 mm at D12.
Coaxial depth from the lamina to the anterior vertebral cortex - chord length
The chord length (CL) at the various levels is shown in [Table 3]. There was no statistically significant difference between the values in the cadaveric and radiological groups. The CL gradually increased caudally from a minimum of 30.4 mm at D1 to a maximum of 43.3 mm at D12. It is important that the CL be measured in the pedicle axis, i.e. in the axis perpendicular to the transverse diameter at the isthmus.
Transverse angle of the pedicle axis with the median sagittal plane (pedicle angle)
The pedicle angle (PA) values are tabulated in [Table 4]. The PA gradually reduced caudally, with a maximum angle of 26.5 0 at D1 and a minimum of 8.1° at T12.
Transverse distance from the pedicle entry point to the center point of the junction of lamina with the transverse process
The pedicle entry point was lateral to this reference point, i.e. the center point of the junction of the lamina to the transverse process. The transverse process (TD) values were higher at T1 and T2 (4.3 and 3.3 mm respectively) and reduced to a minimum value of 2.0 mm at T7 and again increased to a maximum of 2.7 mm at T12 [Table 5],[Figure 2].
Vertical distance from the pedicle entry point to the center point of the junction of lamina with the transverse process
The pedicle entry point was vertical to the reference point. The Vertical distance (VD) values were generally constant at all levels, ranging from a minimum value of 4.1 mm at T5 to a maximum value of 5.0 mm at T12 [Table 5],[Figure 2].
Cadaveric confirmation of the data
Pedicle screw placement was performed on 120 thoracic pedicles using the above data in five unseparated cadaveric specimens. Pedicle entry points were found to be accurate in all five specimens at all levels [Figure 3]. Grade 3 perforation (more than half the diameter of the screw) as per the grading of Ussawakongkiate C was noted bilaterally at T5 and T6 in two cadaveric specimens. However, no injury to the adjacent vascular or neural structures was noted. This could be attributed to faulty transverse angulation of the screws and could have been avoided if the data had been strictly implemented. No medial perforation of the pedicle was noted at any level. Perforation of the anterior vertebral cortex was not encountered.
Pedicle screw rod/plate instrumentation has become a popular and accepted method of thoracic spinal stabilization.  It is considered biomechanically superior to the traditional hook and rod instrumentation. Radiological assistance with a C arm may not be of much value in the lateral view due to the obstruction caused by the rib cage preventing clear visualization of the pedicle and its entry point, especially in the upper thoracic levels. Navigation techniques may be more useful but present their own fallacies and may not be affordable by most institutions in developing countries. Hence, it is imperative that the spinal surgeon has a clear knowledge of the pedicle morphometric anatomy and the landmarks before venturing on thoracic pedicle screw placement. The main morphometric dimensions influencing pedicle screw placement are transverse pedicle diameter, pedicle entry point (TD and VD), transverse pedicle angle and chord length.
The transverse pedicle diameter is an important dimension as this decides the size of the screw to be inserted. However, in all studies, it has been established that the sagittal diameter is always greater than the transverse diameter and hence the latter plays a role in the selection of the screw and not the sagittal diameter. The mean transverse PD at various levels were comparable with the results of Datir et al, whose study was also conducted on Indian subjects.  The mean transverse PD was comparable to other studies. ,,, Twenty-two percent of pedicles from T4 to T6 were less than 4.5 mm in diameter, which was the standard pedicle screw size used. This was in concurrence with other studies. , In the current study, no thoracic pedicle transverse diameter was found to be less than 3.5 mm. Hence, the authors advocate the use of 3.5 mm screws in these areas to avoid pedicle breakage. However, this necessitates the use of "dual diameter rods" if 4.5 mm screws are also used in the same patient at other levels. These rods have a diameter of 3.5 mm in one half and 4.5 mm in the other half. The former fits well into the 3.5 mm screw while the thicker portion fits into the head of screws 4.25 mm and larger in diameter. The surgeon can cut the rod to the required length and use it appropriately. Alternately, the "in-out-in" technique may be used with a 4.5 mm screw, where the method of screw placement involves entering the pedicle in the usual position, penetrating the lateral wall of the pedicle and passing out into the cost-vertebral articulation before penetrating the dorsolateral vertebral body at the base of the pedicle and re-entering the vertebral body. , This is depicted clearly in [Figure 4]. The authors suggest that the technique be individualized for each patient depended upon the level of stabilization. The sagittal pedicle diameter is always greater than the transverse pedicle diameter at all levels, in all studies.
Pedicle entry point
The pedicle entry point in the thoracic spine has been a matter of considerable discussion, with various authors ascribing to different anatomical landmarks as reference points. Roy-Camille et al, suggested that the pedicle entry point in the thoracic spine lies at the junction of the facet joints and the transverse process.  Magerl recommended the junction of the lateral margin of the facet and the midline of the transverse process.  An et al, noted that the pedicle entry point in the upper thoracic spine was at the mid portion and 1 mm below the facet joint.  Hou et al, as well as Datir et al, have reported the junction of the upper border of the transverse process with the vertical line through the lateral part of the inferior facet as the pedicle entry point. , However, the entry point may vary for the various levels of the thoracic spine. The reference point in our study was the center point of the junction of the lamina with the transverse process. We feel that this anatomical landmark can be easily identified by the surgeon. The pedicle entry point was superior and lateral to this reference point at varying levels, as noted in our results. Using this information, accurate screw entry point was achieved in the cadaveric conformation group.
The PA in our study gradually reduces caudally. Our values match the data from the other similar studies. ,, However, the PA values of Datir et al, were lower.  In the study conducted by Zindrick et al, the PA of the lower three thoracic vertebrae (T10, T11 and T12) were found to be approaching zero, and even negative in some cases.  Although the lower three thoracic vertebrae demonstrated a low PA approaching 10 degrees or even lower, there were no instances of negative values of PA in our study.
Sagittal pedicle angle values from the studies conducted by Zindrick et al and Datir et al, show that the pedicles were directed cephalad at all thoracic levels. , This is important in accurate screw placement as inferior migration of the screw may result in injury to the nerve root. However, this measurement was not part of our study.
Chord length determines the length of the pedicle screw to be used. The values obtained in our study were in concurrence with the study of Datir et al, who advocate the safe use of a 25 mm screw at the upper thoracic and 30 mm in the mid thoracic levels.  Although perforation of the anterior cortex of the vertebral body is a potential complication, injury to the great vessels has seldom been reported. Shorter CL of the upper thoracic spine, inadequate fluoroscopic visualization with resultant usage of a longer screw and a shallow transverse pedicle angle of screw insertion are reported as the main reasons for this worrisome complication.  This needs to be kept in mind, especially when surgeons advocate the in-out-in technique of pedicle screw placement in the upper thoracic spine.
The authors have formulated a chart from the knowledge gained from this study to serve as a guide for thoracic pedicle screw placement for spinal surgeons. Direct measurements from the CT scan of the patient when available can be extremely useful. The Pai chart provides the pedicle entry point and also recommends the size of the screw and its angulation at various thoracic levels [Table 6].
The study of thoracic pedicular morphometry is important for the spinal surgeon for pedicle screw placement. Knowledge of the pedicle diameter and the chord length is essential for choosing the appropriate pedicle screw at the various thoracic levels. Pedicle angles and the entry point are important for accurate screw placement and avoidance of complications. The Pai chart can serve as a guide for the spinal surgeon in thoracic screw placement.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]