目的/背景成人发作的静动病(AOSD)与败血症具有相似的临床症状。因此,区分AOSD和脓毒症在做出诊断时提出了巨大的挑战。本研究旨在使用宏基因组下一代测序(mNGS)分析与AOSD和败血症相关的血液微生物群的变化。确定区分AOSD和败血症的潜在生物标志物,并探讨mNGS对这两种病理状况的鉴别诊断价值。方法4例AOSD患者和4例脓毒症患者在风湿免疫科治疗,收集2021年10月至2022年2月徐州医科大学附属医院。通过比较AOSD和脓毒症之间的血液微生物群,分析了这些患者的mNGS诊断记录在物种分类结构和β多样性方面的微生物相关性。使用随机森林算法筛选区分亚组能力最强的生物标志物。结果AOSD患者与脓毒症对照组在性别、年龄方面差异无统计学意义(p>0.05)。总共获得了91个操作分类单位(OTU)。在门的层面上,变形杆菌,子囊菌和担子菌在两组中均存在高丰度(79.76%,14.18%和3.30%vs54.03%,32.77%和5.81%)。在属一级,副流感的丰度,曲霉和Ralstonia是AOSD组中最高的前三(73.88%,10.92%和5.48%),而Ralstonia,曲霉菌和马拉色菌的丰度在脓毒症组中排名前三(48.69%,27.36%和5.52%)。在β多样性分析中,AOSD组和脓毒症组之间的视觉主坐标分析(PCoA)和非度量多维量表(NMDS)均有进展(p<0.05),相似性分析(Anosim)差异不大(p>0.05)。线性判别分析效应大小(LEfSe)表明,酵母菌,Moraxellales,Mucorales,黄原则,酵母菌,不动杆菌,窄食单胞菌,Yarrowia,神秘主义者,约翰逊不动杆菌,Yarrowialipolytica,脓毒症组变形杆菌和嗜麦芽窄食单胞菌的含量更高(p<0.05)。区分AOSD和脓毒症能力最强的前5个变量是乔氏不动杆菌,变形杆菌,痤疮丙酸杆菌,嗜麦芽窄食单胞菌和解脂Yarrowia。结论AOSD的血液微生物与脓毒症不同,mNGS有可能区分AOSD和脓毒症。
Aims/Background Adult-onset Still\'s disease (AOSD) shares similar clinical symptoms with sepsis. Thus, differentiating between AOSD and sepsis presents a great challenge while making diagnosis. This study aimed to analyse the changes in blood microbiota related to AOSD and sepsis using metagenomic next-generation sequencing (mNGS), identify potential biomarkers that distinguish AOSD from sepsis, and explore the diagnostic value of mNGS in differentiation between these two pathological conditions. Methods Clinical data of four AOSD patients and four sepsis patients treated in the Department of Rheumatology and Immunology, The Affiliated Hospital of Xuzhou Medical University between October 2021 and February 2022 were collected. The mNGS diagnostic records of these patients were analysed for microbial correlations in terms of species taxonomic structure and beta diversity by comparing blood microbiota between AOSD and sepsis. The biomarkers with the strongest capability in distinguishing the subgroups were screened using a random forest algorithm. Results There was no statistically significant differences between AOSD patients and sepsis controls in terms of gender and age (p > 0.05). A total of 91 operational taxonomic units (OTUs) were obtained. At the level of phylum, Proteobacteria, Ascomycota and Basidiomycota were present in high abundances in both groups (79.76%, 14.18% and 3.30% vs 54.03%, 32.77% and 5.81%). At the genus level, the abundances of Parainfluenzae, Aspergillus and Ralstonia were the top three highest in the AOSD group (73.88%, 10.92% and 5.48%), while Ralstonia, Aspergillus and Malassezia were ranked as the top three in the sepsis group in term of abundance (48.69%, 27.36% and 5.52%). In beta-diversity analysis, there were advances shown in visual principal coordinates analysis (PCoA) and non-metric multidimensional scaling (NMDS) between the AOSD group and sepsis group (p < 0.05), with little significant differences in the analysis of similarities (Anosim) (p > 0.05). Linear discriminant analysis effect size (LEfSe) showed that Mucoromycota, Saccharomycetes, Moraxellales, Mucorales, Xanthomonadales, Saccharomycetales, Acinetobacter, Stenotrophomonas, Yarrowia, Apophysomyces, Acinetobacter johnson, Yarrowia lipolytica, Apophysomyces variabilis and Stenotrophomonas maltophilia were more enriched in sepsis group (p < 0.05). The top five variables with the strongest capability in distinguishing between AOSD and sepsis were Acinetobacter johnsonii, Apophysomyces variabilis, Propionibacterium acnes, Stenotrophomonas maltophilia and Yarrowia lipolytica. Conclusion The blood microorganisms in AOSD were different from sepsis, and mNGS was potential to distinguish between AOSD and sepsis.