The study was conducted at the Swiss Children's Rehab in Affoltern am Albis, University Children's Hospital Zurich, Switzerland, in accordance with the Declaration of Helsinki and legal and institutional standards. Ethical approval from the cantonal ethics committee was obtained (BASEC 2020-03037). The participants and their parents gave oral consent before applying the short leg cast. The inclusion criteria for the participants were good health, normal child development, and age between 7 and 16 years. Exclusion criteria were open lesions or wounds, the usage of medication (analgesics, sedation), incontinence, orthotics or reduced mobility, abnormal body mass index (BMI, underweight (BMI < 18.5) or overweight (BMI > 25.0)) and loss of sensation or paresis.

Before applying the short leg cast, the children were examined for peripheral sensory deficits. The CMS (circulation, motor, sensory) testing of the legs was carried out before the cast application and routinely performed during the measurements. In addition, we recorded personal data (age, weight, height, sex) and information about regular physical activity (Table 1). We measured vital parameters once before cast application (blood pressure, body temperature, oxygen saturation level, and pulse (Table 1). Furthermore, a physician inspected the child's right leg to examine and document lesions, wounds, or swelling. Pictures were taken of the whole leg, the ankle, the heel, the midfoot, and the sole. The circumference of the lower leg (10 cm below the caput fibulae), the upper ankle joint, and the midfoot were measured before and directly after the cast application to quantify potential swelling or compression. During the measurement, the child was asked to convey comfort level using a visual analog scale (VAS) (34). VAS 0 indicates no pain or discomfort. VAS 10 indicates maximum pain or discomfort. The VAS score was immediately collected after the cast application and each change in body posture. We selected the VAS score as it is a simple and reliable method of assessing pain in children as young as three years (35). A parent was present throughout the measurement. The cast could be removed at any point in case of pain or discomfort.

The compression force distribution inside the cast was recorded using the pressure mapping sensor 9,801 (Tekscan Inc., South Boston, MA, USA). Compression data were recorded at a 50 Hz sampling frequency for 30 s for each body posture. The pressure mapping sensor consists of six sensor strands, each including 16 pressure-sensitive areas (sensels) covering a pressure range of 34 kPa. The maximum values for global compression force includes the data from all 96 sensels inside the short leg cast. To enable an optimal alignment of the sensors covering the body sites prone to PI development, the sensor was cut along the designated lines. The individual sensor strands were placed over anatomical risk areas for developing PIs, including the medial and lateral parts of the heel, the lateral and medial instep areas and the medial and lateral malleolus (Figure 1). The maximum local compression force at the instep, the malleolus lateralis, and the heel area were calculated based on seven sensels covering the corresponding location (Figure 2).

The maximum local compression force at the instep area inside the short leg cast was compared to relative changes in skin physiology parameters. Data were collected at the lateral instep before the application and immediately after removing the short leg cast. Erythema and changes thereof were measured based on light absorption properties of the skin (Mexameter® MX18, Courage + Khazaka electronic GmbH, Cologne, Germany). Elastic properties of the intact skin such as skin firmness, skin elasticity, and skin resilience and changes thereof were quantified based on measurements of the deformation characteristics of the epidermis and stratum corneum exposed to vacuum (Cutometer®MPA580, Courage + Khazaka electronic GmbH, Cologne, Germany). Changes in skin physiology parameters were plotted against the maximum compression observed at the lateral instep.

For statistical analysis, IBM SPSS Statistics for Windows, Version 28.0 (IBM Corp, Armonk, NY, USA) and for data visualization, GraphPad Prism Version 9.4.1 (458), was used (GraphPad Software, LLC, San Diego, CA, USA). For all statistical analyses, α was set at 0.05.

The Shapiro-Wilk test revealed a non-normal distribution for most of the data. Therefore, data are presented as median [interquartile range (IQR)], and non-parametric tests were conducted for the statistical analyses. Differences in global and local compression forces between the three body postures (partial/ full body weight, transfer) were assessed by the Friedmann test, followed by pairwise Wilcoxon tests with a Bonferroni correction to adjust for multiple testing. We quantified the associations between the global compression forces and local compression forces and between local compression forces and changes in skin physiology parameters using Spearman's rank correlations (ρ). We interpreted the magnitude of the Spearman correlations according to Mukaka (2012) (36): 0–0.29 negligible, 0.30–0.49 low, 0.50–0.69, moderate, 0.70–0.89 high, and ≥0.90 very high.

The global maximum compression forces (Figure 3) differed significantly between partial and full body weight conditions and during transfer (Friedmann, p = 0.007). The global compression forces were significantly lower in body postures with partial body weight [9.7 kPa (IQR 3.5–12.8 kPa)] compared to the transfer condition [16.5 kPa (IQR 12.0–16.9 kPa); p = 0.04]. Body postures with full body weight resulted in significantly higher compression forces [55.4 kPa (IQR 47.2–59.6)] compared to partial body weight (p = 0.04) and transfer (p = 0.04).

The local maximum compression forces at the heel (Figure 4, Friedmann, p = 0.008) and at the instep area (Figure 5, Friedmann, p = 0.015) differed significantly between conditions of partial body weight, transfer, and full body weight. No changes in the local maximum compression forces at the malleolus lateralis (Figure 6, Friedmann, p = 0.245) could be measured between partial body weight, transfer, and full body weight conditions. The lowest maximum compression force was observed for partial body weight loading (0.33 kPa [IQR 0.29–1.13 kPa] for the heel and for the lateral instep [0.24 kPa (IQR 0.22–0.38)]. The highest maximum compression force was 27.3 kPa [IQR 17.5–27.6 kPa] for the heel and 2.10 kPa [IQR 0.98–2.25] for the lateral instep.

The global maximum compression forces inside the short leg cast were compared to the maximum local compression forces measured at body sites prone to PI development (the heel, the lateral instep, and malleolus lateralis) during the three body postures (partial/ full body weight loading, and transfer). The global and local maximum compression forces revealed a high correlation for the lateral heel (ρ = 0.82, p < 0.001) and a moderate correlation at the lateral instep area (ρ = 0.51, p = 0.04, Figure 7). We found a low correlation between the global compression forces and the local maximum compression forces at the malleolus lateralis showing a trend towards a statistical significance (ρ = 0.448, p = 0.055).

Regarding skin physiology parameters, data was collected at the lateral instep before the application and immediately after removing the short leg cast. The lateral instep was chosen because the local maximum compression force correlated statistically significantly with the global maximum compression force (Figure 7B).

Relative changes in erythema development (Figure 8A) and in skin firmness (Figure 8B), skin elasticity (Figure 8C), and skin resilience (Figure 8D) were calculated and plotted against the maximum compression forces measured at the lateral instep. Neither the relative change in skin erythema (Figure 8A, Spearman ρ = 0.5, p = 0.391), skin firmness (Figure 8B, Spearman ρ = −0.1, p = 0.873), skin elasticity (Figure 8C, Spearman ρ = 0.1, p = 0.873) nor skin resilience (Figure 8D, Spearman ρ = −0.3, p = 0.624) revealed a statistically significant correlation with maximum compression force at the lateral instep.

Our study measured the global maximum compression force within a short leg cast across various body positions with increased loading. To our knowledge, no previous attempts have been made to conduct experiments involving variable loading within a short leg cast equipped with pressure sensors in a pediatric sample. Our results demonstrated a significant difference in the global maximum compression force between conditions involving partial and full body weight, and during transfer. Notably, we could objectively quantify the increase in compression force inside a short leg cast in a pediatric sample with higher leg loading.

As PIs typically occur over bony prominences, we further analyzed the impact of different body positions with increased loading on the local compression force over anatomical risk areas.

We could show a significant difference in local compression force at the instep area and heel with different body weight loading. Our findings align with a study investigating the main weight-bearing area within a conventional patella tendon-bearing cast based on a pressure-sensitive insole using the Tekscan F-SCAN system. Under full loading conditions, the plantar region, including the heel area, was identified to be the main weight-bearing area (44).

Furthermore, the anterodistal lower leg was identified as the highest total contact cast wall load during full body weight loading (45), which is congruent to our findings in a short leg cast. In contrast, we could not observe any significant changes in local compression force during different body postures at the malleolus lateralis. We hypothesize that the malleolus lateralis experienced a greater exposure to shear forces than compression force.

A significant factor in developing PIs is localized ischemia due to obstruction or occlusion of blood vessels in soft tissues by external pressure (46–50), potentially caused by the application of casts. The critical threshold for obstructing blood flow and inducing ischemia, which has the potential to lead to the development of PIs, was determined at 16 kPa (51). The heel exhibited the highest compression force, reaching up to 27.3 kPa, thus significantly surpassing the critical threshold for obstructing blood flow and potentially causing ischemia. In the case of the digital augmented total contact cast, the maximum peak pressure reached an even higher level, measuring at 128.5 kPa at the rearfoot during walking (33). In a digitally augmented short leg cast using Tekscan FlexiForce A201 sensors, the mean peak force at the plantar surface of the calcaneal tuberosity was measured at 31.1N during full loading (52).

Additionally, we investigated the association between global compression forces and local compression forces measured at anatomical areas prone to PI development across three predefined body postures with full or partial body weight loading and the transfer in between. Our findings revealed a robust and statistically significant correlation between the global and local maximum compression forces at the lateral heel and the lateral instep area. This finding is particularly significant because measuring the global compression force within a short leg cast is technically more straightforward than measuring the local compression force over anatomical areas susceptible to PIs. It can be challenging to accurately position pressure sensors within a cast, specifically over the relevant anatomical area, as they are susceptible to displacement. Placing the pressure sensors directly onto the skin may present issues, as the sensors themselves can contribute to developing PIs.

If the pressure inside the short leg cast reaches a critical level, a potential increase in the local compression force, posing a risk of causing ischemia, can be assumed. It is essential to report that the healthy children involved in this study did not experience any pain or discomfort, even during high pressure that obstructed blood flow. This could indicate that children may be unable to perceive or recognize critically high compression values. On the other hand, the cast was worn for approximately one hour, which may not have been sufficient to cause significant discomfort. Nevertheless, identifying the potential risk of developing PIs would be beneficial to take early preventive actions and avoid any potential side effects. Finally, comparing pressure measurements with other studies is difficult, as they all report different units of measurement, cast techniques, and digital sensors.

Multiple studies have demonstrated that relying solely on pressure measurements at the skin surface is inadequate for preventing PIs, particularly when the PI occurs deep in the tissue, causing subcutaneous damage beneath intact skin (53–55). The skin microclimate, encompassing factors such as temperature, humidity, compression, and shear forces, significantly promotes the development of PIs (56). Changes in the functional skin barrier occur before any visible alterations become clinically apparent (57). Prolonged loading (58) and reduced air convection contribute to the local accumulation of heat and humidity (59, 60). Unloading triggers hyperemia, leading to an elevation in skin temperature (61). Elevated skin temperature can weaken the cohesive strength of the stratum corneum (62) and compromise skin integrity (60).

Therefore, we compared how local compression forces affect changes in skin physiology parameters, including erythema development, skin firmness, skin elasticity, and skin resilience. The data was collected at the lateral instep before applying the short leg cast and immediately after removal. We selected the lateral instep because there was a statistically significant correlation between the local maximum compression force and the global maximum compression force. However, no statistically significant correlation was found between the relative change in skin erythema, skin firmness, skin elasticity, or skin resilience and the maximum compression force at the lateral instep. We suspect this lack of correlation may be attributed to the short time spent in the predefined body position and the overall brief period in which the short leg cast was worn. In a study by Kottner et al., increased erythema indices were observed immediately after loading and 15 min after offloading in a population of healthy adults immobilized in a supine position for 90 and 150 min (58). PIs in subdermal tissues beneath bony prominence typically occur between the first hour and 4 to 6 h after loading in a supine position. Additionally, they proposed that PIs occur sooner in a sitting position due to the greater loading compared to a supine position (63).

It is important to emphasize that our study was a short-term investigation and does not reflect the duration of cast wear in clinical practice, where children are often immobilized in casts for weeks. Therefore, we consider the maximum duration of 104 min cast wearing a limitation.

Additional studies involving longer periods of immobilization and additional sensors to measure compression, shear forces, temperature, and humidity could clarify the risk factors associated with critical conditions. To achieve this, it would be desirable to have a multi-parameter sensing system that can be directly integrated into the cast without compromising its stabilizing properties or increasing the risk of developing PI.

In conclusion, integrating sensors into casts holds great potential for continuously monitoring skin exposure and early recognition of critical conditions in developing PIs. As a result, by combining the accumulation of digital information with individual skin characteristics, it may be possible to create a model that can predict the individual risk of developing PIs.