Long-distance running has become increasingly popular. Cardiovascular adaptations to exercise are relevant to the specific sports and this is also the case in long-distance running. Significant changes regarding inflammatory and endothelial markers along with indices of oxidative stress are observed in marathon and ultra-marathon runners. However, data linking inflammatory marker levels with cardiovascular adaptations to marathon running are limited. The aim of the present study was to describe the cardiovascular adaptations observed in a group of ultra-marathon runners and the association with a series of inflammatory and endothelial markers measured in their plasma. A total of 43 ultra-marathon runners were assessed by echocardiography and a treadmill exercise test. Blood samples were used for tumor necrosis factor-α (TNF-α), asymmetric dimethylarginine (ADMA), interleukin (IL)-6, IL-10, C-reactive protein, creatine phosphokinase (CPK) and oxidative stress indice measurements. Ultra-marathon runners who presented augmented left ventricular (LV) end diastolic diameters >55 mm had higher ADMA values (1.07±0.07 vs. 0.99±0.08 µmol/ml, P<0.01) and lower CPK values (192.5±21.3 vs. 219.1±37.3 mg/dl, P<0.05) compared with those with normal LV diameters. Runners with a moderate and severe abnormal indexed LV mass >131 g/m 2 had statistically significant higher TNF-α values compared with runners, with mildly elevated and a normal LV mass indexed (16.2±1.42 vs. 14.0+1.16 pg/ml, P<0.05). Runners with an abnormal left atrial volume index (LAVI; >29 ml/m 2 ) had higher IL-6 values compared with runners with a normal LAVI (1.09+0.19 vs. 0.99±0.08 pg/ml, P<0.05). ROC curves analysis revealed that ADMA values were able to predict an abnormal LV diameter detected by echocardiography [P<0.05; area under the curve (AUC), 0.763], while TNF-α values could predict an abnormal LV mass in marathon runners (P<0.05; AUC, 0.78). On the whole, the present study demonstrates that, in ultra-marathon runners, cardiovascular adaptations to running are characterized by a specific pattern of changes in inflammatory and endothelial markers, which in turn can be used to predict the occurrence of the observed adaptations.