External factors such as traffic accidents, falling, or bumping into a hard place, can cause bone fracture in animals (12,15). Fractures often occur in small breed dogs rather than in large breed dogs, because the former has relatively thin and weak bones (4). Small dogs require adequate surgical treatment and application of external coaptation, because pain and pressure occur around the limb fracture, and fractures resulting from severe trauma can cause internal tissue damage and shock (23,25).

A surgical approach should involve application of an intramedullary pin, plate and screw to treat patients with bone fractures (23). To improve the patient’s prognosis, keeping the fractured bone alignment close to its original state pre- or post-operation is important (23). Bandages-that is, external coaptation applied to patients-protect the affected area, correct the arrangement of fractured bone, and alleviate the patient’s pain. The Robert Jones bandage is the most widely used method, although other bandage methods are available, such as the Mason metasplint and fiberglass splints (cast) for stronger support. In particular, the casting method serves as an external coaptation method to completely wrap the patient’s legs and support the area (23,25).

In general, bandages that use splints or casts can cause edema, heat sensation, dermatitis, and pressure ulcer due to failure of length control caused by the bandage itself; additionally, necrosis can occur in severe cases. Moreover, in humid conditions, such bandages are poorly ventilated, decelerating the recovery of skin wounds. Replacing bandages applied with casting or splints involves cumbersome disadvantages, and the surgical site cannot be confirmed frequently (8).

Producing orthopedic splints using a 3D printer can reduce these side effects (21). Splints made using 3D printing technology can be customized to a suitable length for patients and movement is minimized to prevent abnormalities, such as chipping (1,21). A splint can be made using various patterns; thus it can be molded into a honeycomb form to facilitate ventilation (1). Furthermore, it can be made detachable, allowing the splint to be customized according to the patient while changing its size and shape as the patient grows (1,6).

For the application of 3D-printed splints in dogs with limb fractures in this study, splints were 3D modeled based on the patient’s biometric data obtained from Chonnam National University Veterinary Medical Teaching Hospital and then stability was analyzed by using a finite element method (FEM) through computer simulation. The FEM provides an approximate interpretation of the result of splitting the object of a target into individual elements. Polylactic acid (PLA) and acrylonitrile-butadiene-styrene (ABS), widely used materials for 3D printing, were utilized to model 3D-printed splints with thicknesses of 1, 3, 5, and 7 mm, and the Peak von Mises stress (PVMS) and maximum displacement (MD) by FEM were measured. Through the above process, it was intended to confirm the splint of the most stable thickness by checking the resistance and degree of deformation according to the load. Afterwards, the stability-tested splint was made and applied to the patient in the clinic.

Three dimensional (3D) splint modeling for FEM

A dog (French Bulldog, 8 month, male and 7.7 kg) with a left tibia fracture due to a traffic accident visited Chonnam National University Veterinary Medical Teaching Hospital, and it was confirmed on the history taking that he had dermatological problems such as pruritus, alopecia, and greasy. It was decided to produce a 3D-pritned splint and apply it to this patient, judging that bandage using the existing splint after surgery could worsen the skin disease. The normal leg (165 mm length), the other side of the fractured leg, was scanned by the 3D scanner (EinScan Pro 2X, Shining 3D Tech. Co., Hangzhou, China), and point data were collected by the scanner program (EXScan Pro 3.4.0.5, Shining 3D Tech. Co., Hangzhou, China). The scanned data were stored as standard template library (STL) files (1,20).

The scan data converted to STL files were transferred to a 3D modeling software program (Meshmixer^TM, Autodesk Inc., San Rafael, CA, USA). The mirror function was utilized to change the scanned leg in the direction of the fractured leg (7), and the Offset function was used to generate curved surfaces spaced 1, 3, 5, and 7 mm from the existing models. The Plane Cut function, parts to be used as splints, were extracted and parts not needed were deleted. The surface was trimmed via a Smooth function and designed with a Discard function. Finally, the 3D modeling was complete by the Make Pattern function, and saved as a STL file (Fig. 1).

Figure 1. Three-dimensional (3D) modeling process of an orthopedic splint using 3D modeling software (Meshmixer^TM).

FEM of 3D printed splint

To confirm the stability of 3D-printed orthopedic splints for application to dogs, PVMS and MD were measured using FEM under compressed load conditions (6,18). For interpretation, SolidWorks (SolidWorks Corporation, MA, USA), a commercial finite element program based on parabolic tetrahedral elements, was applied. In this program, the maximum stress is based on the von Mises-Hencky theory (13,19), and the PVMS is expressed as follows:

$${\sigma }_{vonmises}={\left[\frac{{\left({\sigma }_1-{\sigma }_2\right)}^2+{\left({\sigma }_2-{\sigma }_3\right)}^2+{\left({\sigma }_1-{\sigma }_3\right)}^2}{2}\right]}^2$$

σ1, σ2, σ3 are principal stresses.

MD (L) is measured as follows:

$$L=\sqrt{{\displaystyle \sum _{i=1}^N\frac{{\left(X_i-X_c\right)}^2+{\left(Y_i-Y_c\right)}^2+{\left(Z_i-Z_c\right)}^2}{N}}}$$

Xi, Yi, and Zi are the coordinates of node i; Xc, Yc, and Zc are the coordinates of the model geometric center; N is the total number of nodes in the model.

Table 1 lists the major mechanical properties-Young’s modulus, Poisson’s ratio, and mass density-of the PLA (Premium PLA, Raise 3D Technologies Inc., Irvine, CA, USA) and ABS (Premium ABS, Raise 3D Technologies Inc., Irvine, CA, USA). Young’s modulus and Poisson ratio were measured by using the universal testing machine (STC-USS200, Samyeon Co., Korea), and the mass density was provided by the PLA and ABS manufacturers. Compression load conditions were employed in the FEM (14,18,20). Considering the load transfer process, the bottom of the splint was fixed, and compression loads of 50, 100, 150, and 200 N were applied vertically to the top. Large formation and nonlinear contact analyses were performed by considering the boundary conditions between the elements.

Table 1 . Mechanical properties of 3D printing materials.

	PLA	ABS
Young’s modulus (MPa)	3,450	2,331
Poisson ratio (v)	0.46	0.3
Mass density (kg/m3)	1,205	1,125

PLA, polylactic acid; ABS, acrylonitrile-butadiene-styrene..

Fabrication of 3D-printed orthopedic splint

To reflect the biological data, the modeled STL data were transformed into G-code using a 3D printing slicing program (IdeaMaker 3.6.1, Raise 3D Technologies Inc., Irvine, CA, USA). Before 3D printing, material, temperature, output quality, support, density, and output speed were set. The saved G-code data were stored on USB, and splint was printed in a file form using a 3D printer (RAISE 3D Pro2, Raise 3D Technologies Inc., Irvine, CA, USA) (10).

Analysis of the 3D printed splint by the FEM

Three-dimensional-printed orthopedic splints (165 mm height, with thicknesses of 1, 3, 5, and 7 mm) based on PLA and ABS materials, were designed and analyzed by the FEM. Next, the PVMS and MD values were measured according to the compressive loads (50, 100, 150, and 200 N) applied to the top surface and conditions by the thickness of the splints.

In both PLA and ABS, PVMS appeared on the top surface, and MD was expressed around the Velcro mounting area (Fig. 2). The results were similar for both materials (Table 2). For the 1 mm thickness splint, the deformation was too severe to be analyzed (Fig. 3). For splints with thickness of 3 mm, both PLA and ABS exhibited a yield strength of less than 30% and deformation exceeding 1 mm at compressive loads above 50 N and above 100 N, respectively. At 5 mm thickness splints, both PLA and ABS displayed stable PVMS, as compared to the yield strength, until a force of 100 N was applied. In contrast, PLA indicated a strain exceeding 1 mm when a compressive load of more than 200 N was applied, and the ABS was more than 150 N. For the 7 mm thickness splints, both PLA and ABS demonstrated stable results in PVMS and MD.

Table 2 . Comparisons of PVMS and MD according to the thickness of splints under compressive loads.

Analysis	Compressive load (N)	PLA				ABS
		3 mm	5 mm	7 mm		3 mm	5 mm	7 mm
PVMS (MPa)	50	11.48*	3.94	1.91		11.83*	3.93	1.92
	100	24.18**	7.89	3.85		25.83**	7.99	3.87
	150	38.47***	11.97*	5.80		42.97***	12.19*	5.83
	200	54.78***	16.13**	7.76		64.80***	16.54**	7.82
MD (mm)	50	0.78	0.26	0.13		1.20†	0.39	0.19
	100	1.61†	0.53	0.26		2.52†	0.79	0.38
	150	2.52†	0.80	0.39		4.04†	1.20†	0.57
	200	3.51†	1.08†	0.52		5.86†	1.62†	0.77

PLA, polylactic acid; ABS, acrylonitrile-butadiene-styrene; PVMS, peak von mises stress; MD, maximum displacement..

***Stress greater than yield strength (PLA: >31.24 N, ABS: >28.00 N), **Stress exceeding 50% of yield strength (PLA: >15.62 N, ABS: >14.00 N), *Stress exceeding 30% of yield strength (PLA: >9.37 N, ABS: >8.40 N), †Maximum deformation displacement of >1 mm..

Figure 2. Load and constraint conditions in the FEM by a finite element program (SolidWorks) for PLA-based 5 mm thickness splint at 100 N. The bottom of the splint was fixed (green arrows), whereas the top face was subjected to a compressive load (purple arrows): (A) PVMS occurs near the Velcro attachment area. (B) MD on the top surface of the splint.

Figure 3. One millimeter 3D-printed orthopedic splint: (A) a broken figure of the splint (blue box). (B) magnified cracked parts of the 1 mm 3D-printed orthopedic splint.

Application of the 3D-printed orthopedic splint

Because of the FEM analysis on PVMS and MD, stable 3D-printed splints were produced. Based on the dog’s 3D scan data, the splints were made with a honeycomb hole to enhance ventilation, and Velcro holes were added to facilitate wearing (24).

According to the 3D printer program, the estimated build times for the splints in 3, 5, and 7 mm thickness PLA material were 14.15, 18.93 and 20.75 h, whereas those for the splints in ABS material were 13.38, 18.22, and 20.07 h respectively (Table 3). The estimated weights of the 3, 5, and 7 mm thickness PLA splints were 49.4, 93.3, and 120.3 g, and those of the ABS splints were 44.6, 71.8, and 108.6 g respectively (Table 3).

Table 3 . Temperature, build time and weight according to splint thickness and material.

Splint composition materials	Temperature (°C)			Build time (hours)				Weight (g)
	Filament extraction	Heating layer		3 mm	5 mm	7 mm		3 mm	5 mm	7 mm
PLA	225	45		14.15	18.93	20.75		49.4	93.3	120.3
ABS	250	100		13.38	18.22	20.07		44.6	71.8	108.6

PLA, polylactic acid; ABS, acrylonitrile-butadiene-styrene..

To produce the 165 mm height splint, 3D scanning required 0.5 h and 3D modeling required 0.5-1.0 h. Making splints of PLA and ABS material using the fused deposition modeling (FDM) method based on a layer height of 0.34 mm, side wall of 2, fill density of 15%, and speed of 80 mm/s required 13-20 h. The entire process was completed within 24 h.

Attaching the 3D printed orthopedic splint to dogs directly was difficult because of its strength. Before applying the splint to the dog’s legs, neoprene material was attached to the outer surface of the splint, and Velcro was prepared in holes at the edge of the splint. After confirming that the skin on the surgical incision line has recovered, 3D-printed orthopedic splint was applied to the patient for about three months (Fig. 4).

Figure 4. Completed 3D-printed orthopedic splint and applied it to the patient: (A) PLA material based 3D-printed splint. (B) ABS material based 3D-printed splint. (C) application of the splint to the dog’s tibial fracture. (D) X-ray image of the tibial fractured patient before operation. (E) X-ray image of tibial fractured patient two weeks after operation. (F) X-ray image of tibial fractured patient four months after operation.

Recently, as interest in 3D printers has increased and the market has advanced, applications in the medical field have expanded (17). Patents and journals related to the design and production of human splints have been proposed, but they have not been commercialized because of entry barriers for price and modeling (5,11,22,24). Three dimensional scanning is now available through mobile applications such as Qlon (Eye Cue Vision Technologies LTD, Yokne’am Illit, Israel) or Trnio 3D Scanner (Trnio Inc., San Francisco, CA, USA). In addition, the patent of an FDM-based 3D printer (Stratasys, Rehovot, Israel) has expired (9), lowering the price of the equipment and allowing 3D printers to be manufactured directly using an open source project called RepRap, if necessary.

The main purpose of this study is to develop orthopedic splints for small dogs using 3D scanners and 3D printing technologies that have become easier to access and utilize in response to these changes. Therefore, this study introduces how to create patient-specific splints and evaluate stability using various 3D programs and equipment. Finally, 3D splints targeting small dog leg with tibia fracture were analyzed using the FEM method to evaluate PVMS and MD according to compressive load based on the splint thickness and printing material to derive an appropriate standard of 3D-printed orthopedic splint for small dogs.

Customized systems and structural analysis of splints in a patient of human medicine has been studied (20). However, further studies are required to ensure stability because the physical conditions of dogs are different from humans. The first and second conditions involve the thickness of the splint and compressive load, respectively. The thickness of the splint was set because an increase in the thickness amplifies the weight of the splint, which can affect the movement of the dog. Moreover, various compressive loads were applied to determine whether the dog’s weight could withstand when wearing splints.

The ABS was expected to be superior to PLA in build time and weight, and the filament extraction temperature and heating layer temperature of the former are higher than those of the latter (Table 3). Nevertheless, both the materials took about 0.5 h to prepare for the operation, indicating that their production time was similar. Compared to PLA, ABS was distorted because of severe shrinkage during printing at room temperature.

Under compressive load conditions, splints with thickness of 1, 3, 5, and 7 mm were analyzed by FEM. However, in the 1 mm thickness splint, FEM was not implemented because of severe deformation. Moreover, breakage occurs just as easily as when a hand compressive load is applied to a 1 mm thickness splint (Fig. 3).

Based on the physical properties of PLA and ABS, the results of the FEM analysis of PVMS and MD for both materials displayed similar aspects. The 3 mm thickness splints were clinically unsuitable because of the risk of deformation when subjected to compression loads above 50 N. For the 7 mm thickness splint, PVMS illustrated stable results at less than 30% yield strength for all applied compression loads, but the use of FDM methods can cause longer fabrication times and discomfort because of increased weight. The 5 mm thickness splints were manufactured to confirm stability under a 100 N compressive load.

In general, the peak vertical forces on the leg during the stance phase of the weight-bearing dog are approximately 43% on the forelimb and 28% on the hindlimb (16). Dogs weighing 15-40 kg have a gait speed of 1 m/s, with a load of 50-60% in the forelimb and 30-40% in the hindlimb. The maximum load on a walking dog’s leg is 60% of its weight (2,3).

Splints with a thickness of 5 mm and compression load of ≤100 N are stable. The bearing load on the dog’s legs has also been reported to be up to 60% of its body weight (1,20). Accordingly, a compression load of 100 N was calculated as the patient’s weight at approximately 16.7 kg. In conclusion, the splint with a thickness of at least 5 mm for successful clinical application of 3D-printed orthopedic splints for dogs whose weight under than 16.7 kg is recommended in veterinary medicine.

Finally, in this study, the 3D-printed orthopedic splint produced based on these research results was applied to the patient. Surgical correction for a tibal fracture was performed, and the 3D-printed orthopedic splint was applied for about three months after the skin of the surgical incision line was recovered. It was confirmed that the fracture lined disappeared four months after the surgery (Fig. 4). It is judged that applying the existing splint to this patient could worsen the patient’s skin problem, so it was possible to prevent this by applying the 3D-printed orthopedic splint developed in this study.

This study, used in computer simulation programs, focuses on manufacturing splints in orthopedic animal patients and the analytical results of wearing them by computer simulation. Therefore, to evaluate the wearability of splints in the future, empirical studies are necessary to quantify and verify the pressure, blood circulation, and skin irritation that dogs experience when wearing them. Furthermore, under compressed load conditions, FEM performed an epidemiological analysis of splints by approximating actual shapes with similar shapes on computers, which inevitably vary from actual values. Therefore, comparative experiments are required to compare how the mechanical behavior of the splints varies from the actual dynamic behavior.

In this paper, a study on the application of splints in fracture patients was evaluated. Based on the patient’s biometric data, the stability of the 3D modeled splint was analyzed by FEM via simulation and applied to the patient. 3D-printed splints (5 mm thickness) made of PLA and ABS materials demonstrated stability in PVMS and MD against a 100 N compressvie load, confirming that they were suitable for small dogs under 10 kg. Using the research results of this paper, 3D printing technology can be customized to be patient specific in various forms, such as splints, and can be applied to patients after reviewing stability in advance through modeling. As the contribution of 3D printing technology in the medical field increases, continuous research is needed.

This study was financially supported by Chonnam National University (Grant number : 2021-2429).

The authors have no conflicting interests.