The rise in glucose concentration in saliva samples mixed with test foods monitored using a glucometer: An observational pilot study

Correlations of Salivary and Blood Glucose Levels among Six Saliva Collection Methods

Background: Saliva has been studied as a better indicator of disorders and diseases than blood. Specifically, the salivary glucose level is considered to be an indicator of diabetes mellitus (DM). However, saliva collection methods can affect the salivary glucose level, thereby affecting the correlation between salivary glucose and blood glucose. Therefore, this study aims to identify an ideal saliva collection method and to use this method to determine the population and individual correlations between salivary glucose and blood glucose levels in DM patients and healthy controls. Finally, an analysis of the stability of the individual correlations is conducted. Methods: This study included 40 age-matched DM patients and 40 healthy controls. In the fasting state, saliva was collected using six saliva collection methods, venous blood was collected simultaneously from each study participant, and both samples were analyzed at the same time using glucose oxidase peroxidase. A total of 20 DM patients and 20 healthy controls were arbitrarily selected from the above participants for one week of daily testing. The correlations between salivary glucose and blood glucose before and after breakfast were analyzed. Finally, 10 DM patients and 10 healthy controls were arbitrarily selected for one month of daily testing to analyze the stability of individual correlations. Results: Salivary glucose levels were higher in DM patients than healthy controls for the six saliva collection methods. Compared with unstimulated saliva, stimulated saliva had decreased glucose level and increased salivary flow. In addition, unstimulated parotid salivary glucose was most correlated with blood glucose level (R² = 0.9153), and the ROC curve area was 0.9316, which could accurately distinguish DM patients. Finally, it was found that the correlations between salivary glucose and blood glucose in different DM patients were quite different. The average correlation before breakfast was 0.83, and the average correlation after breakfast was 0.77. The coefficient of variation of the correlation coefficient before breakfast within 1 month was less than 5%. Conclusion: Unstimulated parotid salivary glucose level is the highest and is most correlated with blood glucose level, which can be accurately used to distinguish DM patients. Meanwhile, the correlation between salivary glucose and blood glucose was found to be relatively high and stable before breakfast. In general, the unstimulated parotid salivary glucose before breakfast presents an ideal saliva collecting method with which to replace blood-glucose use to detect DM, which provides a reference for the prediction of DM.

Unstimulated Parotid Saliva Is a Better Method for Blood Glucose Prediction

Objective: Saliva glucose has been widely used in diagnosing and monitoring diabetes, but the saliva collection method will affect saliva glucose concentration. So, this study aims to identify the ideal saliva collection method. Method: A total amount of six saliva collection methods were employed in 80 healthy participants in the morning. Besides, three unstimulated saliva methods were employed in another 30 healthy participants in the morning; in the meantime the blood glucose of these 30 participants was detected with a Roche blood glucose meter. The glucose oxidase method with 2, 4, 6-tribromo-3-hydroxybenzoic acid (TBHBA) as the chromogen has been improved to be suitable for healthy people, through the selection of the optimal pH value and ionic strength of the reaction system. This method was used for the detection of saliva glucose. Results: The improved method obtained absorbance at the wavelength of 520 nm, and the optimized parameter combination was pH 6.5 and 5 mg/dL NaCl. The lower limit of glucose detection was 0.1 mg/dL. Unstimulated saliva glucose concentration was higher than stimulated saliva glucose concentration. Unstimulated parotid saliva glucose concentration was the highest. Besides, unstimulated saliva glucose has a better normal distribution effect. Meantime, it was found that unstimulated parotid saliva was the most highly correlated with blood glucose (R2 = 0.707). Conclusions: the saliva collection method was an important factor that affected saliva glucose concentration. Unstimulated parotid saliva was the most highly correlated with blood glucose, which provided a reference for prediction of diabetes mellitus.