Stimulated Parotid Saliva Is a Better Method for Depression Prediction

Abstract

Background: Saliva cortisol is considered to be a biomarker of depression prediction. However, saliva collection methods can affect the saliva cortisol level. Objective: This study aims to determine the ideal saliva collection method and explore the application value of saliva cortisol in depression prediction. Methods: 30 depressed patients and 30 healthy controls were instructed to collect saliva samples in the morning with six collection methods. Simultaneous venous blood was collected. Enzyme-linked immunosorbent assay was used to determine the cortisol level. The 24-observerrated Hamilton depression rating scale (HAMD-24) was used to assess the severity of depression. Results: The significant differences in saliva cortisol levels depend on the saliva collection methods. The level of unstimulated whole saliva cortisol was most correlated with blood (r = 0.91). The stimulated parotid saliva cortisol can better predict depression. The area under the curve was 0.89. In addition, the saliva cortisol level of the depression patients was significantly higher than the healthy controls. The correlation between the cortisol level and the HAMD-24 score was highly significant. The higher the saliva cortisol level, the higher the HAMD-24 score. Conclusions: All the above findings point to an exciting opportunity for non-invasive monitoring of cortisol through saliva.

Obtaining a Reliable Diagnostic Biomarker for Diabetes Mellitus by Standardizing Salivary Glucose Measurements

“…Finally, samples were transported to the laboratory and centrifuged at −20 °C for future use. The saliva collection process is consistent with that used in our previous study [ 28 , 29 ]. In this study, the whole saliva was collected.…”

Artificial Intelligence in Spinal Imaging: Current Status and Future Directions

“…Sixth, regarding prediction reliability and uncertainty quantification, the current applications of ML in the spinal imaging field (nay, in the entire biomedical informatics field) generally seem to focus on the prediction task, and the model performance is evaluated with such metrics as accuracy, ROC, and sensitivity [ 127 , 128 , 129 , 130 ]. However, the much lower attended reliability and uncertainty of the predictions can be of great significance in medical diagnosis [ 131 ].…”

Neurodegeneration in Cognitive Impairment and Mood Disorders for Experimental, Clinical and Translational Neuropsychiatry

“…Characterized by persistent feelings of sadness, reduced interest, diminished self-worth, and various physical and mental symptoms, it often co-occurs with other conditions, such as anxiety, chronic pain, and neurodegenerative diseases [ 80 , 81 , 82 , 83 ]. In this special issue, three articles explored different approaches for diagnosing and treating depression and its comorbidities [ 84 , 85 , 86 ]. The first article assessed the neuroprotective and swiftly acting antidepressant-like properties of 20 essential oils in mice.…”