Therapeutic efficacy of 3 dimensions printed orthoses on wrist-hand flexor spasticity in post-stroke hemiplegia: a multi-center stratified clinical study

Abstract

Objective

This study mainly aimed to compare the clinical efficacy of 3D-printed orthoses versus low-temperature thermoplastic plate orthoses(LTT)on wrist-hand flexor spasticity post-stroke.

Methods

Patients were stratified into three cohorts based on the Modified Ashworth Scale (MAS) grades (1,1+, and 2), with 60 patients per cohort. Patients in each cohort were randomized into treatment and control groups: MAS1 (3D, n = 23; control, n = 24); MAS1+ (3D, n = 24; control, n = 25); MAS2 (3D, n = 25; control, n = 26). Orthoses were worn 4–8 h daily for 6 consecutive weeks, followed by a 6-week post-intervention follow-up. The primary outcome was MAS scores. Visual analogue scale(VAS), passive range of motion(PROM), Fugl-Meyer Assessment(FMA), and swelling scale (SS) scores were secondary outcomes, assessed at baseline, 6 weeks, and after 12 weeks.

Results

At 6 weeks, significant difference in the MAS scores were observed between groups: effective rates (ERs) were MAS1:71.4% vs. 30.8%, p = 0.006; MAS1+: 63% vs. 37%, p = 0.03. After 12 weeks, significant difference in MAS score changes persisted: ERs were MAS1: 66.7% vs. 33.3%, p = 0.029; MAS1+: 64.7% vs. 35.3%, p = 0.001. Significant difference was observed in the Quebec User Evaluation of Satisfaction with Assistive Technology (QUEST) among the MAS1, MAS1+, and MAS2 cohorts (p < 0.01).

Conclusion

3D-printed orthoses demonstrated superior efficacy to LTT orthoses in reducing mild-to-moderate (MAS 1, 1+) wrist-hand flexor spasticity post-stroke. For severe spasticity (MAS 2), further optimization of material properties and scanning technology may be necessary.

Trial registration

ChiCTR-INR-2,300,070,454 (Registered: 4 December 2023).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12883-025-04497-7.

Introduction

spasticity is one of the most common and debilitating complications after stroke [1], with wrist and hand spasticity being particularly prevalent [2]. According to the classic Brunnstrom guidelines, spasticity peaks during the II-III stage [3] and can disrupt disrupt motor recovery [4]. Once spasticity occurs, it will persists. Therefore, effective management and treatment of the wrist-hand spasticity is a critical area that requires clinical investigation.

Orthotics reduce muscle spindle sensitivity and enhance proprioceptive input through continuous low-load stretching, thereby inhibiting the stretch reflex [5]. This stretching also modulates abnormal collagen fibers proliferation. As a non-invasive treatment modality, orthoses are widely utilized in clinical practice for wrist-hand spasticity [6]. However, due to structural limitations [7]—such as their inability to perfectly align with the wrist, metacarpophalangeal, and interphalangeal joints-often result in inadequate support and limited efficacy in the management of spasticity.

Application of 3 dimensional (3D) printing in rehabilitation orthotics has recently garnered attention, due to its ability to enable precise customization, anatomical accurate, and versatile fabrication. This technology enables the design of personalized orthoses [8]. However, the application of 3D printing has mostly focused on lower limb orthoses. Our preliminary single-center pilot study (n = 40) demonstrated that 3D-printed orthoses significantly outperformed low-temperature thermoplastic (LTT) orthoses in reducing MAS scores, improving joint range of motion, and alleviating pain scores at the 3-week follow-up [9].

Building on these findings, this multi-center trial incorporated spasticity severity stratification and expanded the sample size to validate the clinical efficacy of 3D-printed orthoses and optimize their application in post-stroke rehabilitation.

Methods

Ethical approval

This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Shanghai Xuhui Central Hospital (Approval No. 2022–056). Additionally, the study adhered to the Consolidated Standards of Reporting Trials (CONSORT) guidelines and was registered in the Chinese Clinical Trial Registry (ChiCTRINR-2300070454) on 4 December 2023.

Study design

This multi-center randomized controlled trial enrolled participants from the rehabilitation departments of the Shanghai Xuhui District Central Hospital, Shanghai Eighth People’s Hospital, Lansheng Wanzhong Hospital, and Kanghealthy Community Health Service Center. Following ethical approval, we commenced training and evaluation in the participating centers in February 2023, mainly focusing on upper extremity functional rehabilitation protocols and related evaluation indicators. The first participant was enrolled in April 2023 after clinical trial registration. This study was conducted from February 2023 to November 2024, with spasticity levels stratified according to Modified Ashworth Scale (MAS) grade [10]. All participants provided a signed informed consent.

Setting and participants

The inclusion criteria were as follows: (i) diagnostic criteria for either ischemic stroke or intracerebral bleeding; (ii) occurrence of unilateral paralysis following the initial cerebrovascular incident; (iii) spasticity MAS ≥ 1 in the flexion of the wrist and at the metacarpophalangeal, proximal interphalangeal, and distal interphalangeal joints; (iv) spasticity onset 2–18 months post-stroke; (v) age 35–80 years; (vi) ability to understand and follow instructions; (vii) stable physiological indicators; (viii) consistent medication dosages with the potential to influence muscle stiffness; (ix) signed informed consent.

The exclusion criteria included the following: (i) presence of concurrent pathological condition that impairs motor function, including rheumatoid arthritis, joint abnormalities, and spinal cord injuries; (ii) undergone Botox, alcohol, or phenol injection; (iii) undergone orthopedic surgery on the wrist joint; (iv) patients with local pressure sores or allergy to materials, limiting their ability to wear assistive devices; (v) wrist and hand joint flexion spasticity MAS ≥ 4 grade; (vi) sensory loss on the hemiplegic side.

The recruitment process involved four main phases: (i) the attending physician, thoroughly familiar with the eligibility criteria, identified potential participants and referred them to the principal investigator for further evaluation; (ii) the investigators comprehensively explained the study objectives to the prospective participants, seeking their opinions and concerns; (iii) the eligibility of the participants was evaluated based on the predefined criteria; and (iv) both the participants and their legal guardian or family members provided signed consent, as required.

Patients were assigned into three cohorts based on their MAS scores: A (MAS1), B (MAS1+) and C (MAS2). Based on a 1:1 ratio, participants within each cohort were assigned to either the experimental group (receiving standard rehabilitation therapy combined with 3D-printed orthoses) or the control group (receiving standard rehabilitation therapy along with LTT).

The randomization sequence was generated using a computer, with the sequences sealed in opaque envelopes. The envelopes were sequentially opened by the principal investigator at the time of participant enrollment. Blinded evaluators who were consistently trained before the study performed standardized assessments sequentially across all participating centers. Notably, these evaluators remained blinded throughout the study and did not communicate with one another during the assessments; they were also not permitted to inquire about the nature of the intervention. Orthoses were fabricated by the same group of physicians across all participating centers to ensure consistency.

Intervention

All participants received a 6-week conventional rehabilitation program administered by experienced therapists. This program was delivered five times weekly for 40 min per session and consist of the following: neutral positioning of the limbs, application of stretching techniques to the wrist flexor muscles, and utilization of the proprioceptive neuromuscular facilitation (PNF), Bobath technique, and Rood technique. These interventions targeted enhancement of wrist-hand motor and sensory function. Additionally, occupational therapy activities focused on wrist extension and functional training for wrist-hand movements, with the activities of daily living (ADL) training also included.

The primary difference between the groups was the type of orthoses used: the experimental group used 3D-printed orthoses, while the control group received orthoses fabricated from LTT plates. Devices were worn 4–8 h per day for 6 consecutive weeks, with each wearing period lasting for > 30 min, followed by a resting period of approximately 15 min. Before orthosis application, the caregiver ensured the patient’s wrist and hand flexor muscles were adequately relaxed to prevent soft tissue injury.

Orthoses were fabricated based on the classical resting position protocol [11]:10°–15° wrist dorsiflexion, 30°–45° flexion at metacarpophalangeal (MCP) and proximal interphalangeal (PIP) joints, 10°–15° flexion at distal interphalangeal (DIP) joints, with the thumb positioned in 45° abduction at the carpometacarpal (CMC) joint to maintain functional opposition.

Fabrication and wearing processes of the low-temperature thermoplastic plate orthoses are illustrated in Fig. 1:

Fig. 1.

Low-temperature thermoplastic plate orthose

• The thermoplastic material was trimmed and customized according to the anatomical dimensions of the affected wrist and hand of the patient.

• The thermoplastic plates were subjected to a steady heating of 65–75 °C in a water bath for about 3–5 min to make them pliable for molding.

• Once softened and dried, the pliable plates were molded to conform to 10–15°the dorsiflexion of the wrist.

• The therapist continuously modified the shape of the thermoplastic plate through manual adjustments until aligns to the desired anatomical shape.

• Subsequently, the molded plate was traced with a contour line, followed by cutting and adjustments. The edges were polished, nylon fastener adhered, and then the orthosis adequately modified.

• Patients were provided with detailed orthoses-wearing instructions.

Fabrication and wearing processes of the 3D-printed orthoses are presented in Fig. 2:

Fig. 2.

3D-printed orthoses

The process involved upper limb anatomical remodeling, digital design of the orthotic devices using computer-aided design (CAD) technology, and fabrication through 3D printing.

• During upper limb reconstruction, a high-precision optical scanner (HCP; Creaform Co, Ltd, Canada) was used to provide a detailed visual of the surface geometry of the wrist and hand of the patient.

• The wrist and hand were immobilized using a specialized imaging device for the duration of the procedure to simulate flexor spasticity and produce an appropriate resting position of the upper limb of the patient. The wrist was positioned in approximately 10° to 15° of extension, with the metacarpophalangeal and proximal interphalangeal joints flexed to 30° to 45° and the thumb placed in abduction.

• The orthoses were designed using CAD technology (Unigraphics NX 8.0, Siemens, Germany) using surface stitching, region expansion, clasp creation, and aperture forming processes, based on the scanning model.

• The final CAD model was exported in Stereolithography (STL) format, preparing it for fabrication using 3D printing.

The primary material used for 3D-printed orthoses was a light-curable polymer. Upon completion, the orthoses the patients were invited for a fitting appointment, including instructions on how to wear the orthoses, with ensuring minimal adjustments to ensure a precise and anatomically conforming fit against the skin. Additionally, the patients were comprehensively instructed on how to use the orthoses effectively.

Outcome measures

Wrist flexor spasticity was assessed using the MAS, which ranges from 0 (no rise in muscle tone) to 4 (rigid limb with fixed contractions or extensions), and served as the measure of primary outcome. For statistical analysis, MAS scores of 1 to 4 were reclassified as 2 to 5 to simplify data processing. The efficacy of the intervention was determined by examining the difference in the pre- and post-treatment MAS scores, as well as during follow-up. The detailed benchmarks used were as follows [6]: (i) effective rate (ER) when the initial level declines by at least 1 or drops to level 0; (ii) no response (NR) if the initial level remains unchanged or increases. The ER = number of effective cases/total number of cases × 100%.

The secondary outcomes were as follows:

• Passive range of motion(PROM) [12] of the wrist joint—specifically in flexion, extension, radial deviation, and ulnar deviation—was measured using the corresponding joint angles.

• The motor functions of the affected wrist and hand were evaluated using the wrist and hand sections of the Fugl-Meyer Assessment (FMA) [13] comprising 12 items of wrist stability, mobility, and hand functionality. Items are scored on a three point (0–2) scale with maximum total score of 24 [14].

• Pain in the injured wrist was measured using a visual analogue scale (scored from 0 to 10, with a maximum score = 10, VAS) [15]. During passive wrist movements, the score was recorded at the peak of pain and it ranged from 0 (no pain) to 10 (unbearable pain).

• The extent of the swelling was assessed using advanced 3D scanning methods [16]. The total volume of the hand—from the wrist crease to the tip of the middle finger—was calculated using reverse engineering software (Geomagic Studio 2012). Volumes of the affected side and the healthy side were compared, with the healthy side used as the standard (100%) for relative volume calculations.

• A subjective rating was used to assess the satisfaction levels of the patients with the orthoses. The rating was scored based on the Quebec User Evaluation of Satisfaction with Assistive Technology (QUEST) assessment comprising assessments on the device aspect (including 8 elements addressing size, sturdiness, and wearability) and service aspect (including 10 elements addressing repair and maintenance, ongoing support, and personalization) [17].

All metrics, except for the personal experience rating, were conducted at the beginning and then re-evaluated after 6 weeks and 12 weeks, whereas the subjective sensation score was measured 6 weeks after the initial evaluation.

Statistical analysis

Data analysis was performed using SPSS (version 26.0, SPSS Inc., USA). The MAS scores were presented as count (percentage) and median (interquartile range). For secondary metrics, normally distributed continuous data were presented as mean ± standard deviation (SD), while non-normally distributed continuous data were expressed as median (interquartile range) M (Q1, Q3). Wilcoxon rank sum test or two independent sample t-test was used to compare continuous variables between the two groups. Chi-square test was used to assess the difference in the clinical efficacy of MAS scores between the groups. A p-value of < 0.05 was considered as statistically significant.

Results

Participant characteristics

A total of 180 patients were registered on the website, with 175 meeting the eligibility criteria and 147 completing the study in Fig. 3.The demographic characteristics of the patients are outlined in Table 1.

Fig. 3.

Flow chart illustrating the number of study participants. Group A: experimental group; Group B: control group; T0: baseline; T1: six weeks from baseline; T2: twelve weeks from baseline

Table 1.

Baseline characteristics of the three cohort

	MAS1		MAS1+		MAS2
Participants	Control(n = 24)	3D(n = 23)	Control(n = 25)	3D(n = 24)	Control(n = 26)	3D(n = 25)
Age (years), mean (SD)	67.00 (51.00, 77.25)	72.00 (64.50, 77.00)	74.00(68.00, 79.00)	69.50 (60.75, 77.25)	70.50 (61.00, 77.00)	68.00(60.00, 78.00)
Course disease (month)	9.00 (7.00, 10.00)	9.00 (7.00, 11.00)	8.00 (7.00, 10.00)	6.00 (5.00, 7.00)	7.00 (6.00, 10.50)	7.00 (6.00, 10.00)
Gender, n(%)
male	17 (70.8%)	15 (65.2%)	13 (52.0%)	16 (66.7%)	16 (61.5%)	16 (64.0%)
female	7 (29.2%)	8 (34.8%)	12 (48.0%)	8 (33.3%)	10 (38.5%)	9 (36.0%)
Diagnosis, n(%)
Cerebral infarction	17 (70.8%)	18 (78.3%)	20 (80.0%)	17 (70.8%)	19 (73.1%)	20 (80.0%)
Cerebral hemorrhage	7 (29.2%)	5 (21.7%)	5 (20.0%)	7 (29.2%)	7 (26.9%)	5 (20.0%)
Side of stroke, n (%)
Left	10 (41.7%)	10 (43.5%)	14 (56.0%)	11 (45.8%)	12 (46.2%)	15 (60.0%)
Right	14 (58.3%)	13 (56.5%)	11 (44.0%)	13 (54.2%)	14 (53.8%)	10 (40.0%)
FMA	4.00(3.00,5.00)	4.00(4.00,5.25)	4.00(3.00,4.25)	4.00(4.00,4.00)	1.00(0.00,2.00)	1.50(1.00,2.00)
PROM (median)	45.50 (37.25–62.75)	40.00 (24.75–58.50)	43.00 (31.00–65.00)	48.50 (37.25–62.00.25.00)	50.00 (35.00–68.00)	45.50 (30.25–54.00.25.00)
SS (median)	286 (276–321)	288 (268–328)	308 (253–328)	301.5 (272–335)	297 (269–338)	308 (250–337)
VAS (median)	1.00 (1.00–3.00)	1.00 (0.00–3.00)	2.00 (1.00–5.00)	2.50 (1.00–5.00)	4.00 (3.00–5.00)	4.00 (3.00–5.00)

PROM Passive range of motion, FMA Fugl-Meyer Assessment, VAS Visual analogue scale, SS Swelling scale, MAS Modified Ashworth Scale

In the MAS1 cohort, 47 patients completed the study. Among those initially enrolled, 3 patients voluntarily withdrew, 3 patients experienced changes in clinical condition changed, 1 patient died, 4 were lost to follow-up, and 2 were later found to not meet the inclusion criteria.

In the MAS1 + cohort, 49 patients were enrolled. Among those initially enrolled, 2 patients requested withdrawal, 3 experienced changes in the clinical condition, 4 patients were lost to follow-up, and 2 patients were subsequently found to not meet the enrollment criteria.

In MAS2 cohort, 51 patients were enrolled. Among the enrolled patients, 4 requested withdrawal, 2 patients exhibited changes in the clinical condition, 2 were lost to follow-up, and 1 patient subsequently failed the inclusion criteria.

Primary outcome

The ER of treatment was determined using changes in the MAS scores.Changes in the MAS scores between groups are shown in Table 2.

Table 2.

Comparison of change of MAS

Cohort	group	T1	T2
MAS1	3D (n = 23)	15(71.4%)	14(66.7%)
	Control (n = 24)	6(30.8%)	7(33.3%)
	P	0.006	0.029
MAS1+	3D (n = 24)	17(63.0%)	22(64.7%)
	Control (n = 25)	10(37.0%)	12(35.3%)
	P	0.030	0.001
MAS2	3D (n = 25)	8(36.4%)	10(40.0%)
	Control (n = 26)	14(63.6%)	15(60.0%)
	P	0.115	0.206

T1: six weeks from baseline; T2: twelve weeks from baseline; MAS: Modified Ashworth Scale

In the MAS1 cohort, the ER was 71.4% in the 3D group compared to 30.8% in the control group, with a statistically significant difference observed between the groups (p = 0.006). In the MAS1 + cohort, the 3D group exhibited an ER of 63% compared to 37% of the control group, also yielding a significant difference (p = 0.030).

Following the 12-week follow-up (from T2 to T0), a statistically significant difference was observed between the two groups, with a p = 0.029 in the MAS1 cohort and p = 0.001 in the MAS1 + cohort. In the MAS1 cohort, the 3D group exhibited an ER of 66.7%, compared to 33.3% in the control group. Similarly, in the MAS1 + cohort, the ER was 64.7% in the 3D group and 35.3% in the control group.

Secondary outcomes

Following the 6-week intervention period (T1-T0), significant differences were observed in any of the secondary measures between groups within each cohort (p > 0.05)(Table 3). The median QUEST-T scores for the experimental and control groups were 84 versus 81 for the MAS1 cohort, 85 versus 80 for the MAS1 + cohort, and 74 versus 81 for the MAS2 cohort, respectively.

Table 3.

Group comparison of secondary indicators

	cohort	Time	3D	control	P
VAS	MAS1	T1	2.00(1.00–4.00)	1.00(1.00–4.00)	0.045
		T2	2.00(0.00–2.00)	1.00(0.00–4.00)	0.004
	MAS1+	T1	1.00(0.00–3.00)	2.00(1.00–4.00)	0.004
		T2	1.00(0.00–2.00)	2.00(0.00–4.00)	0.001
	MAS2	T1	2.00(1.00–4.00)	2.00(1.00–5.00)	0.031
		T2	1.00(0.00–2.00)	2.00(1.00–3.00)	< 0.001
SS	MAS1	T1	279 (257–319)	288 (271–320)	0.064
		T2	273(247–306)	290.5 (278–318)	0.000
	MAS1+	T1	288 (266–315)	301 (258–333)	0.073
		T2	279 (259–305)	308 (248–323)	0.006
	MAS2	T1	304 (250–339)	289.5 (270–330)	0.067
		T2	305 (251–336)	283.5 (267–317)	0.005
PROM	MAS1	T1	46.50 (18.00–69.00)	51.00 (16.00–66.00)	0.184
		T2	48.00 (19.00–70.00)	55.00 (15.00–70.00)	0.115
	MAS1+	T1	53.50 (11.00–69.00)	48.00 (8.00–71.00)	0.903
		T2	56.00 (16.00–70.00)	50.00 (9.00–71.00)	0.955
	MAS2	T1	49.50 (10.00–70.00)	47.00 (14.00–70.00)	0.938
		T2	47.00 (10.00–70.00)	50.00 (15.00–70.00)	0.645
FMA	MAS1	T1	1.00 (0.00–2.00)	0.00 (0.00–9.00)	0.633
		T2	1.00 (0.00–2.00)	0.00 (0.00–9.00)	0.275
	MAS1+	T1	1.00 (0.00–5.00)	1.00 (0.00–11.00)	0.208
		T2	1.00 (0.00–5.00)	1.00 (0.00–11.00)	0.019
	MAS2	T1	1.00(0.00–3.00)	0.00 (0.00–2.00)	0.121
		T2	1.00 (0.00–3.00)	0.00 (0.00–2.00)	0.283
QUEST-T	MAS1	T1	84.00 (80.00–88.00)	81.00 (75.00–83.00)	0.000
	MAS1+	T1	85.00 (82.00–87.00)	80.00 (77.00–84.00)	0.000
	MAS2	T1	74.00 (70.00–79.00)	81.00 (76.00–83.00)	< 0.001

VAS Visual analogue scale, SS Swelling scale, PROM Passive range of motion, FMA Fugl-Meyer Assessment, QUEST Quebec User Evaluation of Satisfaction with Assistive Technology, T1: six weeks from baseline; T2: twelve weeks from baseline

After the 12-week follow-up (T2-T0), significant differences were found for all secondary measures between groups (p < 0.01), except the PROM in each cohort. FMA only showed significant differences in the MAS1 + cohort between groups (p = 0.019) (Table 3).

Adverse effects

No participant experienced adverse events. Additionally, all participants exhibited optimal compliance during assessments and administration of the interventions.

Discussion

The primary objective of this study was to evaluate the clinical efficacy of 3D-printed wrist orthoses for managing post-stroke spasticity and to establish practical guidelines for their clinical application. Our results confirmed that 3D-printed orthoses exhibited superior efficacy compared to low-temperature thermoplastic orthoses (LTT) in reducing mild-to-moderate spasticity, achieving higher patient satisfaction. However, Further exploration of 3D-printed materials is warranted to address severe spasticity.

The definition and assessment of spasticity are still debated [18, 19]. The classic definition by Lance describes it as “velocity-dependent hypertonia” due to hyperexcitable stretch reflexes [20]. However, the modern consensus, as proposed by Pandyan, broadens this to “disordered sensorimotor control manifesting as resistance to passive movement“ [21]. This broader perspective encompasses both neurogenic mechanisms (reflex hyperactivity) and peripheral biomechanical adaptations (intrinsic muscle viscoelastic remodeling). Previous studies have shown that reflex-mediated hypertonia peaks at 1–3 months post-stroke [3], whereas chronic spasticity (lasting more than 3 months) is dominated by muscle fiber shortening and collagen deposition [6]. MAS assesses spasticity by quantifying resistance to passive movement, which arises from a combination of neurogenic hyperreflexia and structural alterations in peripheral muscles and connective tissues [17]. Despite ongoing debates regarding its reliability and sensitivity, the MAS remains the most widely used clinical tool in the absence of a superior, validated alternative. Ensuring cross-study comparability, we utilized the MAS to assess the spasticity, supported by evidence of its greater reliability in assessing upper limb spasticity [22, 23].

Following 6 weeks of treatment, the 3D-printed orthoses demonstrated superior spasticity relief compared to LTT orthoses in both the MAS 1 and MAS 1 + cohorts, as measured by the MAS This trend persisted through a 12-week follow-up, with the relief in MAS scores remaining significant. The exceptional clinical efficacy demonstrated by the 3D-printed orthoses on mild-to-moderate spasticity is attributable to its ability to facilitate personalized customization and optimal adaptability.

Previous studies showed that better extension of tendons, ligaments and joint capsules of the wrist-interphalangeal joints can increase proprioceptive input, thereby effectively relieving spasticity [24]. The anatomically conforming design of the 3D-printed orthosis precisely fits the patient’s hand, significantly improving the joint biomechanical status of patients with mild to moderate spasticity. Particularly in the metacarpophalangeal (MCP) and interphalangeal (PIP/DIP) regions, its customized and rigidly supported fit effectively reduces the tension during passive muscle stretching [25].The anatomy-conforming orthosis optimizes pressure distribution across soft tissues, attenuating aberrant afferent input from muscle spindles and consequently suppressing gamma-loop-mediated spastic hyperactivity [26].

In contrast, the clinical efficacy of 3D-printed orthoses for severe spasticity remained comparable to LTT. This phenomenon is attributed to challenges in maintaining a stable wrist position during scanning for patients with severe spasticity, thus limiting the accuracy of the digital capture essential for 3D fabrication. Research has shown that the position of the body has a significant impact on spasticity treament [27]. Thus, orthotic intervention may require progressive strategies tailored to spasticity severity, as the use of an inappropriate orthoses can induce or exacerbate spasticity [28].

Consistent with previous studies [29], our findings demonstrated a statistically significant difference in pain and swelling alleviation for the 3D-printed orthotics group after 12 weeks of follow-up in each cohort. In our study, the orthosis was fabricated with the affected hand in a resting position, thereby eliminating the potential position-related clinical efficacy interference. We fabricated the orthosis with the affected hand in a resting position to eliminate potential interference from position-related clinical efficacy. This action significantly reduces mechanical stimulation, mitigates local inflammation, and ultimately relieves pain [30]. Maintaining the rest position minimizes stress on the joint capsule and ligaments, while optimizing muscle relaxation, thereby facilitating venous and lymphatic return [31]. Currently, there is a limited body of research addressing optimal positioning for orthotic devices, and no consensus has been established regarding standardized sampling or fitting positions. It is important to note that improper positioning may aggravate pain and induce spasticity [28].

During the 12-week follow-up, the MAS1 + cohort showed a difference in FMA scores between the groups. However, no differences were found in other cohorts.This indicates that orthotic intervention may have a limited impact on these specific functional outcomes. While orthotics can reduce spasticity and improve movement in patients, it takes over 13 weeks to observe improvements [32]. Additionally, studies have indicated that the relief of spasticity doesn’t always improve motor function [4]. The positive finding in the MAS1 + cohort may be attributed to a relatively less severe degree of neural damage, which could permit a release of motor function as spasticity is relieved.

The QUEST showed that 3D-printed orthoses had higher satisfaction in MAS1 and MAS1 + cohort. In contrast, patients in the MAS2 cohort exhibited higher satisfaction with traditional orthoses. Conclusively, 3D-printed wrist orthoses are more suitable for mild-to-moderate spasticity, with wrist pain a significant factor that should be considered during postural scanning. For severe wrist and hand spasticity, traditional orthotics showed superior clinical efficacy compared to 3D-printed orthoses. Moreover, five cases of cracking were observed in the 3D-printed orthoses within the MAS2 cohort. These incidents underscore the need for refinement of the 3D printing technology for wrist and hand orthotics, particularly concerning scanning methods and material composition.

Limitations

This study provides valuable insights into the management of PSS. However, it has a few limitations that must be acknowledged. First, limited sample size per MAS cohort (47–51 cases), restricting generalizability—especially for severe spasticity (MAS2). Second, our evaluation relied primarily on clinical rating scales, with a lack of objective quantitative measures. While the MAS scale used in this study is known for its relatively high reliability and validity, the inclusion of more objective indices, such as EMG, was limited by practical constraints during the follow-up.

Authors’ contributions

GYL and YX wrote the main manuscript text, YMW and YZ prepared figures and table. CBT and JS provided patients follow-up and evaluation. SD and YQB were responsible for patient recruitment and making assistive devices. All authors reviewed the manuscript.

Funding

This study was funded by: the Major Project of Shanghai Xuhui District Health Commission (Project No. SHXH292103. 2022 Shanghai Science and Technology Commission (Project No. 22Y31900200). 2023 Xuhui District Health System Peak Discipline Construction Funding Project. Commission Key Disciplinary Construction Project (NO:2024ZDXK0039).

Data availability

Supplementary data mentioned in the text are available to subscribers in online.

Declarations

Ethical approval and consent to participate

This study was implemented according to the Declaration of Helsinki, Good Clinical practice guidelines and the Consolidated Standards of Reporting Trials. This study was approved by the Ethics Committee of Shanghai Xuhui Central Hospital, Shanghai (Approval No. 2022–056).

Consent for publication

Participants and their relatives signed written informed consent for the study.

Competing interests

The authors declare no competing interests.