Recent studies indicate a reduced sensitivity to pain in genetically and experimentally hypertensive rats. A similar finding is also reported for human established hypertension. However, to our knowledge, no data have been reported in borderline hypertension. The aim of this study was to assess in subjects with borderline hypertension the sensory and pain thresholds by a noninvasive pulp-stimulation test performed with a commercial pulp tester that delivered stepwise increased electrical stimuli to four healthy teeth. The data reported are the means of measurements. In order to assess the possible importance of differences in the selection of the controls, two distinct groups of normotensives were chosen, one of volunteers and the other of outpatients. Significantly higher values for the pain and sensory thresholds were observed in borderline hypertensives compared with pooled normotensives. No significant effect on pain perception could be observed for sex, whereas a significant tendency toward higher threshold levels was found for younger subjects. In two normotensive groups, the sensory and pain thresholds were significantly higher in volunteers than in patients. These results suggest that changes in pain perception are present not only in established hypertension but also in borderline hypertension and, moreover, that differences in the selection of normotensive controls can have important influences on the results.