Human saliva is a unique biofluid which can reflect the physiopathological state of an individual. The wide spectrum of molecules present in saliva, compounded by the close association of salivary composition to serum metabolites, can provide valuable information for clinical diagnostic applications through highly sensitive vibrational spectroscopic techniques such as Raman spectroscopy. However, the nature of saliva, in terms of collection and patient-related characteristics, can be considered factors which may strongly affect the Raman spectral profile of salivary samples and disrupt the search for specific salivary biomarkers in the detection of diseases. The main objective of this study was to highlight spectral features associated with the type of collection in an intra- and inter-patient approach. Saliva was collected using both stimulated and non-stimulated approaches from 20 donors, concentrated by centrifugal filtration and further analysed using Raman spectroscopy. The methodology adopted for liquid saliva showed consistency in the qualitative analysis of the groups, confirming the reproducibility of this Raman spectroscopic approach. Using principal component analysis (PCA) and partial least squares – discriminant analysis (PLSDA), non stimulated saliva could be differentiated from stimulated saliva in both intra- and inter-patient analysis, with a classification efficiency of 77 and 87 %, respectively. The bicinchoninic acid (BCA) assay showed a similar trend in terms of total protein concentration, showing a slight increase in stimulated saliva samples. These results are valuable in the process of developing and establishing Raman spectroscopy as a novel diagnostic tool in the future as well as controlling variability, in order to determine specific spectroscopic markers related to a multifactorial disease for diagnostic or follow-up purposes.

人类唾液是一种独特的生物体液，可以反映个体的病理生理状态。唾液中存在的广谱分子，再加上唾液成分与血清代谢物的密切关联，可以通过高灵敏度的振动光谱技术（如拉曼光谱）为临床诊断应用提供有价值的信息。然而，唾液的性质，在收集和患者相关特征方面，可以被认为是可能强烈影响唾液样品的拉曼光谱分布的因素，并破坏在疾病检测中寻找特定唾液生物标志物的过程。本研究的主要目的是突出显示与患者内部和患者间方法中收集类型相关的光谱特征。使用刺激和非刺激方法从 20 个供体收集唾液，通过离心过滤浓缩，并使用拉曼光谱进一步分析。液体唾液采用的方法在组的定性分析中显示出一致性，证实了这种拉曼光谱方法的可重复性。使用主成分分析 (PCA) 和偏最小二乘法 - 判别分析 (PLSDA)，可以在患者内和患者间分析中区分非刺激唾液与刺激唾液，分类效率分别为 77% 和 87%。二辛可宁酸 (BCA) 测定在总蛋白质浓度方面显示出类似的趋势，显示受刺激的唾液样本略有增加。