Abstract
Objective:
We hypothesized that intestinal barrier impairment is a key pathophysiological feature in AD and that the degree of baseline barrier dysfunction, reflected by serum D-lactate levels, predicts the clinical response to Washed Microbiota Transplantation (WMT). This study aimed to test these hypotheses by investigating the association between intestinal barrier biomarkers and AD severity, and their correlation with WMT outcomes.
Methods:
We compared intestinal barrier biomarkers (D-lactate, endotoxin, and diamine oxidase) between 24 AD patients and 23 healthy donors. Additionally, we evaluated the clinical outcomes of 14 AD patients who underwent WMT therapy.
Results:
AD patients exhibited significantly elevated intestinal barrier biomarkers compared to healthy donors (p < 0.01). Following WMT, significant improvements were observed in SCORAD, EASI, and NRS scores (p < 0.05). In exploratory, uncorrected analyses, baseline D-lactate levels showed a significant negative correlation with improvements in SCORAD (R = −0.738, p = 0.037) and NRS scores (R = −0.650, p = 0.012), suggesting that higher pre-treatment levels might predict greater symptom relief. Microbiota analysis revealed a increase in Acidaminococcus and decreases in Ruminococcus_gnavus_group, Flavonifractor, and Norank_f_Oscillospiraceae following WMT.
Conclusion:
This study confirms significant intestinal barrier dysfunction in AD and demonstrates the potential clinical efficacy of WMT. The strong, uncorrected correlations suggest that pre-treatment D-lactate level warrants further investigation as a candidate biomarker for predicting WMT response. The clinical benefits occurred alongside a restructuring of the gut microbiota.
1 Introduction
Atopic dermatitis (AD) is a chronic, relapsing inflammatory skin disease characterized by intense pruritus, eczematous lesions, and a substantial impairment of quality of life (1). The pathogenesis of AD is complex, involving skin barrier dysfunction, immune dysregulation, and as increasingly recognized microbial dysbiosis (2, 3).
While historical research focused on the cutaneous microbiome, a growing body of evidence underscores the pivotal role of the gut microbiota and the bidirectional communication along the gut-skin axis in AD pathophysiology (4, 5). Patients with AD exhibit distinct gut microbial dysbiosis, including alterations in the abundance of genera such as Bifidobacterium, Lactobacillus, and Faecalibacterium (6, 7). This dysbiosis is believed to contribute to systemic inflammation and immune deviation, thereby influencing skin health (8). Supporting this concept, interventions targeting the gut microbiome, such as probiotics and prebiotics, have demonstrated efficacy in alleviating AD symptoms (9).
Fecal microbiota transplantation (FMT) represents a more comprehensive approach to restoring gut microbial homeostasis. Washed Microbiota Transplantation (WMT) is a novel concept and technique derived from FMT. Unlike traditional FMT processes, which rely on manual preparation of fecal samples, WMT automatically enriches microbiota from feces while incorporating a “washing” step. Studies have shown that this washing process removes certain bacterial fragments, pro-inflammatory metabolites, soluble molecules, and viruses, significantly reducing the probability of adverse events such as mucosal barrier damage and fever in patients (10, 11). Preliminary evidence, including work from our team, suggests that WMT can yield clinical benefits in AD patients (12). A critical component of the gut-skin axis is the intestinal barrier, which prevents the translocation of microbes and their products into systemic circulation. The integrity of this barrier can be assessed non-invasively by circulating biomarkers, including D-lactate, diamine oxidase (DAO), and endotoxin, which are elevated in states of increased intestinal permeability (13–15). Intestinal barrier dysfunction has been implicated in AD, yet its relationship with clinical severity and, more importantly, its role in predicting response to microbiota-targeted therapies remain poorly understood (16, 17).
Despite these advances, a significant translational gap persists. Currently, there are no validated biomarkers to predict which AD patients are most likely to respond to interventions like WMT, hindering the development of personalized treatment strategies. Furthermore, the specific impact of WMT on intestinal barrier function in AD has not been systematically evaluated.
Based on this evidence, we posited that intestinal barrier dysfunction is a prevalent and mechanistically important feature in AD patients. Furthermore, we proposed a novel therapeutic hypothesis: the very state of increased intestinal permeability, while detrimental in the chronic phase of AD, might paradoxically facilitate the initial efficacy of WMT by allowing beneficial microbial metabolites derived from the transplanted microbiota to more readily access systemic circulation and exert immunomodulatory effects. Therefore, we specifically hypothesized that: 1. AD patients would exhibit elevated serum levels of intestinal barrier biomarkers (D-lactate, endotoxin, DAO) compared to healthy donors. 2. Higher pre-treatment levels of these biomarkers, particularly D-lactate, would be associated with a greater clinical response to WMT.
2 Subjects and methods
2.1 Research objective
This study comprised two parts. The first part aimed to compare the biochemical markers of intestinal barrier function between AD patients and healthy donors. The second part was an interventional study to evaluate the efficacy of washed microbiota transplantation (WMT) in AD patients over a 4-week period. The pre-specified primary endpoint was the change in the SCORAD score. Secondary endpoints included changes in EASI, NRS, and DLQI scores; alterations in intestinal barrier biomarkers (D-lactate, endotoxin, diamine oxidase); and shifts in gut microbiota composition. Fecal samples were collected before (W0) and 4 weeks after (W1) WMT for microbiota analysis. The established Minimal Clinically Important Difference (MCID) thresholds were: SCORAD ≥8.7, EASI ≥6.6, NRS ≥4, and DLQI ≥4 (18–20).
2.2 Study participants
The study protocol was reviewed and approved by the Ethics Committee of The First Affiliated Hospital of Guangdong Pharmaceutical University [Approval No. [2023] IIT (12)]. Written informed consent was obtained from all participants or their legal guardians (for minor participants) prior to their enrollment in the study.
Study 1: From January 2021 to April 2024, a total of 48 participants under the age of 30 were included in the study: 24 AD patients and 23 fecal microbiota transplantation donors (healthy controls) recruited from the First Affiliated Hospital of Guangdong Pharmaceutical University between January 2021 and April 2024.
Study 2: 14 AD patients who underwent washed microbiota transplantation in the First Affiliated Hospital of Guangdong Pharmaceutical University during the same period were continuously included in the study.
Exclusion criteria: participants with severe gastrointestinal organic lesions such as Crohn's disease, gastrointestinal malignant tumors, no serious psychosomatic diseases such as malignant tumors, autoimmune diseases and other serious systemic organic lesions and mental diseases. Additionally, participants who had used antibiotics or probiotics within the 4-week period preceding the intestinal barrier testing or during the 4-week period surrounding WMT treatment (both pre- and post-procedure) were excluded.
2.3 Research methods
2.3.1 Detection method of biochemical markers of intestinal barrier function
Measurement of intestinal barrier biochemical markers was performed within two days before and four weeks after microbiota transplantation therapy in AD patients. The Intestinal Barrier Function Biochemical Indicator Analysis System (model: JY-DLT) and the corresponding assay kits, manufactured by Beijing Zhongsheng Jinyu Diagnostic Technology Co., Ltd., were used for the detection. The Intestinal Barrier Function Biochemical Indicator Analysis System (model: JY-DLT) and corresponding assay kits, manufactured by Beijing Zhongsheng Jinyu Diagnostic Technology Co., Ltd., were used for the detection. According to the analytical performance data provided by the manufacturer, the assay parameters for each biomarker were as follows: diamine oxidase had a linear range of 5–80 U/L with a limit of blank (LoB) ≤4 U/L; D-lactate had a linear range of 2–50 mg/L with a LoB ≤1 mg/L; endotoxin had a linear range of 10–160 U/L with a LoB ≤8 U/L. The coefficient of variation (CV) for all assays was ≤15%. According to the manufacturer's instructions for the assay kits, the normal reference values were defined as: DAO ≤10 U/L, D-lactate ≤15 mg/L (corrected from U/L, assuming typo based on Table 1), and lipopolysaccharide (endotoxin) ≤20 U/L (21–23).
Table 1
| Characteristic | AD patient (n = 24) | Donor (n = 23) | AD patients receiving WMT (n = 14) |
|---|---|---|---|
| Age (years) | 21.16 ± 4.57 | 17.04 ± 7.96 | 26.85 ± 17.18 |
| Gender (male) | 19 (79.2%) | 14 (60.8%) | 10 (71.4%) |
| Nausea or/and vomiting | 0 | 0 | 0 |
| Decreased or absent bowel sounds | 0 | 0 | 0 |
| Diarrhea | 0 | 0 | 0 |
Demographic characteristics.
The process was as follows: 1. 4 mL peripheral venous blood was collected after fasting for 8 h; 2. The blood samples were placed in a low temperature refrigerator at 4–6 °C, and the conventional standing time was not more than 4 h; 3. DAO, D-lactate and lipopolysaccharide were detected according to the standard process provided by the manufacturer. Patient baseline information and clinical symptoms and signs within 48 h prior to testing were recorded. The report form was automatically generated by the computer.
2.3.2 Donor screening and washed microbiota transplantation (WMT) procedure
In this study, we defined the donor inclusion criteria for this study according to the Nanjing Consensus on Washed Microbiota Transplantation. All fecal microbiota donors were rigorously screened and selected in strict accordance with the Nanjing Consensus on Methodology of Washed Microbiota Transplantation through a multi-stage screening process to exclude individuals at risk of infectious diseases or microbiota-associated disorders. The screening process included initial exclusion of candidates with relevant medical histories and lifestyle risk factors through questionnaire assessment, followed by interview screening conducted by qualified physicians to further evaluate psychological status and potential risks, and comprehensive laboratory testing completed within 3 weeks prior to donation. The laboratory testing encompassed blood tests (screening for human immunodeficiency virus, hepatitis B and C viruses, Epstein–Barr virus, cytomegalovirus, and syphilis) and stool tests (including Clostridioides, difficile, Salmonella, Shigella, Campylobacter and other enteric pathogens). Additionally, a rapid questionnaire was administered on each donation day to exclude any temporary risk factors. Only donors who successfully passed all stages of this screening protocol were deemed eligible for donation.
The treatment procedure for washed microbiota transplantation comprised three main stages: preparation of the bacterial suspension, preparation of the intestinal catheter, and administration of the suspension. First, screened qualified donor feces were mixed with sterile saline in a 1:5 (feces:saline) volume ratio, thoroughly stirred, and allowed to stand for 5 min. The mixture was then filtered through a microporous filter membrane and aliquoted. Subsequently, the sub-packed samples were centrifuged and purified: the centrifuge was run at 2,500 rpm for 3 min at room temperature, and 30–40 mL of sterile saline was added to re-suspend after the clarification liquid layer was discarded. This centrifugation-washing process requires two cycles of operation to complete the sample preparation. After the above samples were washed three times, the supernatant was removed, and the precipitated fecal bacteria and sterile saline were dissolved and diluted in a 1:1 volume ratio and mixed to obtain the bacterial suspension. The bacterial suspension was then administered into the patient's intestinal cavity via a pre-placed digestive tract catheter (upper or lower). Each infusion delivered 120 mL of the bacterial solution. A single course of intestinal microbiota treatment consisted of one infusion per day for three consecutive days.
2.3.3 Statistical analysis
In this study, SPSS 27.0 software was used for statistical analysis. The normality of all continuous data was assessed using the Shapiro–Wilk test. Since the majority of the biochemical marker data and clinical scores were found to be non-normally distributed (p < 0.05 in Shapiro–Wilk test) and/or due to the small sample sizes which limit the reliability of normality assumptions, we chose to use non-parametric statistics and present continuous data as median (interquartile range, IQR). The median and IQR are robust measures of central tendency and dispersion that are not influenced by extreme values or skewed distributions, providing a more accurate representation of our data. The 95% confidence intervals (95% CI) for median differences in all non-parametric tests were estimated using the Hodges–Lehmann method, and “estimated by Hodges–Lehmann method” was clearly indicated in result descriptions and tables to avoid only labeling “95% CI”.
Accordingly, the Mann–Whitney U test was used to assess differences in biochemical indexes between donors and AD patients. In addition to p values, effect sizes were calculated to quantify the magnitude of the observed differences or associations. For the Mann–Whitney U test, the effect size r was calculated as r = Z/√N, where Z is the test statistic and N is the total sample size. For the Wilcoxon signed-rank test, the effect size r was calculated as r = Z/√N, where Z is the test statistic and N is the number of paired observations. Effect sizes were interpreted according to Cohen's criteria: r ≈ 0.10 indicates a small effect, r ≈ 0.30 a medium effect, and r ≈ 0.50 a large effect.
The Wilcoxon signed-rank test was used to evaluate changes in intestinal barrier biomarkers and dermatitis scores before and after WMT treatment. Spearman's rank correlation test was used for correlation analyses, and the 95% CI of its correlation coefficient (rs) was calculated using the BCA Bootstrap method with 1,000 resamples to ensure the robustness of results. A p value less than 0.05 was considered statistically significant. To control for multiple comparisons, a stratified false discovery rate (FDR) correction was applied using the Benjamini-Hochberg procedure within three pre-defined analytical families: (1) the pathophysiology family (3 tests: group comparisons of the three intestinal barrier biomarkers between AD patients and healthy donors), (2) the intervention efficacy family (7 tests: pre- vs. post-WMT comparisons of the four clinical scores and the three biomarkers), and (3) the predictive exploration family (24 tests: all correlation analyses between baseline measures and treatment outcomes).The FDR was controlled independently within each family.
2.3.4 Microbiota sequencing and analysis
Gut microbiota sequencing and analysis were completed by Majorbio Bio-Pharm Technology Co., Ltd. The specific procedures were as follows: First, genomic DNA was extracted, and the extraction quality was verified using 1% agarose gel electrophoresis. Subsequently, specific primers with barcodes were synthesized according to the designated sequencing region. Low-cycle amplification was performed using TransStart Fastpfu DNA Polymerase on an ABI GeneAmp 9,700 PCR instrument, with three technical replicates per sample followed by pooling of the products to minimize amplification bias. The amplification products underwent gel extraction recovery, fluorescence quantification, and were then mixed in proportional ratios for library construction before being sequenced on the Illumina Nextseq 2,000 platform. Following demultiplexing of the sequenced samples, paired-end reads were subjected to quality control, filtering, and assembly based on sequencing quality. The optimized data were subsequently processed using sequence denoising methods such as DADA2 to obtain Amplicon Sequence Variant (ASV) representative sequences and abundance information, followed by taxonomic annotation. To control for variations in sequencing depth across samples and enable comparative analysis of community diversity, all sample data were normalized to the same sequencing depth using rarefaction.
Analysis of the sequenced sample data was conducted as follows: Principal Coordinate Analysis (PCoA) based on the abund_jaccard distance matrix at the phylum and genus levels was performed to visualize inter-sample differences, and Analysis of Similarities (ANOSIM) with 999 permutations was used to assess the statistical significance of intergroup differences, with a p-value less than 0.05 considered statistically significant. The relative abundance of ASV at the phylum and genus levels was determined using the Naive Bayes algorithm, and bar charts and pie charts were generated to visualize the microbial community composition of the two groups (W0 vs. W1). Additionally, to identify bacterial genera showing significant differences before and after WMT treatment, the Wilcoxon signed-rank test was employed for species difference analysis at the genus level.
3 Results
3.1 Demographic characteristics 48 h prior to intestinal biochemical index testing
The donor group comprised 10 females (41.6%) and 14 males (58.4%). The AD patient group comprised 5 females (20.8%) and 19 males (79.2%). The mean age of donors was 21.16 ± 4.57 years, and the mean age of AD patients was 17.04 ± 7.96 years. The mean age of AD patients receiving WMT was 26.85 ± 17.18 years. None of the participants exhibited any relevant gastrointestinal symptoms, including diarrhea, nausea, vomiting, or abnormal bowel sounds, before the detection (Table 1).
3.2 Differences in biochemical indexes of intestinal barrier function between AD patients and healthy donors
Serum levels of D-lactate, endotoxin, and diamine oxidase were significantly elevated in AD patients compared to healthy donors (all p < 0.001), with large effect sizes (r = 0.643, 0.646, and 0.724, respectively). This confirms that AD patients have substantial intestinal barrier dysfunction (Table 2).
Table 2
| Biomarker | Group | Median (IQR) | Hodges–Lehmann Median Difference (95% CI) | Z value | Effect size (r) | p Value | FDR q-value |
|---|---|---|---|---|---|---|---|
| D-lactate (mg/L) | Donor | 5.91 (3.33, 9.59) | −22.5 (−27.53, −8.09) | −4.724 | 0.689 | 2.30 × 10−6 | 2.30 × 10−6 |
| AD patient | 31.44 (12.93, 35.90) | ||||||
| Endotoxin (U/L) | Donor | 5.99 (3.52, 11.77) | −12.83 (−16.04, −10.00) | −5.219 | 0.761 | 1.79 × 10−7 | 5.37 × 10−7 |
| AD patient | 17.26 (13.18, 23.77) | ||||||
| Diamine oxidase (U/L) | Donor | 4.28 (1.10, 8.60) | −13.65 (−17.30, −10.07) | −4.852 | 0.707 | 1.22 × 10−6 | 1.83 × 10−6 |
| AD patient | 18.17 (13.18, 24.87) |
Comparison of AD patients under 30 years old with donors (n = 47).
(n = 47, Donor n = 23, AD patient n = 24) Group comparisons were performed using the Mann–Whitney U test. This family of tests (pathophysiology) comprised 3 hypotheses. All P-values remained statistically significant after False Discovery Rate (FDR) correction for these 3 tests (all q < 0.001).The normal reference ranges of D-lactate, endotoxin and diamine oxidase were <10 mg/L, <15 U/L and <20 U/L, respectively.
3.3 Comparison of biochemical indexes of intestinal barrier function in AD patients before and after one course of WMT treatment
No significant differences were observed in the biochemical indexes of intestinal barrier function in AD patients before and after a single course of WMT treatment (all p > 0.05; Table 3). The changes in diamine oxidase (p = 0.124, r = 0.411) and D-lactate (p = 0.096, r = 0.445) showed medium effect sizes, while the change in endotoxin was negligible (p = 0.826, r = 0.059). These results suggest that a single course of WMT did not produce significant short-term improvement in intestinal barrier function biomarkers (Table 3, Figure 1).
Table 3
| Biomarker | Time point | Median(IQR) | Hodges–Lehmann Median Difference (95% CI) | Z value | Effect size (r) | p Value | FDR q-value |
|---|---|---|---|---|---|---|---|
| D-lactate (mg/L) | W0 | 15.42 (8.13, 27.19) | 6.49 (−0.97, 12.26) | −1.663 | 0.445 | 0.096 | 0.168 |
| W1 | 20.98 (16.10, 30.50) | ||||||
| Endotoxin (U/L) | W0 | 15.19 (14.70, 21.87) | 0.85 (−4.97, 7.25) | −0.219 | 0.059 | 0.826 | 0.174 |
| W1 | 17.92 (16.48, 24.00) | ||||||
| Diamine oxidase (U/L) | W0 | 17.47 (14.70, 22.66) | 2.69 (−1.17, 7.37) | −1.538 | 0.411 | 0.124 | 0.826 |
| W1 | 17.25 (15.67, 23.27) |
Comparison of biochemical indexes of intestinal barrier function in AD patients before (W0) and four weeks after WMT treatment (W1).
n = 14. W0:before treatment; W1:4 weeks after treatment. The Wilcoxon signed-rank test was used for paired comparisons. The tests in this table are part of the pre-specified “intervention efficacy family”, which comprised 7 hypotheses in total (3 biomarkers 4 clinical scores). None of the changes reached statistical significance after FDR correction for these 3 tests (all q > 0.05).
Figure 1
Serum levels of intestinal barrier biomarkers in patients with atopic dermatitis (AD) before and after washed microbiota transplantation (WMT). Paired measurements were taken at baseline (W0) and 4 weeks after WMT (W1). Individual data points are shown with lines connecting paired measurements for each patient. The biomarkers assessed were (A) D-lactate, (B) endotoxin, and (C) diamine oxidase (DAO)
3.4 The changes of rash and AD score in AD patients before and after the treatment of washed microbiota transplantation
Following WMT treatment, significant improvements were observed in SCORAD (p = 0.012, r = 0.891) and EASI (p = 0.018, r = 0.836) scores, with large effect sizes indicating robust improvements in clinical signs. The NRS score also decreased significantly (p = 0.030), albeit with a modest effect size (r = 0.581). In contrast, the improvement in the Dermatology Life Quality Index (DLQI) did not reach statistical significance (p = 0.410) and was associated with a small effect size (r = 0.220; Table 4, Figure 1). These results indicate that while AD severity, skin lesion area, and pruritus were significantly improved after four weeks of WMT, the short-term, patient-reported quality of life did not show a commensurate, statistically significant improvement (Table 4; Figures 2 and 3).
Table 4
| Score | Timepoint | Median (IQR) | Hodges–Lehmann Median Difference (95% CI) | Z value | Effect size (r) | p Value | FDR q-value |
|---|---|---|---|---|---|---|---|
| NRS (n = 14) | W0 | 7.50 (4.75, 9.75) | −1.50(−3.00, −0.50) | −2.173 | 0.581 | 0.030 | 0.105 |
| W1 | 4.50 (1.25, 7.75) | ||||||
| DLQI (n = 14) | W0 | 12.00 (7.00, 21.25) | −5.55(−11.30, −0.150) | −0.824 | 0.220 | 0.410 | 0.478 |
| W1 | 7.00 (3.50, 17.25) | ||||||
| SCORAD (n = 8) | W0 | 68.75 (57.90, 76.32) | −11.67(−22.45, −1.15) | −2.520 | 0.891 | 0.012 | 0.084 |
| W1 | 55.25 (39.42, 70.20) | ||||||
| EASI (n = 8) | W0 | 22.50 (14.25, 41.57) | −1.50(−6.50, 3.00) | −2.028 | 0.717 | 0.043 | 0.100 |
| W1 | 19.95 (11.12, 29.00) |
The changes of AD score before (W0) and four weeks after WMT treatment (W1).
W0: before treatment; W1: 4 weeks after treatment. SCORAD. The Wilcoxon signed-rank test was used for paired comparisons. The tests in this table are part of the pre-specified “intervention efficacy family”, which comprised 7 hypotheses in total (3 biomarkers in Tables 3 and 4 clinical scores in this table).
Figure 2
Clinical scores of patients with atopic dermatitis (AD) before and after washed microbiota transplantation (WMT).Paired measurements were taken at baseline (W0) and 4 weeks after WMT (W1). Individual data points are shown with lines connecting paired measurements for each patient. The clinical scores assessed were (A) NRS (B) (DLQI) (C) SCORAD (D) EASI.
Figure 3
The changes of skin lesions in AD patients before and after WMT treatment. (A) Treatment of forehead rash. (B) Forehead rash after treatment.
3.5 Correlation analysis between intestinal barrier function and dermatitis score in AD patients
No significant correlations were found between pre-WMT levels of D-lactate, endotoxin, or diamine oxidase and pre-treatment SCORAD, EASI, NRS, or DLQI scores in AD patients (all p > 0.05; Table 5). It is important to note that these analyses, part of an exploratory family of tests, did not survive correction for multiple comparisons (FDR-adjusted p > 0.05 for all).
Table 5
| Biomarker before WMT | AD Score before WMT | Spearman's ρ (95% CI) | p Value | FDR q-value |
|---|---|---|---|---|
| D-lactate (mg/L) | NRS (n = 14) | −0.118 (0.468, 0.615) | 0.688 | 1.000 |
| DLQI (n = 14 | −0.042 (−0.752, 0.506) | 0.887 | 0.967 | |
| SCORAD (n = 8) | 0.167 (−0.644, 0.778) | 0.693 | 1.000 | |
| EASI (n = 8) | 0.071 (−0.797, 0.848) | 0.867 | 1.000 | |
| Endotoxin (U/L) | NRS (n = 14) | −0.254 (0.751, 0.357) | 0.381 | 1.000 |
| DLQI (n = 14) | −0.223 (−0.808, 0.408) | 0.444 | 1.000 | |
| SCORAD (n = 8) | −0.2149 (−0.776, 0.468) | 0.610 | 1.000 | |
| EASI (n = 8) | −0.429 (−0.975, 0.640) | 0.289 | 1.000 | |
| Diamine oxidase (U/L) | NRS (n = 14) | 0.145 (−0.468, 0.615) | 0.621 | 1.000 |
| DLQI (n = 14 | 0.053 (−0.560, 0.629) | 0.857 | 1.000 | |
| SCORAD (n = 8) | 0.048 (−0.772, 0.747) | 0.911 | 0.950 | |
| EASI (n = 8) | 0.071 (−1.000, 0.722) | 0.867 | 0.990 |
Correlation analysis between intestinal barrier function before WMT and AD score.
Note. W0: before treatment; W1: 4 weeks after treatment. Note. The tests in this table are part of the pre-specified “predictive exploration family”, which comprised 24 hypotheses in total.
3.6 Analysis of the correlation between intestinal barrier function and the efficacy of WMT in the treatment of AD
In an exploratory analysis of potential predictive biomarkers, we found that pre-treatment D-lactate levels showed a strong, uncorrected correlation with the changes in NRS and SCORAD scores. Specifically, a significant negative correlation was observed between pre-treatment D-lactate level and the NRS score change (r = −0.650, p = 0.012). Similarly, a significant negative correlation was found with the SCORAD score change (r = −0.738, p = 0.037). However, after applying a stratified False Discovery Rate (FDR) correction to the entire family of 24 exploratory correlation tests, these associations were no longer statistically significant. Therefore, these findings, while internally consistent and hypothesis-generating, cannot be considered definitive and require validation in a larger, independent cohort. No other significant correlations, either uncorrected or after FDR correction, were observed between baseline biomarkers and clinical score changes (Table 6).
Table 6
| Biomarker before WMT | Score difference after WMT | Spearman's ρ (95% CI) | p Value | FDR q-value |
|---|---|---|---|---|
| D-lactate (mg/L) | NRS (n = 14) | −0.650 | 0.012 | 0.288 |
| DLQI (n = 14 | −0.286 | 0.321 | 1.000 | |
| SCORAD (n = 8) | −0.738 | 0.037 | 0.444 | |
| EASI (n = 8) | 0.452 | 0.260 | 1.000 | |
| Endotoxin (U/L) | NRS (n = 14) | −0.069 | 0.814 | 1.000 |
| DLQI (n = 14 | 0.236 | 0.417 | 1.000 | |
| SCORAD (n = 8) | −0.190 | 0.651 | 1.000 | |
| EASI (n = 8) | 0.500 | 0.207 | 1.000 | |
| Diamine oxidase (U/L) | NRS (n = 14) | −0.184 (−0.853, 0.435 | 0.528 | 1.000 |
| DLQI (n = 14 | 0.026 (−0.726, 0.720) | 0.929 | 0.929 | |
| SCORAD (n = 8) | −0.310 (−0.921, 0.782) | 0.456 | 1.000 | |
| EASI (n = 8) | 0.095 (−1.000 ,0.926) | 0.823 | 1.000 |
Correlation analysis between intestinal barrier function before WMT and AD score difference after WMT treatment.
The tests in this table are part of the pre-specified “predictive exploration family”, which comprised 24 hypotheses in total.Score difference = 4 weeks after treatment (W1)-before treatment (W0).
3.7 Analysis results of intestinal microbiota
Fecal samples collected before (W0) and after four weeks of WMT treatment (W1) from 11 AD patients underwent bacterial sequencing and were analyzed at the phylum and genus levels. As it was difficult to discern a unified change trend for individual bacterial species (Figure 4). PCoA at the phylum level indicates a significant difference in microbial community structure before and after WMT treatment (R = 0.2363, p = 0.003), while the difference at the genus level is not significant (R = −0.0197, p = 0.533) (Figure 5).Group comparisons (W0 vs. W1) were used to describe overall microbiota changes. At the phylum level (Figure 6), the relative abundances of Bacteroidota, Proteobacteria, and Fusobacteriota showed an increasing trend, rising from baseline values of 39.24%, 4.08%, and 2.45% to 43.01%, 5.14%, and 5.77%, respectively. In contrast, the abundances of Firmicutes and Actinobacteria decreased significantly, declining from 49.29% and 4.28% to 42.64% and 2.69%, respectively. At the genus level (Figure 7), several key alterations were observed. Most strikingly, the abundance of Prevotella increased markedly from 4.89% to 12.91%. Fusobacterium also increased from 2.45% to 5.77%, and Escherichia-Shigella from 0.94% to 2.47%. In contrast, the core genus Bacteroides decreased from 30.25% to 26.36%. Several other genera exhibited decreased abundances: Ruminococcus_gnavus_group declined from 2.36% to 0.57%, Lachnospira from 2.58% to 1.49%, Blautia from 2.90% to 2.06%, and Bifidobacterium from 3.29% to 1.40%. Notably, the beneficial genera Faecalibacterium and Roseburia, known as short-chain fatty acid producers, remained relatively stable, with Faecalibacterium showing a slight decrease from 10.58% to 9.64% and Roseburia from 3.29% to 3.16%.
Figure 4
Bar chart showing bacterial composition at the genus level in AD patients before (W0) and after (W1) WMT treatment.
Figure 5
PCoA at the phylum (A) and genus (B) level. The percentage in parentheses on each axis label (PC1/PC2) represents the proportion of total variation explained by that principal coordinate. A higher percentage indicates that the axis is more important in distinguishing sample differences. The “R” value is derived from PERMANOVA testing, reflecting the extent to which predefined groups explain the observed community variation. A “P” value less than 0.05 indicates statistically significant differences between groups.
Figure 6
Pie charts depicting microbiota composition at the phylum level in AD patients (A) before (W0) and (B) after (W1) WMT treatment.
Figure 7
Pie charts depicting microbiota composition at the genus level in AD patients (A) before (W0) and (B) after (W1) WMT treatment.
Differential abundance analysis at the genus level demonstrated significant alterations in the gut microbiota after WMT, characterized by an increase in Acidaminococcus and decreases in Ruminococcus_gnavus_group, Flavonifractor, and Norank_f_Oscillospiraceae (Figure 8).
Figure 8
Genus-level differential species analysis.the left panel displays the relative abundance of bacterial genera at baseline (W0) and four weeks after WMT treatment (W1), while the right panel shows the 95% confidence intervals for the differences in abundance. Statistical analysis was performed using the Wilcoxon signed-rank test, with a p-value less than 0.05 considered statistically significant.
4 Discussion
Our study demonstrates that patients with AD exhibit significant intestinal barrier dysfunction and that WMT can effectively improve its clinical symptoms. The most novel finding is the strong correlation between pre-treatment intestinal permeability, reflected by serum D-lactate levels, and the clinical efficacy of WMT, suggesting its potential as a predictive biomarker. However, it is critical to emphasize that this correlation, while strong in uncorrected analysis, did not survive strict False Discovery Rate (FDR) correction for multiple comparisons in our exploratory analysis. Therefore, it must be strictly interpreted as a preliminary, hypothesis-generating signal that requires validation in larger, dedicated cohorts.
The baseline barrier dysfunction in these AD patients likely stems from a known pathophysiological cycle: characteristic Th2 immune deviation (e.g., IL-4/IL-13 overexpression) can directly inhibit the synthesis of tight junction proteins like Claudin-1 (24, 25), while other inflammatory mediators may exacerbate epithelial damage (26). We hypothesize that the early clinical benefits of WMT are mediated through rapid immunomodulation rather than immediate structural barrier repair. The pre-existing state of increased intestinal permeability might facilitate the translocation of beneficial microbial metabolites derived from the newly transplanted microbiota into systemic circulation, thereby directly influencing systemic inflammation and skin immunity. This could mechanistically explain why patients with higher baseline D-lactate levels experienced greater symptom improvement in our uncorrected analysis. This concept is supported by studies showing that SCFAs, such as butyrate, can suppress pro-inflammatory pathways and inhibit NF-κB, even in the absence of barrier restoration (27, 28).
Concurrent with clinical improvement, our genus-level differential abundance analysis revealed a functionally coherent restructuring of the gut ecosystem, moving towards a profile with enhanced barrier-protective and anti-inflammatory potential. A pivotal finding was the significant decrease in Ruminococcus_gnavus_group, a genus recognized for its ability to degrade intestinal mucin, a process that can compromise gut barrier integrity and activate immune responses (29). Its abundance is notably elevated in infants with allergic diseases and correlates with inflammation precursors (30).The reduction of this pro-inflammatory, barrier-disrupting genus post-WMT likely represents a crucial step in mitigating the gut-skin axis of inflammation in AD.
This suppression of a detrimental taxon was strategically accompanied by a beneficial enrichment. We observed a marked increase in Acidaminococcus, which can degrade proteins and amino acids to generate SCFAs, notably propionate and butyrate. These metabolites can inhibit the NF-κB pathway, reduce pro-inflammatory cytokines such as IL-6, and promote the secretion of anti-inflammatory IL-10 (31). The co-occurrence of these changes—the suppression of a barrier-disrupting, pro-inflammatory genus (R. gnavus) alongside the enrichment of a SCFA-producing genus (Acidaminococcus)—suggests that WMT orchestrates a targeted restoration of gut ecological function.
The treatment also induced a significant reduction in Flavonifractor, a genus whose role in AD is complex and not yet fully understood (32). Its decrease in our cohort may indicate a beneficial shift away from a dysbiotic state. Similarly, the decline in Norank_f_Oscillospiraceae, a common yet poorly characterized member of the Firmicutes phylum, adds another layer to this ecological restructuring. While its functional contribution remains unclear, its reduction alongside other pro-inflammatory Firmicutes points to a broad, potentially beneficial reorganization within this phylum. The net functional outcome of this new ecosystem is likely more critical than the change in any single genus, potentially creating a microbial community with an enhanced capacity to produce anti-inflammatory metabolites and a reduced capacity to provoke immune activation.
The interpretations of our study must be tempered by its acknowledged limitations. The single-arm, non-blinded design and the modest sample size, particularly in the WMT cohort, preclude definitive causal inferences and limit the generalizability of our findings. The promising correlation between D-lactate and clinical outcomes, while internally consistent, requires independent validation in a larger cohort and did not survive multiple comparison corrections. The 4-week follow-up was sufficient to capture early clinical response but was likely too short to observe the full timeline of microbial and barrier stabilization. Finally, the absence of direct immune or metabolomic measurements means our proposed immunomodulatory mechanism remains speculative and is presented as a testable hypothesis for future research.
In summary, this study confirms significant intestinal barrier dysfunction in AD and demonstrates the potential clinical efficacy of WMT. The strong, uncorrected correlations suggest that pre-treatment D-lactate level warrants further investigation as a candidate biomarker for predicting WMT response. Critically, the dissociation between significant clinical improvement and the absence of short-term barrier restoration reframes our understanding of WMT's mechanism. Our findings suggest that the early efficacy of WMT is likely driven by rapid immunomodulation, mediated by a restructured gut microbiota that suppresses pro-inflammatory taxa and enhances beneficial metabolic potential, rather than by immediate structural repair.
5 Conclusion
In conclusion, this study confirms intestinal barrier dysfunction in atopic dermatitis (AD) and provides preliminary evidence for the efficacy of Washed Microbiota Transplantation (WMT). The strong, uncorrected correlation between pre-treatment D-lactate levels and clinical improvement identifies it as a candidate predictive biomarker, warranting future validation. The dissociation between rapid symptom relief and lack of immediate barrier repair suggests that early WMT efficacy is driven by immunomodulation, potentially facilitated by the pre-existing permeable gut environment and a restructured gut microbiota. These findings advance our understanding of microbiota-targeted therapies for AD.
Statements
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm.nih.gov/, PRJNA1364770.
Ethics statement
The studies involving humans were approved by Ethics Committee of The First Affiliated Hospital of Guangdong Pharmaceutical University,. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants' legal guardians/next of kin. Written informed consent was obtained from the individual(s), and minor(s)' legal guardian/next of kin, for the publication of any potentially identifiable images or data included in this article.
Author contributions
Y-DL: Methodology, Writing – review & editing. H-YZ: Data curation, Methodology, Writing – original draft. YC: Data curation, Writing – original draft. M-MC: Validation, Writing – original draft. L-HW: Funding acquisition, Project administration, Supervision, Writing – review & editing. X-XH: Methodology, Writing – review & editing. W-YD: Data curation, Investigation, Writing – review & editing.
Funding
The author(s) declare financial support was received for the research and/or publication of this article. This study was supported by the Key-Area Research and Development Program of Guangdong Province (No.2022B1111070006), the Medical Scientific Research Foundation of Guangdong Province (No. B2022209), the Scientific Research Projects of Guangdong Bureau of Traditional Chinese Medicine (No. 20221232), and the Guangdong Innovation Research Team for Higher Education (No. 2021KCXTD025).
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declare that no Generative AI was used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1.
Weidinger S Beck LA Bieber T Kabashima K Irvine AD . Atopic dermatitis. Nat Rev Dis Primers. (2018) 4(1):1. 10.1038/s41572-018-0001-z
2.
Chopra R Vakharia PP Sacotte R Patel N Immaneni S White T et al Severity strata for Eczema Area and Severity Index (EASI), modified EASI, Scoring Atopic Dermatitis (SCORAD), objective SCORAD, Atopic Dermatitis Severity Index and body surface area in adolescents and adults with atopic dermatitis. Br J Dermatol. (2017) 177(5):1316–21. 10.1111/bjd.15641
3.
Kong HH Oh J Deming C Grice EA Beatson MA Nomicos E et al Temporal shifts in the skin microbiome associated with disease flares and treatment in children with atopic dermatitis. Genome Res. (2012) 22(5):850–9. 10.1101/gr.131029.111
4.
Rios-Carlos M Cervantes-García D Córdova-Dávalos LE Bermúdez-Humarán LG Salinas E . Unraveling the gut-skin axis in atopic dermatitis: exploiting insights for therapeutic strategies. Gut Microbes. (2024) 16(1):2430420. 10.1080/19490976.2024.2430420
5.
Petersen EBM Skov L Thyssen JP . Role of the gut microbiota in atopic dermatitis: a systematic review. Acta Derm Venereol. (2019) 99(1):5–11. 10.2340/00015555-3008
6.
Šuler Baglama Š Trčko K . Skin and gut microbiota dysbiosis in autoimmune and inflammatory skin diseases. Acta Dermatovenerol Alp Pannonica Adriat. (2022) 31(3):105–9. PMID:
7.
De Andrade P Maria ESJ Carregaro V Sacramento LA Roberti LR Aragon DC et al Efficacy of probiotics in children and adolescents with atopic dermatitis: a randomized, double-blind, placebo-controlled study. Front Nutr. (2022) 8:833666. 10.3389/fnut.2021.833666
8.
Zhang F-M Ng SC He X-X Wu D-C Wu K-C Chiu C-T et al (on behalf of the consensus panel). Nanjing consensus on methodology of washed microbiota transplantation. Chin Med J. (2020) 133(19):2330–2. 10.1097/CM9.0000000000000954
9.
Zhang T Lu G Zhao Z Liu Y Shen Q Chen Y et al Washed microbiota transplantation vs. manual fecal microbiota transplantation: clinical findings, animal studies and in vitro screening. Protein Cell. (2020) 11(4):251–66. 10.1007/s13238-019-00684-8
10.
Deng W-Y Chen W-J Zhong H-J Wu L-H He X-X . Washed microbiota transplantation: a case report of clinical success with skin and gut microbiota improvement in an adolescent boy with atopic dermatitis. Front Immunol. (2023) 14:1275427. 10.3389/fimmu.2023.1275427
11.
De Pessemier B Grine L Debaere M Maes A Paetzold B Callewaert C . Gut-skin axis: current knowledge of the interrelationship between microbial dysbiosis and skin conditions. Microorganisms. (2021) 9(2):353. 10.3390/microorganisms9020353
12.
Johnson CC Ownby DR . The infant gut bacterial microbiota and risk of pediatric asthma and allergic diseases. Transl Res. (2017) 179:60–70. 10.1016/j.trsl.2016.06.010
13.
Okumura R Takeda K . Roles of intestinal epithelial cells in the maintenance of gut homeostasis. Exp Mol Med. (2017) 49(5):e338. 10.1038/emm.2017.20
14.
Leclercq S Matamoros S Cani PD Neyrinck AM Jamar F Stärkel P et al Intestinal permeability, gut-bacterial dysbiosis, and behavioral markers of alcohol-dependence severity. Proc Natl Acad Sci USA. (2014) 111(42):E4485–93. 10.1073/pnas.1415174111
15.
Fukudome I Kobayashi M Dabanaka K Maeda H Okamoto K Okabayashi T et al Diamine oxidase as a marker of intestinal mucosal injury and the effect of soluble dietary fiber on gastrointestinal tract toxicity after intravenous 5-fluorouracil treatment in rats. Med Mol Morphol. (2014) 47(2):100–7. 10.1007/s00795-013-0055-7
16.
Li K-Y Wang J-L Xu Y-Y Gao S-Y Zhang Y-Y He A-Q et al Intestinal barrier disruption in ileal pouchitis after ileal pouch-anal anastomosis in a rat model. Inflamm Bowel Dis. (2017) 23(6):923–31. 10.1097/MIB.0000000000001129
17.
Qing Y Xie H Su C Wang Y Yu Q Pang Q et al Gut microbiome, short-chain fatty acids, and mucosa injury in young adults with human immunodeficiency virus infection. Dig Dis Sci. (2019) 64(7):1830–43. 10.1007/s10620-018-5428-2
18.
Schram ME Spuls PI Leeflang MM Lindeboom R Bos JD Schmitt J . EASI, (objective) SCORAD and POEM for atopic eczema: responsiveness and minimal clinically important difference. Allergy. (2012) 67(1):99–106. 10.1111/j.1398-9995.2011.02719.x
19.
Reich K Thyssen JP Blauvelt A Eyerich K Soong W Rice ZP et al Efficacy and safety of abrocitinib versus dupilumab in adults with moderate-to-severe atopic dermatitis: a randomised, double-blind, multicentre phase 3 trial. Lancet. (2022) 400(10348):273–82. 10.1016/S0140-6736(22)01199-0
20.
Kabashima K Matsumura T Komazaki H Kawashima M , Nemolizumab-JP01 Study Group. Trial of nemolizumab and topical agents for atopic dermatitis with pruritus. N Engl J Med. (2020) 383(2):141–50. 10.1056/NEJMoa1917006
21.
Zhang X Liu D Wang Y Yan J Yang X . Clinical significance on serum intestinal fatty acid binding protein and D-lactate levels in early intestinal injury of patients with sepsis. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. (2019) 31(5):545–50. 10.3760/cma.j.issn.2095-4352.2019.05.005
22.
Li J Zheng J Jin X Zhu K Wang X Zhang H . Intestinal barrier dysfunction is involved in the development of systemic inflammatory responses and lung injury in type A aortic dissection: a case-control study. J Thorac Dis. (2022) 14(9):3552–64. 10.21037/jtd-22-1122
23.
Chunlong D Yuhai W Likun Y et al Clinical roles of plasma D-lactate, endotoxin and diamineoxidase in assessment of intestinal mucosal barrier function post traumatic brain injury. (2019). 10.3760/cma.j.issn.1001-2346.2019.12.015
24.
Xia Y Cao H Zheng J Chen L . Claudin-1 mediated tight junction dysfunction as a contributor to atopic march. Front Immunol. (2022) 13:927465. 10.3389/fimmu.2022.927465
25.
Miake S Tsuji G Takemura M Hashimoto-Hachiya A Vu YH Furue M et al IL-4 augments IL-31/IL-31 receptor alpha interaction leading to enhanced ccl 17 and ccl 22 production in dendritic cells: implications for atopic dermatitis. Int J Mol Sci. (2019) 20(16):4053. 10.3390/ijms20164053
26.
Peng G Tsukamoto S Umehara Y Kishi R Tominaga M Takamori K et al Experimental and clinical evidence suggests that treatment with betacellulin can alleviate Th2-type cytokine-mediated impairment of skin barrier function. Int J Mol Sci. (2022) 23(19):11520. 10.3390/ijms231911520
27.
Nylund L Nermes M Isolauri E Salminen S de Vos WM Satokari R . Severity of atopic disease inversely correlates with intestinal microbiota diversity and butyrate-producing bacteria. Allergy. (2015) 70(2):241–4. 10.1111/all.12549
28.
Elce A Amato F Zarrilli F Calignano A Troncone R Castaldo G et al Butyrate modulating effects on pro-inflammatory pathways in human intestinal epithelial cells. Benefic Microbes. (2017) 8(5):841–7. 10.3920/BM2016.0197
29.
Ahn J-R Lee S-H Kim B Nam MH Ahn YK Park YM et al Ruminococcus gnavus ameliorates atopic dermatitis by enhancing treg cell and metabolites in BALB/c mice. Pediatr Allergy Immunol. (2022) 33(1):e13678. 10.1111/pai.13678
30.
Chua HH Chou HC Tung YL Chiang B-L Liao C-C Liu H-H et al Intestinal dysbiosis featuring abundance of ruminococcus gnavus associates with allergic diseases in infants. Gastroenterology. (2018) 154(1):154–67. 10.1053/j.gastro.2017.09.006
31.
Clementino JR de Oliveira LIG Salgaço MK de Oliveira FL Mesa V Tavares JF et al β-glucan alone or combined with Lactobacillus acidophilus positively influences the bacterial diversity and metabolites in the colonic Microbiota of type II diabetic patients. Probiotics Antimicrob Proteins. (2025) 17(5):3728–42. 10.1007/s12602-025-10491-9
32.
Wang Y Hou J Tsui JC Wang L Zhou J Chan UK et al Unique gut microbiome signatures among adult patients with moderate to severe atopic dermatitis in southern Chinese. Int J Mol Sci. (2023) 24(16):12856. 10.3390/ijms241612856
Summary
Keywords
washed microbiota transplantation, WMT, intestinal barrier, atopic dermatitis, D-lactate
Citation
Lao Y-D, Zhang H-Y, Cai Y, Chen M-M, Wu L-H, He X-X and Deng W-Y (2025) Association of intestinal barrier impairment with symptom severity and washed microbiota transplantation outcomes in atopic dermatitis patients. Front. Allergy 6:1680658. doi: 10.3389/falgy.2025.1680658
Received
06 August 2025
Revised
22 November 2025
Accepted
26 November 2025
Published
09 December 2025
Volume
6 - 2025
Edited by
Indrashis Podder, College of Medicine & Sagore Dutta Hospital, India
Reviewed by
Salman Ahmad Mustfa, AstraZeneca, United Kingdom
Regina Fölster-Holst, University Medical Center Schleswig-Holstein, Germany
Sebastian Criton, Amala Institute of Medical Sciences, India
Updates
Copyright
© 2025 Lao, Zhang, Cai, Chen, Wu, He and Deng.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Li-Hao Wu wulihao888@126.com Xing-Xiang He 13724887928@qq.com Wan-Ying Deng wanyingcindy@outlook.com
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.