Participants

The study used a within-participant, repeated measures design with data collected across multiple sites of Northern Health in Melbourne, Australia. Ethical approval was granted by Austin Health’s Human Research Ethics Committee (HREC/17/Austin/189) with subsequent approval from the College of Science, Health and Engineering Human Ethics Sub-Committee at La Trobe University. Written informed consent was obtained from all participants prior to beginning data collection. A sample size of 16 participants was pre-specified, which provided an 80% likelihood of detecting a clinically important difference between interventions of 100 kPa in peak plantar pressure (standard deviation of 100 kPa and alpha level set at 0.05) [22].

Recruitment took place via direct approach to patients from acute and subacute high-risk foot services within Northern Health. This included a search of electronic patient files, with a preliminary screen to meet selection criteria. A face-to-face or phone contact was then completed with the patient to discuss involvement in the study. Participants were included if they were over 18 years of age, capable of providing informed consent, had a diagnosis of diabetes mellitus, and had an active, plantar neuropathic foot ulcer for at least four weeks.

Wounds were assessed based on the International Working Group on the Diabetic Foot (IWGDF) PEDIS Ulceration Classification criteria [23] and were included if they met: Perfusion Grade 1–2; Depth Grade 1–2; Infection Grade 1; and Sensation Grade 2 (Table 1).

Table 1 Ulcer inclusion criteria based on the IWGDF PEDIS diabetes foot ulceration classification criteria [23]

Participants were deemed ineligible if they: were unable to walk 10 m without the use of a walking aid, had critical limb ischemia as defined by a toe pressure less than 30 mm Hg, an ulcer probing to bone, or infection as marked by the cardinal signs of redness, heat, swelling, pain and purulent exudate [24]. Individuals who did not speak and understand English and who were unable to communicate without an interpreter were excluded to ensure informed consent and adequate communication during the study.

Participant baseline characteristics including age, sex, diabetes duration and type, and amputation history were recorded from electronic records. Amputation history was included as it has been shown that patients with a history of partial foot amputations have high plantar pressures than those without amputations due to biomechanical compensation [25]. DRFU dimensions were recorded on the day of data collection with a sterile ruler (Puritan DM Stick®, Guilford, ME). Body mass index (BMI) was calculated on the day of data collection using an electronic calculator (National Heart Foundation of Australia© BMI calculator), which divides body weight in kilograms by height in metres squared. In the event baseline characteristic details were not available in the participant’s electronic health records, they were confirmed with the participant or local general practitioner.

Offloading conditions

All participants were measured in four conditions, one of which was a control. Selection of these conditions was based on previous research [12]. The conditions included the following (Fig. 1):

(i)

Postoperative footwear with the lining removed (the control),

(ii)

RCW alone,

(iii)

RCW with 20 mm of felt padding adhered to an orthosis (RCW with felt to orthosis),

(iv)

RCW with 20 mm of felt padding adhered to the foot (RCW with felt to foot).

Fig. 1

Conditions assessed in the study: a Postoperative footwear with the lining removed (the control), b RCW alone, c and d RCW with 20 mm of felt padding adhered to an orthosis (RCW with felt to orthosis), e–g RCW with 20 mm of felt padding adhered to the foot (RCW with felt to foot)

For the control condition, participants were measured in a post-operative boot (APB™ All Purpose Boot, DARCO, Huntington, WV) on the limb with ulceration present. This footwear was selected as its design accommodates varying degrees of anatomical deformity, and with the lining removed, it was considered to have limited influence on plantar pressures. For the remaining offloading conditions a RCW was used (ProCare® XcelTrax Standard Tall Walking Brace, DJO Global, Vista, CA). The RCW was only worn on the limb with active ulceration; all participants wore their regular footwear on the contralateral limb for all conditions.

The orthosis used in the study was a prefabricated full length, firm (180 kg/m3) closed cell polyethylene foam device (Foot Science International, Christchurch, New Zealand), which contours the plantar surface of the foot, including the arch and heel. While the insoles can be heat-moulded, they were not for this study. When used in combination with the RCW, the orthosis was placed inside the liner of the RCW.

The felt padding used for the study was made of semi-compressed felt with an adhesive backing (Aetna Felt Corporation, Allentown, Pennsylvania, USA). All felt padding was 20 mm thick with an aperture at the ulcer site where appropriate. The borders of the aperture site under the ulcer were as close as possible to the margins of the ulcer. The edges of the felt were bevelled to avoid unintentional high pressure areas at the margin of the aperture. For the RCW with felt adhered to an orthosis condition, the felt pad dimensions depended on where the DRFU was located. For a hallux or forefoot ulcer, the proximal border of the pad was 1–2 cm distal to the metatarsal bases and the distal border was the distal margin of the orthosis. For a midfoot or rearfoot ulcer, the proximal border of the pad was the anterior margin of the heel and the distal border was the distal margin of the orthosis (an aperture was used for midfoot ulcers, but not for rearfoot ulcers as there was no felt directly under the heel). The orthosis with the felt pad applied was positioned inside the RCW before it was applied to the limb. For the RCW with felt adhered to the foot condition, the proximal border of the pad was the anterior margin of the heel and the distal border was the webspaces of the toes. When necessary, the pad was secured to the foot with hypoallergenic tape (Hypafix, Smith & Nephew, Andover, MA) to ensure the felt did not shift during data collection. The RCW was applied to the limb after the felt pad was adhered to the foot. Foot ulcers were covered with a thin gauze, sterile protective dressing.

Randomisation and blinding

To minimise ordering effects, the order that the offloading conditions were measured was randomised. Participants and investigators were not blinded to the offloading conditions, however the pressure measurements obtained with the pedar-X® were objective, reducing the risk of bias.

Plantar pressure measurement equipment

Plantar pressure data were collected with an in-shoe measurement system, the pedar-X® (Novel, Munich, Germany), which has been found to be valid and reliable [26,27,28,29]. The pedar-X® insoles are 2 mm thick, constructed with 99 capacitive sensors arranged in a grid formation and record peak plantar pressures in kilopascal (kPa) units. The pedar-X® insoles were calibrated using the trublu® calibration system as per manufacturer guidelines (Novel, Munich, Germany). The sampling frequency of the system was 50 Hertz. All plantar pressure measures were obtained in accordance with manufacturer guidelines (Novel, Munich, Germany).

Measurement procedure

Insoles of the pedar-X® system were placed in each offloading intervention as close to the foot as the intervention would allow to standardise the positioning of the pedar-X® insoles and obtain the most accurate plantar pressure measurement. However, due to the nature of the 20 mm felt padding adhered to the foot condition, the pedar-X® insoles were unable to be placed directly against the foot. Instead, they were placed between the felt and RCW, which was based on previously published protocols [20, 22]. Participants were granted five minutes to walk and acclimatise to the device with the pedar-X® in place prior to beginning data collection. Before the initial trial for each condition, pedar-X® insoles were zeroed as per manufacturer guidelines (Novel, Munich, Germany). Participants were then measured whilst walking at a comfortable pace along a pre-measured, flat 10 m walkway. To control for variation in walking speed, each trial was timed and if not within ± 5% of the first trial recorded it was not included, and the participant was required to complete the trial again. To limit the effects of acceleration and deceleration, only the middle three steps of each trial were analysed.

Outcome measures

The primary outcome measure was peak plantar pressure (kPa), which was assessed at the active DRFU site of each participant. This outcome was selected as peak plantar pressure has been shown to be predictive of tissue trauma, ulcer formation, delayed healing and ulcer recurrence [30, 31]. Force–time integral and pressure–time integral were not analysed as they have high interdependency with peak pressure [32, 33].

The secondary outcome measures of contact area (cm2) and contact time (ms−1) were also included. Contact area provided data on whether plantar pressure changes were linked to change in the plantar contact area with the offloading devices tested [20]. Contact time was measured to cross-check consistency of walking speed during trials [20].

Data processing and analysis

Raw plantar pressure data obtained with the pedar-X® system was processed using the Novel multi-mask package (Novel, Munich, Germany). This enabled a single mask to be placed over a specific region of interest for each participant (i.e. the DRFU site). This mask subsequently provided an area of comparison for the four offloading conditions. The secondary outcomes measures of contact area and contact time were processed using a whole foot mask. All statistical analyses were undertaken using the Statistical Package for the Social Sciences (SPSS) Version 23 (IBM Corp, Somers, NY, USA).

Normality of data was confirmed by assessing histograms, skewness and kurtosis values, and Shapiro-Wilks statistical tests [34]. Four outliers were identified, all in peak plantar pressure. Outliers were addressed by replacement with the next non-outlier value. Between-group differences for peak pressure were expressed as mean differences with 95% confidence intervals, and to provide easier comparison, as percentage differences. One-way repeated-measures ANOVAs were used to explore the effects of all offloading conditions on peak plantar pressures (ulcer), contact area (whole foot) and contact time (whole foot). Mauchly’s test was assessed to address the assumption of sphericity, and where it was violated, the Greenhouse–Geisser correction was used for the F-statistic, degrees of freedom and the p-value. Significant ANOVA findings were explored using Bonferroni-adjusted pairwise comparisons. The threshold for statistical significance was set at 0.05.