Keywords
fractures, bone - congenital abnormalities - spine - thoracic vertebrae
Introduction
Systems for correcting spinal deformities mainly use pedicle screws for posterior
anchoring.[1] These systems allow the three-dimensional (3D) correction of deformities, providing
sufficient stability to avoid the use of postoperative immobilization.[2]
[3] Besides, this system allows for even more significant correction of deformities,
especially when compared with hook or hybrid systems.[4]
[5]
[6]
[7] However, pedicle fixation systems have some disadvantages, especially concerning
complications caused by incorrect positioning of the screw inside the pedicle and
exposure of the surgeon to radiation.[8] The incorrect positioning of pedicle screws occurs more frequently in deformities,
whose vertebrae present anatomical changes and due to their 3D positioning.[9]
[10]
The average accuracy of placing pedicle screws freehand, or with fluoroscopy, and
with the aid of navigation is 85.1% and 95.5%, respectively.[11]
[12]
[13]
[14] Frequently, fluoroscopy is used to assist in the insertion of pedicle screws.[14] However, during fluoroscopy, the exposure of the surgeon to radiation is 10 to 12
times greater than in other procedures that use fluoroscopy in segments outside the
spine.[15]
[16]
New alternatives have been developed to improve accuracy and reduce exposure to radiation,
with emphasis on customized guides.[17]
[18]
[19] The advantages of using a customized guide (low-cost) motivated us to carry out
a project for the development of a prototype.
Therefore, the present study aimed to develop and evaluate the use of customized guides
in patients undergoing surgery to correct vertebral deformity with a pedicular fixation
system. These guides are made using 3D printing from spinal models and are developed
to assist in the preparation of the pilot hole in the spinal pedicle.
Methods
The Research Ethics Committee approved the present study (protocol number 3,365,105).
The present study was performed in four patients with spinal deformities who underwent
surgical treatment using a pedicular fixation system.
The demographic data of the patients are shown in [Table 1]. Three patients had idiopathic scoliosis and one patient had kyphoscoliosis. All
patients were female, ranging in age from 11 to 17 years old (mean = 15 years old).
Table 1
|
Age (years)
|
Sex
|
Deformity
|
Levels
|
Cobb Angle
|
|
Patient 1
|
16
|
Female
|
AIS
|
T10-L4
|
65,8°
|
|
Patient 2
|
17
|
Female
|
JIS
|
T3-L3
|
68,1°
|
|
Patient 3
|
11
|
Female
|
Congenital kyphoscoliosis
|
T8-L2
|
57,5° (scoliosis) / 87,3° (kyphosis)
|
|
Patient 4
|
15
|
Female
|
JIS
|
T6-L2
|
64,1°
|
A set of 3D guides was made for each patient. An individual guide was created for
each vertebra programmed to receive pedicle fixation. Along with the guides, a model
of the spine was also made, which helped the 3D orientation of the vertebral structures
([Fig. 1]).
Fig. 1 Illustrative images. (A) photograph of the model of the spine of a patient with congenital deformity; (B) a photo of the surgical guide attached to the model in the position to prepare the
pilot hole.
The 3D guides were made based on preoperative computed tomography (TC) covering the
extension of the vertebral segment programmed to receive the pedicle screws. Computed
tomography was standardized in sections ≤ 1 millimeter to allow greater accuracy in
the anatomical reconstruction of the bone surface.
The preoperative programming to determine bilaterally, in each vertebra, the positioning
of the screw inside the vertebral pedicle, its angulation and length was performed
employing 3D anatomical analysis (ATA) using software (Materialize Brazil, São Paulo,
SP, Brazil). The surgeon guided the position, angulation, and length of the pedicle
screw to be used ([Fig. 2]).
Fig. 2 Photograph of the preoperative 3D anatomical analysis in different angles with the
simulation of the position of the pedicle screws and the fitting of the surgical guide
in the posterior region of the corresponding vertebra.
The guides were made with synthetic material of biocompatible, nonbiodegradable resin,
and were subjected to sterilization at a temperature of 50°C in a Sterrad (Medsteril,
Água Branca, São Paulo, SP, Brazil) device. A specific guide was created for each
vertebra using a 3D printer. Each guide, made for one particular vertebra separately,
consisted of two cylindrical parts that guided the entry point and the preparation
of the pilot hole of the vertebral pedicle by placing the instruments inside it ([Fig. 3]).
Fig. 3 Photograph of the surgical guide for a lumbar vertebra.
During the surgical procedure, the guides were coupled to each vertebra, through their
fit in the spinous process and the opposition of the surface of the guides at the
point corresponding to the projection of the vertebral pedicle on the back of the
vertebra ([Fig. 4])
Fig. 4 Intraoperative image of a surgical guide positioned in the posterior vertebral region,
with an instrument attached to prepare a pilot hole.
With the guide positioned and stabilized, the entry point into the vertebral pedicle
was determined by the introduction of the appropriate instrument within the guide.
Then, the pilot hole was made with probes placed inside, followed by taps and checks
on the vertebral pedicle walls before the insertion of the screws.
To assess the use of the guides, we used the following parameters: technical viability,
precision, and exposure to radiation.
The technical performance of the guide and its use for the desired purpose was considered,
being classified as positive or negative. Therefore, we considered of positive technical
viability the guide that allowed its use according to the desired objectives. On the
other hand, negative technical viability was considered when the guide could not be
used or did not reach the desired goals (inadequate adjustment of the guide in the
posterior vertebral elements, entry point of the perforation without correlation with
the anatomical references, breakage of the guide during its use, failure to couple
the surgical instruments with the guide, inadequacy of the pilot hole observed by
checking the pedicle walls or fluoroscopy).
Accuracy was assessed in the postoperative period using CT. We consider the pedicle
screw to be well-positioned when centralized in the vertebral pedicle, keeping the
lateral and medial walls of the vertebral pedicle integral. If there is a violation
of the lateral or medial wall of the vertebral pedicle, we consider the screw to be
malpositioned.
The exposure to intraoperative radiation was performed by measuring the total time
of use of fluoroscopy and its dose.
The Mann-Whitney nonparametric test was used to analyze the results, and the level
of significance was set at p ≤ 0.05.
Results
We evaluated the total set of 85 vertebral pedicles (56 thoracic and 29 lumbar) in
which the pedicle screws were inserted.
Technical viability was positive in 46 vertebral pedicles (54.1%), of which 25 were
thoracic (54%) and 21 lumbar pedicles (46%). Technical viability was negative in 39
pedicles (45.9%), of which 31 were thoracic (79.5%), and 8 were lumbar (20.5%). Technical
viability was negative due to several factors, such as inadequate fitting of the guide
to the posterior vertebral elements (10 pedicles [11.7%]), the entry point of perforation
without correlation with anatomical references (23 pedicles [27%]), breakage of the
guide during use, failure in the coupling of surgical instruments with the guide (2
pedicles [2.5%]), and inadequacy of the pilot hole observed by checking the pedicle
walls or fluoroscopy (4 pedicles [4.7%]). The negative technical viability was directly
related to the development stages of the customized guide prototype. It was reduced
as a result of surgeries performed with the improvement of the guide prototype.
The evaluation of the accuracy of pedicle screws in which the pilot hole was prepared
with the help of the guide showed that 36 screws were centralized (78.2%), with 17
in the thoracic (36.9%) and 19 in the lumbar spine (41.3%).
In 10 pedicles (21.7%), the screws were not centralized according to what was established
in the preoperative schedule, with violation of the lateral wall in 6 pedicles (13%)
and 4 in the medial (4.3%). The accuracy of the screws in the thoracic spine and concavity
was lower concerning the other vertebral segments.
Here, the positioning of the screws predominated in the thoracic spine and was superior
to the group of vertebral pedicles in which the guide cannot be used. The pilot hole
was prepared with the help of the model, showing its assistance in improving the accuracy.
A noncentralized positioning of the screw was observed in 10 pedicles (21.7%), 8 in
the thoracic (17.4%), and 2 in the lumbar spine (4.3%). The rupture of the lateral
wall was observed in 6 pedicles (13%), 4 of which were thoracic (8.7%), and 2 were
lumbar (4.3%). The rupture of the medial wall was observed in 4 pedicles (8.7%), all
of them in the thoracic spine.
In the 39 pedicles whose pilot holes were prepared without the aid of the guide, the
screws were centralized in 19 pedicles (48.7%), 12 in the thoracic spine (30.8%),
and 7 in the lumbar spine (17.9%). Malposition was observed in 20 screws (51.3%),
18 in the thoracic (46.2%), and 2 in the lumbar spine (5.1%). The rupture of the lateral
wall was observed in 9 pedicles (23%), all of them being thoracic. The separation
of the medial wall was observed in 11 pedicles (28.2%), 10 of which were thoracic
(25.7%), and 1 lumbar (2.5%).
The comparison of the accuracy of the set of pedicles in which the pilot hole was
prepared with and without the guide is shown in [Table 2] and [Fig. 5]. Higher efficiency was observed with the use of guides in the pedicles of the lumbar
vertebrae (p < 0.05). In contrast, in the pedicles of the thoracic spine and in the set of all
pedicles, the accuracy did not show the statistical difference ([Fig. 6]). It must be considered that the nonuse of the drilling guides was related to the
negative technical viability, and that the intraoperative visualization of the model
helped in the preparation of the pilot hole.
Table 2
|
With guide
|
|
|
Without guide
|
|
|
|
Violation
|
|
Central
|
Violation
|
|
Central
|
|
Medial cortical
|
Lateral cortical
|
|
Medial cortical
|
Lateral cortical
|
|
|
T2
|
|
|
|
1
|
1
|
|
|
T3
|
|
|
|
|
|
1
|
|
T4
|
|
1
|
|
2
|
1
|
|
|
T5
|
|
|
2
|
|
2
|
2
|
|
T6
|
|
|
2
|
2
|
1
|
1
|
|
T7
|
|
|
|
2
|
1
|
1
|
|
T8
|
2
|
|
2
|
1
|
|
1
|
|
T9
|
1
|
1
|
2
|
1
|
1
|
2
|
|
T10
|
1
|
|
1
|
|
1
|
3
|
|
T11
|
|
1
|
3
|
1
|
1
|
1
|
|
T12
|
|
1
|
5
|
|
|
|
Fig. 5 Comparison of the accuracy of pedicle screws in the thoracic and lumbar levels in
absolute frequency (number of pedicles), positioned with and without the aid of surgical
guides.
Fig. 6 Accuracy of positioning pedicle screws with and without the use of guides in the
thoracic (category T) and lumbar (category L) spine, by the average number of screws
by number of levels addressed in each vertebral segment (lumbar and thoracic). *p ≤ 0.05.
The general technical feasibility showed statistical significance (p = 0.0089) ([Fig. 7]), and a gradual increase was observed following surgical procedures. Improvement
of prototypes ([Fig. 8]) has been of great help in the correction of complex and severe deformities ([Fig. 9]).
Fig. 7 Technical viability considering all pedicles instrumented with and without the use
of guides. Absolute Frequency: frequency in the absolute number of pedicles addressed.
* p = 0.0089.
Fig. 8 Technical viability with the use of guides for each patient operated at a relative
frequency (percentage).
Fig. 9 Pre (A) and postoperative (B) radiographic and clinical images of a patient with congenital kyphoscoliosis (patient
3), in which the customized guide was used.
Intraoperative radiation exposure ranged from 4.35 millisievert (mSv) to 6.32 mSv
(mean = 5.17 ± 0.72), with radioscopy use time from 180.3 to 207.2 seconds (mean = 190 ± 16.23).
There were no operative and postoperative complications, such as increased bleeding,
neurological injuries, or changes in motor or sensory potential during intraoperative
neurophysiological monitoring.
Discussion
Initially, 3D printing was idealized by Hall[20] in 1986. After that, the technique was improved and introduced as an auxiliary tool
in surgeries, especially in the spine.[19] In the context of spine surgery, it has been used to produce anatomical models,
surgical guides, and implants.[20]
In the present study, we aimed to develop a customized guide prototype and evaluate
its results for the preparation of the pilot hole in the pedicles of the thoracic
and lumbar vertebrae of patients with spinal deformity.
Through the interpretation of our results, it was possible to observe the improvement
of the surgery with the use of the prototype, adjusting and correcting the problems
found, and increasing its technical viability after the performed operations. Changes
in the synthetic composition of the guide, its mechanism of fixation to the posterior
elements of the vertebra, and the best adaptation of the instruments for the preparation
of the pedicle within the guide were the main changes made. Also, problems related
to the technical feasibility of using the guide were more frequent in the pedicles
of the thoracic vertebrae.
The problems related to the fitting of the guides in the vertebrae of patients with
rigid scoliosis and high angular value were also reported by Liu et al.[21] A more significant contact of the guide with the posterior surface of the vertebra
increases the stability of the guide for the preparation of the pilot hole so that
the guides must be made for private use in each vertebra. This observation corroborates
the reports of Berry et al.,[22] showing the inaccuracy of the guides for multiple levels.
The fitting of the guide on the posterior face of the vertebra required extensive
dissection and detachment of the soft parts inserted in the vertebrae. This factor
was also pointed out as being essential for the fitting of the guide in the vertebrae,[21] and could be pointed out as a disadvantage for the use of this type of guide in
procedures of smaller extension. However, in deformities, it is necessary to have
a broad exposure of the vertebra with the disinsertion of the soft parts, so that
the full exposure and detachment does not present a disadvantage for the use of the
guide.
The use of guides increased the precision of screw placement compared with the group
in which the guide was not used, due to the technical unfeasibility and the model
used to help guide the preparation of the pilot hole. In patients with spinal deformity,
the rate of screw malposition varies from 3 to 44.2%, and neurological complications
from 0 to 0.9%.[3]
[13]
[16]
[23]
[24]
[25]
[26]
[27]
[28] The pedicles of the thoracic spine and of the concavity of the curve have the highest
percentage of malposition.[3]
[13]
[16]
[23]
[24]
[25]
[26]
[27]
[28] The results observed in the present study corroborate the reports in the literature,
with the pedicles of the thoracic region having the highest index of malpositioning.
However, despite the noncentralized positioning in 10 pedicles (8 thoracic and 2 lumbar),
there was no damage to the structures adjacent to the pedicle or the need to reposition
or remove implants in any patient.
The accuracy of the use of guides in the thoracic pedicles was 68%, being higher than
the results of the group in which the guide was not used, evidencing the benefit of
its use in the preparation of the pilot orifice.
Indeed, the learning curve and the development of the guide prototype must be considered
when analyzing our results. The results of the accuracy of the last operated patient
showed high technical feasibility and accuracy close to the 3D anatomical analysis
performed in the preoperative period.
The use of the customized guide prototype allowed the reduction of the time of use
of fluoroscopy and, consequently, reduction of exposure to intraoperative radiation.
The exposure of the surgeon during the placement of pedicle screws is between 10 and
12 times greater than that of other procedures outside the spine.[29]
[30] The intraoperative radiation dose in surgeries for vertebral deformities has been
reported to be ∼7.05 mSv. Here, we observed lower values that ranged from 4.35 mSv
to 6.32 mSv (5.17 ± 0.72 mSv), indicating less exposure to intraoperative radiation.
However, the ideal comparison would imply the analysis of similar groups, which was
not possible due to the similarity of heterogeneous samples, so that the comparative
value can only be used as a reference.
Although the technique of preparing the pilot hole and inserting the pedicle screws
without the aid of images or devices has been reported to be safe and with acceptable
accuracy, the use of the pilot hole preparation guides can increase the efficiency
and reduce the amount of intraoperative radiation. The use of guides associated with
the knowledge and experience of the surgeon can make the procedure safer, more accurate,
and reduce the amount of intraoperative radiation. The results presented here are
only related to the development of the guide prototype, indicating that its development
can assist in performing spine surgeries that use the vertebral pedicle as the implant
anchorage site.
Conclusion
The use of guides to prepare the pilot orifice in the vertebral pedicles of patients
with spinal deformity allowed for safe preparation, improving the accuracy of pedicle
screws, and reducing the intensity of intraoperative radiation. This technology has
great potential for clinical use, allowing the placement of pedicle screws in a safer,
more accurate manner, and with less use of intraoperative radiation.
