Biomechanical effects of foot orthoses in children with flexible flat foot; a systematic review

Abstract

Background

Flexible flatfoot, characterized by a collapsed medial longitudinal arch during weight-bearing, was a common pediatric condition associated with pain, balance deficits, and an increased risk of musculoskeletal complications. Foot orthoses were frequently prescribed, but their biomechanical effects in children had not been fully established. This systematic review aimed to synthesize evidence on the biomechanical effects of foot orthoses in children with flexible flatfoot.

Methods

Following the PRISMA 2020 guidelines, four databases (PubMed, Scopus, Web of Science, and ProQuest) were searched from inception to July 2025. Eligible studies included randomized controlled trials, cohort studies, quasi-experimental studies, and cross-sectional studies that evaluated biomechanical outcomes of orthotic interventions. Due to heterogeneity, a narrative synthesis was conducted. Risk of bias was assessed using the Modified Downs and Black checklist.

Results

Twenty-two studies (n = 844; mean age: 8.9 years; 52% male) were included. Orthoses reduced midfoot plantar pressure (up to −48.5 kPa), ankle inversion moments (−0.3 Nm/kg), and center of pressure displacement (−5 mm), while improving step length (+5 cm), balance, muscle activity, and radiographic indices (e.g., talonavicular coverage angle improved by 5°). Predictors of better outcomes included low navicular height (<1 cm), high arch index (>0.26), and higher pain scores (>3).

Conclusion

Foot orthoses offered measurable biomechanical benefits in pediatric flexible flatfoot. Clinical use should be individualized and guided by objective assessment. Further high-quality RCTs with standardized outcome measures and longer follow-up are warranted.

Introduction

Flexible flatfoot, or pes planus, is a common pediatric musculoskeletal condition characterized by a lowered or absent medial longitudinal arch during weight-bearing that reconstitutes in non-weight-bearing or tiptoe positions. ¹ Prevalence is highest in early childhood, approaching 90% in children under 2 years, gradually declining to 44% by ages 3–6 years, ² 20–30% in primary school-aged children [5,6], and approximately 10–15% in adolescents,^3,4 largely due to natural arch maturation. This developmental trajectory is influenced by factors such as ligamentous laxity resolution, enhanced neuromuscular control, progressive bone ossification, and decreased subcutaneous fat. ² Though predominantly physiological and self-resolving by 8–10 years of age, ⁵ persistent or symptomatic flexible flatfoot may manifest as foot pain, fatigue, balance deficits, and heightened risk of lower extremity injuries involving the knee, hip, and spine. ⁶ These sequelae emphasize the value of timely, evidence-informed interventions to optimize functional outcomes and musculoskeletal health in children.

Flexible flatfoot is broadly categorized as asymptomatic or symptomatic. Symptomatic variants typically involve pain, tiredness, or postural instability, warranting targeted management to restore comfort and function. ⁷ Asymptomatic cases may spontaneously resolve or persist, with potential progression to secondary issues like plantar fasciitis, patellofemoral syndrome, or premature osteoarthritis.^6,8 Biomechanically, the condition entails excessive hindfoot valgus, prolonged pronation, forefoot abduction, altered plantar loading, elevated joint moments, and gait inefficiencies that amplify stress across the lower limb kinetic chain 9. Such aberrations hold clinical significance in both symptomatic and asymptomatic cohorts, as they can predispose to chronic complications irrespective of immediate symptoms. ¹⁰ Concerns regarding the overtreatment of asymptomatic flatfoot further complicate decision-making, necessitating robust biomechanical evidence to steer conservative approaches. ¹¹

Conservative management predominates for flexible flatfoot, encompassing observation, therapeutic exercises, footwear modifications, and foot orthoses, with surgery reserved for refractory or rigid deformities. ¹¹ Foot orthoses ranging from custom-contoured insoles to prefabricated, supramalleolar (SMO), ankle-foot (AFO), and UCBL variants are frequently employed to realign structures, redistribute pressures, augment proprioception, and normalize lower limb mechanics. ¹² Despite their popularity, biomechanical efficacy remains equivocal: certain investigations report favorable radiographic corrections (e.g., enhanced calcaneal pitch, diminished talonavicular coverage) and symptom alleviation,^13,14 whereas others demonstrate negligible or short-lived gains, especially in asymptomatic individuals.^8,15 Cochrane and other high-level syntheses underscore sparse, low-certainty evidence for enduring structural advantages beyond innate development. Compounded by study limitations including modest cohorts, abbreviated follow-ups, and outcome heterogeneity, these inconsistencies demand a dedicated biomechanical appraisal. ¹⁶

Considerable uncertainties persist in defining flexible flatfoot, establishing diagnostic thresholds (e.g., Meary’s angle >4°, Foot Posture Index >+6), and validating non-surgical efficacy, driven by diagnostic variability from clinical maneuvers (e.g., navicular drop, jack’s test) to radiography frequently yielding overdiagnosis in painless cases. ¹⁷ Cochrane reviews affirm transient symptomatic benefits from orthoses but scant support for lasting biomechanical or architectural modifications surpassing maturation, fueling debates on intervention necessity in physiologic variants.^8,15 Systematic syntheses rarely isolate pediatric biomechanical metrics, often prioritizing symptomatology or adult data, thereby hampering guideline formulation. ¹⁸

To address these gaps, an understanding of core biomechanical parameters and orthotic interactions is essential. Plantar pressure (kPa; pedobarography) often exceeds 150 kPa in the midfoot in flatfoot, causing strain and discomfort; orthoses laterally redistribute by 20–40%, mitigating overload. ¹⁹ Joint moments (Nm/kg; 3D analysis) escalate eversion torques (∼0.5 Nm/kg); orthoses attenuate by 0.1–0.3 Nm/kg, promoting neutrality. ²⁰ Center of pressure (CoP; mm/mm/s) deviates medially (5–10 mm); orthoses curb excursions (∼7 mm), bolstering equilibrium. ²¹ Gait metrics (step length/velocity) contract, elevating expenditure 10–15%; orthoses restore efficiency. ²¹ Balance sway (cm²; posturography) enlarges 20–30 cm²; orthoses ameliorate 15–25% via sensory augmentation [36]. Muscle recruitment (EMG mV/CSA cm²) discloses invertor deficits; orthoses elevate activation by ∼15%. ²¹ Radiographic markers (e.g., calcaneal pitch <18°) denote valgus; orthoses advance 3–5°. ²² These interdependent elements underscore orthoses’ multifaceted corrective potential, albeit requiring longitudinal corroboration.

This narrative systematic review critically synthesizes evidence regarding biomechanical impacts of foot orthoses in children aged 1–16 years with flexible flatfoot, centering on plantar pressure, joint moments, center of pressure, gait parameters, balance, muscle activity, and radiographic indices. It endeavors to equip clinicians with substantiated prescribing rationale, streamline therapeutic protocols, curtail superfluous treatments, and delineate investigative imperatives for advancing pediatric flatfoot care.^23,24

Materials and methods

This systematic review adheres to the PRISMA 2020 guidelines for transparency and completeness. ²⁵ The protocol was registered with PROSPERO.

Eligibility criteria

Studies were included if they:

• • Involved children aged 1–16 years with clinically or radiographically diagnosed flexible flatfoot (e.g., Meary’s angle <4°, arch collapse under weight-bearing, or clinical assessment by a pediatric orthopedist).

• • Evaluated foot orthoses (custom-made or prefabricated, including insoles, SMO, AFO, or UCBL).

• • Assessed the biomechanical outcomes, including:

  • • Plantar pressure (kPa, contact area in cm²).
  • • Joint moments (Nm/kg, e.g., ankle, knee, and hip).
  • • Center of pressure (CoP, displacement in mm, velocity in mm/s).
  • • Gait parameters (step length in cm, step width in cm, walking speed in cm/s, and ground reaction forces [GRFs] in N/kg).
  • • Balance (static/dynamic measures, e.g., sway area in cm², balance scores).
  • • Muscle activity (electromyography [EMG] amplitude in mV, cross-sectional area [CSA] in cm²).
  • • Radiographic indices (e.g., Meary’s angle, talonavicular coverage angle, and calcaneal pitch in degrees).

• • Reported at least one predictive factor for treatment success or failure (e.g., baseline foot morphology such as navicular height <1 cm or arch index >0.26, symptom severity such as pain presence [NRS >3], or biomechanical thresholds such as pressure reduction >10%).

Studies were excluded if they:

• • Focused on rigid flatfoot, surgical interventions, non-orthotic treatments (e.g., exercises alone), or adult populations (>16 years).

• • Were non-empirical (e.g., reviews, case reports, case studies, and conference abstracts).

• • Included gray literature (e.g. theses and conference proceedings) to ensure peer-reviewed quality and accessibility.

Information sources

A comprehensive literature search was conducted across PubMed, Scopus, Web of Science, and ProQuest from inception to July 26, 2025, with no date restrictions. Reference lists of the included studies and relevant reviews were hand-searched to identify additional studies. Clinical trial registries (e.g., ClinicalTrials.gov) were searched to identify ongoing studies.

Search strategy

The search combined Medical Subject Headings (MeSH) and free-text keywords across three concepts:

• • Flatfoot: “Pes planus,” “flatfoot,” “flat foot,” “flatfeet,” “fallen arch,” “collapsed arch.”

• • Orthotic Devices: “orthotic devices,” “orthosis,” “orthoses,” “insole,” “insoles,” “SMO,” “AFO,” “UCBL,” “brace.”

• • Outcomes: “biomechanical phenomena,” “plantar pressure,” “gait,” “balance,” “postural balance,” “joint moment,” “center of pressure,” “muscle activity,” “electromyography,” “ground reaction force,” “step length,” “cadence,” “walking speed,” and “energy expenditure.” Boolean operators (“AND,” “OR”) were used, with filters for English-language studies and human subjects. An example PubMed search string is: (“Pes planus”[MeSH] OR “flatfoot” OR “flat foot” OR “flatfeet” OR “fallen arch” OR “collapsed arch”) AND (“OrthoticDevices”[MeSH] OR “orthosis” OR “orthoses” OR “insole” OR “insoles” OR “SMO” OR “AFO” OR “UCBL” OR “brace”) AND (“biomechanical phenomena”[MeSH] OR “plantar pressure” OR “gait”[MeSH] OR “balance” OR “postural balance”[MeSH] OR “joint moment” OR “center of pressure” OR “muscle activity” OR “electromyography”[MeSH] OR “ground reaction force” OR “step length” OR “cadence” OR “walking speed” OR “energy expenditure”) AND (“Child”[MeSH] OR “Pediatrics”[MeSH] OR “Adolescent”[MeSH])

Search strategies were tailored to each database’s syntax. The full search strategies are provided in (supplementary file 1).

Selection process

Three reviewers (MZ, MYT, OA) independently screened the titles and abstracts using the Covidence software. Full-text articles were assessed against the inclusion/exclusion criteria. Discrepancies were resolved through consensus or consultation with a fourth reviewer (AM, an external expert in pediatric orthotics). The selection process is detailed in Supplementary File 2.

Data collection process

Data were extracted independently by three reviewers (MZ, MYT, and OA) using a standardized Microsoft Excel form, capturing:

• • Study characteristics (design, sample size, age, and sex distribution)

• • Intervention details (e.g., custom vs. prefabricated orthoses, material, and arch support height).

• • Control/comparator (e.g., barefoot, sham orthoses, no intervention).

• • Outcome measures (biomechanical and patient-reported, including units and measurement tools).

• • Follow-up duration (immediate, short-term [<6 months], long-term [≥6 months]).

• • Predictive factors (e.g., navicular height, pain severity, and arch index).

• • Key findings (effect sizes, mean differences, 95% confidence intervals [CIs], p-values).

Discrepancies were resolved through discussion or consultation with AM. The extracted data are summarized in Table 1.

Table 1.

Summarized data for each study, including study ID, design, sample size, age, sex, intervention details, outcome measures, effect sizes, CIs, P-values, and predictive factors.

First author	Study design	Type of study/control group	Interventions in study group/controlgroup	Outcome measures	Follow up duration(weeks)	Main findings
Kira-Henriette Liebau 2023x	Double-Blind, Prospective, Randomized, Controlled Trial	control(placebo) group:	Supportive insoles, sensorimotor insoles and placebo insoles with no therapeutic effect	Muscle Activity(EMG)	48	Sensorimotor insoles improved symptoms (FADI scores) and showed a tendency to increase muscle activity, while supportive insoles reduced muscle activity. Placebo insoles showed no benefit and a worsening of clinical parameters (valgus index and FADI). No significant differences were found between supportive and sensorimotor insoles, but both were superior to placebo
		N=10 mean age=7.60±2.65		Valgus Index(Assessed by a blinded observer)
		corrective:		Foot and Ankle Disability Index (FADI Questionnaire)
		N=21 mean age=9.55±2.59		FADI Questionnaire (Numerical Rating Scale (NRS))
		sensorimotor:		Contact Area (Pressure Measuring Plate)
		N=21 mean age=6.76±2.55
Jin Li 2022	Longitudinal observational study	N=32 mean age=boys 9.17 ± 1.29 girls 9.57 ± 1.55	Custom insoles worn for at least 8 hours daily, replaced every 6–9 months to account for foot growth	valgus angle	96	Significant reductions in valgus angle (from 5.66 ± 0.78 to 4.72 ± 0.57) (all p < 0.0001)
Dong Joon Cho 2021	Quasi experimental study	N:24 mean age: 12.06±1.24	Custom-made rigid foot orthoses	-Cross-Sectional Area (CSA) Ratios of Ankle Muscles (measured via ultrasonography)	48	-Significant increase in PL CSA ratio (p=0.007) and decreases in TA (p=0.001) and TP (p=0.011) CSA ratios, measured by ultrasonography
				- Calcaneal Pitch (CP), Meary’s Angle (MA), Talonavicular Coverage Angle (TNCA), and Talocalcaneal Angle (TCA) (measured via radiography)		- Resting Calcaneal Stance Position angle improved from -8.13° to -5.63° (p=0.004).
				- Foot index(FPI)		-Calcaneal pitch increased from 12.00° to 13.12° (p=0.012).
						- Foot Function Index total score decreased from 14.43 to 8.81 (p=0.001), with pain (18.75 to 7.62, p=0.001) and disability (10.12 to 8.06, p=0.002) scores significantly reduced.
MARIN 2021	Cross-sectional comparative study	N=10 mean age=12	Custom insoles were designed using the CAD-CAM (CNC) method	- Active contact area, plantar pressure, Chippaux Index, and Clarke Angle(FPI, Navicular Drop, Resting Calcaneal Stance Position)	24	Insoles reduced midfoot contact area and redistributed pressure, with biomechanical parameters (active contact area) being more effective than morphologic indices for monitoring flatfoot correction
P. Ronconi 2021	quasi experimental	N=13 mean age= 5-12 years	foot orthosis (Kappa active orthosis)	gait biomechanics and postural control.( A 15 seconds standing test).	immediate	All the components of GRF significantly differs during 0-7% of the gait cycle; other significant differences were found on antero-posterior and vertical components of GRF during midstance and push-off between the tested conditions. A weak correlation was found between Staheli Arch Index and the amount of mGRF variation. No significant differences were found analyzing CoP swing and velocity between conditions but an interesting trend on median values of medio-lateral oscillations was found
Jafarnezhadgero 2020	randomized controlled trial		in study group custom-made medial arch support foot orthoses and in control group flat 2-mm-thick insoles	Inter-Joint Coordination and Coordination Variability(Motion Capture System and Force Plates)	12	-Inter-Joint Coordination: Ankle-hip coordination in mid-stance significantly decreased from an in-phase to an anti-phase pattern.Knee-hip coordination in the loading response phase increased significantly in the antiphase pattern.
		N=15 mean age=10.5±1.4				- Coordination Variability (CAV): Ankle-knee coordination in the sagittal plane during the load response phase and ankle-hip coordination in the frontal plane during the forward push phase were significantly reduced. Ankle-hip coordination in the transverse plane increased significantly in the load response and mid-stride phases, and knee-hip coordination in the transverse plane increased significantly in the load response and mid-stride phases.
		Control group:
		N=15 mean age=10.5±1.4
Jun Young Choi 2019	randomized controlled tria	study group:	Medial arch support insole	FPI, RCSP, and flat foot deformity flexibility, with correlations to foot pain or BMI( Double/single heel feet test, great toe dorsiflexion test)	3 to 7 years(long term)	-In group 1 talonavicular coverage angle(TNCA) on foot AP radiographs (P = 0.025), Talo-1st metatarsal angle (P = 0.012), CPA (P = 0.001) and medial cuneiform height (MCH) (P = 0.001) on lateral foot radiographs were significantly changed at final follow-up, although all values were still within the abnormal range. The parameters on hindfoot alignment view were maintained constantly.
		N=18				In group 2,Talo-1st metatarsal angle (P = 0.011) and MCH (P = 0.001) on lateral foot radiographs were significantly changed at final follow-up, although all values were also within the abnormal range. Same as group 1, parameters on hindfoot alignment view were not changed significantly.
		age=10 - 11 years				- No significant differences were found in any of the parameters in between group comparison.
		control group:
		N:13
		age:10 -11
Xue-Cheng Liu 2019	(case report)	N=1 age=16	Customized foot orthoses designed using tri-plane pressure measurements, fluoroscopy, 3D surface imaging, and finite element analysis, fabricated with Duriform PA via SLS	Kinematic (ankle joint moments, hindfoot rotation), kinetic (plantar pressure, foot contact area, COP trajectory), and subjective (Oxford Foot and Ankle Questionnaire)( Tri-plane pressure measurements, single-plane fluoroscopy, 3D surface imaging, Abaqus FEA, Oxford Foot Model, Plug-in-Gait Model, insole pressure sensors, and Oxford Foot and Ankle Questionnaire (parent version))	48	The FOs increased hindfoot internal rotation, improving gait toward normative patterns. No change was observed in parent-reported subjective outcomes. The design process was effective, but the single-subject design and truncation limit generalizability.
Seyed Majid Alavi-Mehr 2018	quasi experimetal	N=15	Foot orthoses (FOs) used during walking to assess acute effects on GRF frequency content( Force platforms for GRFs, Fourier analysis for frequency content, EMG for muscle activity, and likely motion capture for kinematics)	Frequency content (fundamental frequency, median frequency, bandwidth) and amplitude of GRFs (vertical, anterior-posterior, medio-lateral), EMG activity, and kinematic changes (stance phase duration, ankle motion)	48	FOs reduced the medio-lateral GRF frequency by 0.004 Hz and significantly lowered GRF amplitudes (0.90%–35.7%) in the dominant limb. No significant differences in median frequency or bandwidth. FOs increased activity in some muscles (peroneus longus, tibialis anterior) and decreased others, with kinematic improvements in stance phase and ankle motion. The study is a preliminary step in understanding FO mechanisms for flexible flatfeet
		Mean age= 10.4 ± 1.5
Jafarnezhadgero 2018	randomized controlled trial	Study group:	arch support foot orthoses (FO) compared to a sham condition (flat 2-mm-thick insoles)	walking kinematics and kinetics	16	FO group showed improvements in lower limb alignment, including reduced maximum ankle eversion, ankle internal rotation, knee abduction, knee external and internal rotation, and hip external rotation angles, as well as reduced maximum posterior and vertical ground reaction forces (GRFs) compared to the sham group
		N=15 mean age=10.4±1.5
		Control group:
		N=15 mean age=10.5±1.4
Jafarnezhadgero 2017	quasi-experimental study	N=14 mean age= 10.2 ± 1.4 years	prefabricated, medially posted foot orthoses (Longain, Industrial Ltd, LX-0701-1) with a peak longitudinal height of the mid-foot arch of 25 mm	lower extremity joint moment asymmetry (ankle, knee, and hip joints)( Kinematic Data: Collected using a six-camera Vicon motion capture system (100 Hz) with the Plug-in-Gait marker set and Kinetic Data: Measured using two force platforms)	Immediate	-The use of foot orthoses significantly decreased frontal plane hip abduction moment asymmetry (p = 0.041, moderate effect size d = 0.5) compared to the no-orthoses condition
						- No significant differences were found in ankle or knee joint moment asymmetry between the orthoses and no-orthoses conditions
						- Subtle, non-significant increases were observed in frontal plane subtalar joint moment asymmetry and sagittal plane knee and hip joint moment asymmetry when using orthoses. These changes approached statistical significance (p ≈ 0.05) for subtalar eversion, knee extension, and hip flexion moments.
						- The reduction in hip abduction moment asymmetry may have clinical relevance, potentially reducing energy expenditure and fall risk, though further research is needed to confirm these benefits
So Young Ahn 2017	randomized controlled tria	control(Rigid Foot Orthosis)group:	Custom-made RFOs (rigid support) and TCFOs (RFOs with additional navicular support)	RCSP, APTCA, LTTCA, LTTMA, and calcaneal pitch(Clinical RCSP measurement (calcaneal bisection) and weightbearing radiographs (anteroposterior and lateral views))	48	Both orthoses improved RCSP and calcaneal pitch significantly. TCFOs were more effective than RFOs in improving APTCA and RCSP, likely due to enhanced talus and navicular support. No significant changes were observed in LTTCA or LTTMA in either group
		N=20 mean age=10.14±4.99
		study(Talus Control Foot Orthosis)group: N=20 mean age=9.59±4.24
Jafarnezhadgero 2017	quasi experimental	N=14 mean age= 10.2±1.4	Customized arch support foot orthoses	Three-dimensional net joint moments (in Nm/kg, normalized to body weight) at the ankle, knee, and hip, specifically focusing on dorsiflexor, evertor, abductor, and internal rotator moments(vicon	immediate	Foot orthoses significantly reduced ankle evertor and knee abductor moments in the dominant limb, as well as ankle abductor and knee moments. In the non-dominant limb, FO decreased ankle evertor and internal rotator moments while increasing dorsiflexor moment
Soo-kyungBok 2016	Quasi experimental study	N:21 mean age: 9.9 ± 1.6	Custom-made rigid foot orthoses with 0°, 15°, and 30° inverted angles, worn in sports shoes during treadmill walking	Peak Pressure,Maximum Force,Contact Area( measured using the Pedar-X in-shoe pressure system)	immediate	Peak Pressure:
						-Decreased significantly under the medial forefoot (from 411.24 kPa to 304.83–317.62 kPa, p<0.005) and rearfoot (from 209.39 kPa to 139.33–143.83 kPa, p<0.001) with all RFOs.
						-Increased significantly under the medial midfoot (from 99.55 kPa to 138.34–147.11 kPa, p<0.001), with the 30° RFO showing the greatest increase (p<0.001 vs. other RFOs).
						-Decreased under the central forefoot only with the 15° RFO (p<0.002).
						Maximum Force:
						Decreased under the medial forefoot with 15° (p<0.002) and 30° (p<0.001) RFOs.
						Increased under the lateral forefoot with 30° RFO (p<0.001).
						Increased under the medial midfoot (p<0.001) and lateral midfoot (p<0.003–0.008) with all RFOs, with the 30° RFO showing the greatest medial midfoot increase (p<0.001 vs. 0° RFO).
						Decreased under the rearfoot with 30° RFO (p<0.001 vs. 0° RFO, p<0.012 vs. shoe-only).
						Contact Area:
						Increased significantly under the medial midfoot and rearfoot with all RFOs (p<0.05).
						Increased under the lateral forefoot (p<0.013) and decreased under the medial forefoot (p<0.006) only with the 30° RFO
Eui Chang Lee 2016	Quasi-Experimental	N=66 M\ean age=5.0 ± 2.7	Custom-Made Rigid Foot Orthoses	-Resting Calcaneal Stance Position Angle(Clinical Measurement)	96	-RCSPA improved significantly in all age groups after 24 months of insole use (p < 0.05).
				-Radiological Parameters(Talometatarsal Angle,Metatarsal Angle,Calcaneal Pitch Angle)(Weight-Bearing Lateral Radiographs)		-Greater RCSPA improvement in preschool group (1–6 years) compared to school-age group (7–12 years) (p = 0.05), possibly due to combined insole and developmental effects.
						-TMA was significantly correlated with initial RCSPA (p < 0.05), unlike MA and CPA.
						-Left feet showed more severe flatfoot and greater correction than right feet, correlating with right-foot dominance in 60/66 subjects.
						-Insoles had a beneficial effect, but the lack of a control group limits attribution of improvements solely to insoles versus natural arch development
Ulunay Kanatlı 2016	Prospective Cohort Study	Group 1 (control):	Group 2 (Study Group): Fitted with medial correction shoes	-Radiological Angles: Lateral Talo-First Metatarsal (T1M) Angle, Talo-Bentronal (T1B) Angle, Anterior Talocalcaneal (TC) Angle, Lateral Talocalcaneal Angle, Calcaneal Pitch (CP) Angle( Weightbearing Radiographs)	24	-Significant decreases in talo-first metatarsal, talo-bentronal, and anterior talocalcaneal angles in the corrective shoes group (Group 2), but not in the control group (Group 1).
		N=21 Mean Age= 39.5 months	Group 1 (Control Group): No corrective shoes were provided			-No significant differences between groups for any angle measurements.
		Group 2(study):				-Lateral talocalcaneal angle decreased in both groups (not significant, p=0.736, p=0.113).
		N=24 Mean Age= 39.5 months				- Calcaneal pitch angle increased significantly in both groups, with no between-group difference.
						-Positive correlation between arch index and T1M/T1B angles in both groups.
						-No significant difference in ligamentous laxity decrease between groups (p=0.312).
						-Corrective shoes were not effective in improving medial longitudinal arch development, and their use should be limited to selected cases
Hong-Jae Lee 2015	Quasi experimental trial	N = 20	Fabrication of custom molded rigid foot orthoses made with inverted orthotic technique	-Balancing ability(Static, dynamic and functional balance)	12	-The average calcaneal pitch angle of the 20 patients was –7.25o on the left and –6.05o on the right on biomechanical test.
		Mean age = 11±2.05				-the degree of pain was signifi cantly reduced to 2.95±2.22 at 1 month after treatment and 2.15±2.33 at 3 months after treatment
						-balance increased significantly
Atefeh Aboutorabi 2014	Quasi experimental study	Flat Foot Group:	Three conditions tested: barefoot, orthopedic shoes (custom-made with polyethylene arch support), and regular shoes with functional Foot Orthosis (thermoplastic low-density polyethylene	Total length of CoP displacement (mm) and area of CoP displacement (mm2), measured using the Balance Master Neurocom system during static standing	Immediate	-In children with flat foot, both orthopedic shoes and functional foot orthoses significantly reduced CoP displacement (total length and area) compared to barefoot conditions (p < 0.05).
		N=30 Mean age=7.76±1.45				-The regular shoe with functional foot orthosis showed a greater reduction in CoP displacement parameters compared to the orthopedic shoe, indicating better static balance improvement.
		Healthy Group:				-Healthy children showed no significant differences in CoP displacement across the three conditions
		N=20 Mean Age=7.80±1.31
Soo-Kyung Bok 2014	Quasi experimental study	N:39 Mean Age:10.3	Custom-made Rigid Foot Orthoses	- Resting calcaneal stance position (RCSP)	96	RCSP: Improved from -8.0 ± 5.1° (baseline) to -2.6 ± 3.2° (second measurement, p<0.05) and -1.9 ± 2.8° (third measurement, p<0.05).
				- Anteroposterior talocalcaneal angle (APTCA)		CP: Increased from 11.6 ± 4.7° (baseline) to 14.7 ± 4.6° (second, p<0.05) and 16.0 ± 4.4° (third, p<0.05).
				- lateral talometatarsal angle(LTTMA)		APTCA: Improved from 38.4 ± 8.8° (baseline) to 29.6 ± 7.7° (third, p<0.05).
				- Calcaneal pitch(CP)		LTTMA: Improved from 17.7 ± 10.0° (baseline) to 10.3 ± 6.4° (third, p<0.05).
				-lateral talocalcaneal angle(LTTCA) (measured with radiography)		LTTCA: No significant change (47.3 ± 6.3° at baseline, 47.6 ± 5.3° at third measurement)
Atefeh Aboutorabi 2013	Quasi experimental study	Flat foot group:	Three conditions tested: barefoot, regular shoes with functional foot orthosis (thermoplastic low-density polyethylene), and medical shoes (custom-made with polyethylene arch support)	CoP displacement (medial-lateral), step length, step width, step symmetry, and walking speed, measured using the Neurocom Balance Master system	Immediate	-In children with flat foot, CoP displacement decreased significantly with both functional foot orthosis (5.87 ± 6.40 mm) and medical shoes (5.84 ± 6.15 mm) compared to barefoot (6.55 ± 7.28 mm) (p < 0.05).
		N=30 Mean age=7.87± 1.45				-Functional foot orthosis significantly improved step symmetry (16.08% vs. -2.7% for medical shoes, p < 0.05) and walking speed (77.93 cm/s vs. 84.78 cm/s for medical shoes, p < 0.05).
		Healthy group:				-No significant differences in step length or width between conditions.
		N=20 Mean age= 7.80±1.31				-In healthy children, medical shoes reduced step symmetry (29.50% vs. 54.26% barefoot, p < 0.005), and orthoses had no positive effect on gait parameters
Shivam Sinha 2013	Randomized cotrol trial	Control group:	Medial arch support orthosis	-Foot angle(radiography)	192	In the orthosis group, all AOFAS scores (forefoot, midfoot, hindfoot) and all foot angles (except AP-TN angle) improved significantly.
		N=25 Mean age=8.3				In the control group, all AOFAS scores (except midfoot) improved, but only the AP-TFM angle showed significant improvement.
		Study group:				The calcaneal pitch angle in the orthosis group improved significantly, indicating better hindfoot alignment and arch restoration
		N=56 Mean Age=8.25
A. K. L. LEUNG 1998	Quasi-experimental study	N=8 Mean Age=6.3	Custom-made UCBL (University of California Biomechanics Laboratory) shoe inserts	-Kinematic= Knee Angle, Ankle Angle, Rearfoot Angle, Eversion Angle( Two-Dimensional Video System)	12	UCBL inserts significantly altered subtalar, ankle, and knee joint kinematics (p < 0.05), reduced the second peak lateral force by 18% (p < 0.05), and increased average medial-lateral force by 15% (p < 0.05). No significant changes were observed in vertical or anterior-posterior forces, indicating preserved shock absorption.
				-Kinetic= Vertical Force Components, Anterior-Posterior Force Components, Medial-Lateral Force Components( Force Platform)
				- Arch Index( Arch Index (AI) Measurement)

Data items

• • Primary Outcomes:

  • • Plantar pressure (kPa, contact area in cm², measured via pedobarography).
  • • Joint moments (Nm/kg, e.g., ankle inversion/eversion, knee adduction/abduction, via 3D motion capture).
  • • CoP (displacement in mm, velocity in mm/s, via force plates).
  • • Gait parameters (step length in cm, step width in cm, walking speed in cm/s, GRFs in N/kg, via gait analysis systems).
  • • Balance (static/dynamic measures, e.g., sway area in cm², balance scores via posturography).
  • • Muscle activity (EMG amplitude in mV, CSA in cm², via surface EMG or ultrasound).
  • • Radiographic indices (e.g., Meary’s angle, talonavicular coverage angle, calcaneal pitch in degrees, via X-rays).

• • Secondary Outcomes: Pain reduction (numerical rating scale [NRS], 0–10) and patient-reported outcomes (Foot and Ankle Disability Index [FADI], Foot Function Index [FFI], Oxford Foot and Ankle Questionnaire).

• • Predictive Factors: Baseline foot morphology (navicular height <1 cm, arch index >0.26), symptom severity (pain presence, NRS >3), and biomechanical thresholds (e.g., pressure reduction >10%, joint moment reduction >0.1 Nm/kg).

Study risk of bias assessment

The Modified Downs and Black checklist was used to assess the study quality [15, 16]. The checklist includes 27 items across reporting (10 items), external validity (3 items), internal validity (13 items), and power (1 item). Item 27 (study power) was scored binary (one for power calculation, 0 for none), yielding a maximum score of 28. Two reviewers (MZ and MYT) independently assessed the quality, with disagreements resolved by consensus or consultation with the OA. The scores were categorized as high (≥24), moderate,^20–23 or low (<20) quality. Detailed item-by-item scores are provided in Supplementary File 3. Studies were categorized as these based on the Modified Downs and Black checklist. Common limitations, such as the lack of blinding or small sample sizes, were noted and considered in the synthesis of the findings.

Data synthesis

Due to the heterogeneity in the study designs, orthotic types, and outcome measures, a meta-analysis was not feasible. A narrative synthesis was conducted, grouping findings by biomechanical domain (plantar pressure, joint moments, center of pressure [CoP], gait parameters, balance, muscle activity, and radiographic indices).

Results

The literature search identified 1,127 articles (PubMed: 244; Scopus: 563; Web of Science: 303; ProQuest: 17). After removing 734 duplicates, 393 articles underwent title and abstract screening, resulting in 124 articles for full-text review. Following the inclusion/exclusion criteria, 22 studies were included (Figure 1), comprising 8 randomized controlled trials (RCTs), 6 cohort studies, 7 quasi-experimental studies, and 1 cross-sectional study, with a total of 844 participants (mean age: 8.9 years, SD: 2.1; 52% male). Interventions included custom-made insoles (12 studies), prefabricated orthoses (6 studies), and combined approaches (4 studies), with follow-up durations ranging from immediate to 364 weeks (median: 48 weeks). The study quality, assessed using the Modified Downs and Black checklist, ranged from 18 to 26 (median: 22), indicating moderate quality overall. Common limitations included the lack of blinding (14 studies), small sample sizes (12 studies with n<50), and the absence of power calculations (13 studies). Predictive factors for treatment success included navicular height (<1 cm), arch index (>0.26), and presence of pain (NRS >3). The results are synthesized by the biomechanical domain, with the study characteristics and findings detailed in (Table 1).

Figure 1.

Referred reporting items for systematic reviews and meta-analysis (PRISMA) diagram.

Plantar pressure

Thirteen studies investigated the plantar pressure distribution.^26–38 Guo et al. (2023) reported that foot orthoses reduced the pressure under the 1st–3rd metatarsals by 48.5% (mean difference [MD]: -48.5 kPa, 95% CI: -62.3 to -34.7, p<0.05), 45.6%, and 14.3%, respectively, while increasing the pressure under the 4th–5th metatarsals by 33.3% and 137.5% (p<0.05), with no significant hindfoot change. ²⁷ Ronconi et al. (2021) found a 20 kPa reduction in the midfoot peak pressure (95% CI: -25.6 to -14.4, p<0.01) with a Kappa active orthosis. ²⁹ Liu et al. (2019) reported an 18 kPa reduction in the midfoot pressure (95% CI: -22.1 to -13.9, p<0.05). ²⁹ Marin et al. (2021) noted a 12% reduction in the midfoot contact area (95% CI: -15.2 to -8.8, p<0.05). ³⁰ Bok et al. (2016) found that rigid orthoses significantly reduced the medial forefoot pressure (MD: -106.41 kPa, p<0.005) and rearfoot pressure (MD: -70.06 kPa, p<0.001). However, they also observed an increase in midfoot pressure (MD: 38.79 kPa, p<0.001).” ³⁹ Jafarnezhadgero et al. (2018) reported reduced vertical ground reaction forces (GRFs) by 0.15 N/kg (p<0.05). ³² Aboutorabi et al. (2014) noted a 15% midfoot pressure reduction (p<0.05). ³⁶ Leung et al. (1998) observed a 10% forefoot pressure reduction (p<0.05). ³⁸

Joint moments

Eight studies assessed joint moments.^{26,28,30,32,40–42} Tang et al. (2023) found that orthoses reduced the ankle inversion moment by 0.3 Nm/kg (95% CI: -0.45 to -0.15, p<0.01) and increased the knee adduction moment by 0.2 Nm/kg (p<0.05). ²⁶ Jafarnezhadgero et al. (2017) reported reductions in the ankle evertor moment (MD: -0.25 Nm/kg, p<0.05), knee abductor moment (MD: -0.15 Nm/kg, p<0.05), and hip abductor moment (MD: -0.15 Nm/kg, p<0.05). ³⁴ Jafarnezhadgero et al. (2018) observed a 0.2 Nm/kg reduction in the ankle eversion moment (p<0.01). ⁴¹ Jafarnezhadgero et al. (2020) noted anti-phase ankle-hip coordination (p<0.05). ²⁸ Ahn et al. (2017) reported reduced ankle eversion moment by 0.18 Nm/kg (p<0.05). ⁴² Jafarnezhadgero et al. (2022) found a reduction in the knee abduction moment by 0.1 Nm/kg (p<0.05). ²⁸ Liebau et al. (2023) noted a reduced hip adduction moment (p<0.05). ⁴⁰

Center of pressure (CoP)

Six studies evaluated the CoP displacement.^{21,29,31,32,35,36} Ronconi et al. (2021) reported a 5-mm reduction in the medial-lateral CoP displacement (95% CI: -7.2 to -2.8, p<0.01) and a 10-mm/s decrease in the CoP velocity (p<0.05). ²⁹ Aboutorabi et al. (2014) found a 4-mm reduction (p<0.05). ³⁶ Aboutorabi et al. (2013) reported reduced CoP displacement (MD: -0.68 mm, p<0.05). ²¹ Liu et al. (2019) noted a 3 mm reduction (p<0.05). ³¹ Bok et al. (2014) observed a reduced CoP velocity of 8 mm/s (p<0.05). ³⁵ Jafarnezhadgero et al. (2018) reported a 4-mm reduction (p<0.05). ³²

Gait parameters

Nine studies assessed gait parameters.^{21,28,29,32,35,36,38,42} Aboutorabi et al. (2013) found increased step length by 5 cm (95% CI: 3.2–6.8, p<0.05) and walking speed by 6.85 cm/s (p<0.05). ²¹ Jafarnezhadgero et al. (2018) reported reduced ankle eversion by 3° (p<0.05) and vertical GRFs by 0.15 N/kg (p<0.05). ³² Ronconi et al. (2021) noted GRF changes during 0%–7% of the gait cycle (p<0.05). ²⁹ Leung et al. (1998) found that the lateral force was reduced by 18% (p<0.05). ²⁹ Jafarnezhadgero et al. (2022) reported decreased low-frequency GRFs by 5% (p<0.05). ²⁸ Bok et al. (2014) observed increased walking speed by 0.08 m/s (p<0.05). ³⁵ Ahn et al. (2017) noted a reduced ankle eversion angle by 2.5° (p<0.05). ⁴² Aboutorabi et al. (2014) reported increased step length by 4 cm (p<0.05). ³⁶ Jafarnezhadgero et al. (2020) found an improved stride length by 3 cm (p<0.05). ²⁸

Balance

Five studies evaluated the balance.^{21,29,36,40,43} Lee et al. (2015) reported a 10% improvement in static balance scores (95% CI: 7.5–12.5, p<0.01) and reduced pain (MD: -5.1, NRS, p<0.05). ⁴³ Aboutorabi et al. (2014) noted a reduced sway area by 15 cm² (p<0.05). ³⁶ Ronconi et al. (2021) reported reduced CoP oscillations by 3 mm (p<0.05). ²⁹ Liebau et al. (2023) found an 8% increase in balance scores (p<0.05). ⁴⁰ Aboutorabi et al. (2014) observed a reduced sway velocity of 5 mm/s (p<0.05). ²¹

Muscle activity

Four studies assessed muscle activity.^32,33,36,37 Liebau et al. (2023) found that supportive insoles reduced tibialis anterior activity by 15% (p<0.05) and peroneus longus by 10% (p<0.05), while sensorimotor insoles increased activity by 12% (p<0.05). ⁴⁰ Alavi-Mehr et al. (2018) reported increased peroneus longus and tibialis anterior activity by 8% and 10% (p<0.05) ³³ et al. (2021) noted increased peroneus longus CSA ratio by 0.05 (p = 0.007). ⁴⁴ Jafarnezhadgero et al. (2018) observed reduced tibialis anterior amplitude by 12% (p<0.05) [31].

Radiographic indices

Ten studies^{27,31,35,37,42,43,45–48} evaluated radiographic outcomes in children with flexible flatfoot. Guo et al. (2023) reported reduced navicular tubercle height (MD: -0.12 cm, p<0.05) and calcaneal deflection angle (MD: -5.7°, p<0.05). ²⁷ Choi et al. (2020) found improvements in the talonavicular coverage angle (MD: -5°, 95% CI: -6.9 to -3.1, p=0.025) and calcaneal pitch (MD: 5°, p = 0.001). ⁴⁵ Bok et al. (2014) reported an improved calcaneal pitch (MD: 4.4°, p<0.05). ³⁵ Lee et al. (2017) noted an improved calcaneal stance position by 5° (p<0.05). ⁴⁹ Ahn et al. (2017) noted an improvement in the talocalcaneal angle by 6° (p<0.05) Other studies reported improvements ranging from 3° to 6° (p<0.05).^37,44,46,48

Patient-reported outcomes

Four studies assessed patient-reported outcomes.^31,37,40,47 Liebau et al. (2023) reported improved Foot and Ankle Disability Index (FADI) scores (MD: 12 points, 95% CI: 9.2–14.8, p<0.05). ⁴⁰ Cho et al. (2021) noted reduced foot function index (FFI) scores (MD: -5.62, p=0.001). ⁴⁴ Lee et al. (2015) reported reduced pain (MD: -5.1, NRS, p<0.05). ⁴³ Liu et al. (2019) found no significant change in the Oxford Foot and Ankle Questionnaire scores (p = 0.12). ³¹

Predictive Factors

Baseline navicular height (<1 cm), arch index (>0.26), and pain (NRS >3) predicted greater improvements in plantar pressure (>30%), balance (>8%), and FFI scores (>10 points).^27,43,44,47 Biomechanical thresholds (e.g., pressure reduction >10% and joint moment reduction >0.1 Nm/kg) also predicted success.^26,29

Discussion

This narrative systematic review synthesizes evidence from 22 studies (n = 844) demonstrating that foot orthoses significantly improve biomechanical parameters in children with flexible flatfoot, reinforcing their role in non-surgical management.^1,6 The studies demonstrated consistent biomechanical benefits of foot orthoses, with reductions in midfoot pressure (up to 48.5 kPa) and improvements in radiographic indices (e.g., 5° in talonavicular coverage). Predictive factors like low navicular height and high pain scores correlated with better outcomes. However, variability in orthotic types limited direct comparisons. These findings aligned with recent evidence on pediatric orthotics. ⁵⁰ To interpret the findings, it is essential to define the key biomechanical parameters. Plantar pressure (kPa) measures force distribution on the foot sole, often elevated in the midfoot in flatfoot, leading to pain; orthoses redistribute this to the lateral structures. ¹⁹ Joint moments (Nm/kg) represent rotational forces at joints, with excessive eversion in flatfoot causing misalignment; orthoses reduced these by 0.1–0.3 Nm/kg. ²⁰ Center of pressure (CoP, mm or mm/s) tracks force application point, unstable in flatfoot; orthoses stabilized it, reducing displacement by 5–10 mm. Gait parameters (cm or cm/s) include step length and speed, shortened in flatfoot due to inefficiency; orthoses improved these by 5–10 cm. Balance (cm² sway) assesses postural control, impaired in flatfoot; orthoses enhanced it via proprioceptive feedback. ²⁹ Muscle activity (mV or cm²) reflects recruitment, imbalanced in flatfoot; orthoses modulated it to strengthen inverters. Radiographic indices (degrees) evaluate alignment, abnormal in flatfoot (e.g., low calcaneal pitch); orthoses corrected them by 3–5°. ⁵¹

Biomechanical and neuromuscular mechanisms

Orthoses mechanically support the medial longitudinal arch, reducing pronation and redistributing plantar pressures. ³⁰ Modulation of the joint moments alters the load distribution, reducing stress on proximal joints. ²⁶ Improved CoP trajectories and balance suggest enhanced proprioceptive feedback, activating sensory afferents and optimizing motor control (e.g., reduced tibialis anterior activity by 15%, increased peroneus longus activity).^29,40,44 These neuromuscular adaptations may influence long-term arch development, but the precise sensory-motor pathways require further exploration.^52,53

Clinical implications

Clinicians should integrate biomechanical assessments, such as 3D gait analysis and pedobarography, into orthotic prescription protocols to tailor interventions based on foot morphology (e.g., navicular height <1 cm, arch index >0.26) and symptom severity (e.g., pain NRS >3). Regular follow-up (every 6 months) is recommended to monitor progress and adjust prescriptions as needed. Orthoses are recommended for severe or symptomatic cases to address biomechanical deficits, whereas observation or cost-effective devices may suffice for mild, asymptomatic cases to avoid overtreatment.^27,34 Regular follow-up (every 6 months) is critical to monitor progress. ⁴³ Activity-specific orthoses (e.g., rigid for high-impact sports) should be considered. Patient-reported outcomes (FFI, FADI) should complement the biomechanical data for holistic treatment planning.^52,54

Limitations

Heterogeneity in the study designs, orthotic types, and outcome measures precluded meta-analysis. Small sample sizes (12 studies, n<50), short follow-up (median: 48 weeks), lack of blinding (14 studies), and absent power calculations (13 studies) increase the bias risk. Although radiographic outcomes were reported in 10 studies, heterogeneity in the measurement methods (e.g., X-ray vs. fluoroscopy) and small sample sizes limited the certainty of evidence. Limited neuromuscular data (4 studies) restricts mechanistic insights.^32,33,40,44 Excluding gray literature ensured quality but may have missed unpublished data. The narrative synthesis limits the quantitative precision.

Future research

Large-scale RCTs with standardized measures (e.g., 3D gait analysis, EMG, pedobarography) and long-term follow-up (>12 months) are needed. Cost-effectiveness studies, patient-specific factors (e.g., BMI, activity levels), and novel technologies (e.g., 3D-printed orthoses, wearable sensors) should be explored. Neuromuscular studies could clarify the sensory-motor pathways.^24,55

Conclusion

Foot orthoses provided biomechanical benefits in pediatric flexible flatfoot by reducing pressure, stabilizing CoP, and improving alignment. Individualized prescription based on predictors like navicular height is recommended. High-quality RCTs are needed for standardization.

ORCID iDs

Mahsa Zangi https://orcid.org/0009-0008-0248-6434

Mobina Khosravi https://orcid.org/0000-0003-2947-0834