The possible association between low-level lead exposure and blood pressure (BP) remains debated. The purpose of this review was: (1) to determine whether the available studies in humans support a positive association, in particular at lower exposure levels (blood lead concentration < 1 mumol/l), and (2) to explore whether animal studies and the proposed pathophysiological mechanisms are supportive of a positive and causal association between lead exposure and hypertension. A meta-analysis of 23 studies included 33,141 subjects recruited from the general population in 13 surveys and from occupational groups in 10 studies. In all but four studies the results had been adjusted for age, and most studies also considered additional confounders. The association between BP and blood lead was similar in both sexes. In all 23 studies combined, a two-fold increase in blood lead concentration was associated with a 1 mm Hg rise in the systolic pressure (CI 0.4-1.6 mm Hg; P = 0.002) and with a 0.6 mm Hg increase in the diastolic pressure (CI 0.2-1.0 mm Hg; P = 0.02). Of 21 animal studies, one was carried out in dogs, one in pigeons and the remainder in various rat strains. In 15 studies, in which the lead dose in drinking water or food exceeded 1 p.p.m. the association between BP and exposure was found to be positive in seven, inconsistent in three, absent in four and negative in one. Of the six studies at lower exposure levels (< or = 1 p.p.m.), five found a pressor effect attributable to lead. Whether the lead doses in the animal studies are equivalent to the human exposure levels and to what extent one can extrapolate from genetically heterogeneous animals to humans, remains doubtful. If a causal relation between lead exposure and hypertension exists, the proposed mechanisms may include interference of lead with ion transport across cell membranes, interactions with calcium homeostasis and calcium-mediated processes, direct vasomotor actions and the potentiation of sympathetic stimulation. Interference of lead with the balance between the renin-angiotensin-aldosterone and the kallikrein-kinin systems and impairment of renal function are unlikely to be implicated. On balance, the published evidence suggests that there can only be a weak positive association between BP and lead exposure. The latter relation, which is barely visible at the horizon of epidemiological observation, may not be causal in nature and is unlikely to entail any public health implication in terms of hypertension-related complications.