Study setting {9}
The study will be conducted at the Anandaban Hospital currently run under The Leprosy Mission Nepal. The Anandaban hospital lies in southern part of Lalitpur district of Nepal, approximately 20km south of Kathmandu in the Bagmati province. Established in 1957 by The Leprosy Mission England & Wales, it is the tertiary leprosy referral hospital in Nepal, with 110 beds and provides specialist tertiary leprosy care for approximately 6,000 patient visits annually from all over Nepal and the Northern part of India. It also provides general medical care for the local population.
Eligibility criteria {10}
The inclusion criteria are as follows:
- Patients with a chronic foot ulcer of at least 6 weeks duration due to leprosy neuropathy
- ≥18 years of age.
- Ulcer surface area between 2 and 20 cm2
- Ulcer is clean, dry, and free from infection.
- Patient can provide informed consent.
The exclusion criteria are as follows:
- Any significant medical condition, laboratory abnormality, or psychiatric illness that would prevent the participants from participating in the study (e.g. diabetes or diabetic ulcer, HIV, chronic Hep B, chronic Hep C or TB patients under active treatment).
- Ulcer with surface area <2cm2 and >20cm2.
- Untreated high blood pressure > 150 mm HG systolic
- Haemoglobin less than 9 gm/dL or platelets < 100x103/ul
- Patient requires skin graft.
- Pregnant or breast-feeding.
- Patients with Erythema Nodosum Leprosum (ENL) or a leprosy reaction under steroid treatment.
- Any wound that has clinical microbial infections.
- A patient who has returned to the hospital having previously been a participant in the trial.
Who will take informed consent {26a}
A local research fellow trained in Good Clinical Practice (GCP) will screen all admissions for eligibility. Eligible patients will be provided with a Patient Information Leaflet and verbal information as necessary about the study from a research fellow in local languages. Written informed consent will be sought the following day, once eligibility has been confirmed. Thumb or fingerprints will be requested in lieu of a signature if necessary. Translated consent forms have been back-translated according to the WHO methodology (15) for quality assurance purposes.
Additional consent provisions for collection and use of participant data and biological specimens {26b}
Consent includes the option to give permission for the data collected to be used in future research. Participants can reject this option and still take part in the TABLE trial.
Intervention
Intervention description {11a}
The intervention uses the participant’s own blood to prepare strips of leukocyte and platelet-rich fibrin matrix (L-PRF) to be applied to the ulcer bed. The process is described in detail elsewhere (16) but briefly, up to 80ml of participants’ blood will be collected twice per week at the time of dressing change. The blood will be centrifuged to obtain the fibrin matrix gel, which will be compressed and applied to the ulcer bed and then covered with a Vaseline gauge dressing. Blood collection from participants, centrifugation procedure and application of L-PRF will be done in a minor operating room following aseptic technique. All participants (including those in the control group) will be given iron and folic acid tablets. Both control and intervention groups will receive routine twice-weekly dressing changes by trained nurses or paramedics until their ulcers are healed (complete re-epithelization) up to a maximum of 70 days. Any missed dressing change sessions will be noted but not treated as a deviation from the protocol.
In the event that a participant has more than one ulcer, the largest ulcer will be selected as the index ulcer for analysis purposes before randomisation. However, all of the participant’s ulcers will receive the same treatment. Thus, if the participant is in the intervention group they will receive L-PRF treatment on all of their ulcers. Eligible patients will be offered entry in the trial at the point where their clinician judges them suitable for treatment i.e. when the lesion is clear of any debris or infection.
Activity measurement
All participants will be invited to wear a pedometer (Model: Mi Smart Band 5, Model: XMSH10HM) on the ankle of their non-affected limb (or non-index case affected limb) which they will wear from the first dressing change until 42 days (the point where cross-over may occur) or discharge, whichever comes first. This will act as a proxy measure of weight bearing and enable us to monitor activity across intervention and control groups and thereby evaluate whether the level of activity is similar across groups.
Explanation for the choice of comparators {6b}
The comparator is usual care. Participants in the control group will receive usual care of twice-weekly standard saline dressings only. The clinical care of these participants will be identical to the intervention participants.
Criteria for discontinuing or modifying allocated interventions {11b}
Details of any concomitant illness or medication (present at start of the trial) will be recorded at trial entry. If any change influences the participant’s eligibility to continue in the trial, the local Principal Investigator will be informed and a decision to continue with the intervention will be made in the participant’s best interest. The intervention may also be discontinued at participant’s request. Participants who withdraw will receive usual care.
Strategies to improve adherence to intervention {11c}
None.
Relevant concomitant care permitted or prohibited during the trial {11d}
Given that weight bearing and physical activity are adversely associated with healing rates of plantar ulcers, all participants will be encouraged to rest during the trial.
Provisions for post-trial care {30}
Discharge information will be noted along with the participant and a family member’s contact details. Each trial participant will receive a cell phone and contact details to use in the event of any difficulties. Any readmissions at Anandaban Hospital, or any other hospitals for treatment of the same ulcer (the ‘trial ulcer’) will be recorded (both dates and duration).
Outcomes {12}
The main end-points will be:
- Rate of healing based on two observations per week (cm2 per unit time).
- Time to complete re-epithelialization (up to a maximum of 70 days).
Both end-points will be analysed with and without adjustment for baseline characteristics (trial ulcer area and participants’ age).
Secondary end-points:
- Generic quality of life (QoL) measured fortnightly using the EQ-5D 3L. This scale has previously been used and validated in a Nepali population (17). As far as we are aware, there is no valuation tariff in the Nepalese population, although this may change by the time that the analysis starts. Therefore, we propose not to prespecify the tariff but instead select the tariff at a later date. We note that tariffs are currently available for nearby countries including Sri Lanka (18) and China (19, 20).
- Longer-term outcomes measured at six-month follow up from randomisation will be proportion with:
- Recurrence of treated ulcer;
- Appearance of new ulcer;
- Anatomical changes in the limb;
- Days hospitalised prior to discharge and total (to include any readmission due to leprosy/ulcers) by 6 months.
- Health economic information:
- Number of visits to any healthcare facility from discharge to the end of follow-up at 6 months.
- Data on the time taken to change dressings at the twice weekly changes.
Rate of healing will be assessed from blindly assessed photographs (Figure 2). However, as complete healing may trigger discharge, we will monitor for cases discharged before the blinded assessor in Birmingham has noted complete re-epithelization.
Level of activity (step count) will be collected daily and monitored across both intervention and control groups until 42 days (the point where cross-over may occur) or discharge, whichever comes first – see analysis below.
Participant timeline {13}
See Figure 1 for the participant’s timeline through the trial.
Sample size {14}
Sample size is based on the two primary outcomes: rate of healing and time to complete re-epithelialization. On the latter, we assume that 70% of ulcers will heal within 42 days with standard care (21). Further, assuming that the intervention will increase this proportion to 90% and hazards are constant and proportional, for a two-sided test of the hazard ratio with a type I error of 5% and statistical power of 80% and a 1:1 allocation ratio, 47 individuals are required in each group. To allow for withdrawals and right censoring we aim to recruit 65 patients in each group. We expect rate of healing to provide yet more precise estimates.
Recruitment {15}
Eligible individuals are identified by the clinical research team and invited to participate by the local research fellow (see eligibility criteria).
Assignment of intervention: allocation
Sequence generation, concealment mechanism and implementation {16a, 16b, 16c}
Participants will be enrolled sequentially and randomly allocated (1:1) to undergo L-PRF treatment or usual care using a “digital sealed envelope” method (22). An allocation table will be generated remotely by the trial statistician at The University of Birmingham to allocate participants in a 1:1 ratio at the level of the individual over the course of the trial. A random number generator will be used to generate a random sequence of the numbers between 1 to N inclusive. A permuted block randomisation method will be used by randomly selecting blocks of size 2, 4, 6, or 8 in order to maintain balance between the numbers allocated to each of the two groups. The generated table will be uploaded into the REDCap software to be used for participant enrolment. Access to the allocation table will be restricted. When a participant’s details are submitted, the trial arm and a unique study number will be assigned and revealed to the local clinician so that the randomised group that the participant is assigned to cannot be altered.
Assignment of interventions: blinding
Who will be blinded {17a}
The Nepal research team, the database managers in Birmingham, the clinical staff carrying out dressing changes in the room designated for this purpose and participants themselves will be aware of participants’ randomly assigned group. Ward staff will not be informed. Researchers in Birmingham involved in the data analyses will be blinded to treatment allocation (Figure 2).
Procedure for unblinding {17b}
There is no requirement for an emergency unblinding procedure.
Data collection and management
Plans for assessment and collection of outcomes {18a}
Demographic data (age and sex), clinical data (number and size of ulcers), concurrent diseases and quality of life (QoL) information (measured using the EQ-5D 3L) will be collected for all consenting participants before randomisation (Figure 1). Data on QoL will be collected fortnightly from baseline until discharge. Ulcers will be assessed twice weekly (see below) during the participant’s hospital stay, to a maximum of 70 days, and then again at a six month follow-up visit. Steps taken will be collected daily until 42 days (the point where cross-over may occur) or discharge, whichever comes first. Follow up data collection at six months will require an outpatient appointment or a home visit (Figure 3). Data, including photographs of ulcers, will be collected on electronic tablets using the Research Electronic Data Capture (REDCap) system by local researchers.
Ulcer measurements
Standardised photographs (23) will be taken twice weekly during inpatient stay dressing changes for participants in both intervention and control groups. Photographs will be obtained in Nepal using two different methods (Figure 4). The photographs will be transferred to the University of Birmingham and the ulcer dimensions measured in three ways. Two observers will be trained to take these measurements. Both observers will be blinded to the participant’s allocated treatment. All photographs from a given participant will be assigned to the observers at random, separately for the two methods of photography. So that the measurements are not all relegated to the end of the study, they will be made in batches of ten participants reaching completion of their baseline treatments (at complete re-epithelization or 70 days from randomisation). A proportion (20%) of all ulcer photographs will be measured by both observers to test inter-rater reliability. These photographs will be selected at random.
The date at which complete re-epithelization took place will be determined by the local clinician. Photographs will be taken at the point of complete re-epithelialization and during the follow up visit.
ARANZ® SilhouetteStar photography
For the first two measurement methods, a photograph will be taken using the ARANZ® SilhouetteStar device (camera) and the image synchronised to the participant’s trial number and saved to the secure ARANZ® SilhouetteCentral SQL server. This will be assessed by a designated observer at the University of Birmingham. The validated ARANZ 3D wound measurement software tool will then be used to measure the ulcer in two waysThe first measurement will be made using a manual tracing method whereby the ulcer boundary will be ‘hand’ drawn in the ARANZ® software. The software will then calculate the ulcer dimensions based on this outline. The second measurement will utilise an automatic tracing feature in the software. In this instance, the observer will delineate a region of interest (extending beyond the boundary of the ulcer) and the software will automatically locate the boundary and calculate the wound dimensions.
Digital tablet
The final measurement will use a photograph taken using the in-built camera in the tablet devices (Samsung Galaxy Tab S6). The photograph will be taken perpendicular to the ulcer. For calibration purposes, a 3cm size clean paper ruler with date and participant’s trial identification number will be placed in the photograph frame above or below the ulcer but at the level of the skin. The photograph will uploaded to the server at the University of Birmingham and evaluated digitally by a designated observer in Birmingham using the PictZar™ Digital Planimetry Software (24) with an electronic PUSH Tool (National Pressure Injury Advisory Panel (NPIAP) at https://npiap.com/page/PUSHTool). The observer will delineate an area of interest by manually ‘painting’ the ulcer area with colour using a computer mouse. The software will then calculate the ulcer dimensions based on this profile.
All observations will be made blind to treatment allocation. The ‘closed loop’ ulcer assessment process is shown in Figure 2. The changes in surface area can then be determined between each observation using three methods of assessment (Figure 4).
Plans to promote participant retention and complete follow-up {18b}
Each trial participant will receive a cell phone to enable the research team to contact them regarding their six-month follow up.
Data management {19}
Each participant will be allocated a unique participant identification number which will be used on all electronic documents and photographs. Data will be collected by researchers in Nepal and entered on to the REDCap platform hosted on a secure server at the University of Birmingham. Ulcer photographs taken using the ARANZ® device will be saved to the secure SilhouetteCentral SQL server. Data will be encrypted and access will be password restricted.
Confidentiality {27}
All collected data will remain confidential. All data will be stored in accordance with GCP and General Data Protection Regulations 2018 (GDPR), and the Data Protection Act 2018.
Plans for collection, laboratory evaluation and storage of biological specimens for genetic or molecular analysis {33}
None
Analyses and inference
Statistical methods for main endpoints {20a}
Time to healing
Time to healing will be analysed using a Cox proportional hazards model with and without adjustment for baseline characteristics (trial ulcer area and participants’ age) allowing for right-censoring. For the rate of healing we will define the outcome ulcer size in cm2 at each time point and include in the model time since admission, treatment status, and their interaction. We will analyse this model using a linear mixed-effects model with participant-level random effects and both with and without adjustment for participant characteristics. Given there are multiple primary outcomes (two outcomes, with and without adjustment, for three types of ulcer assessment) we will adjust reported p-values for multiple testing using a stepdown method, which provides an efficient means of controlling the family-wise error rate (25). We will derive the approximate distributions of the test statistics to perform the stepdown procedure using a permutation test approach, by simulating 10,000 re-randomisations of the individuals (26).
Measurements
The above analyses of healing rate and time of complete re-epithelialisation will be made separately for each method of measurement. Photographs will be assessed blind to allocated treatment by two trained observers at the University of Birmingham (Figure 2). Inter-rater reliability of the measurements will be made on 20% of ulcer photographs and assessed using interclass correlation coefficients.
Measurements are based on the ARANZ ® or tablet based cameras (see Figure 2). The ARANZ® images are each measured in two ways: 1) manual tracings and 2) automatic tracings of wounds using a camera specifically designed to obtain standardised photographs of ulcers for scientific purposes. The distance of the camera from the ulcer surface is controlled by triangulation of three laser lights. The images taken by the camera in the tablet are calibrated digitally by reference to a measurement ruler in the frame. The areas (cm2) are then generated within the different software packages. We will measure, agreement between the methods of measurement:
- The Push tool vs. the automated ARANZ® method
- The automated vs. manual ARANZ® method
Interim analyses {21b}
An interim analysis will be conducted when at least 49 participants have been followed up for a minimum of 42 days. The rationale for this analysis is the detection of a ‘penicillin like’ benefit or statistically significant negative effect of the treatment on either of the two main endpoints. A statistical threshold of 0.01, one-sided, (0.02 two-sided) will be used for either of the two main endpoints. In the event that more than 10% of control participants cross-over from one arm to another at 42 days, we will consider performing a complier average causal effects analysis to the 70 day outcome.
Missing data {20c}
We will analyse by intention to treat. In the (extremely unlikely) event that some participants withdraw before complete healing and that rate differs between groups, a sensitivity analysis will be applied. We will explore the patterns and extent of any missing data, particularly those relating to the two main endpoints. We do not plan to impute missing values, but may consider the use of multiple imputation or other strategies within the sensitivity analysis if necessary.
Methods for additional analyses {20b}
Quality of life
Quality of life (QoL) will be analysed by calculating and comparing area under the curve (AUC) across intervention and controls. Baseline results will be triangulated with clinical observations to avoid bias and to determine how differences in healing rates correspond to differences in quality of life end-points.
Activity measurement
We will also compare average daily step count between treatment and control groups as a simple difference in means (t-test). Since one group may stay longer in hospital than the other and since there may be an interaction between rate of healing and step count, we will compare step counts over periods pre-set at 7, 14 and 42 days.
Economic evaluation
An economic evaluation of L-PRF will be undertaken. The model will estimate the impact of L-PRF on net population health in terms of daily and quality adjusted life years (QALYs/DALY) over a long-term time horizon. The effectiveness of the treatment will be reflected by estimating the statistical relationship between the primary end-point in the efficacy trial and HRQoL (health-related quality of life) based on participants’ completion of the generic quality of life instruments. This analysis will also reflect other characteristics of participants that are potentially prognostic and predictive of the efficacy of L-PRF. The analysis will provide a careful assessment of how the cost-effectiveness of L-PRF varies according to its local acquisition cost, as well as other parameters that could vary between localities. Full uncertainty analysis will be undertaken to establish whether there is sufficient evidence to support local funding of L-PRF at a given acquisition cost or whether additional evidence generation is necessary and worthwhile. We shall develop the model before the data to populate it are available and our work-plan will included a review of the relevant literature in countries with both high and low incomes.
Reporting
Where a participant withdraws from the intervention as a whole, no further data will be collected. Where the participant withdraws from the randomised intervention but agrees to contribute data, they will be followed-up to the point of discharge as an inpatient. The trial will be reported in line with the CONSORT (Consolidated Standards of Reporting Trials) Standards (27, 28).
Plans to give access to the full protocol, participant level data and statistical code {31c}
The full protocol, non-identifiable participant level data and statistical code may be available for sharing once the trial has ended. All requests will be approved by the Chief Investigator (CI), Professor Richard Lilford ([email protected]).
Oversight and monitoring
Composition of the co-ordinating centre and trial steering committee {5d}
The trial will be overseen by the Trial Management Group (TMG) and a Trial Steering Committee (TSC). The TMG will be responsible for the day-to-day management of the trial. It includes individuals at the University of Birmingham (Chief Investigator, Trial Manager, Trial Co-ordinator, Clinical Trials Unit management staff) and Anandaban Hospital, The Leprosy Mission Nepal (Principal Investigator, local Project Manager, and patient representatives). The TMG will meet monthly by teleconference. The Trial Steering Committee provides overall supervision of the trial and will ensure that it is conducted in accordance with the principles of Good Clinical Practice and other relevant regulations. Meetings will be scheduled before enrolment and after each meeting of the Independent Data Monitoring Committee and more frequently during the analysis phase. The TSC includes an independent chair and members with clinical expertise.
Composition of the Data Monitoring Committee, its role and reporting structure {21a}
The Data Monitoring Committee will review safety and efficacy data during the active phase of the trial. They will advise on the continued recruitment of trial participants. They will meet either by teleconference or face-to-face. This committee consists of an independent chair, a statistician and members with clinical and methodological expertise.
Adverse event reporting and harms {22}
The Principal Investigator in Nepal, Dr Indra Napit, is responsible for recording all adverse events (AEs) and reporting any serious adverse events (SAEs) to the Chief Investigator and University of Birmingham Clinical Trials Unit (BCTU) within 24 hours of becoming aware of such an event. A SAE form will be available on the data collection tablets and a database of any events will be maintained. The Trial Management Group, Chief Investigator and the BCTU will review any SAE forms. The Trial Steering Committee will periodically review all safety data and liaise with the Independent Data Monitoring Committee regarding any safety issues. Any deaths will be reported to the Sponsor irrespective of whether the death is related to the disease progression, the intervention or an unrelated event.
Frequency and plans for auditing trial conduct {23}
The trial is audited and monitored by the Sponsor, the University of Birmingham.
Plans for communicating important protocol amendments to relevant parties (e.g. trial participants, ethical committees) {25}
Any protocol amendment will be reported to the Trial Management Committee for approval. The Sponsor, University of Birmingham Biomedical and Scientific Research Ethics Committee and the Nepal Health and Research Council will subsequently be notified.
Dissemination plans {31a}
We will publish and disseminate through the usual academic channels, including peer reviewed journals and at academic conferences. Our dissemination plans include close liaison with The Leprosy Mission to engage affected people and communities and to ensure that the results are disseminated widely. Other dissemination plans include: bite-sized research reports in lay format; public announcements in communities in LMICs; policy briefings; print and online media; the Chief Investigator’s News Blog (680+ subscribers); institutional and professional social media accounts and websites.