There is no consensus on the definition of an exaggerated increase in systolic blood pressure (SBP) during exercise. The aim was to explore a potential threshold for exercise SBP associated with increased risk of coronary heart disease (CHD) in healthy men, using repeated exercise testing. 2,014 healthy Caucasian male employees were recruited into the Oslo Ischemia Study during early 1970s. At follow-up seven years later, 1,392 men were still considered healthy. A bicycle exercise test at 100W workload was performed at both visits. Cox regression analyses were performed with increasing cutoff levels of peak SBP during 100W workload (SBP100W) from 160 mmHg to 200 mmHg, adjusted for cardiovascular risk factors and physical fitness. Participants with SBP100W below cutoff level at both baseline and first follow-up were compared to participants with SBP100W equal to or above cutoff level at both visits. Compared to participants with SBP100W below all cutoff levels between 165-195 mmHg, CHD risk was increased amongst participants with SBP100W equal to or above cutoff at all levels. There was no evidence of a distinct threshold level for CHD risk, and the relation between SBP100W and CHD appears linear. When investigating exercise SBP at moderate workload measured at two exercise tests in healthy middle-aged Caucasian men, there is increasing risk of coronary disease with increasing exercise SBP independent of SBP at rest. The association is linear from the low range of exercise SBP and there is no sign of a distinct threshold level for increased coronary disease risk. Participants were divided into groups defined by SBP100W at Visit 1 and Visit 2 (Figure S1). Group 1 includes participants with SBP100W below cutoff level at both Visit 1 and Visit 2. Group 3 includes participants with SBP100W equal to or above cutoff level at both Visit 1 and 2, and Group 2 includes participants with SBP100W equal to or above cutoff level at only one of the two visits, regardless of which. To investigate the risk of CHD, we performed Cox regression analyses adjusted for age, resting SBP, total serum cholesterol, smoking status and family history of CHD. We also performed additional analyses adjusting for physical fitness. Physical fitness was defined as total workload during the exercise test measured in kilojoules divided by body weight. Due to the increases in cutoff levels, the number of participants in each group successively changed in each analysis, as Group 1 expanded and Group 3 diminished.