Objective: Abnormal functioning of the metabolic cascade of cytosolic phospholipase A2 (cPLA2) and cyclooxygenase-2 (COX-2) enzymes may underlie attenuated skin flush response to niacin (a water-soluble B vitamin) and disturbances of glucose and lipid metabolism. Individuals with schizophrenia frequently exhibit attenuated niacin-induced skin flushing, and there is an increased rate of dyslipidemias and diabetes in this illness. We investigated whether an attenuated niacin flush response might be associated with plasma lipid and glucose concentrations in patients with schizophrenia and schizoaffective disorder. To our knowledge, this is the first study to investigate the possible association between niacin skin flush response and plasma lipid and glucose concentrations. Patients and methods: Forearm skin response to patches containing niacin of 0.1M, 0.01M, 0.001M and 0.0001M concentrations in 78 patients was rated using the method of volumetric niacin response (VNR). Fasting plasma total cholesterol, LDL cholesterol, HDL cholesterol, triglyceride, and glucose levels were measured using an Olympus AU640 autoanalyzer (Olympus, Tokyo, Japan). Results: Niacin skin flush response did not significantly predict total cholesterol, LDL cholesterol, HDL cholesterol, or glucose levels (P > 0.05). We found that higher plasma triglyceride levels were associated with higher total VNR values (P < 0.05) and that total VNR accounted for approximately 14% of the triglyceride levels variability (R2 change = 0.140). We also observed higher triglyceride levels in patients with positive niacin skin flush response (visual score ≥ 2) compared to in patients without niacin skin flushing reaction (visual score < 2) at the 5-minute interval at niacin concentrations of 0.1 and 0.01M, and at the 10- and 15-minute intervals at niacin concentration of 0.001M (P < 0.05). Conclusion: The attenuated niacin skin flush response is a statistically significant predictor of plasma triglyceride levels in patients with schizophrenia and schizoaffective disorder, but it does not predict other investigated metabolic parameters. Furthermore, the VNR satisfactorily contributes to plasma triglyceride concentration.