1. Plasma levels of vasopressin were found to be significantly higher in 29 patients with malignant hypertension than in 106 normotensive control subjects (13 ± 2 ng/l vs 5.8 ± 0.2 ng/l, P < 0.001).
2. No correlation was apparent between blood pressure and circulating vasopressin.
3. An acute, incremental infusion of vasopressin (0.125, 0.25 and 0.5 ng min−1 kg−1) in nine patients with mild-moderate essential hypertension produced only minor changes in blood pressure despite achieving plasma levels spanning those seen in malignant hypertension.
4. Despite a small increase in pressor sensitivity to vasopressin in hypertensive subjects it is thus unlikely that an acute vasoconstrictor effect of this peptide is important in the pathogenesis of malignant hypertension.