Bleach baths for atopic dermatitis: A systematic review and meta-analysis including unpublished data, Bayesian interpretation, and GRADE.

BACKGROUND: Bleach bathing is frequently recommended to treat atopic dermatitis (AD), but its efficacy and safety are uncertain. OBJECTIVE: To systematically synthesize randomized controlled trials (RCTs) addressing bleach baths for AD. METHODS: We searched MEDLINE, EMBASE, CENTRAL, and GREAT from inception to December 29, 2021, for RCTs assigning patients with AD to bleach vs no bleach baths. Paired reviewers independently and in duplicate screened records, extracted data, and assessed risk of bias (Cochrane version 2) and GRADE quality of evidence. We obtained unpublished data, harmonized individual patient data and did Frequentist and Bayesian random-effects meta-analyses. RESULTS: There were 10 RCTs that enrolled 307 participants (median of mean age 7.2 years, Eczema Area Severity Index baseline mean of means 27.57 [median SD, 10.74]) for a median of 6 weeks (range, 4-10). We confirmed that other trials registered globally were terminated. Bleach baths probably improve AD severity (22% vs 32% improved Eczema Area Severity Index by 50% [ratio of means 0.78, 95% credible interval 0.59-0.99]; moderate certainty) and may slightly reduce skin Staphylococcal aureus colonization (risk ratio, 0.89 [95% confidence interval, 0.73-1.09]; low certainty). Adverse events, mostly dry skin and irritation, along with itch, patient-reported disease severity, sleep quality, quality of life, and risk of AD flares were not clearly different between groups and of low to very low certainty. CONCLUSION: In patients with moderate-to-severe AD, bleach baths probably improve clinician-reported severity by a relative 22%. One in 10 will likely improve severity by 50%. Changes in other patient-important outcomes are uncertain. These findings support optimal eczema care and the need for additional large clinical trials. TRIAL REGISTRATION: PROSPERO Identifier: CRD42021238486.

Responsiveness to Change of the Morphea Activity Measure in Pediatric Patients.

Importance: Detecting activity of morphea can be complex but is crucial for adequate treatment and outcome assessment. The Morphea Activity Measure (MAM) was recently validated, but its responsiveness to change in disease activity has not been studied. Objective: To evaluate the internal and external responsiveness of MAM to changes in disease activity in pediatric patients. Design, Setting, and Participants: This multicenter prospective, longitudinal prognostic study was performed from October 2021 to January 2023 at 4 pediatric referral centers in North America. Consecutive pediatric patients with morphea who were available for data collection at baseline and at a follow-up visit at least 3 months later were studied. Exposure: Patient demographics, clinical characteristics, and measurements of disease activity collected at baseline and the subsequent visit. Main Outcome and Measures: Responsiveness of MAM to disease activity according to the modified Localized Scleroderma Severity Index (mLoSSI), the Physician Global Assessment (PGA), and a patient and parent global assessment (PtGA) was analyzed using mean and percentage change, standardized effect size, and standardized response mean (SRM) from baseline to follow-up 3 or more months later. Differences between patients whose activity improved vs did not improve were evaluated using the Mann-Whitney U test. The correlation between percentage change in MAM score and mLoSSI, the PGA, and the PtGA was calculated using Spearman rank correlation. Results: A total of 43 patients (mean [SD] age at onset, 7.11 [3.18] years; 26 [60.5%] female) were included. The mean change and percentage change in MAM score were significantly larger in those whose disease activity improved by the PGA (mean: -18.75 [95% CI, -31.92 to -5.57] vs 2.73 [95% CI, -1.97 to 7.45]; percentage: -108.08% [95% CI, -155.21% to -60.95%] vs -24.11% [95% CI, -81.22% to 32.99%]) and by mLoSSI (mean: -24.15 [95% CI, -41.89 to -6.41] vs -1.30 [95% CI, -8.50 to 5.70]; percentage: -172.06% [95% CI, -263.68% to -80.45%] vs -21.57% [95% CI, -48.13% to 4.97%]) than in those whose activity did not change. The SRM of MAM was significantly different between groups for both measures; the responsiveness was large in those whose activity decreased by the PGA (-0.75 [95% CI, -1.29 to -0.22]) and mLoSSI (-0.97 [95% CI, -1.69 to -0.25]) and none to small in those whose activity did not change by the PGA (0.11 [95% CI, -0.08 to 0.30]) or mLoSSI (-0.05 [95% CI, -0.34 to 0.23]). Percentage change in MAM score correlated strongly and significantly with change in mLoSSI (ρ = 0.69; P < .001) and PGA (ρ = 0.65; P < .001), but there was no correlation with change in the PtGA (ρ = 0.26; P = .09). Conclusions and Relevance: In this prognostic study, MAM was found to be internally and externally responsive to changes in disease activity. Further evaluation in mixed cohorts of all ages and specialties is needed.