BACKGROUND: orthostatic hypotension in elderly people is often attributed to diminished afferent baroreflex sensitivity, but this has not been demonstrated. We examined the hypothesis that postural change in blood pressure is related to baroreflex sensitivity, independent of the confounding effect of baseline blood pressure. METHODS: we studied 25 active, untreated elderly subjects free of postural symptoms (mean age 70 +/- 1 years): 16 with hypertension (clinic blood pressure 194 +/- 6/98 + 3 mmHg) and nine normotensive controls (clinic blood pressure 134 + 3/77 + 3 mmHg). We assessed baroreflex sensitivity from the heart rate and blood pressure responses to the Valsalva manoeuvre and a pressor and depressor stimulus (bolus phenylephrine injection or sodium nitroprusside infusion respectively). Subjects were then passively tilted to 60 degrees and maximum changes in systolic blood pressure, heart rate, forearm blood flow and forearm vascular resistance recorded. RESULTS: maximum change in systolic blood pressure with head-up tilt was correlated with supine systolic blood pressure (r = 0.60, P = 0.001). Maximum change in systolic blood pressure with orthostasis was greater in the hypertensive subjects (45 +/- 4 mmHg versus 29 +/- 6, P = 0.04) and the heart rate increment was less (16 +/- 2 bpm versus 24 +/- 4, P = 0.02). The increase in forearm vascular resistance with tilt was similar in the two groups (47 +/- 11 versus 38 +/- 7 units, P = 0.52). All three methods of assessing baroreflex sensitivity showed a reduction in the hypertensive subjects (all P < or = 0.02). Lower values of baroreflex sensitivity were related to greater falls in systolic blood pressure with tilt, after adjustment for the baseline level of systolic blood pressure. CONCLUSIONS: we found a relationship between baroreflex sensitivity and the systolic blood pressure fall with orthostasis, even after adjustment for prevailing systolic blood pressure. Despite equivalent changes in forearm vascular resistance with tilt, greater falls in systolic blood pressure were seen in hypertensive subjects than in normotensive controls, due in part to an inadequate baroreflex-mediated heart rate response. The postural fall in blood pressure often observed in elderly hypertensive subjects may be related to the reduced baroreflex sensitivity seen in this condition.