CASE REPORT

J Neurosurg Spine 26:513–518, 2017

The utility of 3D printing for surgical planning and

patient-specific implant design for complex spinal
pathologies: case report
Ralph J. Mobbs, MD, FRACS,1–3 Marc Coughlan, MBBS, FRACS,2
Robert Thompson, MBus, BInfoTech, DipAppSci,4 Chester E. Sutterlin III, MD,5–7 and
Kevin Phan, BSc1–3
NeuroSpine Surgery Research Group (NSURG), Prince of Wales Private Hospital, Sydney; 2Prince of Wales Private Hospital,
1

Sydney; 3University of New South Wales, Sydney; 4Anatomics Pty. Ltd., Melbourne, Australia; and 5Department of Neurosurgery,
University of Florida, Gainesville; 6Spinal Health International, Inc., Longboat Key, Florida; and 7ProCRO Pty. Ltd., Sydney, New
South Wales, Australia

OBJECTIVE There has been a recent renewed interest in the use and potential applications of 3D printing in the assis-
tance of surgical planning and the development of personalized prostheses. There have been few reports on the use of
3D printing for implants designed to be used in complex spinal surgery.
METHODS The authors report 2 cases in which 3D printing was used for surgical planning as a preoperative mold, and
for a custom-designed titanium prosthesis: one patient with a C-1/C-2 chordoma who underwent tumor resection and
vertebral reconstruction, and another patient with a custom-designed titanium anterior fusion cage for an unusual con-
genital spinal deformity.
RESULTS In both presented cases, the custom-designed and custom-built implants were easily slotted into position,
which facilitated the surgery and shortened the procedure time, avoiding further complex reconstruction such as har-
vesting rib or fibular grafts and fashioning these grafts intraoperatively to fit the defect. Radiological follow-up for both
cases demonstrated successful fusion at 9 and 12 months, respectively.
CONCLUSIONS These cases demonstrate the feasibility of the use of 3D modeling and printing to develop personal-
ized prostheses and can ease the difficulty of complex spinal surgery. Possible future directions of research include the
combination of 3D-printed implants and biologics, as well as the development of bioceramic composites and custom
implants for load-bearing purposes.
https://thejns.org/doi/abs/10.3171/2016.9.SPINE16371
KEY WORDS 3D printing; model; surgery; simulation; implant; design; chordoma; lumbar fusion; cervical

T
he reconstruction of the spine and skull has a rich cations of 3D printing in several avenues, including manu-
and longstanding history, spanning at least 5 mil- facturing of anatomical models,4,6 customized prosthetics
lennia to include attempts in which a range of ma- and implants, and tissue and organ fabrication, as well as
terials has been used, from fruit shells, to sheep scapulae, drug delivery vehicles and discovery.7
to man-made plastics.3,14 There has been a recent renewed There have been reports of the application of 3D print-
interest in the use and potential applications of 3D printing ing for preoperative surgical planning;4,6 however, there
in these realms, as a source of customizable patient-spe- have been few reports on implant design for complex spi-
cific implants for the nuances of each patient’s individual nal surgery. With computer 3D modeling software and the
anatomy. Medical applications for 3D printing are expand- patient’s neuroimaging data, a mold of the patient’s anato-
ing at a rapid pace and are expected to revolutionize health my as well as a corresponding patient-specific implant can
care.12 Currently there is innovative research into the appli- be fabricated. We report 2 cases in which 3D printing was

ABBREVIATIONS CAD = computer-aided design; C.S.I.R.O. = Commonwealth Scientific and Industrial Research Organisation; Oc = occiput.
SUBMITTED March 29, 2016. ACCEPTED September 13, 2016.
INCLUDE WHEN CITING Published online January 20, 2017; DOI: 10.3171/2016.9.SPINE16371.

©AANS, 2017 J Neurosurg Spine Volume 26 • April 2017 513

Unauthenticated | Downloaded 03/21/24 04:18 PM UTC

R. J. Mobbs et al.

used for surgical planning as a preoperative mold, and for Part 2: Implant Design Process
a custom-designed titanium prosthesis: one patient with A transoral anterior approach was planned, which
a C-1/C-2 chordoma who underwent tumor resection and provides an extradural route to the upper cervical spine
vertebral reconstruction, and another patient with a cus- without brainstem or spinal cord retraction around the
tom-designed titanium anterior fusion cage for an unusual craniocervical junction. Given the complex anatomy and
congenital spinal deformity. challenge of surgical treatment, 3D printing technol-
ogy was used for surgical planning by constructing an
Case Reports anatomically accurate 3D plastic model of the patient’s
craniocervical anatomy. The surgeon (R.J.M.) and Ana-
Case 1 tomics Pty. Ltd. worked together on 3D computer-aided
A 63-year-old man presented to our institution with a design (CAD) software (Fig. 2A and B) and, using the
3-month history of neck and shoulder pain. The cervical plastic models, to customize a range of potential implant
CT and MRI studies demonstrated an osteolytic lesion in- models to fit the defect after tumor resection. The custom
volving the C-2 vertebral body and the anterior C-1 arch titanium-printed prosthesis was manufactured by Com-
(Fig. 1). The lesion occupied these 2 vertebrae without monwealth Scientific and Industrial Research Organisa-
lateral extent, or growth into the vertebral foramen, but tion (C.S.I.R.O.) (Fig. 2D) based on the resection margins
extended up to the anterior epidural space at C-2. A CT- calculated from 3D modeling.
guided transoral biopsy revealed chordoma. The surgeon correlated the medical imaging data with
Tumor removals at the craniocervical junction have a 3D plastic model of the patient’s spine. A second 3D
historically been challenging and highly morbidity-pro- plastic model incorporating the proposed tumor resec-
ducing procedures. There have been only limited attempts tion was made. A semitransparent plastic model of the
at tumor resection and structural reconstruction in the anatomy was also created, with the existing posterior fixa-
area below the occiput (Oc), given the narrow operative tion highlighted in red (Fig. 2C). The red screw coloration
corridor and depth of surgical anatomy. The prognosis for combined with the semitransparency of the bone in the
chordoma located at C-1 and C-2 also is poor, with these model confirmed the position of the existing posterior fix-
2 vertebrae contributing primarily to rotation of the head. ation screws with respect to the proposed resection.
Without treatment, tumor progression will result in brain- Implant design then commenced, using global dimen-
stem, cranial nerve, and spinal cord compression, and sions implied by the resection and surgeon-provided con-
progressive instability at the craniocervical junction with cept sketches. Plastic implant prototypes were then con-
intractable pain, with eventual quadriplegia and death.5 structed for a trial fit on the resected anatomy models to
There were a number of stages to performing this pro- test various implant designs (Fig. 2C). These prototypes
cedure. were inspected by the surgeon. Further implant design
refinement was achieved. The fixation screw trajectories
Part 1: Posterior Fusion, Oc–C3 were determined by the surgeon using the plastic model
Initial occipitocervical (Oc–C3) fusion was performed and implant prototypes. The fixation screw dimensions
(Stryker OASYS) posteriorly for stability, prior to the
planned anterior tumor resection. In addition, the poste-
rior fusion was performed so that exact planning of the tu-
mor resection and patient-specific implant reconstruction
could be accomplished due to fixed anatomy joining the
skull base to the upper cervical spine and to C-3. Neuro-
navigation was possible because the skull and upper cervi-
cal spine were fixed in relative space to provide accuracy
during the procedure for navigation.

FIG. 2. A: CAD modeling of implant to reconstruct defect from tumor

excision. B: Posterolateral view demonstrating prior occipitocervical
fusion, and lines demonstrating planned trajectory for screw fixation for
FIG. 1. Left: Sagittal T2-weighted MR image demonstrating chordoma the 3D implant. C: Plastic model of implant with planned screw trajecto-
(arrows). Right: Sagittal T1-weighted MR image demonstrating chor- ries into clivus and C-3. D: Final model prior to sterilization and surgery.
doma (arrows). Figure is available in color online only. Figure is available in color online only.

514 J Neurosurg Spine Volume 26 • April 2017

Unauthenticated | Downloaded 03/21/24 04:18 PM UTC

 Utility of 3D printing for implant design

were then calculated in software and reported to the sur-

geon for prearrangement of implant fixation. The screw
trajectories were then built into the implant to aid in screw
placement (Fig. 2B–D).
The final implant designs were exported electronically
to and built by C.S.I.R.O. The 3D-printed parts proceeded
to quality checks. A semitransparent plastic model, im-
plant template, and screw trajectory indicators were sup-
plied and sterilized to assist intraoperatively.

Part 3: Anterior Transoral Approach, Tumor Resection, and 3D

FIG. 3. Postoperative CT (left) and postoperative radiograph (right)
Implant Insertion demonstrating the 3D-printed titanium implant.
General anesthesia was performed via tracheostomy.
The patient was positioned supine. The head was fixed in
a Mayfield clamp, and a BrainLab reference frame (Brain- Case 2
Lab) was attached for neuronavigation. A planned Le Fort A 52-year-old woman presented with an 18-month his-
midface split was not performed because an adequate ap- tory of intractable back pain and unilateral leg pain. The
proach could be achieved via the transoral corridor. An patient was treated conservatively for 2 years with weight
incision was made in the mucosa of the midline gingiva loss and physical therapy. Her CT and MRI studies dem-
sulcus to expose the anterior nasal spine. The submucosal onstrated an unusual vertebral anomaly, with a congeni-
nasal floor was dissected along the palate, and the nasal tal hemivertebra at L-5, segmental kyphosis with loss of
septum was cut and freed from the palatal bone. A mid- lordosis, and degenerative changes. The opinion of the
line palate split was performed to allow access to the cli- surgeon (M.C.) was that a standard fusion performed us-
vus for the reconstruction phase of the procedure. A linear ing a commercially available implant would not provide
incision of the posterior oral mucosa was made to allow adequate anatomical restoration and support, based on the
surgical field access to the clivus, craniocervical junction, patient’s unique anatomy.
and C-2 and C-3 vertebral bodies.
A C-2/C-3 discectomy was performed to identify the Operative Approach
inferior extent of the chordoma. To remove the superior An anterior and posterior fusion approach was planned
extent of the tumor, the anterior C-1 arch was resected and for the patient. The preoperative CT and MRI studies were
division of the apical and alar ligaments was performed. used to construct a 3D model of the patient’s anatomy as
The lateral borders of the chordoma were identified using well as to design a customized 3D-printed titanium im-
CT-based navigation, and en bloc resection of the lesion plant (Fig. 4). A standard L-4/L-5 anterior approach was
was achieved. performed, and the level of pathology was confirmed
After the en bloc resection, reconstruction and anterior with radiographs prior to the disc removal, with endplate
fixation was completed with the custom-designed 3D- preparation. The custom 3D-printed titanium implant
printed titanium cage filled with bone matrix and 2-screw was packed with the bone graft (Allograft, Ausbiotech-
fixation into the clivus and 2 screws into the C-3 vertebra nologies) and inserted, with a precision fit. Anterior plate
(Fig. 3). All screw lengths were predetermined based on fixation was implemented to secure the motion segment,
the preoperative 3D and plastic modeling, and screw an- with additional posterior percutaneous pedicle screws for
gulation was built into the prosthesis to ease screw inser- further stabilization of the L-4/L-5 segment. Radiography
tion. Following successful reconstruction, the mucosa and was used to confirm correct placement, and antibiotic ir-
palate split was closed, and the patient was transferred to rigation was used prior to closure (Fig. 5B and C).
the intensive care unit. The total length of the procedure
was 16.5 hours, with 480 ml of blood loss. Patient-Specific Implant Design
Postoperatively, the patient had temporomandibular The following design criteria were developed for the
joint dysfunction secondary to prolonged stretching of the required patient-specific interbody fusion implant. 1. Al-
oral cavity during the procedure. The prolonged opera-
tion duration also resulted in an upper esophageal stric-
ture. These complications were attributed to the lengthy
period of oral retraction. The patient recovered speech by
8 weeks postoperatively; however, he required percuta-
neous endoscopic gastrostomy to enable feeding for the
initial 3 months following surgery. Radiotherapy target-
ing the lateral aspect of the tumor resection margin was
performed at 6 months postsurgery. There were no com-
plications with this intervention. The patient had normal
phonation and swallowing function at his 9-month follow-
up. At the 9-month follow-up, flexion and extension ra- FIG. 4. Left: Plastic model of hemivertebra with planned 3D-printed
diographs demonstrated no movement and thus probable prosthesis. Right: Prosthesis prior to sterilization and implantation.
fusion. Figure is available in color online only.

J Neurosurg Spine Volume 26 • April 2017 515

Unauthenticated | Downloaded 03/21/24 04:18 PM UTC

R. J. Mobbs et al.

FIG. 5. A: Preoperative midsagittal CT. B: Postoperative anteroposterior radiograph. C: Postoperative lateral radiograph.

low maximum strength to support the weight of the pa- At 12-month follow-up, significant improvements were
tient while simultaneously providing a large central empty found in terms of axial back (from preoperative 10 to post-
space for bone graft delivery and containment. 2. Ensure operative 0), right leg (from preoperative 2 to postopera-
minimal support material for 3D printing to reduce post- tive 0), and left leg (from preoperative 2 to postoperative 0)
processing by being a structure that is “self-supporting” visual analog scale scores. The Oswestry Disability Index
during the building process. 3. Provide maximum sparse- also improved significantly, from 68% preoperatively to
ness to ensure the best possible postoperative imaging 0% postoperatively. The 12-month radiological follow-up
(Fig. 4 right). (Fig. 6) demonstrated solid mature fusion with no failure
The surgeon, in consultation with representatives of the of fixation and no subsidence.
involved 3D implant design and manufacture companies
(Anatomics, ProCRO), specified that the implant should
effect a lordosis change of 6°. The implant was then de- Discussion
signed around this altered geometry to meet the design Treatment of complex spinal pathologies such as a
criteria. The final design was used to manufacture the im- craniocervical junction chordoma or complex spinal con-
plant (RMIT University, Advanced Manufacturing Pre- genital deformity requires meticulous surgical technique
cinct, Carlton, Australia). The 3D-printed parts were then and considerable preplanning.13,18 In the case of the upper
checked for postprocessing as in Case 1 (Fig. 4). cervical chordoma, this particular procedure is associated

FIG. 6. Upper Row: Case 1. Nine-month follow-up radiographs showing no movement and likely fusion. Lower Row: Case 2.
Twelve-month radiographic follow-up demonstrating solid mature fusion, no failure of fixation, and no subsidence. Figure is avail-
able in color online only.

516 J Neurosurg Spine Volume 26 • April 2017

Unauthenticated | Downloaded 03/21/24 04:18 PM UTC

 Utility of 3D printing for implant design

with a high failure and morbidity risk, and such prolonged Science and society. Experts warn against bans on 3D
procedures are particularly high risk in the elderly.10 The printing. Science 342:439, 2013
difficulty and challenge of such complex spinal surgeries 2. Anderson JR, Thompson WL, Alkattan AK, Diaz O,
Klucznik R, Zhang YJ, et al: Three-dimensional printing of
can be anticipated and relieved by preoperative planning. anatomically accurate, patient specific intracranial aneurysm
The introduction of 3D printing has provided a means of models. J Neurointerv Surg 8:517–520, 2016
accurately depicting the patient’s surgical anatomy, as well 3. Barker FG: Repairing holes in the head: a history of
as the ability to fashion the custom-designed implants and cranioplasty. Neurosurgery 41:999, 1997 (Letter)
prostheses that match more closely the unique anatomy of 4. D’Urso PS, Askin G, Earwaker JS, Merry GS, Thompson
any particular patient and pathology.12–16 In both present- RG, Barker TM, et al: Spinal biomodeling. Spine 24:1247–
ed cases, the custom-designed and custom-built implants 1251, 1999
were easily slotted into position, which facilitated the sur- 5. Harwick RD, Miller AS: Craniocervical chordomas. Am J
Surg 138:512–516, 1979
gery and shortened the operating time, avoiding further 6. Izatt MT, Thorpe PLPJ, Thompson RG, D’Urso PS, Adam
complex reconstruction such as the harvesting of rib or CJ, Earwaker JWS, et al: The use of physical biomodelling in
fibular grafts and the fashioning of these grafts intraop- complex spinal surgery. Eur Spine J 16:1507–1518, 2007
eratively to fit the defect. 7. Klein GT, Lu Y, Wang MY: 3D printing and neurosurgery—
These 2 cases, which we approached by preoperative ready for prime time? World Neurosurg 80:233–235, 2013
planning using 3D-printed models as well as custom-de- 8. Mao K, Wang Y, Xiao S, Liu Z, Zhang Y, Zhang X, et al:
signed implants, are part of a larger paradigm of personal- Clinical application of computer-designed polystyrene
models in complex severe spinal deformities: a pilot study.
ized medicine. As an alternative to standard off-the-shelf Eur Spine J 19:797–802, 2010
implants, 3D-printed implants can be customized to the 9. Martelli N, Serrano C, van den Brink H, Pineau J, Prognon
unique anatomy of the patient. Although such develop- P, Borget I, et al: Advantages and disadvantages of
ments are exciting, this surgery has not disseminated 3-dimensional printing in surgery: A systematic review.
widely due to the high costs and the requirements of so- Surgery 159:1485–1500, 2016
phisticated software and machinery not easily accessible 10. Molina CA, Ames CP, Chou D, Rhines LD, Hsieh PC,
to the majority of surgeons. Zadnik PL, et al: Outcomes following attempted en bloc
Charles Hull first proposed the 3D printing process in resection of cervical chordomas in the C-1 and C-2 region
versus the subaxial region: a multiinstitutional experience. J
1986.1 Since that time, the 3D printing market has explod- Neurosurg Spine 21:348–356, 2014
ed, becoming increasingly more commonplace and lower 11. Namba K, Higaki A, Kaneko N, Mashiko T, Nemoto
in cost.9 The application of 3D printing in medicine and S, Watanabe E: Microcatheter shaping for intracranial
surgery may offer several benefits, including customiza- aneurysm coiling using the 3-dimensional printing rapid
tion of medical products and implants, improved preop- prototyping technology: preliminary result in the first 10
erative planning, reduced operative duration and surgi- consecutive cases. World Neurosurg 84:178–186, 2015
cal complexity, and accurate placement of implants and 12. Schubert C, van Langeveld MC, Donoso LA: Innovations
prostheses, as well as enhanced productivity. Particularly in 3D printing: a 3D overview from optics to organs. Br J
Ophthalmol 98:159–161, 2014
in the context of complex spinal and cranial surgery, 3D- 13. Song ZL, Feng CK, Chiu FY, Liu CL: The clinical
printed neuroanatomical models can be particularly help- significance of rapid prototyping technique in complex spinal
ful by providing an accurate representation of cranial deformity surgery—case sharing and literature review. J
nerves, vessels, and skull architecture and their relation- Musculoskeletal Disord 4:88–93, 2013
ships, which may be difficult to interpret using standard 14. Tan ET, Ling JM, Dinesh SK: The feasibility of producing
2D radiographic imaging.2,8,11,16,17 Accurate 3D models can patient-specific acrylic cranioplasty implants with a low-cost
help the surgeon plan and rehearse a safe surgical corri- 3D printer. J Neurosurg 124:1531–1537, 2016
15. Waran V, Narayanan V, Karuppiah R, Owen SLF, Aziz T:
dor or approach preoperatively, which may ease the diffi- Utility of multimaterial 3D printers in creating models with
culty and lower the operative duration of complex surgical pathological entities to enhance the training experience of
cases. neurosurgeons. J Neurosurg 120:489–492, 2014
Although the technique is promising, current barriers 16. Weinstock P, Prabhu SP, Flynn K, Orbach DB, Smith E:
include commercial costs of 3D printing of prostheses as Optimizing cerebrovascular surgical and endovascular
well as limited access to high-quality 3D printers. The procedures in children via personalized 3D printing. J
proliferation and increasing quality of low-cost desktop Neurosurg Pediatr 16:1–6, 2015
3D printers may be a game changer in the future, and it 17. Wurm G, Lehner M, Tomancok B, Kleiser R, Nussbaumer
K: Cerebrovascular biomodeling for aneurysm surgery:
may facilitate lower-cost, custom-designed implants and simulation-based training by means of rapid prototyping
prostheses. Several published reports have demonstrated technologies. Surg Innov 18:294–306, 2011
the feasibility of producing patient-specific cranioplasty 18. Yang JC, Ma XY, Xia H, Wu ZH, Ai FZ, Zhang K, et
implants using low-cost desktop 3D printers coupled with al: Clinical application of computer-aided design-rapid
cheap and widely available acrylic bone cement.14 Possible prototyping in C1–C2 operation techniques for complex
future directions of research include the combination of atlantoaxial instability. J Spinal Disord Tech 27:E143–
3D-printed implants and biologics, as well as the develop- E150, 2014
ment of bioceramic composites and custom implants for
load-bearing purposes.
Disclosures
References Robert Thompson is an employee of Anatomics Pty. Ltd., who
1. American Association for the Advancement of Science: were involved in the design and manufacture of the 3D-printed

J Neurosurg Spine Volume 26 • April 2017 517

Unauthenticated | Downloaded 03/21/24 04:18 PM UTC

R. J. Mobbs et al.

implants reported in this study. Chester E. Sutterlin III is a repre- the article: Mobbs, Thompson, Sutterlin, Phan. Critically revising
sentative of and has ownership in ProCRO Pty. Ltd., which was the article: Mobbs, Coughlan, Phan. Reviewed submitted ver-
involved in the design and manufacture of a 3D-printed implant sion of manuscript: Mobbs, Coughlan, Phan. Study supervision:
reported in this study. Dr. Mobbs is a consultant for Stryker, Mobbs.
Kasios Biomaterials, and A-Spine Asia.
Correspondence
Author Contributions Ralph J. Mobbs, NeuroSpineClinic, Ste. 7, Level 7 Prince of
Conception and design: all authors. Acquisition of data: all Wales Private Hospital, Randwick, Sydney NSW 2031, Australia.
authors. Analysis and interpretation of data: all authors. Drafting email: ralphmobbs@hotmail.com.

518 J Neurosurg Spine Volume 26 • April 2017

Unauthenticated | Downloaded 03/21/24 04:18 PM UTC