Education
3D printing provides a useful method for creating orbital models with similar rigidity as bone for anatomical education and surgical training [
56,
57]. Traditional anatomy education with cadavers may be limited by access, cost, and social stigma. 3D-printed orbital models were created by Adams et al. based on 3D designs recreated from cadaveric prosections [
58]. However, these designs can also be achieved through the use of magnetic resonance imaging (MRI) and CT studies [
59]. Orbital anatomy is complex, difficult to conceptualize externally, and may vary significantly between patients; furthermore, the consequences of violating the intracranial space during surgery include meningitis, cerebrospinal fluid leak, and intracranial hemorrhage. Trainees can hone skills such as lateral wall decompression in a safe and easily reproducible context [
56]. Soft tissues are more difficult to model due to less crisp delineation on imaging and lack of conducive printing materials [
60]. Soft tissue has been more challenging to print with realistic quality, but can still be of educational value [
57,
58]. Adjacent structures such as the pterygopalatine fossa can be modeled as well [
61].
Repair of Bone Defects
Patient-specific orbital models can be used for orbital surgical planning of bone defects. Models allow for preoperative familiarization of patient-specific anatomy and enable the surgeon to select the optimal location and material for implants and screws [
7,
62]. They have also been shown to improve patient understanding [
63].
Accurate sizing of an implant to repair an orbital fracture is technically challenging and typically requires multiple intraoperative implant modifications [
64]. Even with experienced surgeons, postoperative complications such as globe malposition can occur [
65]. Implant templates can also be printed by virtually mirroring the unaffected orbit on the normal side [
66]. Both CT and MRI have been shown to create anatomical models accurately [
67,
68]. Templates can be sterilized for intraoperative use without losing their shape [
69,
70].
Optimal implant sizing through 3D printing helps prevent globe misalignment, strabismus, blood loss, and other postoperative complications [
70‐
73]. Preoperative sizing reduces surgical time and limits fatigue of the implant material due to fewer adjustments [
74]. It also reduces the risk of misplacement and damage to adjacent structures, such as positioning an implant too posterior and impinging the optic nerve [
75]. Furthermore, this approach can be combined with intraoperative navigation systems to guide optimal placement [
76,
77].
The orbital literature concerning 3D-printed bony defect templates includes numerous reports of successful use in the orbital floor, medial wall, and lateral wall [
67,
78‐
81]. Chai et al. published a study of 127 fracture repairs done with the aid of 3D-printed skull models [
70]. Oh et al. reported successful outcomes without complication in 104 patients [
82]. Tel et al. described 14 orbital floor fractures repaired via a transmaxillary endoscopic approach with CT-guided 3D prints used to fit autologous bone into the defects [
71].
Complex, multi-wall fractures have been addressed via this approach as well. Combined inferior and medial wall fractures pose a particular challenge given destruction of the inferomedial orbital strut. Kim et al. used 100 cadaver skulls to create a digital model of a standardized inferomedial orbital strut used in 3D printing of polycaprolactone implants [
83]. Interlocking puzzle-piece implants can be used to bridge two or more implants across missing landmarks in cases of adjacent wall fractures; doing so may reduce incisional burden by enabling piece-by-piece placement through the same incision [
84‐
87].
Several comparative studies have shed light on the impact of this technology in clinical practice. Fan et al. compared 29 cases of fracture repair using 3D-printed technology with 27 cases addressed with conventional methods. They found that patients in the 3D printing group had statistically significantly better implant fit, shorter surgical duration, and better outcomes at their postoperative assessment [
88]. Kozakiewicz et al. showed an improvement in binocular single vision and globe motility in a retrospective study of 12 fractures treated with 3D-printed templates and 12 treated with conventional implant manual manipulation [
89]. Sigron et al. found that ten patients treated with pre-bent implants based on printed models had lower volume difference between their two orbits and a 42% decrease in surgery time [
90]. In a retrospective study of 82 patients, Kim et al. found that the 3D printing group had a smaller bony defect, shorter gap between implant and bony edge, and smaller angle between implant and natural bony contour [
91]. Incorporating 3D printing into orbital fracture repair has been found to reduce surgical time by as much as 50% [
70,
92].
3D printing has also been used in the bony reconstruction of congenital craniofacial malformations, including orbital defects such as hypertelorism [
62,
93]. It has been used to characterize the location of an intraorbital cyst in a patient microphthalmia and design an orbital floor implant [
94]. In such cases, 3D-printed models can be particularly useful for explaining the bony deformities and surgical repair to patients and families [
62].
In addition to using printed models as a template to size implants, custom patient-specific implants can be produced directly using 3D printing technology. Implants have been produced by casting polymethyl methacrylate in printed injection molds [
95]. Implants can also be printed themselves. Several authors have reported successful repairs using printed polyetheretherketone [
77,
96,
97]. Titanium implants can be printed as well [
76,
77,
98‐
100]. Kitabata et al. printed a rigid titanium floor implant with reduced risk of buckling, but note that the process took 1 day and was more expensive than a traditional flexible titanium implant [
101]. Interlocking puzzle-piece implants have also been printed [
77,
85,
87]. Printing of implants results in a more rigid titanium implant; it also produces softer edges than those of traditional implants which are cut from titanium mesh, thereby reducing the risk of soft tissue entrapment [
85]. In a series of 34 orbital fractures, custom-fit 3D-printed implants showed a significant advantage over pre-bent implants in terms of volume difference between the operative and unaffected orbit [
102].
Limitations to 3D printing for the repair of bony implants include cost and time, with directly printed implants taking as long as 5 days to produce [
85]. Manipulation of preformed implants may still be required in complex fractures with missing landmarks and unpredictable final orbital shape or to avoid neurovascular structures [
64,
103]. While most printing materials can withstand sterilization, melting point is a concern for some [
104]. There are regulatory concerns as well. Custom-made implants are not subject to the same rigorous process of regulatory approval as mass-produced medical devices, which may create liabilities for institutions which print and use them [
105,
106].
Repair of Anophthalmic Sockets
Production of an ocular prosthesis typically entails the impression of hydrocolloid in the patient’s orbit, then fitting a scleral shell into this space [
107]. A prosthesis can be printed based on a CT scan of a molded wax model [
108]. However, the molding process is uncomfortable for the patient and can shift the soft tissues and decrease accuracy [
107]. Several authors have demonstrated methods of virtually modeling of the patient’s orbit for prosthesis printing. 3D-printed scleral shells can be printed from either CT or corneoscleral topography scans and undergo standard post-fitting modifications such as iris art and acrylic finish [
109‐
111]. Two authors have described a method for avoiding the manual artwork by printing a shell and layering on an iris with a sublimation transfer method [
112,
113]. The resulting product met safety standards including cytotoxicity and tissue reactivity and had strength testing comparable to that of conventional prostheses [
112]. Huang et al. printed a custom-designed surgical guide to optimize the placement of titanium osseointegrated implants for retention of an orbital prosthesis [
114].
3D printing can streamline the production of ocular implants (the space-occupying item behind the visible prosthesis). A series of ten patients underwent implantation of a 3D-printed sphere following evisceration. None of the patients developed systemic or local toxicity, infection, inflammation, extrusion, or exposure [
115]. Implant migration can cause poor prosthesis fit and cosmesis, often necessitating a secondary dermis fat graft. Dave et al. used a CT-guided virtual skull model to 3D-print a secondary implant to recenter an inferotemporally displaced primary implant [
116].
Following enucleation, a conformer must be placed to prevent tissue contraction, symblepharon, and forniceal shortening while the tissues heal. Poor conformer fit prevents proper healing and may cause conformers to fall out or be removed by patients. Custom conformers have been printed for this purpose [
117]. Serial 3D printing has also been used to produce successive socket conformers in orbital expansion therapy for patients with microphthalmia [
118].
Weisson et al. custom-printed orbital exenteration prostheses based on facial topography scans for three patients who had stopped wearing their initial prosthesis due to issues with fit, color, or degradation. The exenteration prosthesis can integrate with a separately produced ocular prosthesis [
119]. Similarly, a 3D-printed orbital mold can be used for casting of a polydimethylsiloxane exenteration prosthesis [
120].