Biomechanical evaluation of 3D-printed joint-type orthosis for hallux valgus

Abstract

Background:

Orthoses play an important role in the conservative treatment of hallux valgus (HV) with different therapeutic effects. In this study, a new HV orthosis was developed using three-dimensional (3D) printing technology. In addition, its kinematic effect was evaluated using motion analysis.

Methods:

Seventeen participants with an HV angle of >20° were included in the study. The first metatarsophalangeal abduction angle before and after the orthosis was measured statically. Subsequently, dynamic first metatarsophalangeal abduction, dorsiflexion angle and ground reaction force with and without the orthosis were recorded and calculated during walking using a Vicon motion analysis system and force plates. The patients’ comfort scales were determined after the motion analysis.

Results:

The angular corrections of the orthosis in the first metatarsophalangeal abduction were 14.6° and 6.3° under static and dynamic conditions, respectively. Reduced hallux dorsiflexion was observed with the orthosis in the early stance phase. However, no significant changes in ground reaction forces were observed.

Conclusion:

The results of our study confirm the potential of the 3D-printed HV orthosis in the static and dynamic correction of deformities while ensuring patient comfort with minimal impact on hallux kinematics, suggesting the potential of our design for long-term use.

1. INTRODUCTION

Hallux valgus (HV) is one of the most common foot and ankle diseases, characterized by the deformity of the first metatarsal medial deviation, first metatarsophalangeal joint valgus angulation, and pronation of the first ray.^1,2 HV prevails in 23.0% to 35.7% of the general population. Of which, its percentage is higher in women and increases with age.^3,4 Common symptoms of HV include pain over the medial eminence of the first metatarsal head, metatarsalgia, and difficulty in wearing shoes.¹ From a biomechanical perspective, patients with HV show dysfunction of the great toe during walking and increased pressure in central forefoot compared with healthy individual.⁵ In addition, HV has negative effects on gait parameters, including speed and stride length, which increase with the severity of the deformity.⁶

Although surgery can directly correct HV, the complication rate is up to 50%.^7,8 However, patient satisfaction does not correlate with improvement in radiographic parameters after surgery.⁹ In particular, months are needed to relieve postoperative pain and swelling.¹⁰ Therefore, conservative treatment is another important measure. With the aim of realigning the hallux and relieving symptoms, conservative HV treatment can be divided into two main categories: active training to strengthen the abductor hallucis muscle and passively rebalancing the soft tissue tension on both sides of the hallux with orthosis or taping.

Among them, foot orthoses can provide immediate biomechanical support to correct the HV angle, reduce plantar pressure and prevent progression.^11–13 However, the wide range of designs and materials for HV orthoses leads to inconsistent therapeutic effects.¹⁴ At the same time, the corrective effect and comfort of custom-made HV orthoses may vary due to the different methods of obtaining the foot model and the variety of technical procedures. In addition, some conventional HV orthoses have some disadvantages in terms of orthotic breakage, are expensive and uncomfortable.

Recently, three-dimensional (3D) printing technology has been recognized as a fast and cost-effective approach for the development of medical devices. This technique could be better integrated into standardized design and manufacturing, demonstrating the potential for consistent therapeutic effects. In addition, the 3D-printed device has the advantage that it can be customized for the patient. However, previous studies have evaluated orthopedic treatment mainly based on clinical and radiographic assessments without a dynamic examination.^14–16 This study therefore develops a 3D-printed HV orthosis and evaluates its immediate static and dynamic therapeutic effects. The first metatarsophalangeal abduction angle before and after the orthosis was measured statically. Subsequently, dynamic first metatarsophalangeal abduction, dorsiflexion angle, and ground reaction force (GRF) with and without the orthosis were recorded and calculated during walking using a VICON motion analysis system and force plates. The patients’ comfort scales were determined after the motion analysis.

2. METHODS

The study was approved by the Institutional Review Board of National Yang Ming University (YM107023E). All participants gave informed consent before the study. The inclusion criteria were as follows: (1) age of >20 years and (2) HV angle of >20°. The exclusion criteria were as follows: recent use of foot orthoses or orthosis within 1 month, surgical history of foot and ankle fractures/reconstructions/osteotomies, and medical history related to connective tissues. Seventeen participants (12 women and five men; age: 24.47 ± 6.9 years; height: 162.7 ± 6.7 cm; weight: 61.5 ± 16.2 kg; body mass index [BMI]: 23 ± 4.7 kg/m²) were included in this study.

2.1. Fabrication of the 3D-printed orthosis

The primary contours of the great toe and medial forefoot of the participants were first obtained by Sense-RS (3D Systems Corporation, Washington, DC, USA) in a nonweight-bearing status. To maintain the correct position of the great toe and allow prolonged stretch to the soft tissue on the lateral side, the orthosis was divided into two parts. First, the lever fulcrum was modeled according to the medial aspect from the midpoint of the medial arch to the first metatarsophalangeal joint as shown in Fig. 1A. Second, a loop was designed as the correction part to fit the middle portion of the great toe with proximal extension on the medial side. To allow sagittal plane movement, a hinge joint located at the first metatarsophalangeal joint rotation center with a target HV angle of 10° was designed to connect these two parts. A software Meshmixer (Autodesk Inc., San Francisco, CA) was implemented to edit the 3D foot model and export an acceptable stereolithography (STL) file format for the 3D printer. To slicing the 3D foot model, a software Flash Print slicing software (FlashForge Corporation, ZheJiang, China) was used to process the STL file. Eventually, a 3D printer Finder (FlashForge Corporation) with fused deposition modeling technology was used to print the orthosis out. The thickness of the orthosis was uniform and its thickness was 3 mm. The printing material used was semirigid thermoplastic polyurethane (TPU) and the printing parameters were as follows: Thickness of 0.18 mm per layer, infill density of 100%, printing temperature of 210°C and printing speed of 30 mm/s. The reason for choosing the semirigid TPU material was that it could provide flexibility in dorsiflexion of the first metatarsophalangeal joint and a correct HV angle. Additionally, the TPU material was used to fabrication of 3D-printed insole.¹⁶ Once the orthosis was fabricated, a self-adhesive nylon strap was attached to the widest part of the fulcrum (Fig. 1B). Since a nonarticulating 3D-printed HV orthosis¹⁷ was developed in a previous study, which caused a foreign body sensation when walking, an articulating HV orthosis was developed in the study to avoid this problem. In other words, the novelty of the HV orthosis was that it made it easy to walk in the sagittal plane and correct the HV angle.

Fig. 1.

A, Design of the 3D-printed orthosis model. The red arrow indicates the lever fulcrum. The blue line indicates the splitting of the hallux. The shape of the middle part was obtained to form the correction loop. B, HV correction. θ and θ’ are the abduction angle of the first metatarsophalangeal joint before and after wearing the orthosis, respectively. 3D = three-dimensional.

2.2. Experimental procedures

To compare the static and dynamic correction effects of the developed orthosis, this study recorded the static abduction angle of the first metatarsophalangeal joint before and after wearing the orthosis. This angle was defined as the intersection angle between the morphological long axis of the first metatarsal and proximal phalangeal bone under a weight-bearing status (Fig. 1B). A Vicon motion analysis system (Oxford Metrics, Oxford, UK) with eight cameras capturing at 100 Hz was implemented to collect dynamic correction effect and hallux motion. Thirteen reflective markers were placed on the individual limb, as shown in Fig. 2. Hallux angle was measured using previous developed model.¹⁸

Fig. 2.

Locations of the reflective markers for the static and dynamic correction analysis.

The calculations of kinematics and kinetics in the motion analysis were the same as in our previous study.¹⁷ MATLAB software (MathWorks Inc., Natick, MA, USA) was used to analyze the movements. To familiarize themselves with the orthosis, participants were asked to walk at their speed on a 6-m walkway. After a 5-minute practice walk, participants’ 3D movement data were recorded for at least five walking trials. The motion analysis was performed for the participants with (3DP_orthosis) and without (Barefoot) orthosis. To detect the change in GRF, the three AMTI force plates (Advanced Mechanical Technology Inc., MA) sampling at 1000 Hz were recorded during one gait cycle. Two peak GRF values (Fz1, Fz2) and the anteroposterior force were analyzed to determine whether the orthosis affected walking. The magnitude of the GRF component was normalized to the body weight (BW) of each participant. After the dynamic experiments, participants were asked to rate their wearing comfort on a scale of 0 to 10. A score of 5 means that the participant felt normal and without any significant feelings. The values 0 and 10 mean “very painful” and “comfortable”, respectively. The entire experimental schedule is shown in Fig. 3.

Fig. 3.

Flowchart about measurement of experiment. 3D = three-dimensional; HV = hallux valgus.

2.3. Statistical analyses

SPSS Statistics 20.0 (IBM Corporation, Armonk, NY) was used for the statistical analysis. The static and dynamic angular data, and comfort scales are presented as the mean ± SD. Paired sample t test was performed to compare the difference between the Barefoot and 3DP_orthosis data. The significance level was set to 0.05.

3. RESULTS

3.1. Abduction angle of the first metatarsalphalangeal joint

The average static abduction angle of the first metatarsophalangeal joint under Barefoot was 29.7 ± 5.7°, which significantly decreased to 15.5 ± 3.4° under 3DP_orthosis. In the motion analysis, the peak first metatarsophalangeal joint abduction angle was 17.8 ± 8.4 and 11.5 ± 6.7 under the Barefoot and 3DP_orthosis conditions, respectively, indicating significant differences. Meanwhile, the average first metatarsophalangeal joint abduction angle was significantly lower under 3DP_orthosis at 90% to 100% of the stance phase (Fig. 4).

Fig. 4.

Variations in the abduction angle of the first metatarsophalangeal joint during the stance phase of the gait. +: adduction; −: abduction. *Significant difference (p< 0.05).

3.2. Dorsiflexion angle of the first metatarsophalangeal joint

The maximum dorsiflexion angle of the first metatarsophalangeal joint was 43.8 ± 12.6° and 29.7 ± 12.3° under the Barefoot and 3DP_orthosis conditions, respectively, indicating a significant difference. In the gait cycle, the dorsiflexion angle was significantly higher under the Barefoot conditions at 0% to 10% of the stance phase (Fig. 5).

Fig. 5.

Variations in the dorsiflexion angle of the first metatarsophalangeal joint during the stance phase of gait. +: dorsiflexion; −: plantarflexion. *Significant difference (p < 0.05).

3.3. Ground reaction force

The anteroposterior force was 0.15 ± 0.02 and 0.15 ± 0.03 BW under the Barefoot and 3DP_orthosis conditions, respectively. The first peak of the vertical force (Fz1) was 1.01 ± 0.03 BW in both conditions, and the second peak of vertical force (Fz2) was 1.06 ± 0.04 BW under Barefoot and 1.06 ± 0.05 BW under 3DP_orthosis. All results revealed no statistical difference between the Barefoot and 3DP_orthosis conditions (Table 1).

Table 1.

Maximum GRF of the Barefoot and 3DP_orthosis conditions

	Anteroposterior force (BW)	Vertical force Fz1 (BW)	Vertical force Fz2 (BW)
Barefoot	0.15 ± 0.02	1.01 ± 0.03	1.06 ± 0.04
3DP_orthosis	0.15 ± 0.03	1.01 ± 0.03	1.06 ± 0.05

BW = body weight; GRF = ground reaction force.

3.4. Comfort scale

A score of 5 means the feeling of participant without wearing orthosis. Only a few patients felt that the front ring edge of the HV orthosis was slightly uncomfortable. As a result, the comfort scale for wearing the HV orthosis was reduced to 4.05 ± 1.74.

4. DISCUSSION

The results of the study showed that the orthosis developed can effectively correct the HV angle both statically and dynamically. The GRF, especially the anteroposterior force and the vertical force Fz2 for the barefoot and 3DP_orthosis conditions, showed that the orthosis had no negative impact on the effectiveness of forefoot function during walking (Table 1). Although the dorsiflexion of the first metatarsophalangeal joint was significantly lower under 3DP_orthosis, the deviations were only observed in the early stance phase. Since the hallux was free and without ground contact in the early stance phase, the reduced dorsiflexion only represents the restriction due to the orthosis and not the restriction of the range of motion. Satisfactory results were obtained for the comfort scale. Four of 17 participants reported above-average discomfort after the experimental procedures.

Although treatment with HV orthoses has been widely proven in terms of deformity correction, symptom relief¹⁴ and gait biomechanics,¹⁹ there have been few reports on 3D-printed HV orthoses. Therefore, our team developed a 3D-printed HV orthosis¹⁷ and conducted an experiment on its effect on deformity correction, kinematics, and comfort. Using the same materials, the previous orthosis was molded in one piece with a hollow design under the first metatarsophalangeal joint to allow walking, which restricts the movement of the first metatarsophalangeal joint. Although the results were convincing, the lack of freedom in the first metatarsal joint raises concerns about the comfort and side effects of joint stiffness as well as the longevity of the orthosis after repeated use. In this study, we developed an orthosis with a hinged joint in the center of rotation of the first metatarsal joint to increase the range of motion and comfort. With the current hinged orthosis, the participants’ maximum dorsiflexion angle of the first metatarsophalangeal joint was 47.1° in the barefoot group and 37.5° in the orthotic group. With the previous one-piece orthosis,¹⁷ the maximum dorsiflexion angle of the first metatarsophalangeal joint of the participants was 43.8° in the barefoot group and 29.7° in the orthosis group. The normal dorsiflexion angle of the first metatarsophalangeal joint in healthy subjects was 40°.²⁰ This meant that the 3D-printed orthosis could also provide the dorsiflexion angle of the first metatarsophalangeal joint and correct the HV angle. Although the comfort scale was satisfactory, the present orthosis (comfort scale: 4.05) performed worse than our previous design (comfort scale: 4.5).¹⁷ This can be attributed to three reasons. Firstly, a greater degree of freedom may increase the friction between the orthosis and the hallux. Secondly, the allowed range of motion of the hallux was in a relatively corrected position, neglecting the reamed joint surface. Meanwhile, the stretching of the previously contracted soft tissues, such as the lateral joint capsule and the abductor hallucis muscle, may increase during joint movement. Third, the center of rotation of the first metatarsophalangeal joint was determined only by anatomical landmarks. This distortion could lead to excessive crunching during walking. Therefore, in our future design, a precise measurement of the center of rotation of the joint and a soft additive filling between the orthosis and the foot should be added.

Few reports have discussed the effects of HV on GRF. Farzadi et al²¹ compared the GRF of 90 HV patients and 30 healthy subjects, and patients with higher deformity tended to have higher peak vertical forces. Rotational deformities have a medium positive correlation with GRF. Kim²² analyzed the GRF in HV patients with different types of orthoses and under barefoot conditions. The late anteroposterior maximum and first vertical peak forces increased significantly when a hard type of orthosis was worn. In this study, no significant GRF difference was found with the 3D_orthosis and barefoot walking, which is due to the design to realign the deformity in the coronal plane without rotational correction and the softer material of the orthosis.

Nevertheless, there are some limitations to this study that should be noted. First, we only examined the kinematics of the hallux, not the entire lower limb. The orthosis may affect other parts of the lower limb. For example, a previous study reported that pelvic motion during walking improved after wearing orthoses in HV patients.²³ Second, we obtained a foot model in a non–weight-bearing condition. Under weight-bearing, there may be a mismatch between the orthosis and the foot, resulting in distortion. Third, we did not evaluate the long-term effects. Although previous studies have reported correction of HV deformity after long-term orthosis,^12,24 patients may still experience discomfort. In addition, the angle of the HV orthosis was originally designed to be 10°. However, the recruited participants had an angle of more than 20°, so the discomfort and durability varied according to different HA angles of the patients. In this study, we developed a customized dynamic HV orthosis by 3D printing. Good corrective effects were observed under static and dynamic conditions. However, the influence of the orthosis on sagittal hallux range of motion and GRF during walking is limited. Although the comfort scale results were satisfactory, precise registration of the center of rotation of the first metatarsophalangeal joint might be needed to improve our design.

ACKNOWLEDGMENTS

This study was partly supported by the grant from Veterans General Hospitals and the University System of Taiwan Joint Research Program (VGHUST113-G7-2-1).