It has been suggested that exercise intensity is of importance in the regulation of increase in pro-inflammatory molecules, but there is still a debate about the effect of duration on these molecules. Therefore, the effect of exercise duration on the serum concentrations of interleukin-6 (IL-6), tumour necrosis factor-alpha (TNF-alpha), soluble form of intercellular adhesion molecule-1 (sICAM-1), and matrix metalloproteinase-9 (MMP-9) was studied in 22 half-marathon (HM) and 18 marathon (M) male amateur runners who completed their exercise task in 1.8 +/- A 0.2 (mean +/- A standard deviation) and 3.6 +/- A 0.4 h, respectively (thus, average speed was 11.7 +/- A 1.5 and 11.9 +/- A 1.8 km h(-1), respectively). Blood was sampled 2 days before, 15 min after, and 28 h after the race. IL-6, TNF-alpha, and MMP-9 always increased immediately after exercise, but the increase was larger (P < 0.05) in M versus HM (a dagger IL-6: 31 +/- A 24 vs. 5 +/- A 4 pg ml(-1); a dagger TNF-alpha: 1.7 +/- A 1.9 vs. 0.5 +/- A 0.8 pg ml(-1); MMP-9: 288 +/- A 216 vs. 145 +/- A 128 ng ml(-1), respectively). sICAM-1 also increased with exercise, but similarly in M and HM (20 +/- A 40 vs. 23 +/- A 32 ng ml(-1), respectively). Only sICAM-1 remained elevated 28 h post-exercise in both HM and M, while IL-6, TNF-alpha, and MMP-9 returned to pre-exercise levels. Competitive HM and M races induce significant increases in IL-6, TNF-alpha, sICAM-1, and MMP-9 concentrations. As HM and M runners performed the competition with similar absolute intensity, the difference in response between the groups suggests that exercise duration is of importance in the regulation of IL-6, TNF-alpha, and MMP-9, but not sICAM-1 concentrations in response to prolonged running.