Casual blood pressure (BP) can predict the development of cardiovascular morbidity and mortality, but the correlations between its values and the subsequent occurrence of such complications are low. This may depend on different individual resistance to the damage produced by hypertension. However, it may also depend on the recognized inability of causal BP to reflect accurately the 24-h mean and profile BP. In order to test the latter hypothesis, 24-h BP was recorded intra-arterially (Oxford method) in 108 hospitalized subjects with essential hypertension ranging from mild to severe. The 24-h means and standard deviations (i.e. variabilities) for systolic, mean and diastolic BP obtained by computer analysis of the BP tracing were related to the rate and severity of target-organ damage (TOD) assessed by clinical examination and quantified according to a predetermined score. The results confirmed that 24-h BP may be variably different from cuff BP among subjects. For nearly any value of cuff BP, subjects in whom the 24-h mean BP was low had a lower prevalence and severity of TOD than those in whom the 24-h mean BP was high (P less than 0.01). Furthermore, for nearly any level of 24-h mean BP, subjects in whom the 24-h BP variability was low had a lower prevalence and severity of TOD than those in whom the 24-h BP variability was high (P less than 0.05). These findings demonstrate that the severity of hypertension is more closely related to 24-h mean BP than to cuff BP values.(ABSTRACT TRUNCATED AT 250 WORDS)