Background It has been hypothesized that the level of end-systolic wall stress (σ_m) is a feedback signal that regulates the level of hypertrophy. Thus, low levels of σ_m may signify inappropriate hypertrophy. Methods To characterize left ventricular (LV) structure and systolic function in hypertensive subjects with low levels of σ_m, we studied 763 patients. LV function was studied by midwall stress-shortening analysis. Partition values for σ_m were derived from a separate group of normal subjects, and the study population was divided into low stress (group I, n = 136), high stress (group III, n = 157), and intermediate stress group II (n = 470). LV chamber and myocardial function were characterized by relating shortening at the endocardium and at the midwall, respectively, to stress. Results As expected, group III patients had the highest values for systolic blood pressure and LV cavity size but the lowest values for wall thickness and relative wall thickness. Surprisingly, however, there were no significant differences among stress groups with regard to age or body mass index. Contrary to the hypothesis that low levels of stress are indicative of excessive hypertrophy, there were no significant differences among the 3 groups with regard to LV mass or any form of LV mass index. Furthermore, despite lower mean values for afterload, group I patients had significantly lower values for midwall shortening, and this finding was indicative of reduced myocardial function; stress-shortening plots demonstrated that 28% of group I patients fell below 95% CI compared with 10% of group II and only 5% of group III patients. Conclusions Hypertensive subjects with low values for σ_m have more concentric LV geometry (higher relative wall thickness) and, on average, reduced myocardial function. (Am Heart J 2002;143:546-51.)