Study setting: Auckland city has a temperate climate and is a large urban centre of 1.6 million people. Primary school health clinics serve the more socioeconomically disadvantaged areas within two of the three district health boards in the Auckland region: Auckland and Counties Manukau. These health clinics provide primary care, including free skin and throat infection management, to students aged 5–13 years.17,18 The school nurses running the clinics are invited to participate in this study to provide a potential eligible population of ~ 10,000 enrolled students. Due to their over representation of socio-economic disadvantage, over 90% of children in the schools are of Māori or Pacific island ethnicity and there is a high rate of impetigo.
Eligibility Criteria
Children meeting the eligibility criteria are identified and their caregivers contacted to explain the study and obtain consent. Caregivers must provide verbal informed consent before any study procedures occur. This is then followed by written informed consent (see Appendix 1 for sample of Informed Consent Form).
Inclusion criteria:
Children eligible for the trial must comply with all of the following prior to randomisation:
-
Enrolled in one of the participating school clinics
-
Mild-to-moderate impetigo
-
Aged 5–13 years.
Exclusion criteria:
Severe impetigo requiring oral antibiotics; defined as extensive lesions (> 3 lesions or > 5% body surface area), presence of cellulitis or fever > 38.5C
Children who are immunocompromised
Known allergy to study drugs
Current use, or use within the previous 5 days, of topical or oral antimicrobials
Commencement of antimicrobials for other reasons during the trial period
Failure to obtain informed consent for randomisation or withdrawal of consent
Excluded children continue with treatment according to the existing school health clinic standard operating procedures (SOP).
Intervention
All lesions are cleaned with saline and scabs gently removed. For the group randomised to fusidic acid, 2% fusidic acid ointment (DP Fusidic Acid, Douglas Pharmaceuticals Ltd, Auckland, NZ) is applied topically and for the hydrogen peroxide arm, 1% hydrogen peroxide cream (Crystaderm, AFT Pharmaceuticals, Auckland, NZ) is applied topically. In both cases an adequate amount to cover each lesion is used and dressing(s) then applied. A tube of appropriate topical medication is supplied for the child and/or caregivers to continue applications twice daily for 5 days with dressing changes. Participants allocated to simple hygiene measures receive no medication but a dressing is applied following the cleaning of the lesion(s). All participants are provided with supplies to allow them to clean and redress the wound twice daily for 5 days. Low adherent wound pads are used so as not to interfere with the wound healing process. In all groups, scabies is treated if present.
Modifications
All adverse effects, including pain, itch or allergy to study medication, will be reported; the study medication must be withdrawn, and the patient changed to routine treatment as per the SOP of the health clinic. If clinical deterioration while on study medication is identified by the school nurse, the study medication may be withdrawn at the nurse’s discretion. In this case the patient will be changed to routine treatment as per the standard operating procedures of the health clinic.
Adherence
Face to face reminders of adherence are provided by nurses at both day 0 and day 2 visits and adherence over the trial period will be assessed on days 2 and 7. Sticker charts are provided for each participant to encourage adherence.
Participant Timeline
On the first visit (day 0) demographic data, inclusion and exclusion criteria and verbal consent are obtained. All lesions are cleaned and the single largest lesion is photographed using a digital camera and a bacteriological swab is taken from the same lesion as the photograph. The patient is then randomised and the appropriate treatment is commenced and continued for five days. Two days after enrolment into the trial (day 2), a safety check is performed by the school nurse to assess for rapid worsening of the impetigo or for adverse effects. The safety check can be performed between days 2-4 if necessary to allow for day 2 falling on a weekend. Seven days after commencing the trial (day 7), the participant is re-assessed by the nurse. A second set of photographs and repeat bacterial swab is taken from the same lesion as originally documented. Both child and caregiver complete a verbal questionnaire. Caregivers can withdraw a participant from the study at any stage.
Sample size
On the assumption of non-inferiority between hygiene measures and topical fusidic acid and between topical hydrogen peroxide and fusidic acid, and a predicted efficacy of fusidic acid of 80%; we require 160 patients in each group. This provides 80% power and a one-sided α of 0·05 to show non-inferiority (10% margin) between each group and topical fusidic acid. In order to allow for 10% loss to follow up and subsequent exclusions, recruitment of 178 participants to each group is required. Subsequent exclusions and loss to follow up are defined as participants not available for follow up at day 7 or when two digital images are not available to assess.
Recruitment
School health teams within recruiting schools will follow existing SOPs. Community health workers attend every class 3-5 days each week during the school term and ask children to self-identify any skin infections. These children are then reviewed by the school nurse. If the lesions are confirmed as impetigo then eligibility for participation is assessed.
Allocation concealment mechanisms and implementation
Randomisation is implemented by block randomisation within each participating school, sequence allocation code was written in R.19 Children are randomly allocated (1:1:1) to topical fusidic acid, topical hydrogen peroxide or simple hygiene measures. Participating schools are randomised separately to limit bias caused by excess recruitment to any individual arm within a single school. Allocation is performed by school nurses using the pre-generated codes contained in sealed, opaque, sequentially numbered envelopes. Both the participant and school nurse are aware of treatment allocation due to the appearance of the study medications (e.g. hydrogen peroxide has a silvery sheen) or lack of study medication in the simple hygiene measure arm. However, investigators, photograph reviewers and laboratory staff are blinded to allocation.
Data Collection Methods
Demographic characteristics
Baseline characteristics of participants are collected at the time of randomisation and reported per randomisation group. The following demographic characteristics will be reported: age, gender, ethnicity, weight, history of pre-existing skin disease, allergies, and location of most severe lesion.
Primary outcome methods
The primary outcome of this trial is treatment success assessed by comparison of digital photographs taken at day 0 and 7 or clinical deterioration based either on digital images or on nursing assessment leading to discontinuing trial medication if second image is not available. To aid in standardisation, the primary outcome uses digital photographs that can be assessed retrospectively by blinded reviewers. The single largest lesion is photographed prior to commencing treatment. Using a digital camera at a distance of 15cm with an adhesive paper tape measure and unique study ID placed next to the lesion. Three digital images are taken at each data collection visit and all are submitted to the study investigators. The single best quality image is then selected by the lead investigator (SP) for outcome assessment. Pairs of images are presented to three individual assessors who are blinded to intervention arm. The method of image presentation is randomised such that assessors are unaware of which image was taken first and which second. The outcome will be treatment success if the images are considered healed or improved, and treatment failure if they are the same or worse or if the reviewers cannot determine the outcome based on the digital images. Where there is discordance between assessor opinions, the majority opinion of two of three reviewers will be considered the correct assessment. Patients removed from the trial protocol by school nurses or general practitioners before seven days due to clinical deterioration will also be considered treatment failures. This process has been standardised previously in another large RCT of treatment of impetigo in remote access locations; different from our multicentre urban setting.20
Clinical assessments
Clinical assessments are standardised and recorded on the existing skin assessment database used for school health clinics. A written description of the location of the lesions and identification of the primary lesion being used for assessment is recorded. A paper template is also provided to allow nurses the option to mark on a diagram the location of the lesions.
Seven days after commencing the trial (day 7), the participant is re-assessed by the school nurse, who also records whether the lesion has improved. Both child and caregiver are asked to comment on their satisfaction with treatment and any adverse events related to the medication, including itch, pain or allergy and school records are checked for absence over the prior seven days.
Microbiology
All children have a bacteriological dry cotton swab taken from the most severe lesion at presentation and at day 7. Swabs are cultured onto blood agar and any clinically significant growth is reported. European Committee on Antimicrobial Susceptibility Testing (EUCAST) susceptibility method and criteria are used for S. aureus susceptibility testing. Susceptibility is tested to commonly used skin and soft tissue antimicrobials; fusidic acid, flucloxacillin, erythromycin, clindamycin and co-trimoxazole. If MRSA is identified then extended susceptibility is performed, including mupirocin and tetracyclines. No susceptibility testing is performed on S. pyogenes.
Outcomes
Primary Outcome:
The primary outcomes of this trial is treatment success assessed by comparison of digital photographs or clinical deterioration based either on digital images or on nursing assessment leading to discontinuing trial medication.
Secondary outcomes
- Clinical success will be compared between group. This will be defined by
- nursing opinion that the impetigo has improved at day seven
- participant and/or caregiver opinion that the impetigo has improved at day seven
- Microbiological secondary outcomes are eradication of S. pyogenes and/or S. aureus on day 7 and development of antibiotic resistance on day seven compared to baseline
- Educational impact is assessed by comparison of school absence over the seven days of the trial period
- Adverse events will be compared across study arms.
Retention
On successful completion of the trial protocol a supermarket voucher worth NZ$20 will be offered to the family as koha; a thank you gift.
Data Management
Trial data is stored using a study identification number on a password protected access database maintained on a secure network. This database is also used to randomise images and presents anonymised pairs of photographs to the graders for analysis, and records the outcomes of the grading.
Statistical analyses
For baseline data, dichotomous variables will be summarised as proportions of patients in each treatment group, differences between groups will be assessed using a Chi-square statistic, and where small cell sizes (less than 5) are present, a Fisher’s exact test will be used. Continuous variables with an underlying normal distribution will be summarised as mean and standard deviation and differences between groups will be assessed using a Students t-test. Otherwise distributions will be either transformed if suitable and t-tests performed with reporting of geometric means. Otherwise distributions will be reported as median and interquartile range and differences in groups assessed using a Wilcoxon rank non-parametric tests with Hodges-Lehmann estimates and 95% confidence intervals.
Analysis will be performed after completion of recruitment. Both hydrogen peroxide and simple hygiene groups will be compared independently with fusidic acid with 95% confidence interval. Non-inferiority will be defined as a treatment success rate of no more than 10% below that of the fusidic acid success rate. An intention to treat and per protocol analysis will be performed using all patients with available primary outcome data. Patients without primary outcome data or for whom caregivers withdrew consent to participate will not be included in the analysis.
To investigate predictors of treatment success, backwards stepwise random-effects logistic regression will be performed on a priori and other variables identified as different in baseline characteristics between randomisation groups.
Data Monitoring
A Data Safety Monitoring Board (DSMB) has been convened. An interim safety analysis will be performed after recruitment of 150 participants. The DSMB will have unblinded access to all data and will discuss the outcome of the analysis with the trial steering committee. The unblinded data, apart from the outcome of the DSMB decision, will not be made available to the authors prior to the completion of the trial and unblinding.
Harms
Any adverse event will be reported; these are defined as any untoward medical occurrence in a subject without regard to the possibility of a causal relationship after entry into the study and until the completion of the study. At day two and day seven of the trial, a safety check is performed by the school nurse to assess for rapid worsening of the impetigo or for adverse effects. If these occur then the participant can be withdrawn from the trial at the discretion of the nurse.
Auditing
Regular visual review of the data will be performed by the lead investigator (SP) for completeness and quality of the data.