Decoding disorders of follicular occlusion: results from a clinico-epidemiological and immuno-microbiological study of 81 patients
Description
This cross-sectional study involved 81 patients with a recurrent history of multiple blind boils/nodules/cysts, at least one polyporous comedone with or without sinuses/bridging scars over specific sites. Serum cytokine levels and lesion/perilesional skin microbiome analysis were done. Skin biopsies were done in 48 patients to exclude follicular Dowling-Degos disease. Clinical, histopathological, cytokine and skin microbiome profiles were correlated among each other and also among three disease groups ( Hidradenitis suppurativa, combined disease group and Unclassified group). Acne conglobata and pilonidal sinus disease group were not included in comparison due to smaller sample size. Data normality was checked using the Shapiro-Wilk test. Categorical variables were presented as numbers and percentages. Quantitative data with normal distribution were expressed as means ± SD, while nonnormal distribution data were presented as median with interquartile range (25th and 75th percentiles). For categorical variables, Fisher's exact test and Chi-square test were employed. The Kruskal-Wallis test was employed to analyse quantitative variables between three or more independent groups. A p-value of <0.05 was considered statistically significant. Correlation coefficient (r) interpretations were: -1 to 0 indicated negative correlation, 0 denoted no correlation, and 0 to +1 represented positive correlation. Table 1 includes the skin microbiome details .Lesional 41(50.6%) and perilesional 61(75.3%) skin swabs showed no growth in the majority of patients. Most common organism grown in lesional skin was Staphylococcus epidermidis (MR CoNS) [16 patients (19.8%)] and in perilesional skin, it was Staphylococcus epidermidis (MS CoNS) [6 patients (7.4%)]. Apart from these, different microbiome appears to be involved in follicular occlusion disorders. On clinico-epidemiological correlation, nodules and blind boils were more common in younger patients and early disease stages, decreasing with age and disease duration. Other clinical, histopathological, cytokine profile or microbiological pattern doesn’t show any correlation with age, gender, duration of disease and total number of involved sites. The unclassified group showed a higher cyst count (55) than HS (2) with significant p value of 0.011 while total bridging scars were more common in HS (34) and combined groups (34) than in the unclassified group (8) with p value of 0.000. The unclassified group also had more polyporous comedones (476) compared to HS (106) with p value of 0.017.TNF alpha levels with increased disease duration whereas IL 17A levels lacked significant change. The correlation between cytokine levels and clinical morphologies, number of sites and histology (antler pattern and infiltrates) lacked statistical significance. Presence of pitted scars over multiple sites and histological antler pattern showed a positive correlation (phi coefficient: 0.324).
Files
Steps to reproduce
In this descriptive study considering the range at which 95% of the sample will fall, including 95% confidence interval and 5% margin of error and the population to choose from; the minimum sample size was expected to be 78.This cross-sectional observational study was conducted in 81 patients attending Dermatology and General Surgery OPD of All India Institute of Medical Sciences Jodhpur. After informed written consent patients were included according to proposed inclusion criteria. It involves patients presenting with ≥2 blind boils or nodules or cysts and≥1 polyporous comedones with or without sinuses and bridging scars in any of the sites (axilla, groin, buttocks, natal cleft, back, chest, face, scalp) withrecurrent and chronic course (>3 months with >3 recurrences). All were evaluated by two independent dermatologists to label them with one of the known follicular occlusion disorders if possible. Serum IL-17A and TNF alpha levels were measured via an ELISA kit. Skin microbiome was evaluated by bacterial culture, and samples were collected from lesional and perilesional skin from inflammatory lesions (nodules/blind boils/sinuses) using sterile cotton swabs. Skin biopsy was taken from 48 cases of suspected Dowling degos disease. Clinical, histopathological, cytokine and skin microbiome profiles were correlated. Data normality was checked using the Shapiro-Wilk test. Categorical variables were presented as numbers and percentages. Quantitative data with normal distribution were expressed as means ± SD, while nonnormal distribution data were presented as median with interquartile range (25th and 75th percentiles). For categorical variables, Fisher's exact test and Chi-square test were employed. The Kruskal-Wallis test was employed to analyse quantitative variables between three or more independent groups. Data entry was done using a Microsoft Excel spreadsheet, and the final analysis was done with Statistical Package for Social Sciences (SPSS) software. A p-value of <0.05 was considered statistically significant. Correlation coefficient (r) interpretations were: -1 to 0 indicated negative correlation, 0 denoted no correlation, and 0 to +1 represented positive correlation.