Crural and Plantar Fasciae Changes in Chronic Charcot Diabetic Foot: A Cross-Sectional Ultrasound Imaging Study—An Evidence of Fascial Continuity

Crural fascia (CF) and plantar fascia (PF) are biomechanically crucial in the gait and in the proprioception, particularly in the propulsion phase of the foot during the gait cycle and in the dissipation of forces during weight-bearing activities. Recent studies have revealed an association between increases in PF thickness and diabetes. The purpose of this study was to measure and compare by ultrasound (US) imaging the thickness of the CF and PF at different regions/levels in chronic Charcot diabetic foot patients (group 1) and in healthy volunteers (group 2). A cross-sectional study was performed using US imaging to measure the CF with Pirri et al.’s protocol and PF with a new protocol in a sample of 31 subjects (15 patients and 16 healthy participants). The findings for CF and PF revealed statistically significant differences in the poster region of CF (Post 1: group 1 vs. group 2: p = 0.03; Post 2: group 1 vs. group 2: p = 0.03) and in PF at two different levels (PF level 1: group 1 vs. group 2: p < 0.0001; PF level 2: group 1 vs. group 2: p < 0.0001). These findings suggest that chronic Charcot diabetic foot patients have CF and PF thicker compared to healthy volunteers. The US examination suggests that fascial thicknesses behavior in these patients points out altered fascial remodeling due to diabetes pathology and biomechanical changes.


Introduction
Diabetes mellitus (DM) is the "major metabolic epidemic" of the 21st century, and its prevalence continues to increase worldwide [1]. The International Diabetes Federation (IDF) reported that there are over 460 million adults in the world affected by this disease and that this number is destined to increase further [1]. Consequently, its complications increase in terms of prevalence with large growth in economic expenditure [1]. Among the latter, the diabetic foot is one of the main factors of morbidity and mortality associated with diabetes [2,3]. Indeed, it is estimated that 50% of hospitalizations related to diabetes are caused by consequential foot problems (infection, ulceration, osteomyelitis, etc.), halving survival at 1 and 5 years compared to other diabetic patients without the aforementioned alterations [4]. In general, the diabetic foot and its intra-and extra-hospital treatment are responsible for up to 20% of the economic expenditure for diabetes [5].
Despite having an important impact on the disability of diabetes patients, the mechanism of diabetic foot chronicity has not yet been understood in a complete way [6]. Factors associated with the pathogenesis of diabetic foot are complex and multifactorial but fundamentally involve the interaction of extrinsic biomechanical forces with intrinsic structural

Participants and Clinical Assessment
A total sample of 31 subjects was recruited and divided into two groups: "group 1" comprised 15 subjects with chronic Charcot diabetic foot; and "group 2" comprised 16 healthy subjects, from October 2018 and June 2021. Based on the following criteria, the inclusion criteria for group 1 participation consisted of some parameters: patients with a clinical and radiographic diagnosis of diabetes complicated by chronic phase Charcot neuro-osteo-arthropathy (Eichenholtz stage 3) at the level of the foot, monoliteral or bilateral, confirmed by upright radiography, evaluated by experienced orthopedic surgeon. The exclusion criteria for group 1 included age > 75 years old, Charcot diabetic foot operated to correct deformities, previous orthopedic surgery of the lower limb, active foot ulcers, rheumatic and connective tissue diseases, patients with pancreas transplant, whose antidiabetic therapy has, therefore, been suspended, neoplastic patients. The healthy normoglycemic participants were recruited among relatives of doctors of the department and the hospital staff. The exclusion criteria for group 2 encompassed individuals with a documented medical history involving lower extremities surgery, foot deformities, pain in the lower limbs, a history of fracture of the lower extremities, fibromyalgia, balance disorders, and systematic disease, such as rheumatological conditions and diabetes, among others.
The subjects underwent a US examination to evaluate the US thickness of PF and CF. The recruitment of participants was carried out by an orthopedic physician specializing in diabetic foot conditions possessing over a decade of experience in the field.

Ultrasound Examination Measurements
Utilizing a high-resolution device (Edge II, Sonosite, FUJIFILM, Inc. 21919, Lexington, WA, USA) equipped with a probe frequency range of 6-15 MHz and boasting a screen resolution of 1680 × 1050 pixels, US images were obtained at the foot and the leg regions/levels following a predefined US scanning procedure. The US assessments were performed by a physician who specialized in physical and rehabilitation medicine, possessing 7 years of experience in skeletal muscle US examination and US examination of fasciae. A standardized protocol was developed and employed to evaluate the PF bilaterally, while for the CF, a protocol previously published by Pirri et al. [33] was used, excluding the assessment of the anterior level 3, posterior level 3, and lateral levels. "The US system was set to a conventional speed of ultrasound (c = 1540 m/s) commonly used in diagnostic US systems, operating in B-mode and providing a depth of 30 mm; to ensure optimal scans and minimize surface pressure, the sonographer applied an appropriate amount of gel. The probe was positioned on the skin with light pressure to avoid tissue compression while maintaining stable contact for consistent imaging" [34][35][36][37]. The sonographer followed the same protocol to ensure consistent quantification of each point in the PF and CF. The US beam was maintained perpendicular to the PF and CF to mitigate the anisotropy that typically affected them. The power and overall gain of the US machine were adjusted to optimize visualization of the fascial layers and obtain high-quality scans. The resulting US images were frozen and captured.
The sonographer used the short axis for the leg, according to Pirri et al. [33], whereas the PF used the longitudinal axis because, in the two topographical regions, they are the best axis to visualize and follow landmarks correlated with the fascial layers' visualization imaging used by Pirri et al. [32]. A specific protocol for the PF was defined: PF: the patient was relaxed in the prone position with the foot hanging freely over the edge of the examination table, maintaining the foot perpendicular to the leg and toes pointing down. The US transducer was placed longitudinally over the center arch of the foot. The US examination was performed at two levels: (level 1) at the calcaneal insertion of the PF up to 2 cm from it; (level 2) in the middle third of the PF at 4-5 cm from the calcaneal insertion. For this purpose, the probe was moved in proximal-distal direction ( Figure 1). The scans were taken on the long axis, paying close attention to maintaining the same structure in the center of the US monitoring image and keeping the probe perpendicular. PF: the patient was relaxed in the prone position with the foot hanging freely over the edge of the examination table, maintaining the foot perpendicular to the leg and toes pointing down. The US transducer was placed longitudinally over the center arch of the foot. The US examination was performed at two levels: (level 1) at the calcaneal insertion of the PF up to 2 cm from it; (level 2) in the middle third of the PF at 4-5 cm from the calcaneal insertion. For this purpose, the probe was moved in proximal-distal direction ( Figure 1). The scans were taken on the long axis, paying close attention to maintaining the same structure in the center of the US monitoring image and keeping the probe perpendicular. At the conclusion of each assessment, all US images from every scan were saved and acquired. The measurement of fascial thickness was conducted using ImageJ analysis software (available online: https://imagej.nih.gov/iJ/, accessed in 5 March 2022). Each individual image was divided into three sections, and within each section, three points with the highest visibility were identified and measured. To mitigate the potential impact of thickness fluctuations, three equally spaced points were measured across the image, and the resultant values were averaged for further analysis. Moreover, the same procedure was repeated three different times to calculate the reliability of the measurements.

Statistical Analysis
Statistical analysis was performed using GraphPad PRISM 8.4.2. (GraphPad Software Inc., San Diego, CA, USA), and a p < 0.05 was always considered as the limit for statistical significance. The resulting effect size was calculated by G Power 3.1. (Universität Düsseldorf: Psychologie) and interpreted according to Cohen's kappa as small (d = 20), medium (d = 0.50), and large (d = 0.80) [47]. Based on a first pilot study, the sample size calculated for both CF and PF was 7 subjects for the group, as the effect size was, respectively, for CF thickness d = 2 and for PF thickness d = 3.6, with α err prob = 0.05 and power: 1-β err prob = 0.95. Nevertheless, we could include a sample of 31 subjects in our group, a minimum of 15 subjects for the group.
The normality assessment was carried out using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Descriptive and clinical statistics were calculated for both groups separately, including measures of central tendency and their dispersion ranges using mean At the conclusion of each assessment, all US images from every scan were saved and acquired. The measurement of fascial thickness was conducted using ImageJ analysis software (available online: https://imagej.nih.gov/iJ/, accessed on 5 March 2022). Each individual image was divided into three sections, and within each section, three points with the highest visibility were identified and measured. To mitigate the potential impact of thickness fluctuations, three equally spaced points were measured across the image, and the resultant values were averaged for further analysis. Moreover, the same procedure was repeated three different times to calculate the reliability of the measurements.

Statistical Analysis
Statistical analysis was performed using GraphPad PRISM 8.4.2. (GraphPad Software Inc., San Diego, CA, USA), and a p < 0.05 was always considered as the limit for statistical significance. The resulting effect size was calculated by G Power 3.1. (Universität Düsseldorf: Psychologie) and interpreted according to Cohen's kappa as small (d = 20), medium (d = 0.50), and large (d = 0.80) [47]. Based on a first pilot study, the sample size calculated for both CF and PF was 7 subjects for the group, as the effect size was, respectively, for CF thickness d = 2 and for PF thickness d = 3.6, with α err prob = 0.05 and power: 1-β err prob = 0.95. Nevertheless, we could include a sample of 31 subjects in our group, a minimum of 15 subjects for the group.
The normality assessment was carried out using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Descriptive and clinical statistics were calculated for both groups separately, including measures of central tendency and their dispersion ranges using mean and standard deviation (SD) to describe parametric data. Differences in US-estimated thickness of CF and PF across regions/levels were statistically analyzed by one-way analysis of variance (ANOVA) followed by Sidak's multiple comparison test.
Finally, a comparative analysis between the chronic Charcot diabetic foot patient's group and the healthy control group was performed using an unpaired Student's t-test. In addition, Pearson's test was employed for both groups to evaluate the correlation between the descriptive variables and US thicknesses.
Moreover, a two-way intra-class correlation coefficient (ICC 3,k) type C was used to assess the intra-rater reliability. The ICC values were interpreted as poor when below 0.5, moderate when between 0.5 and 0.75, good when between 0.75 and 0.90, and excellent when above 0.90 [48].

Results
A total of 31 subjects (17 females and 14 males) participated in this study. The descriptive data of the two groups are summarized in Table 1. In regards to the characteristics of the chronic diabetic foot patients (group 1), only 2 cases out of 15 were affected by Diabetes Mellitus 1 (DM1) (13%), and 7 cases out of 15 were insulin dependent. The average duration of diabetes from diagnosis was 18.33 ± 12.15 years (range 6-47 years), while the glycemic control in only 1 out of 15 cases was adequate (5.99%). Eight out of 15 (53%) cases were affected by bilateral chronic Charcot diabetic foot (of the remaining case, four were right and three left). Regarding blood pressure, only 2 out of 15 (13%) patients were normotensive, and the remaining 13 out of 15 (87%) patients were affected by arterial hypertension. Finally, 7 out of 15 patients (47%) were affected by diabetic retinopathy (DR), and among them, 3 (20%) were also affected by diabetic nephropathy (DN), which was not detected in the absence of retinopathy (Tables 2 and 3).   The clinical assessments regarding peripheral neuropathy, peripheral arteriopathy, and other risk factors (cigarette smoke and alcohol consumption) of group 1 are reported in Table 4.

Group 2 (Healthy Volunteers)
In the healthy volunteers, at anterior levels of the leg, the CF had, respectively, a mean US thickness of 0.72 ± 0.14 mm (Ant 1) and 0.76 ± 0.14 mm (Ant 2); while in the posterior region, CF had, respectively, a mean of US thickness of 0.97 ± 0.2 (Post 1) and 1.02 ± 0.30 (Post 2) (Table 8). Moreover, the US thickness of PF was, respectively, at level 1 (proximal) of 1.8 ± 0.57 while at level 2 (middle third) of 1.03 ± 0.42 mm (Table 8).

Ultrasound Measurements of Crural and Plantar Fasciae: Comparison between Group 1 and Group 2
According to Sidak's test, the comparisons between the different regions/levels of the CF and PF between group 1 and group 2 showed a statistically significant difference in the US thickness: Post 1 (group 1 vs. group 2: p = 0.03), Post 2 (group 1 vs. group 2: p = 0.03), PF level 1 (group 1 vs. group 2: p < 0.0001) and PF level 2 (group 1 vs. group 2: p < 0.0001) ( Table 9 and Figures 2-4).

Correlation Ultrasound Thicknesses and Years of Diabetes
Regarding Table 10, there was no detected statistically significant correlation betwe the duration of diabetes in years and US thicknesses of Ant 1, Ant 2, PL level 1, and le 2, whereas Post 1 and Post 2 showed both a statistically significant correlation with t

Correlation Ultrasound Thicknesses and Years of Diabetes
Regarding Table 10, there was no detected statistically significant correlation between the duration of diabetes in years and US thicknesses of Ant 1, Ant 2, PL level 1, and level 2, whereas Post 1 and Post 2 showed both a statistically significant correlation with the years of diabetes, respectively, for Post 1 (r = 0.3875, p = 0.0344) and for Post 2 (r = 0.5089, p = 0.0041) (Table 10 and Figure 5).

Correlation Ultrasound Thicknesses and HbA1c
Regarding Table 11, there was no detected statistically significant correlation between the duration of diabetes in years and US thicknesses of Ant 1, Ant 2, Post 1, Post 2, and PL level 1, whereas PL level 2 showed a statistically significant correlation with the HbA1c (r = −0.4115, p = 0.0239) (Table 11 and Figure 6).

Correlation Ultrasound Thicknesses and HbA1c
Regarding Table 11, there was no detected statistically significant correlation between the duration of diabetes in years and US thicknesses of Ant 1, Ant 2, Post 1, Post 2, and PL level 1, whereas PL level 2 showed a statistically significant correlation with the HbA1c (r = −0.4115, p = 0.0239) (Table 11 and Figure 6).

Correlation Ultrasound Thicknesses and Neuropathy Disability Score (NDS)
Regarding Table 12, there was no detected statistically significant correlation between the NDS and US thicknesses of Ant 1, Ant 2, Post1, PL level 1, and level 2, whereas Post 2 showed a statistically significant correlation with the NDS (r = 0.5779, p = 0.0008) (Table 12 and Figure 7).

Correlation Ultrasound Thicknesses and Neuropathy Disability Score (NDS)
Regarding Table 12, there was no detected statistically significant correlation between the NDS and US thicknesses of Ant 1, Ant 2, Post1, PL level 1, and level 2, whereas Post 2 showed a statistically significant correlation with the NDS (r = 0.5779, p = 0.0008) (Table 12 and Figure 7).

Intra-Rater Reliability
In addition, the intra-reliability was reported as good and excellent. The results for the CF and PF were as follows: Ant 1 (group 1:  (Table 13). Table 13. Intra-rater reliability of the ultrasound fascial thicknesses measurements within different regions/levels of group 1 and group 2.

Intra-Rater Reliability
In addition, the intra-reliability was reported as good and excellent. The results for the CF and PF were as follows: Ant 1 (group 1:  (Table 13). Table 13. Intra-rater reliability of the ultrasound fascial thicknesses measurements within different regions/levels of group 1 and group 2.

Discussion
Based on our current knowledge, this study may be stated as the first study detailing the CF and PF thicknesses in chronic Charcot diabetic foot patients compared with healthy volunteers. As has been reported by other studies examining PF, the PF was easily visu-alized in the longitudinal axis at the calcaneal insertion, appearing as a multilayer, linear, hyperechogenic layers below the subcutaneous tissue [49], while no study studied it at the level of the middle third of the sole of the foot. Moreover, for the first time, this study evaluated the CF in chronic Charcot diabetic foot patients.
The study's primary aim was to investigate the differences in CF and PF thicknesses at different regions/levels in chronic Charcot diabetic foot patients compared with healthy volunteers. An analysis of our results on the CF and PF thicknesses showed that in group 1, in the posterior region of the leg at Post 1 and Post 2 levels of the CF, the latter was thicker than in group 2, showing statistical differences (Post 1: group 1 vs. group 2: p = 0.03; Post 2: group 1 vs. group 2: p = 0.03) ( Table 9, Figures 2 and 3).
Moreover, an analysis of our results on the PF showed that in chronic Charcot diabetic foot patients (group 1), at two different levels, it was thicker than group 2 (PF level 1: Group 1 vs. Group 2: p < 0.0001; PF level 2: Group 1 vs. Group 2: p < 0.0001) ( Table 9 and Figures 2 and 4).
In light of these findings, the CF and PF tend to be thicker in chronic Charcot diabetic foot patients. They remodeled over time in response to repetitive stresses and diabetes pathology [25]. An increase in the CF thickness leads to a reduction in the ankle's range of motion (ROM) [33,49], limiting its mobility and altering the gait [50] and the load distribution on the foot [51]. Furthermore, the involvement of CF and PF in transmitting forces within the lower limb is crucial [29,33]. It is worth noting that these structures can easily undergo significant alterations in terms of their thickness, stiffness, and impaired movement. They tend to remodel themselves in debilitated tissue that has become dense and fibrotic due to the effects of AGEs' action [25]. These findings provided further confirmation, as supported by previous research [50,52], that changes in tissue, particularly in the fasciae [25], occur at an early stage in the progression of diabetes. Abate et al. [52] reported that in 51 patients with DM2, diagnosed less than a year prior, compared to 18 healthy volunteers, early fascial tissue changes with microvascular complications. Giacomazzi et al. [50] demonstrated, in a population similar to that of our study, how the PF thickness at calcaneal insertion increases concurrently with the degree of impairment of the nervous structures of the foot. In addition, all of the patients in group 1 showed values of NDS and NSS scores consistent with the diagnosis of neuropathy. A total of 30.4% of chronic Charcot diabetic foot patients showed vasculopathy. Only 13% of patients were affected by DM1, whereas 47% of all patients required insulin therapy. These data are in line with the published data about this type of diabetic foot [8][9][10][11]. Additionally, Fede et al. [26] demonstrated that in females, "the fascia becomes enriched in collagen-I, with low hormone levels, becoming more rigid during menopause". According to this evidence, the greater number of women in menopause in the two groups studied could cause further fascial remodeling.
Furthermore, the correlation between the years of diabetes and CF US thickness of the poster region of the leg, respectively, for Post 1 (r = 0.3875, p = 0.0344), for Post 2 (r = 0.5089, p = 0.0041) ( Table 10 and Figure 3), and between Post 2 and NDS (r = 0.5779, p = 0.0008), could be explained by the fact that the proximal progression of diabetes leads to involvement of CF and the latter becomes densified/fibrotic, consequently increasing its thickness [25] and altering their proprioception, with fascia richly innervated [53]. These observations could be confirmed surgically by the effectiveness of release intervention at the level of the myotendinous junction of the medial gastrocnemius [54], which could work on two fronts: (1) to reduce the tension on the Achilles tendon; (2) to hold CF, not foreseeing the surgical incision of the latter. The results have also confirmed, as has been demonstrated by other previous studies [22,23], that PF US thickness has increased in diabetic patients at calcaneal insertion; while no study studied it at the level of the middle third of the sole of the foot, this study for the first time demonstrated that also at this level there is an increase in the PF thickness, confirming that diabetes affects the whole plantar fascia and fasciae [25]. Moreover, the negative correlation between HbA1c and PF level 2 (middle third of the plantar surface) (r = −0.4115, p = 0.0239) could be explained by the fact that HbA1c is a punctual estimate of the glycaemic trend over a limited period of time, while, conversely, the PF thickness provides a more extensive representation of the progress of the disease, as the fascia presents a degenerative process lasting for years, resulting in more stable than glycaemic control. The collapse of the plantar arch typical of chronic Charcot diabetic foot could lead to a distribution of the load, such as compromise of the plantar fascia, leading to progressive thinning [25]. US examination could be revealed as a crucial tool to follow up with the patient and to intercept and prevent the progressive changes of diabetes, being portable and economical imaging. The outcomes have affirmed, mirroring previous investigations, that there exists a dependable and commendable level of intra-rater reliability in the US assessment when evaluating the deep fascia. This is particularly true for sonographers who possess optimal technical expertise in US assessment and a profound understanding of fascial anatomy [34].
This work represents the initial investigation that we are aware of, aiming to analyze and compare the thickness of the CF and PF in various regions/levels using US imaging in individuals with chronic Charcot diabetic foot conditions and compare them with those of healthy volunteers. In the future, extensive longitudinal studies involving a substantial number of patients will contribute significantly to our understanding of the underlying mechanisms behind diverse thickness patterns. Furthermore, US examination has the potential to reveal early changes in the fascia that cannot be detected during regular clinical examinations. Ultimately, defining CF and PF thickness in different regions/levels among these patients would enable a more precise and targeted approach to treatments and therapies. The reduction in tensions generated by proximal alterations to the foot could lead to indirect benefits also distally, with potential improvement in the biomechanics of gait and reduction in pressure in non-physiological load points. All that could reduce the risk of the most dramatic diabetic foot complication, ulceration [25].

Limitation of Study
The limited power of the study makes it impossible to statistically analyze the prevalence of the US findings and explain their possible causes, prognostic significance, and therapeutic implications. Additionally, the US examination of CF and PF morphology heavily relies on the skill of the sonographer and the proper setting of the US device. Furthermore, the non-differentiation by sex and the non-blinding do not allow for generalizing the results; a large blinded study would be necessary to better contribute to our knowledge of the pathophysiology of different thickness patterns.

Conclusions
The US permits an optimal visualization of the fascial layers in patients with chronic Charcot diabetic foot patients, opening the road for a more in-depth comprehension of fascial changes in chronic Charcot diabetic foot. In addition, it may reveal changes, not only in plantar fascia but also in crural fascia, not highlighted by normal clinical examination. Some of these changes still need to be investigated further as they have not been fully described yet. In summary, the findings of the study confirmed that in patients with chronic Charcot diabetic foot, the PF is thicker at both its insertion point in the calcaneus and its middle third. Additionally, the CF was found to be thicker in posterior regions/levels compared to healthy volunteers. The observed thickness patterns of CF and PF in these patients suggested abnormal remodeling of the fascia due to the presence of diabetes and biomechanical alterations.
Funding: This research received no external funding.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki and approved by Ethical Committee for clinical trials in the province of Padova (approval no. 3513/AO/15, study approved on 28 January 2016).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available upon request from the corresponding author. The data are not publicly available due to privacy.