Exercise training or higher levels of physical activity are known to exert anti-inflammatory effects. CD14+CD16+ monocytes are potent producers of inflammatory proteins, and elevated levels of these "inflammatory" monocytes have been implicated in disease development. Little is known about the influence of exercise training on this cell population. On the basis of their physical activity pattern, male and female subjects, 65-80 years old, were assigned to a physically active (PA; n=15) or inactive (PI; n=15) group. The PI group performed 12 weeks (3 days/week) of endurance (20 min at 70-80% heart-rate reserve) and resistance exercise training (eight exercises, two sets at 70-80% of one repetition maximum). Subjects in the PA group maintained their habitual activity level. Flow cytometry was used to determine monocyte phenotype and monocyte TLR4 expression. ELISAs were used to measure whole blood, LPS-stimulated TNF-alpha production, and serum C-reactive protein (CRP). At baseline, the PA group had a lower percentage of CD14+CD16+ monocytes and lower unstimulated production of TNF-alpha than the PI group. CD14+CD16+ monocyte percentage and 1 ng/ml LPS-stimulated TNF-alpha production were reduced after the PI group underwent 12 weeks of exercise training. PI subjects also had higher TLR4 expression on classical monocytes, but there were no significant exercise training-induced changes in monocyte TLR4 expression. The PA group had significantly lower serum CRP than the PI group. Physical activity was associated with lower CD14+CD16+ monocyte percentage and LPS-stimulated TNF-alpha production. Exercise training-induced reductions in CD14+CD16+ monocytes may contribute to the anti-inflammatory effects of exercise training.