Decoding disorders of follicular occlusion: results from a clinico-epidemiological and immuno-microbiological study of 81 patients

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 &lt;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).本项横断面研究（cross-sectional study）纳入81例存在复发性多发性潜性脓肿/结节/囊肿病史的患者，所有患者均存在至少1个多孔粉刺（polyporous comedone），伴或不伴窦道/桥接瘢痕，好发于特定部位。研究检测了受试者的血清细胞因子水平，并对皮损及皮损周边皮肤进行了微生物组分析。对48例患者实施皮肤活检以排除毛囊性道林-德戈斯病（Dowling-Degos disease）。对临床、组织病理学、细胞因子及皮肤微生物组特征进行两两关联分析，并将上述特征与3个疾病组——化脓性汗腺炎（Hidradenitis suppurativa，HS）、合并疾病组、未分类组——进行关联比较。由于样本量较小，聚合性痤疮（Acne conglobata）与藏毛窦疾病组未纳入本次对比分析。采用夏皮罗-威尔克检验（Shapiro-Wilk test）验证数据正态性。分类变量以例数及百分比形式呈现；符合正态分布的定量数据以均数±标准差（mean ± SD）表示，非正态分布定量数据则以中位数（四分位数间距，25th~75th百分位数）表示。分类变量的比较采用费希尔确切概率法（Fisher's exact test）与卡方检验（Chi-square test）；对于3组及以上独立样本的定量变量分析，则采用克鲁斯卡尔-沃利斯检验（Kruskal-Wallis test）。以P<0.05为差异具有统计学意义。相关系数（r）的判读标准为：-1~0为负相关，0为无相关，0~+1为正相关。表1展示了皮肤微生物组的相关细节。多数患者的皮损处皮肤拭子（41例，占50.6%）与皮损周边皮肤拭子（61例，占75.3%）培养结果为阴性。皮损处最常见的分离菌株为表皮葡萄球菌（甲氧西林耐药凝固酶阴性葡萄球菌，MR CoNS），共计16例（19.8%）；皮损周边皮肤最常见的分离菌株为表皮葡萄球菌（甲氧西林敏感凝固酶阴性葡萄球菌，MS CoNS），共计6例（7.4%）。除此之外，毛囊闭锁性疾病的发生还涉及其他多种微生物组特征。临床-流行病学关联分析显示，结节与潜性脓肿多见于年轻患者及疾病早期阶段，随患者年龄增长与病程延长，其发生率逐渐降低。其余临床、组织病理学、细胞因子特征或微生物学模式与患者年龄、性别、病程及受累部位总数均无显著相关性。未分类组的囊肿计数（55个）高于HS组（2个），差异具有统计学意义（P=0.011）；桥接瘢痕总数则在HS组（34个）与合并疾病组（34个）中更为常见，显著高于未分类组（8个），差异具有统计学意义（P<0.001）。未分类组的多孔粉刺数量（476个）亦多于HS组（106个），差异具有统计学意义（P=0.017）。肿瘤坏死因子-α（TNF-α）水平随病程延长呈升高趋势，而白细胞介素-17A（IL-17A）水平无显著变化。细胞因子水平与临床皮损形态、受累部位数量及组织病理学特征（鹿角状模式与浸润模式）之间的相关性均无统计学意义。多部位凹陷性瘢痕的存在与组织病理学鹿角状模式呈正相关（phi系数=0.324）。