Abstract：
Mastitis is one of the most impacting diseases in dairy farming, and its sensitive and specific detection is therefore of the greatest importance. The clinical evaluation of udder and mammary secretions is typically combined with the milk Somatic Cell Count (SCC) and often accompanied by its bacteriological culture to identify the causative microorganism. In a constant search for improvement, several non-enzymatic milk proteins, including milk amyloid A (M-SAA), haptoglobin (HP), cathelicidin (CATH), and lactoferrin (LF), have been investigated as alternative biomarkers of mastitis for their relationship with mammary gland inflammation, and immunoassay techniques have been developed for detection with varying degrees of success. To provide a general overview of their implementation in the different dairy species, we carried out a systematic review of the scientific literature using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guidelines. Our review question falls within the type \"Diagnostic test accuracy questions\" and aims at answering the diagnostic question: \"Which are the diagnostic performances of mastitis protein biomarkers investigated by immunoassays in ruminant milk?\". Based on 13 keywords combined into 42 searches, 523 manuscripts were extracted from three scientific databases. Of these, 33 passed the duplicate removal, title, abstract, and full-text screening for conformity to the review question and document type: 78.8% investigated cows, 12.1% sheep, 9.1% goats, and 6.1% buffaloes (some included more than one dairy species). The most frequently mentioned protein was M-SAA (48.5%), followed by HP (27.3%), CATH (24.2%) and LF (21.2%). However, the large amount of heterogeneity among studies in terms of animal selection criteria (45.5%), index test (87.9%), and standard reference test (27.3%) resulted in a collection of data not amenable to meta-analysis, a common finding illustrating how important it is for case definitions and other criteria to be standardized between studies. Therefore, results are presented according to the SWiM (Synthesis Without Meta-analysis) guidelines. We summarize the main findings reported in the 33 selected articles for the different markers and report their results in form of comparative tables including sample selection criteria, marker values, and diagnostic performances, where available. Finally, we report the study limitations and bias assessment findings.
摘要：
乳腺炎是奶牛养殖中影响最大的疾病之一，因此，其灵敏和特异的检测是最重要的。乳房和乳腺分泌物的临床评估通常与乳体细胞计数（SCC）相结合，并且通常伴随其细菌学培养以鉴定致病微生物。在不断寻求改进的过程中，几种非酶乳蛋白，包括乳淀粉样蛋白A（M-SAA），触珠蛋白（HP），cathelicidin(CATH),和乳铁蛋白（LF），已被研究作为乳腺炎的替代生物标志物与乳腺炎症的关系，和免疫测定技术已经开发用于检测，并取得了不同程度的成功。为了概述它们在不同乳制品品种中的实施情况，我们使用PRISMA(系统评价和Meta分析的首选报告项目)指南对科学文献进行了系统评价.我们的评论问题属于“诊断测试准确性问题”类型，旨在回答诊断问题：“哪些是反刍动物乳中通过免疫测定研究的乳腺炎蛋白质生物标志物的诊断性能？”。基于13个关键词组合成42个搜索，从三个科学数据库中提取了523份手稿。其中,33通过了重复删除，title,abstract,全文筛选是否符合审查问题和文件类型:78.8%的调查奶牛，12.1%绵羊，9.1%山羊，和6.1％的水牛（有些包括一种以上的乳制品）。最常提及的蛋白质是M-SAA(48.5%),其次是HP(27.3%),CATH（24.2%）和LF（21.2%）。然而,就动物选择标准而言，研究之间存在大量异质性(45.5%)，指数测试（87.9%），和标准参考检验(27.3%)导致收集的数据不适合荟萃分析，一个共同的发现说明了病例定义和其他标准在研究之间标准化的重要性。因此,结果根据SWiM(无Meta分析的综合)指南给出。我们总结了33篇选定文章中针对不同标记的主要发现，并以比较表的形式报告了其结果，包括样本选择标准，标记值,和诊断性能，可用的地方。最后,我们报告了研究局限性和偏倚评估结果.
