Abstract
Background
The current 8-week regimen of rifampicin and clarithromycin for Buruli ulcer (BU) is suboptimal. High-dose rifampicin could shorten treatment and improve outcomes. We evaluated whether a 4-week high-dose regimen could improve time to clearance of viable Mycobacterium ulcerans.
Methodology
In this open-label, individually randomised trial conducted in Ghana, participants with PCR-confirmed BU were assigned (1:1) to either high-dose oral rifampicin (20mg/kg) with clarithromycin (15mg/kg) for 4 weeks (HR) or standard-dose rifampicin (10mg/kg) with clarithromycin (15mg/kg) for 8 weeks (SR). All wounds were dressed with DACC-coated dressings. The primary outcome was time to clearance of viable M. ulcerans assessed by 16S rRNA qPCR up to week 20.
Findings
Between 26th November 2021 and 31st July 2024, 42 participants were randomised (HR:23, SR:19), which was short of the target (n = 112). The mean time to clearance was 4.9 weeks (SE 1.4) in the HR arm versus 7.0 weeks (SE 1.9) in the SR arm, with an adjusted mean difference of -0.5 weeks (95%CI -5.0 to 4.1, p = 0.835). No recurrences occurred in either arm. Paradoxical reactions were observed in 0/21 participants in the HR arm vs. 5/15 in the SR arm (adjusted OR 0.19, 95%CI 0.00-1.35, p = 0.102). Secondary infection rates were similar between groups.
Conclusion
High‑dose rifampicin combination for 4 weeks was well tolerated in this population. A nominally lower proportion of paradoxical reactions was observed in the high-dose rifampicin arm compared with standard therapy. The trial did not find evidence of significantly shorter time to microbiological clearance or healing compared with standard therapy. Larger trials are needed to determine efficacy and safety.
Trial Registration
Pan African Clinical Trials Repository, trial registration number: PACTR202011867644311 (https://pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=14534).
Author summary
Buruli ulcer (BU) is a neglected tropical disease that causes destructive skin lesions and long-term disability, particularly in rural communities. The current standard treatment, eight weeks of rifampicin and clarithromycin, is effective but lengthy, which can reduce adherence and increase the burden on patients and health services. This study explored whether a high-dose rifampicin regimen could offer a safe, shorter, and more convenient alternative. In a randomized clinical trial conducted in Ghana, we compared a 4-week high-dose rifampicin regimen with the standard 8-week treatment. We assessed how quickly the bacteria causing BU were cleared from the wounds of affected individuals and monitored safety outcomes. The results showed that the shorter, high-dose regimen was not superior but was well tolerated and achieved similar bacterial clearance and healing times compared to standard treatment although we enrolled fewer participants than planned and the intervention and control groups were not evenly balanced. Notably, patients receiving the high-dose regimen experienced fewer paradoxical reactions, a common inflammatory complication during treatment. These findings suggest that high-dose rifampicin may offer a promising strategy to shorten BU treatment time. However, larger studies are needed to confirm its efficacy, safety and potential to shorten treatment duration while maintaining good outcomes.
Introduction
Buruli ulcer (BU) is a neglected tropical disease (NTD) caused by Mycobacterium ulcerans (M. ulcerans). While the mode of transmission remains unknown, BU is endemic in over 30 countries globally, with the highest burden in rural West Africa, including Ghana. Children under 15 years are most commonly affected, though individuals of any age may develop the disease [1].
Initial lesions present as painless subcutaneous nodules or dermal plaques, which enlarge over days to weeks and eventually ulcerate. Without treatment, ulcers can expand to cover entire limbs or the trunk, causing extensive disfigurement. Patients typically remain systemically well, but secondary bacterial infections may occur. Access to health facilities in rural regions is often limited, and advanced disease is common due to fear, stigma, skepticism about allopathic medicine, and the economic burden on poor families [2–4]. The World Health Organization (WHO) NTD roadmap aims to control BU by 2030, and a key priority is the development of shorter, more cost-effective treatments [5].
Over the past two decades, antibiotic therapy has improved healing and reduced recurrence rates to below 2%. However, the current WHO‑recommended regimen comprising oral rifampicin (10 mg/kg) plus clarithromycin for 8 weeks [6] remains suboptimal. Many affected individuals must travel long distances to complete treatment, and healing times vary considerably, even among patients with similar lesions and irrespective of the regimen used [7–10]. Approximately two-thirds of patients heal within 25 weeks of treatment initiation while the remaining one-third may take up to a year to heal [7,9]. Additional challenges with the current rifampicin-clarithromycin regimen are the relatively long 8-week duration, adverse effects and drug interactions especially in older populations.
Rifampicin, the backbone of BU treatment [11] is a potent inducer of drug-metabolizing enzymes and can reduce the effectiveness of various medicines, creating management challenges in some patients. The current 10 mg/kg dose is derived from tuberculosis (TB) guidelines. Emerging evidence indicates that higher‑dose rifampicin is safe and well tolerated [12,13], and several large‑scale TB trials are evaluating this strategy [14,15]. In a study using high dose rifampin among individuals with tuberculous meningitis, no evidence of survival benefit from high-dose rifampin was observed. However, drug-induced liver injury occurred in 8.0% of the participants in the high-dose group and 4.4% of those in the standard-dose group, but no deaths from drug-induced liver injury occurred [15]. Preliminary data in BU suggest that higher doses may accelerate clearance of viable M. ulcerans enabling regimen shortening [16]. Some data suggest that standard rifampicin/clarithromycin-containing regimens may result in suboptimal antibiotic exposures, particularly among patients with higher body weight, potentially contributing to treatment failure [17]. These findings highlight the importance of considering drug exposure as a determinant of treatment response and support further evaluation of dose optimization strategies. Collectively, these data suggest that high-dose rifampicin is likely to be a safe and potentially more efficacious approach for BU, but it has not yet been evaluated in a clinical trial.
A complementary strategy to address both high mycobacterial load and secondary wound infections is the use of dialkylcarbamoyl chloride (DACC)-coated dressings. These dressings irreversibly bind bacteria on wound surfaces, removing them when changed, thereby reducing bacterial load without antimicrobials. In other clinical contexts, DACC dressings have been shown to reduce bacterial load more effectively than silver‑impregnated dressings [18], and to reduce wound size reduction [19]. Compared with vaseline gauze, the standard wound dressing in most of West Africa, DACC dressings may offer superior outcomes.
High‑dose rifampicin (20 mg/kg) with DACC dressings could accelerate clearance of viable mycobacteria, reduce the risk of paradoxical reactions and secondary infections, and improve overall patient outcomes compared with standard‑dose rifampicin (10 mg/kg) plus DACC. Paradoxical reactions are inflammatory responses that can cause significant pain, swelling, and wound deterioration, often hindering the healing process and requiring additional clinical management [20]. A shorter, more effective regimen would also be more convenient for affected individuals, potentially improving adherence, reducing costs for affected persons, their families and healthcare systems, and ultimately contributing to WHO’s 2030 BU control targets. The aim of this study was to determine whether high‑dose oral rifampicin (20 mg/kg) combined with DACC dressings improves the time to clearance of viable mycobacteria from BU wounds compared with standard‑dose oral rifampicin (10 mg/kg) combined with DACC dressings.
Methods
Ethics statement
This was an open-label phase 3 prospective, randomised controlled trial. The trial was sponsored by the London School of Hygiene and Tropical Medicine (LSHTM) and executed in collaboration with the Kwame Nkrumah University of Science and Technology (KNUST) and the Noguchi Memorial Institute of Medical Research (NMIMR). The trial protocol [21] was approved by the Ghana Health Service Ethics Review Committee (GHS-ERC: 003/01/21), KNUST (CHRPE/AP/030/21), LSHTM (reference number, 22912), and the FDA [FDA/CT/2110 and FDA/CT/2110 (ext 1)] The study was registered on the Pan African Clinical Trials Repository with trial registration number PACTR202011867644311 (https://pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=14534).
Written informed consent was obtained from all participants. Written parental or guardian consent was obtained for all under 18 and assent for those aged 13–17 years. A Data and Safety Monitoring Board (DSMB) operating independently of the trial was constituted to oversee the conduct of the trial. The DSMB reviewed the trial protocol and plan for analysis at the start of the trial and annually throughout the recruitment period to conduct its oversight role.
Participants
Participants were recruited from 26th November 2021–25th January 2024 from two treatment centres in Ghana with extensive expertise in BU care: Pakro Health Centre and Wassa Akropong Municipal Hospital. These facilities are located in districts endemic for BU. Eligible participants were those presenting with a Buruli nodule, plaque, or ulcer, with or without associated oedema, aged between 5 and 80 years old, for whom a sample collected from their lesion was positive for M. ulcerans DNA by PCR for the IS2404 repetitive genetic element. Children younger than 5 years were excluded primarily for safety reasons, as high-dose rifampicin was not part of the standard treatment regimen at the time of the study. In addition, the distinct pharmacokinetic characteristics and smaller body size of very young children may necessitate age-specific dosing formulations. Furthermore, the assessment of subjective outcomes, such as pain, and the maintenance of reliable follow-up and treatment adherence may be more challenging in this age group. Individuals were also excluded if pregnant or if either rifampicin or clarithromycin was contraindicated. Complete eligibility criteria are provided in the published trial protocol (S1 File) [21].
Randomisation, allocation and treatment
Participants enrolled in the study were randomly individually assigned in a 1:1 ratio to one of the two treatment groups. A computer-generated random allocation sequence was prepared at the LSHTM using block randomization, stratified by site, BU Category (Category I-III), and specific randomization numbers placed in sealed, opaque envelopes. This process was conducted securely, with access restricted to prevent any possibility of viewing the allocation sequence, and was maintained by the study coordinator.
The control regimen was the WHO–recommended standard combination of rifampicin (10mg/kg) and clarithromycin (15mg/kg) given daily for 8 weeks. The intervention group was given rifampicin (20mg/kg) and clarithromycin (15mg/kg) daily for 4 weeks. Participants in both groups had their ulcers dressed with DACC-coated dressing and changed every 48 hours until the lesion healed. Adherence to trial medication was monitored through directly observed therapy supervised in the clinic and at home by a domiciliary treatment monitor. A time allowance was in place for missed doses. Community-based surveillance volunteers (CBSVs) within study districts supported participant follow-up and adherence to trial medications and dressings. Additional details are provided in the published protocol [21].
Trial procedures
Every participant attended 15 scheduled review visits at Weeks 2, 4, 6, 8, and monthly thereafter until reaching 52 weeks after starting antibiotic treatment. Participants were monitored for at least 20 weeks, and any visits beyond this period were part of standard care practices. Participants samples were collected according to the assessment schedule (S1 Table).
At enrolment, we collected demographic data. At enrolment and follow-up, clinical, cost, health-related quality of life data, and photographs of lesions were collected. Data on pain assessment were recorded using a Visual Analogue Scale (VAS). Wound diameter (including the region of induration) and area were recorded by digital photography and measured by computer planimetry using a Silhouette Wound Imaging System (ARANZ, Wellington, New Zealand) at each scheduled clinical visit.
Swabs or fine needle aspirate (FNA) samples were processed for PCR IS2404 using previously published methods in the laboratories that currently provide PCR confirmation for routine BU cases at Kumasi Centre for Collaborative Research in Tropical Medicine (KCCR) and NMIMR [22]. Samples were collected before and during treatment to assess the persistence of viable mycobacteria. Swabs were taken from the undermined edges of ulcerated lesions, or FNA samples from the centre of non-ulcerative lesions (nodule, plaque, and oedema) at weeks 2, 4, 8, 12, 16 and 20 (if lesions were not healed) for the combined M. ulcerans 16S rRNA reverse transcriptase/IS2404 real-time qPCR [23].
For the combined 16S rRNA reverse transcriptase/IS2404 real-time qPCR assay to assess viability of M. ulcerans, swabs and FNAs were stabilized in 500µl of RNA protect (Qiagen) and subjected to combined DNA/RNA extraction by the All-Prep DNA/RNA extraction kit (Qiagen). RNA was reverse-transcribed to cDNA. DNA and cDNA were subjected to IS2404 qPCR and 16S rRNA real time PCR respectively [23]. The median turn-around time for IS2404 qPCR assay was 5 days as previously published [24].
Wounds were assessed clinically for secondary infection at each time point. In the context of a new purulent discharge, a single swab was taken for microbiological assessment at KCCR and NMIMR Laboratories using previously published methods [25,26].
Transportation cost of participants between their homes and the health facilities was reimbursed.
Blinding was not possible for clinical research staff administering drugs. A skilled wound assessor blinded to the allocation evaluated wound images for assessment of healing to avoid bias and maximize the validity of results. Clinical and laboratory data were collected by staff blinded to the study arm.
Outcomes
The primary outcome was the time to clearance of viable mycobacteria, evaluated up to week 20. The time to clearance was based on the duration from randomisation (which was same day on which antibiotic treatment was started) until the participant had no viable organisms, as assessed by the combined 16S rRNA reverse transcriptase/IS2404 real-time qPCR assay. Participants with healed wounds or no evidence of viable mycobacteria by week 20 were considered to have had a favourable outcome.
Secondary outcomes included the proportion of participants achieving complete healing at 20 weeks, the proportion of participants with recurrence, the incidence of paradoxical reaction, and the incidence of secondary infection. Incidence of recurrences, paradoxical reaction, and secondary infection were assessed at 12 months following randomization. Definition of terms
Complete healing was defined as re-epithelialization of the ulcerated area with stable scar formation.
Recurrence was defined as the appearance of a new Buruli ulcer lesion within 12 months of completing antibiotic treatment, after complete healing of the initial lesion, and not considered a paradoxical reaction. This was assessed by the presence of viable micro-organisms detected by M. ulcerans culture or the combined M. ulcerans IS2404-16S rRNA assay.
Paradoxical reactions were defined as:
an initial improvement in the clinical appearance of lesion upon antimicrobial treatment of BU followed by deterioration of the lesion or its surrounding tissues
or the appearance of a new lesion(s)
or examination of excised tissue from the clinical lesion showing evidence of an intense inflammatory reaction consistent with a paradoxical reaction.
Secondary infections were evaluated by trained clinicians using a standardized checklist that assessed the presence of pain, erythema, warmth, swelling, and purulent discharge to identify signs suggesting secondary bacterial infection as previously described [27]. A secondary infection was defined as the occurrence of new purulent discharge within a BU lesion in the presence of clinical signs of infection and accompanied by microbiological evidence of infection.
Statistical methods
It was estimated that 50 participants in each arm would be required to have 80% power to detect a 2-week reduction in mean time to M. ulcerans clearance using statistical testing at the 5% level of significance. This estimate was based on previous studies showing a mean time to clearance of viable mycobacterium with standard dose oral rifampicin of 9 weeks (3 weeks standard deviation) [23].This sample size would also be sufficient to detect a 27% absolute increase in the proportion of participants with healed wounds by 20 weeks, assuming 50% would heal at 20 weeks in the standard dose arm. Allowing for a 10% rate of loss-to-follow-up, the study aimed to recruit a total of 112 participants.
Statistical analysis was based on complete cases and guided by an analysis plan written according to the published trial protocol [21]. The analysis was conducted on a modified intention-to-treat (mITT) population of individuals who were PCR positive for M. ulcerans at enrolment, based on the randomised treatment allocation regardless of intervention received or whether it was wholly or partly taken.
Baseline demographic and other characteristics of participants were summarized using the mean and standard deviation for continuous variables, and counts with proportions for categorical variables. A generalized linear model assuming a normally distributed outcome was used to estimate mean differences in time to clearance of viable mycobacteria between intervention and control groups with associated 95% confidence intervals (CI). A modified Poisson regression model was used to estimate risk ratios with 95% confidence intervals for wound healing. Recurrence of BU, incidence of paradoxical reaction, and incidence of secondary infection were analyzed using a binomial regression or, for rare outcomes, exact logistic regression. Safety was assessed in all randomized participants receiving at least one treatment dose. Information on adverse events, serious adverse events (SAEs), and deaths was collected and reported in both arms of the trial.
Results
The results of the study are reported according to the Consolidated Standards of Reporting Trials (S1 Checklist). Recruitment to the trial was stopped on 31st July 2024, on advice of the DSMB and Trial Steering Committee (TSC) due to slow recruitment. Between 26th November 2021 and 31st July 2024, a total of 52 individuals were screened for inclusion into the study. In total, 42 individuals who met the inclusion criteria were randomized to receive either HR+DACC (23 participants) or SR+DACC (19 participants). Two participants in the SR+DACC arm were lost to follow-up at 20 weeks, and a further two participants in each arm were lost to follow-up by week 52. (Fig 1).
Fig 1. CONSORT Flow Diagram.
Participants in the HR arm were on average older, included more female participants, were slightly heavier, had lower levels of education, and were more likely to be in some form of employment. The distribution of participants was similar across arms in terms of the form of lesion (with ulcers being most common), stage of ulcer, and study sites (Table 1).
Table 1. Characteristics of participants at recruitment.
| Characteristic | HR+DACC N = 23 |
SR+DACC N = 19 |
Total N = 42 |
|---|---|---|---|
| Age in years, mean (SD) | 35.9 (22.2) | 19.6 (14.2) | 28.5 (20.5) |
| Sex, n (%) Female Male |
9 (39.1) 14 (60.9) |
5 (26.3) 14 (73.7) |
14 (33.3) 28 (66.7) |
| Weight in kilograms, mean (SD) | 50.4 (14.8) | 40.3 (16.4) 1 missing |
46.0 (16.2) |
| Highest level of education, n (%) | |||
| None | 9 (39.2) | 1 (5.3) | 10 (23.8) |
| Primary | 14 (60.9) | 15 (79.0) | 29 (69.1) |
| Secondary | 0 (0.0) | 3 (15.8) | 3 (7.1) |
| Primary occupation, n (%) | |||
| Unemployed | 2 (8.7) | 1 (5.3) | 3 (7.1) |
| Student | 6 (26.1) | 10 (52.7) | 16 (38.1) |
| Farmer | 9 (39.1) | 4 (21.1) | 13 (31.0) |
| Trader | 2 (8.7) | 0 (0.0) | 2 (4.8) |
| Miner | 2 (8.7) | 1 (5.3) | 3 (7.1) |
| Other | 2 (8.7) | 3 (15.8) | 5 (11.9) |
| Number of lesions, n (%) | |||
| Single | 16 (69.6) | 12 (63.2) | 28 (66.7) |
| Multiple | 7 (30.4) | 7 (36.8) | 14 (33.3) |
| Form of lesions, n (%) | |||
| Nodule | 1 (4.4) | 0 (0.0) | 1 (2.4) |
| Plaque | 0 (0.0) | 1 (5.3) | 1 (2.4) |
| Oedema | 1 (4.4) | 1 (5.3) | 2 (4.8) |
| Ulcer | 21 (91.3) | 17 (89.5) | 38 (90.5) |
| Stage of Buruli ulcer, n (%) | |||
| Category I | 4 (17.4) | 3 (15.8) | 7 (16.7) |
| Category II | 3 (13.0) | 4 (21.1) | 7 (16.7) |
| Category III | 16 (69.6) | 12 (63.2) | 28 (66.7) |
| Site, n (%) | |||
| Pakro Health Centre | 10 (43.5) | 8 (42.1) | 18 (42.9) |
| Wassa Amenfi East | 13 (56.5) | 11 (57.9) | 24 (57.1) |
HR+DACC: high-dose oral rifampicin with dialkylcarbamoyl chloride-coated dressings, the intervention group.
SR+DACC: standard-dose oral rifampicin with dialkylcarbamoyl chloride-coated dressings, the control group.
N: group size.
n: number of individuals in subgroup.
SD: standard deviation.
The mean time to clearance of viable M. ulcerans was 4.9 weeks in the HR arm versus 7.0 weeks in the SR arm, a crude difference of -2.1 weeks (95%CI -6.6 to 2.4), p-value 0.366. After adjusting for age, sex, stage of BU, and trial site, the mean difference was -0.5 weeks (95%CI -5.0 to 4.1, p-value 0.835).
Overall, there was a pattern suggesting more rapid microbiological clearance of M. ulcerans RNA in the HR arm. Participants in the HR achieved complete observed RNA negativity earlier, whereas those in the SR arm showed slower and less consistent clearance and remained RNA-positive in one tested sample at week 20 (Table 2).
Table 2. 16S rRNA positivity by study timepoint.
| Study week | HR+DACC, n (%) | SR+DACC, n (%) |
|---|---|---|
| 0 | 7/23 (30.43) | 5/19 (26.32) |
| 2 | 6/22 (27.27) | 6/19 (31.58) |
| 4 | 3/20 (15.00) | 4/18 (22.22) |
| 6 | 3/20 (15.00) | 5/17 (29.41) |
| 8 | 4/17 (23.53) | 3/16 (18.75) |
| 12 | 0/17 (0) | 3/14 (21.43) |
| 16 | 2/15 (13.33) | 2/15 (13.33) |
| 20 | 0/14 (0) | 1/14 (7.14) |
*Numerator represents the proportion of participants testing positive at the time point; denominator represents the total number of participants contributing a sample for testing at the time point. Positivity (%) = Number of RNA-positive samples/ Number of samples tested at that time point multiplied by 100.
Within 20 weeks of follow-up, wound healing occurred in seven of the 23 participants in the HR arm, and four of the 17 in the SR arm, a crude relative improvement of 29% (RR 1.29, 95%CI 0.44 to 3.77, p-value 0.637). After adjustment, the risk ratio was 0.88 (95%CI 0.25 to 3.13, p-value 0.849). No participants in either arm experienced recurrence of BU. Paradoxical reaction was observed in none of the participants in the HR arm [0% (0/21)] and five participants in the SR arm [33.3% (5/15)], a crude odds ratio of 0.08 (95%CI 0.00 to 0.65, p-value 0.016), which after adjustment was 0.19 (95%CI 0.00 to 0.35, p-value 0.102). Amongst participants followed up for up to 52 weeks, secondary infections occurred in 13 of the 21 participants in the HR arm, and 10 of the 15 participants in the SR arm, corresponding to a crude risk ratio of 0.93 (95%CI 0.56 to 1.56, p-value 0.770) and an adjusted risk ratio of 0.86 (95%CI 0.49 to 1.50) (Table 3).
Table 3. Effects of the intervention on primary and secondary outcomes.
| Outcome | HR+DACC | SR+DACC | Unadjusted effect (95% confidence interval) p-value |
Adjusted1 effect (95% confidence interval) p-value |
|---|---|---|---|---|
| Assessed up to 20 weeks 2 | N = 23 | N = 17 | ||
| Time in weeks to clearance of viable M. ulcerans, mean (SE) | 4.9 (1.4) 1 missing |
7.0 (1.9) 1 missing |
MD -2.1 (-6.6 to 2.4) p = 0.366 |
MD -0.5 (-5.0 to 4.1) p = 0.835 |
| Wound healing, n (%) | 7 (30.4) | 4 (23.5) | RR 1.29 (0.44 to 3.77) p = 0.637 |
RR 0.88 (0.25 to 3.13) p = 0.849 |
| Assessed up to 52 weeks 3 | N = 21 | N = 15 | ||
| Recurrence of Buruli ulcer within 12 months, n (%) | 0 (0.0) | 0 (0.0) | _ | _ |
| Paradoxical reaction within 12 months, n (%) | 0 (0.0) |
5 (33.3) | OR 0.08 (0.00 to 0.65) p = 0.016 |
OR 0.19 (0.00 to 1.35) p = 0.102 |
| Secondary infection within 12 months, n (%) | 13 (61.9) | 10 (66.7) | RR 0.93 (0.56 to 1.53) p = 0.770 |
RR 0.86 (0.49 to 1.50) p = 0.591 |
HR+DACC: high-dose oral rifampicin with dialkylcarbamoyl chloride-coated dressings, the intervention group.
SR+DACC: standard-dose oral rifampicin with dialkylcarbamoyl chloride-coated dressings, the control group.
N: group size.
n: number of individuals where different from overall group size.
SE: standard error.
p: p-value.
MD: mean difference.
OR: odds ratio.
RR: risk ratio.
1Adjusted for age, sex, stage of BU and trial site, except effect on paradoxical reaction which is only adjusted for age, sex and site due to low event counts.
2Excludes two participants in the SR+DACC arm who are lost to follow-up before week 20.
3Excludes two participants in the HR+DACC and a further two in the SR+DACC arm in addition to (2) above who are lost to follow-up before week 52.
No concomitant medications were prescribed or administered by the study team in either arm during the study. However, some participants reported the use of non-study medications, mainly occasional analgesics and ‘over-the-counter immune boosters’, which were documented in the Case Report Form (CRF) as concomitant medications and monitored throughout the study. All reported medications were reviewed for potential interactions with the investigational products, and no clinically relevant contraindications were identified; therefore, these medications were not considered to have affected study treatment or outcomes.
All reported adverse events were mild and resolved without additional treatment or intervention (Table 4). A total of 28 adverse events were reported among 23 participants across the two study arms during the study period: 15 events in 13 participants in the SR+DACC arm and 13 events in 10 participants in the HR+DACC arm. More than one adverse event was reported by four participants in the SR+DACC arm and two participants in the HR+DACC arm.
Table 4. Summary of adverse reactions.
| Adverse Events | HR+DACC n (%) | SR+DACC n (%) | Total n (%) |
|---|---|---|---|
| Participants with ≥1 adverse event | 10 | 9 | 19 |
| Total adverse events reported | 16 (52) * | 15 (48) * | 31 (100) |
| Serious adverse events | 0 | 0 | 0 |
| Non-serious adverse events | 16 (100) | 15 (100) | 31 (100) |
| Mucocutaneous/Dermatological manifestations | 9 (56) | 5 (33) | 14 (45) |
| Skin rash | 1 (6) * | 0 | 1 (3) |
| Body itch | 1 (6) * | 0 | 1 (3) |
| Itch on wound surface | 7 (44) * | 5 (33) * | 12 (39) |
| Constitutional symptoms | 3 (19) | 1 (7) | 4 (13) |
| Fever | 0 | 1 (7) * | 1 (3) |
| Dizziness | 2 (13) * | 0 | 2 (6) |
| Headache | 1 (6) * | 0 | 1 (3) |
| Gastrointestinal symptoms | 2 (13) | 0 | 2 (6) |
| Diarrhoea | 1 (6) | 0 | 1 (3) |
| Abdominal/Epigastric pain | 1 (6) * | 0 | 1 (3) |
| Urine discolouration | 1 (6) * | 7 (47) * | 8 (26) |
| Insomnia | 1 (6) | 0 | 1 (3) |
| Anorexia/Loss of appetite | 0 | 2 (13) * | 2 (6) |
HR+DACC: high-dose oral rifampicin with dialkylcarbamoyl chloride-coated dressings, the intervention group. SR+DACC: standard-dose oral rifampicin with dialkylcarbamoyl chloride-coated dressings, the control group. n: number of adverse events reported by participants in each arm. Percentages for specific adverse events are calculated using the total number of adverse events in each treatment arm as the denominator (HR+DACC: n = 16 adverse events; SR+DACC: n = 15 adverse events).
A total of 31 adverse events were reported in 19 participants. In the HR+DACC arm, four participants experienced multiple adverse events. One participant reported itchiness on the surface of the wound at week 4, followed by dizziness and epigastric pain at week 6; another participant experienced dizziness and headache concurrently at Week 4; a third participant reported body itch at Week 2 and two episodes of itch on the wound surface at weeks 6 and 12; and a fourth participant experienced urine discolouration and an itchy rash concurrently at week 2.
In the SR+DACC arm, two participants experienced multiple adverse events during follow-up. One participant reported recurrent urine discolouration at weeks 2, 4, and 12; and another participant experienced anorexia at week 2, fever and anorexia concurrently at Week 4, and itch on the surface of the wound at weeks 16 and 28.
Discussion
In this randomized trial, we found a trend of faster time to clearance of viable M. ulcerans in the short-course high-dose arm (4.9 weeks vs. 7.0 weeks) with no evidence of a difference after adjustment. Wound healing rates at 20 weeks were similar between groups. Paradoxical reactions were observed only in the standard-dose arm (26.3% vs. 0%). Although paradoxical reactions occurred less frequently in the high-dose rifampicin arm, there was no evidence of a difference in their occurrence between the study arms and recurrence rates were zero in both groups. The trial was stopped early due to slow recruitment, which precludes definitive conclusions on the superiority of the high-dose rifampicin-based regimen, but it was well tolerated and potentially useful.
Our results are consistent with the possibility that a high-dose short course regimen could achieve comparable bacterial clearance to standard long-course therapy. While the adjusted difference was not statistically significant, the fact that the shorter regimen cleared the pathogen at least as rapidly as the longer regimen challenges the necessity of the full 8-week course [6]. Our findings add to the growing body of evidence suggesting that shorter-course antibiotic treatment regimens for Buruli ulcer may be feasible. Previous studies have reported the effectiveness of treatment durations shorter than the standard 8-week regimen in selected patients [17,28,29]. Our study provides additional evidence from a high-burden endemic setting that may support future evaluation of shortened treatment strategies.
The notably lower incidence of paradoxical reactions in the high-dose cohort is another important observation. Paradoxical reactions have been observed in 8–12% of Buruli ulcer patients in Africa during or after antibiotic treatment [30]. Patients experiencing PR tend to have a high bacterial load in baseline lesion samples and a greater likelihood of testing positive for M. ulcerans culture [31]. The marked reduction of paradoxical reactions in the HR arm may be attributed to shorter antigenic stimulation due to more rapid bacterial clearance in the high dose arm. Although the loss of statistical significance after adjustment, likely due to the small sample size and zero events in one arm, tempers any definitive conclusion, this finding constitutes an important indication in support of the potential benefits of the shorter regimen.
Consistent with evidence in tuberculosis [32,33], the safety profile of the high-dose regimen was reassuring. High-dose rifampicin was well tolerated, with no serious adverse events attributed to the higher dose. The types and frequency of adverse effects were similar between the two treatment arms, consistent with the extensive safety data from TB studies where similar or higher doses of rifampicin are used [33]. However, given the younger median age of our high-dose rifampicin cohort, these findings may not be generalizable to older endemic populations - notably those in Australia, where the median age of affected individuals approaches 60 years.
Our findings strongly support the need for a larger, adequately powered non-inferiority trial to definitively determine whether a 4-week high-dose regimen is as effective or superior to the current WHO recommended 8-week course. If confirmed in larger trials, a high-dose, short-course rifampicin regimen could have substantial clinical and public health benefits. Such a regimen would be particularly valuable in rural endemic regions where access to care is limited and prolonged treatment imposes a significant burden on affected individuals, families, and health systems, directly aligning with the WHO’s 2030 roadmap for NTDs.
Supporting information
(PDF)
(PDF)
(PDF)
Acknowledgments
We wish to thank the individuals and communities for their participation in the work of the Skin Health Africa Research Programme (SHARP).
Disclaimer
The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.
Additional members of the SHARP Collaboration
Armauer Hansen Research Institute, Ethiopia (AHRI)
Endalamaw Gadisa: Funding acquisition
Addis Ababa University (AAU)
Mirgissa Kaba: Funding acquisition
Kumasi Centre for Collaborative Research (KCCR)
Bodicea Fosua Andoh, Philemon Boasiako Antwi: Project administration
Noguchi Memorial Institute for Medical Research
Eric Koka, Collins Ahorlu, Daniel Okyere, Edmund Kwaku Ocloo: Project administration
London School of Hygiene and Tropical Medicine (LSHTM)
Saba Lambert: Funding acquisition
Esther Amon: Project administration
Data Availability
All relevant data have been included in the manuscript and accompanying files.
Funding Statement
The Skin Health Africa Research Programme (SHARP) was funded by the National Institute for Health and Care Research (NIHR) under its Research and Innovation for Global Health Transformation (RIGHT) Programme [Grant Reference Number NIHR200125]. MM is supported by an NIHR Research Professorship (NIHR304258). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
References
- 1.Amoako YA, Agbavor B, Frimpong M, Agbanyo A, Loglo AD, Oppong MN, et al. Clinical epidemiology of Buruli ulcer disease in Ghana, 2006–2024. Clin Exp Dermatol. 2025. doi: 10.1093/ced/llaf326 [DOI] [PubMed] [Google Scholar]
- 2.Ahorlu CK, Koka E, Yeboah-Manu D, Lamptey I, Ampadu E. Enhancing Buruli ulcer control in Ghana through social interventions: a case study from the Obom sub-district. BMC Public Health. 2013;13:59. doi: 10.1186/1471-2458-13-59 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Asiedu K, Etuaful S. Socioeconomic implications of Buruli ulcer in Ghana: a three-year review. Am J Trop Med Hyg. 1998;59(6):1015–22. doi: 10.4269/ajtmh.1998.59.1015 [DOI] [PubMed] [Google Scholar]
- 4.Hailemichael Y, Novignon J, Owusu L, Okyere D, Mtuy T, Alemu AY, et al. The role of economic factors in shaping and constituting the household burden of neglected tropical diseases of the skin: Qualitative findings from Ghana and Ethiopia. Soc Sci Med. 2024;356:117094. doi: 10.1016/j.socscimed.2024.117094 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ending the neglect to attain the sustainable development goals: a road map for neglected tropical diseases 2021-2030. World Health Organization; 2021. [Google Scholar]
- 6.Phillips RO, Robert J, Abass KM, Thompson W, Sarfo FS, Wilson T, et al. Rifampicin and clarithromycin (extended release) versus rifampicin and streptomycin for limited Buruli ulcer lesions: a randomised, open-label, non-inferiority phase 3 trial. Lancet. 2020;395(10232):1259–67. doi: 10.1016/S0140-6736(20)30047-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Nienhuis WA, Stienstra Y, Thompson WA, Awuah PC, Abass KM, Tuah W, et al. Antimicrobial treatment for early, limited Mycobacterium ulcerans infection: a randomised controlled trial. Lancet. 2010;375(9715):664–72. doi: 10.1016/S0140-6736(09)61962-0 [DOI] [PubMed] [Google Scholar]
- 8.Sarfo FS, Phillips R, Asiedu K, Ampadu E, Bobi N, Adentwe E, et al. Clinical efficacy of combination of rifampin and streptomycin for treatment of Mycobacterium ulcerans disease. Antimicrob Agents Chemother. 2010;54(9):3678–85. doi: 10.1128/AAC.00299-10 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Phillips RO, Sarfo FS, Abass MK, Abotsi J, Wilson T, Forson M, et al. Clinical and bacteriological efficacy of rifampin-streptomycin combination for two weeks followed by rifampin and clarithromycin for six weeks for treatment of Mycobacterium ulcerans disease. Antimicrob Agents Chemother. 2014;58(2):1161–6. doi: 10.1128/AAC.02165-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Vincent QB, Ardant M-F, Adeye A, Goundote A, Saint-André J-P, Cottin J, et al. Clinical epidemiology of laboratory-confirmed Buruli ulcer in Benin: a cohort study. Lancet Glob Health. 2014;2(7):e422-30. doi: 10.1016/S2214-109X(14)70223-2 [DOI] [PubMed] [Google Scholar]
- 11.Converse PJ, Almeida DV, Tyagi S, Xu J, Nuermberger EL. Shortening Buruli Ulcer Treatment with Combination Therapy Targeting the Respiratory Chain and Exploiting Mycobacterium ulcerans Gene Decay. Antimicrob Agents Chemother. 2019;63(7):e00426-19. doi: 10.1128/AAC.00426-19 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Dian S, Yunivita V, Ganiem AR, Pramaesya T, Chaidir L, Wahyudi K, et al. Double-Blind, Randomized, Placebo-Controlled Phase II Dose-Finding Study To Evaluate High-Dose Rifampin for Tuberculous Meningitis. 2018. 10.1128/AAC [DOI] [PMC free article] [PubMed]
- 13.Aggarwal AN, Agarwal R, Dhooria S, Bhavani PK, Prasad KT, Sehgal IS, et al. Efficacy and safety of daily drug regimens containing high-dose versus standard-dose rifampicin for treating pulmonary tuberculosis: Systematic review and meta-analysis. Lung India. 2026;43(3):309–22. doi: 10.4103/lungindia.lungindia_532_25 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Velásquez GE, Brooks MB, Coit JM, Pertinez H, Vargas Vásquez D, Sánchez Garavito E, et al. Efficacy and Safety of High-Dose Rifampin in Pulmonary Tuberculosis. A Randomized Controlled Trial. Am J Respir Crit Care Med. 2018;198(5):657–66. doi: 10.1164/rccm.201712-2524OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Meya DB, Cresswell FV, Dai B, Engen N, Naidoo K, Ganiem AR, et al. Trial of High-Dose Oral Rifampin in Adults with Tuberculous Meningitis. N Engl J Med. 2025;393(24):2434–46. doi: 10.1056/NEJMoa2502866 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Omansen TF, Almeida D, Converse PJ, Li S-Y, Lee J, Stienstra Y. High-Dose Rifamycins Enable Shorter Oral Treatment in a Murine Model of Mycobacterium ulcerans Disease. 2019. 10.1128/AAC [DOI] [PMC free article] [PubMed]
- 17.O’Brien DP, Friedman ND, Walton A, Hughes A, Athan E. Risk Factors Associated with Antibiotic Treatment Failure of Buruli Ulcer. Antimicrob Agents Chemother. 2020;64(9):e00722-20. doi: 10.1128/AAC.00722-20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Stanirowski PJ, Kociszewska A, Cendrowski K, Sawicki W. Dialkylcarbamoyl chloride-impregnated dressing for the prevention of surgical site infection in women undergoing cesarean section: a pilot study. Arch Med Sci. 2016;12(5):1036–42. doi: 10.5114/aoms.2015.47654 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Stanirowski PJ, Bizoń M, Cendrowski K, Sawicki W. Randomized Controlled Trial Evaluating Dialkylcarbamoyl Chloride Impregnated Dressings for the Prevention of Surgical Site Infections in Adult Women Undergoing Cesarean Section. Surg Infect (Larchmt). 2016;17(4):427–35. doi: 10.1089/sur.2015.223 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Agbavor B, Agbanyo A, Loglo AD, Antwi PB, Ackam N, Adjei J, et al. Clinical and microbiological predictors of healing in Buruli ulcer disease. J Clin Tuberc Other Mycobact Dis. 2024;34:100415. doi: 10.1016/j.jctube.2024.100415 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Amoako YA, Agbanyo A, Novignon J, Owusu L, Tuffour J, Asante-Poku A, et al. Buruli-RifDACC: Evaluation of the efficacy and cost-effectiveness of high-dose versus standard-dose rifampicin on outcomes in Mycobacteriumulcerans disease, a protocol for a randomised controlled trial in Ghana. NIHR Open Res. 2023;2:59. doi: 10.3310/nihropenres.13332.2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Marion E, Hycenth N, Vedithi SC, Robbe-Saule M, Donkeng V, Ganlonon L-M, et al. A combined effort of 11 laboratories in the WHO African region to improve quality of Buruli ulcer PCR diagnosis: The “BU-LABNET”. PLoS Negl Trop Dis. 2022;16(11):e0010908. doi: 10.1371/journal.pntd.0010908 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sarpong-Duah M, Frimpong M, Beissner M, Saar M, Laing K, Sarpong F, et al. Clearance of viable Mycobacterium ulcerans from Buruli ulcer lesions during antibiotic treatment as determined by combined 16S rRNA reverse transcriptase /IS 2404 qPCR assay. PLoS Negl Trop Dis. 2017;11(7):e0005695. doi: 10.1371/journal.pntd.0005695 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Agbanyo A, Agbavor B, Gyabaah S, Oppong MN, Dornu O, Antwi PB, et al. Implementation of an integrated control programme for neglected tropical diseases of the skin in Ghana: The essential role of the laboratory. J Clin Tuberc Other Mycobact Dis. 2024;35:100442. doi: 10.1016/j.jctube.2024.100442 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ackam N, Akenten CW, Boadi AO, Agbanyo A, Abass KM, Amofa G, et al. Antimicrobial resistant bacteria isolated from Buruli ulcer lesions in Ghana. PLoS Negl Trop Dis. 2025;19(5):e0013140. doi: 10.1371/journal.pntd.0013140 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ackam N, Opoku-Boadi A, Agbavor B, Adjei JK, Agbanyo A, Oppong MN, et al. Bacterial diversity in Buruli ulcer lesions in Ghana. J Clin Tuberc Other Mycobact Dis. 2024;36:100468. doi: 10.1016/j.jctube.2024.100468 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Boadi AO, Akenten CW, Thye T, Ackam N, Nkrumah-Appau M, Oppong M, et al. Characterization of secondary bacterial infections in Buruli ulcer disease in Ghana: a focus on antimicrobial resistant Staphylococcus aureus. BMC Infect Dis. 2026;:10.1186/s12879-026-13986–0. doi: 10.1186/s12879-026-13986-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Cowan R, Athan E, Friedman ND, Hughes AJ, Mcdonald A, Callan P, et al. Mycobacterium ulcerans treatment – can antibiotic duration be reduced in selected patients?. PLoS Negl Trop Dis. 2015. doi: 10.1371/journal.pntd.0003503 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Nienhuis WA, Stienstra Y, Abass KM, Tuah W, Thompson WA, Awuah PC, et al. Paradoxical responses after start of antimicrobial treatment in Mycobacterium ulcerans infection. Clin Infect Dis. 2012;54(4):519–26. doi: 10.1093/cid/cir856 [DOI] [PubMed] [Google Scholar]
- 30.Frimpong M, Agbavor B, Duah MS, Loglo A, Sarpong FN, Boakye-Appiah J, et al. Paradoxical reactions in Buruli ulcer after initiation of antibiotic therapy: Relationship to bacterial load. PLoS Negl Trop Dis. 2019;13(8):e0007689. doi: 10.1371/journal.pntd.0007689 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Haigh KA, Twabi HH, Boloko L, Namale PE, Lutje V, Nevitt S, et al. Efficacy and safety of higher dose rifampicin in adults with presumed drug-susceptible tuberculosis: an updated systematic review and meta-analysis. EClinicalMedicine. 2024;77:102857. doi: 10.1016/j.eclinm.2024.102857 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Seijger C, Hoefsloot W, Bergsma-de Guchteneire I, Te Brake L, van Ingen J, Kuipers S, et al. High-dose rifampicin in tuberculosis: Experiences from a Dutch tuberculosis centre. PLoS One. 2019;14(3):e0213718. doi: 10.1371/journal.pone.0213718 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kannabiran BP, Arthanari J, Bhaskar A, Narayanan MKS, Inbaraj LR. Efficacy & safety of high-dose rifampicin in pulmonary tuberculosis: A systematic review & meta-analysis. Indian J Med Res. 2025;161(5):449–60. doi: 10.25259/IJMR_1673_2024 [DOI] [PMC free article] [PubMed] [Google Scholar]