The Relationship Between Foot Posture, Validated Foot and Ankle Tests, and Hypermobility in Paediatric Population: A Cross‐Sectional Study

ABSTRACT

Background

Previous studies have investigated flattened foot arches associated with joint hypermobility using the Beighton scale. The absence of lower limb items below the knee in the Beighton scale questions the relevance of this relationship. Addressing this query, a new validated test has been used to assess joint hypermobility, the Foot and Ankle Flexibility Index (FAFI). Utilising the FAFI, the intention was then to relate it against known reliable morphofunctional foot and ankle tests, and also a validated paediatric physical activity test.

Methods

A total of 205 healthy children, aged 5 to 10 years, participated in this cross‐sectional study, which included the Lunge Test, Rest Calcaneal Stance Position (RCSP), Foot Posture Index (FPI) and 6 Minute Walking Test (6MWT).

Results

Significant correlations were found between FAFI and the test variables: RCSP (r = 0.334), Age (r = −0.254), FPI (r = 0.252), and 6MWT (r = −0.240). The multivariable linear regression of the hypermobility component according to FAFI presented an R ² value of 24.9%.

Conclusions

This study suggests that younger children and those with greater foot and ankle hypermobility have a more pronated foot stance, a greater range of ankle dorsiflexion, and a decreased walking speed/distance. Given the public health implications of unnecessary attention to many paediatric flatfeet presentations, these findings increase clinical clarity, using the new and validated FAFI.

Abbreviations

6MWT

6 Minute Walking Test

BMI

Body mass index

FAFI

Foot and Ankle Flexibility Index

FPI

Foot Posture Index

RCSP

Rest Calcaneal Stance Position

PFF

Paediatric flexible flatfoot

1. Introduction

Paediatric flatfoot is the most frequently seen condition in paediatric orthopaedic clinics [1, 2], cited as being 90% of clinic visits for foot problems concerning flatfeet [3]. The concern of parents relates to the future possibilities of disabling repercussions as children grow into their adult life [4]. Such concern continues in spite of the fact that the prevalence of flat feet is shown to decline with age, viz. flat feet at 2 to 6 years is 37%–59.7%, flat feet at 8 to 13 years 4%–19.1% [5, 6, 7].

Paediatric flexible flatfoot (PFF) is identified with a normal medial longitudinal arch when not weight‐bearing, and a flattened arch with weight‐bearing, often accompanied by hindfoot valgus and forefoot supination [8]. It is well documented that the flat foot is complex, involving multiple foot and ankle joints across all cardinal body planes. Hence, validated tools, such as the Foot Posture Index (FPI), which evaluates foot morphology in three dimensions, are valuable [9]. The FPI is obtained from the sum of six criterion scores (−2 to +2) for a scaled total value from −12 points (highly supinated) to +12 points (very pronated) [10].

Previous studies have concluded that bones and connective tissues (ligaments, tendons, joint capsules) are most important in maintaining the medial arch of the foot, with simultaneous neuromuscular control [11, 12, 13]. However, there is also evidence indicating that neither the intrinsic nor the extrinsic muscles maintain the arch of the foot [14]. Furthermore, in the study by Wenger and Leach [15], the morphology of the arch was attributed to weight‐bearing stress in the soft tissues.

The Beighton scale is a generalised joint hypermobility test, regarded as ‘positive’ with scores of six or more on a nine‐item scale [16]. Both Lin et al. [17], El et al. [2] have shown positive correlation between hypermobility and PFF, with more hypermobility children predisposed to flatter feet. However, Tsai v et al. [18] did not find a correlation between higher Beighton scores and flatter, but used footprints rather than foot posture for assessments.

The Beighton scale does not include any foot or ankle items, providing little specificity regarding the relationship between flexible flat feet and foot and ankle hypermobility. This problem, together with the lack of consensus for measuring foot and ankle function, continues to limit good quality studies [19].

The new Foot and Ankle Flexibility Index (FAFI) enables evaluation of foot and ankle hypermobility, with more specificity for foot and hypermobility [20].

Further tests, viz., the Lunge Test [21], Resting Calcaneal Stance Position (RCSP) [22], and the 6 Minute Walk Time (6MWT) [23] also display good reliability. The 6MWT has been previously explored regarding foot types. It was found that the type of foot, as assessed using the validated FPI, does not determine differences in walking ability in youths [24]. In contrast, a positive correlation has been found with the strength of the long flexor of the finger and physical capacity [25]. Further, it has been found that in subjects with generalised joint hypermobility, walking distance was decreased [26].

2. Aims and Objectives

The aim of this study was to examine our hypothesis that children with higher FPI scores will have a higher FAFI.

It is currently unknown whether the newly developed FAFI has an interpretable relationship with other robust and validated tests, for example RCSP, Lunge test.

Hence, the study objectives were to compare FAFI with the Lunge and RCSP tests, and also with the 6MWT, as a functional test to assess exercise tolerance and endurance as possible predictive factors.

3. Materials and Methods

3.1. Ethics Statement

This study fully complied with the Declaration of Helsinki regarding ethical principles for medical research involving human subjects and was approved by the Ethics Committee of San Antonio de Murcia Catholic University [ce112104].

3.2. Study Design

It is a cross‐sectional observational study, in which the Strengthening Reporting of Observational Studies in Epidemiology (STROBE) criteria will be followed.

3.3. Participants

The sample was the same as that recruited for the earlier FAFI validation investigation [20]. In this current study, other variables were explored in the same cohort, not to repeat, but to enlarge and contrast any results. A total of 205 children aged between 5 and 10 years participated in this cross‐sectional study. All measures occurred between January and June 2022. The participants were evaluated at the San Francisco de Asís school in Lorca, Murcia (Spain).

The inclusion criteria were age 5 to 10 years, no foot pain at the time of evaluation. Informed consent was obtained from parents or guardians after a detailed explanation of the purpose and study procedure. The exclusion criteria were the presence of congenital foot or ankle anomalies, cerebral palsy, foot or lower extremity surgery, and genetic, neurological, or muscular diseases.

3.4. Procedure

Prior to the evaluation of the participating children, demographic data was obtained for sex, age, weight, and height.

Children were barefoot, wearing comfortable sportswear. The children were instructed to relax whilst joint mobility measurements were painlessly performed. The two examiners were podiatrists with at least a year of experience with a paediatric gait screening program. All measures were performed on both feet of each participant.

Foot position was assessed using the FPI. The six FPI criteria were evaluated (palpation of the head of the talus, supramalleolar and inframalleolar curvature, frontal plane of the calcaneus position, talonavicular prominence, medial arch congruence, forefoot abduction/adduction), and rated on a scale of −2, −1, 0, 1, 2, and then summed for total score (−12, more supinated, to +12, more pronated) for each foot. The inter‐examiner reliability for the FPI in the paediatric population has reached a consistent weighted Kappa value (Kw = 0.86), in a sample of children aged 5 to 16 years [27].

Foot and ankle hypermobility was assessed with the FAFI, which has been validated in children. A cut‐off point of 4/6 was used for hypermobility positivity. This test has excellent intra‐rater (ICC = 0.96) and inter‐rater reliability (ICC = 0.89) [20].

The RCSP was standardly assessed [22], using the bisector of the calcaneus and the ground perpendicular. Intra‐rater reliability for RCSP in the paediatric population has shown weighted Kappa (Kw = 0.61 to 0.90) [28, 29].

The Lunge test was used to assess ankle dorsiflexion range of motion. The maximum angle of advancement of the tibia with respect to the vertical was measured using a digital inclinometer (Smart Tool) in contact with the anterior surface of the tibia. An earlier study demonstrated excellent inter‐examiner reliability (Kw = 0.97) [21].

The 6MWT is a reliable and valid functional test to assess exercise tolerance and endurance. The 6MWT was performed according to the standardised protocol [23]. The children performed this test wearing trainers. This test is validated in healthy children and has an intraclass coefficient (95% CI) of 0.94 (0.89–0.96) [30].

3.5. Statistical Analysis

All statistical analyses were performed using SPSS version 29 [IBM SPSS Statistics SPSS Inc., 2022]. An exploratory data analysis was performed that included descriptive statistics for age, gender, and BMI.

The normality of data distribution (Kolmogorov–Smirnov) and the homogeneity (Levene) in both samples directed the descriptive statistics to characterise the sample. A paired samples t‐test was used to compare FAFI scores between left and right sides.

A descriptive and frequency analysis of the variables was performed, and the means and standard deviation were determined. Relationships between continuous variables were explored with Pearson's correlation coefficient for normally distributed continuous data.

Finally, a linear regression model was obtained to evaluate the predictors of foot and ankle hypermobility as a continuous measure, according to the components of age, calcaneus position, ankle lunge range, foot type, and physical activity. The significance level was set at p < 0.001.

4. Results

A convenience sample of 196 children was included in the final analysis, as nine children were excluded due to pain during the tests. The average age of the participants was 7.6 (5 to 10) years; 60.2% were girls and 40.8% were boys. The mean body mass index (BMI) was 17.8 kg/cm².

Concerning laterality, there were no significant differences between the scores of the left and right sides for FAFI. The difference in means was 3.43 on the left and 3.44 on the right (p < 0.01). Thus, at random, the left limb of the 196 children was chosen for subsequent analyses.

Table 1 shows descriptive statistics for foot and ankle hypermobility, foot posture, ankle dorsiflexion range, and physical activity.

TABLE 1.

Descriptive statistics for the study sample, addressing foot and ankle hypermobility, foot posture (FPI, RCSP), ankle dorsiflexion lunge range, physical activity, and BMI.

N = 196	Mean	SD
FAFI	3.15	2.11
FPI	3.80	2.77
Lunge test (°)	50.32	6.59
RCSP (°)	4.42	2.54
6MWT (m)	463.39	55.52
Weight (kg)	28.90	8.48
Height (m)	1.26	0.10
BMI (kg/cm2)	17.79	3.24

Abbreviation: SD, Standard Deviation.

Bivariate analyses were performed to determine the relationships between hypermobility and foot posture components, ankle dorsiflexion, physical activity, and age.

Our results found a relationship between more pronated feet and greater foot and ankle hypermobility (r = 0.252; p < 0.001). We also found a relationship between foot and ankle hypermobility and RCSP, where more pronated feet showed greater hypermobility (r = 0.334; p < 0.001). Our results show a relationship between foot and ankle hypermobility and the Lunge test (r = 253; p < 0.001). The RCSP/FAFI correlation of 0.334 (p < 0.001) is notable (Table 2). Also, the relationship between BMI and the distances covered in the 6MWT and FAFI was calculated, obtaining statistically non‐significant data (Table 3).

TABLE 2.

Inter‐item correlation matrix for study variables found significant relationships between FAFI and FPI/RCSP (representing foot posture), FAFI and Ankle dorsiflexion lunge range.

N = 196		Age	FPI	Lunge test	RCSP	6MWT
FAFI	Pearson correlation	−0.254**	0.252**	0.253**	0.334**	−0.240**
	p	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001

^**Significance for each variable p < 0.001.

TABLE 3.

Inter‐item correlation matrix for study variables found a non‐significant relationship between BMI and 6MWT distances, BMI and FAFI.

	N = 196
		FAFI	6MWT
BMI	Pearson correlation	−0.119	−0.109
	p	0.097	0.129

Post hoc analysis yielded a power of 0.95 for this 5‐predictor model (Table 4). The multivariable linear regression of the hypermobility component according to FAFI presented an R ² value of 24.9%. There was no collinearity in the model (maximum variance inflation factor (VIF) 1.65 and minimum tolerance of 0.6) and the residuals were independent (Durbin‐Watson, 1.92), using the parameters age, lunge test, RCSP, FPI, and 6MWT. [Given laterality equivalence, at random the right side was used for post hoc analyses].

TABLE 4.

Multivariate linear regression for the prediction of hypermobility using FAFI against study parameters yielded R ² value (24.9%), directing further exploration toward other possibly relevant parameters.

				p	95% CI
	B	SE	b		Lower	Upper
FPI	0.061	0.066	0.075	0.358	−0.069	0.191
Lunge test	0.074	0.023	0.215	0.002	0.027	0.120
RCSP	0.260	0.072	0.292	0.000	0.118	0.403
6MWT	−0.007	0.003	−0.170	0.009	−0.012	−0.002
Age	−0.231	0.109	−0.148	0.036	−0.446	−0.015

5. Discussion

The main objective of this study was to compare the newly validated foot and ankle hypermobility tool (FAFI) with other known, commonly used, and robust tests viz. RCSP, Lunge test, and the 6MWT.

This study has also agreed with several previous investigations in finding a relationship between age and hypermobility [31]. Studies utilising the Beighton test as a measure of hypermobility found an inverse correlation with age [32, 33, 34], as have investigations using the LLAS test [35]. Thus, as children's age increases, general hypermobility of lower limbs decreases.

We found the same inversely proportional relationship between age and FAFI (r = −254; p < 0.01), such that increased age is associated with less foot and ankle hypermobility.

PFF is a common in children and a common concern for parents. Many studies illustrate that a flat arch is common in children at birth, with a normal arch developing during the first decade of life [36]. The PFF is multifactorial, with associated factors including ankle equinus, heel valgus [37] or joint hypermobility [17].

It is notable that methods to determine and categorise foot posture differ greatly between studies.

Investigations where the inclusion criteria were flattened foot arches and hypermobility was assessed using the Beighton score observed that the flattest feet were associated with the greatest hypermobility [2, 17]. However, Tsai et al. [18], using the Staheli Plantar Arch Index as an inclusion method for flat feet, found no relationship with hypermobility. The association between Beighton score and FPI has been established, finding that the most flat/pronated feet occurred with more generalised joint hypermobility [35, 38, 39, 40, 41]. In contrast, Hawke et al. [35] found no association between the FPI foot type and hypermobility in the lower limbs, using LLAS (which correlates with the Beighton score).

This investigation differs from most previous studies by using the validated FPI to assess foot posture instead of the subjectivity of arch type or indistinct soft tissue expansion with the Staheli Plantar Arch Index. The FPI gives importance to the three‐dimensionality of the foot, as opposed to footprint‐based measures. Similarly, the specificity of the FAFI test is unlike the Beighton test, which has predominance in the upper limb and no lower limb items below the knee.

Overall, our results found the following relationships for hypermobility:

• –more pronated feet and greater foot and ankle hypermobility in younger children, reducing with age
• –greater foot and ankle hypermobility with more everted/pronated RCSP
• –increased foot and ankle hypermobility with increased Lunge test range.

This last finding is in contrast with Hawke et al. [35] where the Lunge test was poorly associated with hypermobility, via both LLAS and Beighton score. It is pertinent to note that the LLAS is derived from the Beighton score, and that Carlos M. et al. [20] have now found a very strong relationship between FAFI and LLAS.

The 6MWT has emerged as a common submaximal test in clinical settings, which is simple and easy to use, where minimal coordination skills are required. The 6MWT has been studied in various childhood pathologies such as chronic respiratory diseases [42], cerebral palsy [43], equinus foot [44], neuromuscular diseases such as Charcot–Marie‐Tooth and Duchenne muscular dystrophy [45] for the evaluation of gait impairment. Our results found a relationship between hypermobility and 6MWT; therefore, it is necessary to include variables for assessing muscle fatigue to determine whether children with greater foot and ankle hypermobility may experience more fatigue and consequently walk shorter distances. Children with greater foot and ankle joint hypermobility walked shorter distances. Other possible confounding variables include how body weight may influence physical activity and foot and ankle hypermobility. For this reason, the study calculated these factors, obtaining data that suggested that the lower 6MWT values are not influenced by BMI. It was also found that there is no statistically significant relationship between foot and ankle hypermobility and BMI. Similarly, Marc et al. [26] found that young dancers with increased generalised joint hypermobility had reduced physical capacity as assessed with the 6MWT.

The results of the present study regarding prediction using FAFI as the dependent variable—with predictors FPI, Lunge Test, RCSP, 6MWT, and age—yielded a low R ² value (24.9%), suggesting the need for further exploration of other potentially relevant factors.

The strength of our study is that FAFI is a new and validated tool in children, and hence can now be used in preference to other hypermobility tests for the paediatric foot and ankle. This study has verified the relationship between hypermobility and foot type, giving rise to the new hypothesis that normal reduction of the paediatric flat foot may be due to reduced foot and ankle hypermobility.

Currently, a lot of (undue) attention is given to paediatric foot posture, and often yields unnecessary use of healthcare resources and the unjustified use of foot orthoses for paediatric flatfoot presentations, despite best evidence to the contrary [46]. We believe it is crucial to obtain validated and reliable tests and to continue their use in research, discarding outdated tests, which can mislead findings and misdirect public health practices.

The study also has limitations, being cross‐sectional, and therefore inappropriate to make assumptions about causal correlations between the factors analysed. The sample included predominantly Caucasian children who were asymptomatic and aged 5 to 10 years; hence, to date, the use of FAFI, whilst validated, has limited bounds of application. Also, the study has some limitations because of its nature of being a cross‐sectional study, which does not allow making assumptions about causal correlations between the analysed factors.

Our future research will compare FAFI with ankle inversion, eversion, dorsal flexion, and plantarflexion strength to relate ankle strength with 6MWT and hypermobility. Future research should explore the relationship between ankle strength and variables related to foot and ankle hypermobility, as well as physical activity, as already commenced [25]. It is considered important to conduct further research in other age groups to compare the cut‐off point. Additionally, studies should also be conducted in other types of populations to strengthen the generalisability of findings. Conducting cohort studies that follow individuals from childhood to adolescence would be an excellent way to understand the evolution of foot and ankle hypermobility. These longitudinal studies would allow us to observe how FAFI values change over time and their relationship with morphofunctional variables of the foot and ankle.

6. Conclusion

This study has found specific relationships between paediatric foot and ankle hypermobility with age, foot type, ankle range, and walking distance capacity.

The findings of this study suggest that children with greater foot and ankle hypermobility have a flatter/pronated foot posture, a greater range of ankle dorsiflexion, and a decreased physical ability to walk.

Our now hypothesized relationship between foot posture and hypermobility requires more research.

Ethics Statement

This study fully complied with the Declaration of Helsinki regarding ethical principles for medical research involving human subjects and was approved by the Ethics Committee of San Antonio de Murcia Catholic University [ce112104].

Consent

Informed consent was obtained from parents or guardians after a detailed explanation of the purpose and study procedure.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data may be available upon reasonable request; data forms part of ongoing works.