We present two illustrative cases demonstrating utilization of 3-D generated prostheses for limb salvage in the treatment of end-stage talar pathology complicated by infection. While the use of 3-D printing to create custom implants is emerging, little has been written regarding its use to create custom antibiotic cement spacers to enhance reconstructive and salvage solutions for the foot and ankle.
Antibiotic-laden cement is widely used across orthopaedics. Most often, it is used as a bridge in an infected revision/reconstructive procedure acting as a physical space holder and as an antibiotic carrier directly to the affected tissues. Literature discussing cement spacers for permanent implantation in the foot and ankle is limited to case reports and short case series. Hong et al reported a successful outcome of modified hindfoot fusion and midfoot antibiotic cement spacer with overlying adipofascial flap coverage in a 53 year old woman with severe Charcot destruction and ulceration.  Forefoot osteomyelitis treated similarly had encouraging outcomes as reported by Melamed et al in their series of 23 patients.  The cement spacer was molded in situ to fill the defect and was left permanently in nearly 50% of patients. The locations of retained spacers included the hallux interphalangeal joint, the 1st metatarsal head, and the metatarsophalangeal joint. The overall success rate was 91.3% at an average of 21 months, defined by the authors as resolution of infection without the need for amputation. Elmarsafi and colleagues reached similar conclusions regarding the use of cement spacers for limb salvage in diabetic patients with osteomyelitis of the mid and forefoot. They documented a 66.7% salvage rate in 30 high-risk diabetic patients with an average follow up of 52 months and concluded that cement spacers are safe and durable and provide a viable definitive reconstructive option. 
Nonanatomic cement spacers for the tibiotalar joint have been reported as definitive treatment with mixed results. Ferrao et al described their experience treating postoperative ankle infections after total ankle replacement (TAR) or arthrodesis. Their series included 6 patients with infection after TAR and 3 patients after arthrodesis that were either medically unfit for further surgery or preferred to retain their spacer. Two patients required below knee amputation for recurrent infection while the remaining seven patients with retained cement spacers were able to perform basic activities of daily living with minimal discomfort at an average of 20.1 months (range 6–62 months). They noted a tendency for spacer loosening and migration, although no patient demonstrated excessive bone loss at final follow up. The spacers were shaped by hand in situ, and the authors emphasized the importance of limiting cement protrusion anterior to the tibia to avoid pressure on the skin. Lee and colleagues reported on functional outcomes after cement arthroplasty as the primary salvage procedure for ankle joint destruction. The authors followed 16 patients with intractable infection, nonunion, and/or a large bone defect or tumor requiring a cement spacer for an average of 39 months (range 14–100). Two patients demonstrated subluxation and osteolysis at 10 and 78 months. Mean AOFAS and VAS scores improved significantly from 39 pre-operatively to 70 post-operatively and from 8 to 3, respectively. Nine of 16 patients were able to walk continuously for more than an hour. The authors molded the cement by hand in situ, adding screws in some cases to augment stability and maintaining overall mechanical alignment of the joint manually while the cement cured. They concluded that primary cement arthroplasty is a viable option for elderly or low-demand patients.
Custom antibiotic cement spacers have long been utilized in the treatment of prosthetic joint infection (PJI) in total hip and knee replacement.[7, 8, 9, 10] While a complete review is beyond the scope of this work, more advanced spacer designs used in hip and knee surgery may demonstrate similar benefit for the foot and ankle. Articulating spacers have been associated with improved outcomes compared to static spacers in revision total knee replacement (TKR) for infection, possibly by preserving physiologic motion of the soft tissue envelope prior to reimplantation. Patients with articulating spacers for PJI after total hip replacement (THR) had better functional scores than patients awaiting primary THR but lower than patients after successful primary THR in one Canadian study. Several studies have described the use of antibiotic spacers as long-term or definitive treatment in the THR, TKR, and total shoulder replacement (TSR) literature.[13, 14, 15, 16] Currently, there are no prefabricated articulating TAR spacers on the market, but the body of hip and knee literature on this topic is invaluable as revision TAR technology advances.
There are few investigations in the foot and ankle literature reporting increased customization of cement spacers for staged reconstruction of the tibiotalar joint after infection or trauma.[17, 18] Short et al described their technique for intra-operative fabrication and placement of an articulating antibiotic cement spacer through an anterior approach in a patient with an infected TAR. The implant provided 10 degrees of dorsiflexion and 15 degrees of plantarflexion. Extrapolating from the arthroplasty literature, the authors suggested that articulating spacers for the foot and ankle might have similar advantages over static spacers such as delivering high doses of antibiotics locally while minimizing soft tissue contractures and maintaining some joint range of motion.[11, 18] In another case report, Huang and colleagues contoured a cement spacer by hand in a case of traumatic complete talar extrusion. They used the available extruded talus to guide their back table fabrication of a replica antibiotic cement talus. This closely-matched replica talus was implanted and an external fixator applied in order to preserve leg length, joint space, and hindfoot alignment.17 Three months later, the patient underwent successful staged tibiocalcaneal fusion with femoral head interposition allograft.
Only one case report has documented the use of an anatomically-contoured total talus cement spacer as definitive treatment after total talus extrusion. Chiu and colleagues initially treated a 30-year old man for traumatic loss of the talus with an external fixator and a hand-molded antibiotic spacer intended to fill the void left by the talus. Due to persistent instability, wound drainage, and infection, they produced an anatomically-molded spacer in the shape of the patient’s native talus by using a CT scan of the contralateral talus to 3D print a negative mold for the affected talus. Stability improved immediately after insertion of the anatomic spacer, and the patient was allowed full weightbearing approximately 5 weeks after removal of the external fixator. Infection was eradicated, and at 14 months’ follow up the patient was able to walk for 15 minutes without pain and had a 15-degree arc of motion at the ankle.
There are additional pragmatic considerations of custom printing, mainly cost and time sensitivity. Production of custom 3-D implants and molds may incur additional cost as compared to more traditional treatment options. However, this must be weighed in relation to the potential financial and personal costs incurred by the patient and health system of repeated surgeries, complications and lifetime prosthetic fitting if limb salvage is unachievable. For the illustrative cases described in this report, the total cost for fabrication of the custom produced cement molds was comparable to the cost of prefabricated knee or hip spacer molding kits currently available. While most conditions can be temporized by other surgical procedures until a custom 3-D implant is produced, the luxury of time may not be available in all scenarios. The total time, from the initiation of the design phase through final production of our custom cement molds, was less than 5 days. This turnaround time was expedited by utilizing the previously submitted CT imaging and with fluid communication between the surgeon and the design team.
Finally, the use of 3D-printed antibiotic spacers should be considered only when no other viable, time-tested options for limb-salvage remain. Although no supportive literature currently exists, joint preservation in the long-term is unlikely given the coefficient of friction between the cement/articular cartilage interface. While it is unclear whether custom 3-D printed cement spacers will be viable long-term options, we present it as a feasible option for acute limb-salvage. Furthermore, eradication of infection while maintaining normal joint space congruity may allow for future reconstructive options in some patients where amputation may seemingly be the only available immediate treatment option.
Custom 3D-printed antibiotic spacers have the potential to become an important solution for limb salvage in-particular for the tibiotalar joint. While further investigations are required, anatomically matched talar cement spacers may allow for improved joint kinematics and decreased osteolysis while allowing for the local delivery of antibiotics and preservation of joint space and function. Furthermore, increased use as definitive treatment without the need for secondary reconstruction may be possible due to their anatomic design which may improve function and pain relief as compared to nonanatomic spacers. As utilization of TAR is increasing, there will be an increased need for revision and salvage solutions due to implant loosening and infection. Complications of post-traumatic conditions of the hindfoot such as osteonecrosis and infection may require similar solutions. Utilization of currently available 3-D printing technology as applied to production of anatomic cement spacers should be considered to enhance limb salvage. The bony defects resulting from TAR failure or post-traumatic collapse can lead to unique anatomy where preserving bone is paramount for future revision and reconstruction.
3-D printing has the potential to be a useful tool for temporary or long term solutions in these scenarios. While joint preservation in the long-term is unlikely given the coefficient of friction between the cement/articular cartilage interface, this technique may allow for limb-salvage or temporization until comorbid conditions may be optimized to allow for other reconstructive procedures. Further studies are required to demonstrate cost-effectiveness in the setting of patient outcomes.