Fasting salivary glucose levels is not a better measure for identifying diabetes mellitus than serum or capillary blood glucose levels: comparison in a Ghanaian population.

We compared the diagnostic performance and correlation between salivary, serum and capillary blood glucose of diabetes and non-diabetes patients. Early detection of diabetes mellitus (DM) contributes to the prevention of complications and management. This case-control study was conducted among a total of 138 participants comprising 79 newly diagnosed diabetes patients (cases) and 59 non-diabetes patients (controls). Fasting salivary glucose (FSLG), fasting serum glucose (FSEG) levels and fasting capillary whole blood glucose (FCWBG) level were assayed for each participant. The mean FSLG, FSEG and FCWBG levels were significantly higher among the cases compared to controls (p < 0.0001). There was a significant mean difference between the levels of FSLG vs. FSEG (p < 0.0001) and FSLG vs. FCWBG (p < 0.0001) but not levels of FSEG vs. FCWBG (p > 0.05) in both cases and controls. A positive correlation was observed between FSLG and FSEG (r = 0.89; p < 0.0001) and FCWBG (r = 0.87; p < 0.0001). At the cut-off value >6.8 mmol/l for FSEG, a sensitivity of 99%, specificity of 100.0% and area under the curve (AUC) of 98.8% was observed for predicting DM while a sensitivity of 80%, specificity of 95% and AUC of 91.0% was observed for FSLG at a cut-off value >0.5 mmol/l. At the cut-off value >6.9 mmol/l for FCWBG, a sensitivity of 100.0%, specificity of 100.0% and AUC of 100.0% was observed for predicting DM. Fasting salivary glucose (FSLG) levels increased with increasing blood glucose levels. However, it does not generate enough diagnostic and predictive accuracy compared to capillary whole blood glucose which less invasive.