Background. Immune parameters (IP) have been extensively studied to distinguish between latent tuberculosis (LTBI) and active tuberculosis (TB). Objective. To determine the IP associated with LTBI, compared to active TB and individuals not infected by M. tuberculosis published in literature. Methods. We conducted a systematic search using Google Scholar and PubMed databases, combining the MeSH terms latent tuberculosis, Mycobacterium tuberculosis, cytokines, and biological markers, with the free terms, biomarkers and cytokines. Spanish, English, and Portuguese articles comparing the concentration of IP associated with LTBI, either in plasma/serum or in vitro, in adults and nonimmunocompromised versus individuals with TB or without M. tuberculosis infection between 2006 July and 2018 July were included. Two blinded reviewers carried out the searches, read the abstracts, and selected the articles for analysis. Participants’ information, diagnostic criteria, IP, detection methods, and biases were collected. Results. We analyzed 36 articles (of 637 abstracts) with 93 different biomarkers in different samples. We found 24 parameters that were increased only in active TB (TGF-α, CSF3, CSF2, CCL1 [I-309], IL-7, TGF-β1, CCL3 [MIP-1α], sIL-2R, TNF-β, CCL7 [MCP-3], IFN-α, fractalkine, I-TAG, CCL8 [MCP-2], CCL21 [6Ckine], PDGF, IL-22, VEGF-A, LXA4, PGE2, PGF2α, sCD163, sCD14, and 15-Epi-LXA4), five were elevated in LTBI (IL-5, IL-17F, IL-1, CCL20 [MIP-3α], and ICAM-1), and two substances were increased among uninfected individuals (IL-23 and basic FGF). We found high heterogeneity between studies including failure to account for the time/illness of the individuals studied; varied samples and protocols; different clinical classification of TB; different laboratory methods for IP detection, which in turn leads to variable units of measurement and assay sensitivities; and selection bias regarding TST and booster effect. None of the studies adjusted the analysis for the effect of ethnicity. Conclusions. It is mandatory to harmonize the study of immune parameters for LTBI diagnosis. This systematic review is registered with PROSPERO CRD42017073289.

中文翻译：

---

潜伏性结核感染免疫参数研究的研究设计差距：系统评价。

背景。免疫参数 (IP) 已被广泛研究以区分潜伏性肺结核 (LTBI) 和活动性肺结核 (TB)。客观。为了确定与 LTBI 相关的 IP，与文献中发表的活动性 TB 和未感染结核分枝杆菌的个体相比。方法。我们使用 Google Scholar 和 PubMed 数据库进行了系统搜索，结合 MeSH 术语潜伏性结核病、结核分枝杆菌, 细胞因子, 和生物标志物, 用自由术语, 生物标志物和细胞因子。西班牙语、英语和葡萄牙语的文章比较了2006 年 7 月至 2018 年 7 月期间成人和非免疫功能低下者与患有 TB 或未感染结核分枝杆菌的个体中与 LTBI 相关的 IP 浓度，无论是血浆/血清中还是体外。两名不知情的审稿人进行搜索、阅读摘要并选择文章进行分析。收集参与者的信息、诊断标准、IP、检测方法和偏见。结果。我们分析了 36 篇文章（共 637 篇摘要），在不同样本中有 93 种不同的生物标志物。我们发现仅在活动性结核病中增加的 24 个参数（TGF- α, CSF3, CSF2, CCL1 [I-309], IL-7, TGF- β1 , CCL3 [MIP- 1α ], sIL-2R, TNF- β , CCL7 [MCP-3], IFN- α , fractalkine, I-TAG、CCL8 [MCP-2]、CCL21 [6Ckine]、PDGF、IL-22、VEGF-A、LXA4、PGE2、PGF2 α、sCD163、sCD14 和 15-Epi-LXA4），五个在 LTBI 中升高（IL-5、IL-17F、IL-1、CCL20 [MIP-3 α] 和 ICAM-1），两种物质在未感染个体中增加（IL-23 和碱性 FGF）。我们发现研究之间存在高度异质性，包括未能考虑所研究个体的时间/疾病；不同的样本和协议；结核病的不同临床分类；IP 检测的不同实验室方法，进而导致测量单位和检测灵敏度的变化；以及关于 TST 和助推器效应的选择偏差。没有一项研究调整了种族影响的分析。结论。必须协调对 LTBI 诊断的免疫参数的研究。该系统评价已在 PROSPERO CRD42017073289 注册。