Rapid Prototyping for Patient-Specific Surgical Guides in Orthopaedics: A Systematic Literature Review

Review Article
Proc IMechE Part H:
J Engineering in Medicine
2016, Vol. 230(6) 495–515
Ó IMechE 2016
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0954411916636919
pih.sagepub.com
Rapid prototyping for patient-specific
surgical orthopaedics guides: A
systematic literature review
Diana Popescu and Dan Laptoiu
Abstract
There has been a lot of hype surrounding the advantages to be gained from rapid prototyping processes in a number of
fields, including medicine. Our literature review aims objectively to assess how effective patient-specific surgical guides
manufactured using rapid prototyping are in a number of orthopaedic surgical applications. To this end, we carried out a
systematic review to identify and analyse clinical and experimental literature studies in which rapid prototyping patient-
specific surgical guides are used, focusing especially on those that entail quantifiable outcomes and, at the same time, pro-
viding details on the guides’ design and type of manufacturing process. Here, it should be mentioned that in this field
there are not yet medium- or long-term data, and no information on revisions. In the reviewed studies, the reported
positive opinions on the use of rapid prototyping patient-specific surgical guides relate to the following main advantages:
reduction in operating times, low costs and improvements in the accuracy of surgical interventions thanks to guides’ per-
sonalisation. However, disadvantages and sources of errors which can cause patient-specific surgical guide failures are as
well discussed by authors. Stereolithography is the main rapid prototyping process employed in these applications
although fused deposition modelling or selective laser sintering processes can also satisfy the requirements of these
applications in terms of material properties, manufacturing accuracy and construction time. Another of our findings was
that individualised drill guides for spinal surgery are currently the favourite candidates for manufacture using rapid proto-
typing. Other emerging applications relate to complex orthopaedic surgery of the extremities: the forearm and foot.
Several procedures such as osteotomies for radius malunions or tarsal coalition could become standard, thanks to the
significant assistance provided by rapid prototyping patient-specific surgical guides in planning and performing such
operations.
Keywords
Patient-specific guide, rapid prototyping, orthopaedic instrumentation, computer-aided surgery
Date received: 1 October 2015; accepted: 3 February 2016
Introduction
Rapid prototyping (RP) is a group of manufacturing
processes that can build physical objects directly from
three-dimensional (3D) virtual model data in an addi-
tive way, which is to say, by superimposing layers of
material one on top of the other. Other terms used for
this kind of processes are layer manufacturing, layer
fabrication, solid freeform fabrication and layer-by-
layer fabrication. Since 2013, the standard name, addi-
tive manufacturing (AM) has been defined as ‘the pro-
cess of joining materials to make objects from 3D
model data, usually layer upon layer, as opposed to
subtractive manufacturing technologies’.1 However, the
analysis carried out for this article showed that the
term RP is used more often in the medical literature, as
the advantages offered by such processes are in direct
correlation to surgical guides’ individualisation, that is,
the concept of a prototype.
Personalised healthcare is becoming an increasingly
salient approach to medicine,2 and advances in both
medical and technical fields are being focused on pro-
viding solutions for medical interventions and devices
Politehnica University of Bucharest, Bucharest, Romania
Orthopaedics, Clinical Hospital Colentina, Bucharest, Romania
Chelariu Clinic, Bacau, Romania
Corresponding author:
Dan Laptoiu, Orthopaedics, Clinical Hospital Colentina, Sos. Stefan cel
Mare, 19-21, code 020125 Sector 2, Bucharest, Romania.
Email: [e-mail]
at University College London on May 27, 2016pih.sagepub.comDownloaded from

tailored to the patient’s bone morphology and individ-
ual needs. Patient-specific surgical guides (PSGs) are
part of this approach, thus representing a subject of
interest for both surgeons and engineers.
The use of RP processes to manufacture patient-
specific surgical guides (RP-PSGs) specially designed
for a specific patient can be traced back in 1997, when
Van Brussel et al.3 reported the design, manufacture
and use of the first individualised template for drilling
trajectories when inserting pedicle screws into the verte-
bra of a human spine. In 1998, Radermacher et al.4
employed the stereolithography (SL) process to obtain
the physical prototype of a PSG for pelvic osteotomies.
Berry et al.5 also engaged in studies of image processing
methods, data conversion and manufacturing using the
selective laser sintering (SLS) process. Ever since then,
the number of such applications has been increasing
thanks to a greater awareness of the advantages offered
by RP processes in manufacturing objects with compli-
cated geometrical features (freeform features). This is
the case of PSGs designed to exactly fit the patient’s
anatomical structures, thus supporting increased preci-
sion in a number of surgical procedures.
The improvement in the accuracy of screw implanta-
tion techniques and other orthopaedic surgical proce-
dures has been possible through the use of radiological
examination during surgery. The irradiation during
such procedures is higher in percutaneous surgery,
where, in order to make as small as possible incisions,
the use of interventional radiology is required when
identifying anatomical landmarks. Therefore, in recent
years, intra-operative navigation systems have been
developed and implemented in operating theatres in
order to visualise the patient’s anatomical structure
without resorting to interventional radiology. This
approach is based on pre-operative image acquisition
and on the use of an anatomical landmarks system cali-
brated at the beginning of the surgical intervention.
However, performing this calibration process is a com-
plicated task. Moreover, the static landmarks estab-
lished at the beginning of surgery do not always
maintain the same position throughout the surgical
procedure, which obviously leads to imprecision. In
this context, PSGs can represent an alternative when
used for guiding surgical actions such as drilling, tap-
ping, cutting or axis alignment, since they transfer to a
physical object (called guide, jig or template) the
planned trajectories required in order to prepare the
bones for the implantation and fixation of screws, rods
and plates.6 These guides can also help to identify the
correct position/orientation of instruments or substi-
tute for an instrument, as is the case in total knee
arthroplasty (TKA) applications,7 for instance.
The personalisation of surgical guides implies coop-
eration between engineers and medical specialists8 when
obtaining patient’s scan data, modelling the anatomical
areas of interest, planning the surgery/tool trajectories,
designing the guide, choosing the material and manu-
facturing process, building the guide, and sterilising
and utilising it. The flow is similar for all PSGs, but
what differs is the design of the guides, which is dic-
tated by the patient’s anatomy, the type of intervention
and surgical approach, as well as by the anatomical
landmarks selected by the surgeon in the pre-operative
stage. Other requirements, such as transparency9 or
multi-level guides10,11 for spinal applications, may also
be considered convenient solutions in some medical
cases.
However, despite reported cases of orthopaedic sur-
gical operations in which PSGs manufactured via RP
processes have been used, to the best of our knowledge
literature reviews have been conducted only for
TKA7,12,13 and cranio-maxillofacial applications.14 No
systematic information is yet available for other ortho-
paedic applications such as spinal surgery or correc-
tions of malunions or deformities presented in
literature.
In this context, this review addresses these applica-
tions, the information in the article being organised so
that to answer four questions. This approach represents
a modality to organise the information found when
querying the literature databases, the questions provid-
ing a structured and systematic manner to analyse the
data presented in RP-PSGs literature studies. Thus,
these questions can be considered as filters applied for
screening the literature in the field. Moreover, these are
the questions that engineers and surgeons are first ask-
ing themselves when decide to start build and use such
devices.
These questions are as follows:
Q1. Has there been a significant constant increase in
the use of RP-PSGs in the last couple of years?
Q2. What is the reported efficiency (in terms of accu-
racy, operation time saving and costs) of RP-PSGs use
in orthopaedic surgery?
Q3. Is there a preferred RP process for manufacturing
such guides and, if so, why?
Q4. What are the main aspects taken into account dur-
ing the RP-PSGs design process?
The answers offered by our article aim to provide the
basis for a more accurate understanding of the use,
design and manufacturing process, on one hand, and of
the advantages and pitfalls of RP-PSGs in several ortho-
paedic surgery interventions, on other hand. Thus,
knowledge can be gathered and hopefully used for fur-
ther studies and enhancements in the field by providing
possible suggestions for improvements in PSGs and for
their use in other types of surgical applications.
Materials and methods
The systematic literature review was conducted based
on the flow presented in Figure 1.
In December 2014, we searched the computerised
databases of medical journals (in the following order:
496 Proc IMechE Part H: J Engineering in Medicine 230(6)
at University College London on May 27, 2016pih.sagepub.comDownloaded from

PubMed, Springer, Elsevier and SearchMedica) for the
following types of English-language studies: clinical
trials, comparative studies, evaluation studies, journal
articles, meta-analysis, reviews, systematic reviews and
validation studies.
The timeframe we selected for the search was 2005–
2014. Although initially we intended to focus only on
research carried out after 2009, we subsequently rea-
lised that a review of a longer period (10 years) would
be more appropriate if we were to identify a growing
trend. We thought to consider 2009 as our reference
point, because this was the year when the patent for the
fused deposition modelling (FDM) process expired,
bringing about a ‘democratisation’ of the field and a
substantial increase in the number of applications and
manufactured parts, which could also have an impact
on the use of RP in medical applications.
Combinations of the following keywords were used
in the search on PubMed: ‘rapid prototyping’ and
‘guide’, ‘template’, ‘patient-specific instrumentation’,
‘cutting’, ‘drill’ and ‘osteotomies’ (all fields).
Combinations of words containing the terms: ‘additive
manufacturing’, ‘jigs’ and ‘3D printing’ were also used,
but they did not generate new results, that is, results
other than those presented below.
Thus, we identified 344 potentially eligible studies, as
follows:
 Rapid prototyping AND guide: 95 results;
 Rapid prototyping AND cutting: 23 results;
 Rapid prototyping AND drill: 34 results;
 Rapid prototyping AND patient-specific instru-
mentation: 27 results;
 Rapid prototyping AND osteotomies: 72 results;
 Rapid prototyping AND template: 93 results.
The searches on SearchMedica, Springer and
Elsevier produced mostly duplicate results of those
from PubMed, while additional references were
obtained through examination of the bibliography of
the articles using a shortlist obtained after two screen-
ings, as we detailed below.
In order to eliminate duplicates, the results were fil-
tered, first using the functionalities offered by PubMed
and then by ordering all the search outcomes by author.
We were left with 224 papers to provide our focus.
The next step was to apply exclusion criteria. We
decided to remove from the systematic literature review
papers on RP general applications in medicine and
papers that related RP processes to implants manufac-
turing or scaffolds fabrication, as these were not ger-
mane to our questions and selected field of interest
(orthopaedic surgery). In addition, studies on the use of
RP-PSGs in TKA and cranio-maxillofacial applications
were also excluded, given the existence of new and
Figure 1. Flowchart for studies selection.
Popescu and Laptoiu 497
at University College London on May 27, 2016pih.sagepub.comDownloaded from

comprehensive surveys. Nor were studies presenting
TKA surgical guide applications from companies such
as Smith and Nephew, Biomet and Zimmer included in
the survey in order to avoid presentation of potentially
biased information, and also because they are not offer-
ing too many technical details on the design aspects and
manufacturing process. Moreover, Thienpont et al.15
already presented a comprehensive survey (Europe and
worldwide) on patient-specific instrumentation for
TKA (including RP guides), based on information from
orthopaedic companies (2011–2012 volumes of sales).
Therefore, we focused our review only on the cases pre-
sented in the literature as we consider that they can pro-
vide documented technical information on different
aspects of the development, use and evaluation pro-
cesses of this type of surgical guides.
Papers with English titles that came up in the search
results, but which were written in other languages
(German, Chinese and Japanese) were also eliminated
from our list. We assumed that it was more than likely
that the results laid out in these papers had also been
published in English in various journals.
In the end, 52 studies were retained for further anal-
ysis. When filtering titles and abstracts, we noticed the
very large number of RP applications in manufacturing
PSGs for dental applications, an area better developed
than that of orthopaedic surgery.
In the next stage of filtering, the articles were
reviewed in full by both authors independently, in
order to establish mutual agreement, especially given
that their separate fields of specialisation (engineering
and medicine) might determine different perspectives
on and understandings of the same subject. In addition,
a supplementary search was carried out using the refer-
ences found in the 52 articles, thereby generating nine
further papers relating to the subject.
Thus, a total of 61 studies presenting applications of
RP processes in the manufacturing of PSGs for ortho-
paedic surgery, in both clinical cases and experiments
on cadavers, were selected for ‘List 1’. However, not all
of these articles contained sufficient information or
outcomes (i.e. radiological data for PSG usefulness
assessment, information on guide design, information
on computed tomography (CT) scanning protocols,
types of RP process and materials) to answer our
questions. Furthermore, some of the articles presented
information or preliminary information that was later
repeated, in a more detailed manner, in other papers or
book chapters. For these reasons, 23 papers were elimi-
nated from ‘List 1’, and the remaining 38 articles
(named as ‘List 2’) are discussed separately in section
‘Discussion’. The articles in ‘List 2’ were divided into
two broad categories: RP-PSGs for spinal surgery
applications (discussed in section ‘RP-PSGs for spinal
surgery applications’) and RP-PSGs for other general
orthopaedic surgery applications (osteotomies, tumour
resections, etc., discussed in section ‘RP-PSGs for gen-
eral orthopaedic surgery applications’).
In conclusion, the articles in ‘List 1’ were used to
answer Q1 and Q2, and, in part, the other questions,
while the articles in ‘List 2’ provided a detailed exami-
nation of specific RP-PSGs design and manufacturing
criteria, diverse methods and approaches, reported effi-
ciency, thereby providing answers to Q3 and Q4.
Results
The articles from ‘List 1’ contained sufficient informa-
tion to allow us to plot the charts presented in Figures 2
and 3. Figure 2 shows the distribution of the number of
RP-PSGs orthopaedic surgery studies per year, provid-
ing an answer to the first question, while Figure 3 pre-
sents the number of RP-PSGs applications in spinal
surgery.
Table 1 presents the distribution of studies on anato-
mical areas, based also on information from ‘List 1’.
Table 2 summarises the data of several studies
regarding the use of RP-PSGs for spinal surgery appli-
cations, with the studies used to extract information
were those from ‘List 2’. The individualised RP tem-
plates for these applications are employed both for fix-
ing vertebra using screws and for guiding other
orthopaedic procedures – osteotomies being the most
complex, because they frequently require multi-planar
corrective incisions of the bone followed by assembly
and fixation. Thus, the RP-PSGs are employed to loca-
lise entry points and to transfer the pre-planned tap-
ping and drilling tool trajectories from computer
simulation of correction to real surgery.
Figure 2. Number of RP-PSGs studies in orthopaedic surgery.
498 Proc IMechE Part H: J Engineering in Medicine 230(6)
at University College London on May 27, 2016pih.sagepub.comDownloaded from

Table 3 summarises data from studies focused on
different applications of RP-PSGs over the extremities
(femur, forearm, tibia, radius, cubitus, etc.).
Tables 2 and 3 also contain columns (called Results)
with information on how RP-PSGs are evaluated. Both
qualitative opinions and quantitative methods/mea-
surements are presented by the authors of the reviewed
studies. Therefore, these columns contain diverse data
on the research outcomes, time of intervention when
using PSGs or PSGs’ costs (where available).
Figure 4 indicates the number of clinical studies as
opposed to other studies (clinical/cadavers, cadavers,
sawbones) for each main application category under
consideration: spine applications (10 clinical, 1 clinical/
cadavers, 1 synthetic spine/animal tests, 9 on cadavers),
other applications (13 clinical, 1 experimental/saw-
bones, 1 clinical/experimental and 2 experimental).
These data were retrieved from the papers in ‘List 2’.
Corroborating this information with that presented in
Tables 2 and 3, one can observe that for all the applica-
tions, the number of clinical studies is higher (23–15)
although for spinal surgery, the number of clinical trials
as opposed to cadaver studies is almost the same.
Figure 5 shows data on the type of RP process used
in each article under review, with this information
being extracted from the studies in ‘List 2’. In the case
of spinal surgery applications, the RP process used in
manufacturing PSGs are as follows: SLS – two clinical
studies by Merc et al.,10,11 3D printing – one cadaver
study, FDM – two cadavers’ studies and PolyJet – two
clinical studies. The other 14 applications use SL
process.
Discussion
The expiration of the FDM process patent in 2009 led
to a noticeable increase in the number of RP parts man-
ufactured in different fields and for different purposes.
In addition, the cost of the manufactured parts
decreased, while there was a dramatic increase in the
overall number of machines sold and the number of
published scientific studies and articles containing gen-
eral information. However, in regard to the medical
field, we considered that the publicity surrounding RP
has a greater impact than the ‘democratisation’ of
access to such processes, as the most part of the medical
prototypes are built using SL machines (which continue
to be expensive) by companies that provide services in
the RP field, rather than directly by hospitals.
Table 1. Anatomical regions interested in the retrieved studies – List 1.
Anatomical region Number of
studies
Details
Arm/shoulder 2 Debarre et al., 16 Tricot et al. 17
Wrist/forearm 20 Mahaisavariya, 18 Murase et al., 19 Oka et al., 20,21 Murase et al.,22 Hsieh et al., 23
Stockmans, 24 Zhang et al., 25 Oka et al.,26,27 Miyake et al., 28–30 Kataoka et al., 31
Kunz et al., 32 Schweizer et al., 33 Takeyasu et al.34 and Omori et al.35,36
Hand 1 Imai et al. 37
Spine 28 Berry et al., 38 D’Urso et al.,39 Owen et al., 40 Ryken et al., 41,42 Zhang et al., 43 Lu
et al., 44–46 Zhang et al., 25 Lu et al., 47,48 Wu et al., 49 Takemoto et al., 50 Lu et al.,51
Kawaguchi et al., 52 Ma et al., 53 Hu et al., 54 Fu et al., 55 Ferrari et al., 56 Sugawara
et al., 57 Merc et al.,11 Hu et al., 58,59 Merc et al.,10 Tomnic et al., 60 Kaneyama et al. 9
and Li et al.61
Thorax 1 Yang et al. 62
Hip/thigh 6 Pressel et al., 63 Hung et al., 64 Bellanova et al., 65 Chai et al.,8 Cartiaux et al. 66 and
Blakeney et al.67
Foot 1 De Wouters et al.68
Tibia 2 Dobbe et al.69,70
Figure 3. Studies per year on RP-PSGs for orthopaedics (spine vs other applications) – List 1.
Popescu and Laptoiu 499
at University College London on May 27, 2016pih.sagepub.comDownloaded from

Table 2. Synthetic data on RP-PSGs for spine surgery applications – studies in List 2.
Study Cadaver/
clinical
Single/
multi-level
Spine zone No. of
screws
RP process PSG design approach Results
Merc et al.11 Clinical
9 patients
Multi-level
2, 3 levels
Lumbo-sacral 54 SLS, polyamide Cylinders connected left-
right across the spinous
process, fitting the dorsal
parts of the facet joints
Comparison: 9 patients in RP-PSGs group
and 10 patients in the freehand group
Post-op CT scans evaluation. Measurements:
cortex perforation – significantly lower for
RP-PSGs group
screw length violation – less frequent for RP-
PSGs, but not significantly
Disadvantages: precise strip of soft tissue,
difficulty to avoid the tipping of the guide in
the transversal plane
Merc et al.10 Clinical
11 patients
Multi-level
2, 3 levels
Lumbo-sacral 72 SLS, polyamide Post-op CT scans evaluation. No cortex
violation. 26% of screws implanted
inaccurately, but strongly. Despite the high
rate of inaccurate screws insertion, authors
considered that RP multi-level guides provide
satisfactory accuracy and ‘a precise screw
placement with low pedicle perforation
incidence and a clinically unimportant mistake
factor’
Lu et al. 46 Clinical
9 patients
Single Cervical 19 SL, acrylate resin Negative/inverse of C2
laminar and spinous
process
Post-op CT scans evaluation. No bony breach
during RP-PSGs use.
Mean time between placing the guide and
inserting the screw: 1–2 min.
Around 16 h to manufacture each RP guide.
Price for each model of vertebra and guide:
US$ 20
Disadvantages: learning curve, precise soft
tissue removal, any movements between
bones affect implant accuracy.
‘Promising alternative for C2 laminar screw
placement’
Lu et al. 47 Clinical
25 patients
Single Cervical 84 SL, acrylate resin Negative of the postural
surface of the vertebrae
Implant positioning assessment using x-rays
and CT scans.
82 screws rated as grade 0, 2 screws as grade
1.
RP-PSG is easy to use, ensures ‘highly
accurate cervical pedicle screw placement’.
Same disadvantages as in previously
mentioned studies
(continued)
500 Proc IMechE Part H: J Engineering in Medicine 230(6)
at University College London on May 27, 2016pih.sagepub.comDownloaded from