Abstract
Adrenergic and non-adrenergic vasopressor agents can be used to correct hypotension in shock states. For a similar increase in arterial pressure, these agents may be associated with different haemodynamic, metabolic, endocrinological or immunological effects. But how relevant are these differences? Do these affect the outcome of patients with shock? Large-scale randomized trials comparing the effects of different vasopressor agents are scarce. Data on potential alternatives, and especially vasopressin, are even more scarce. Over-interpretation of the data, and especially of data obtained in subgroups, is common. Analysis of subgroups may be useful to address mechanisms and to raise hypotheses. However, subgroup analysis is often biased by confounding factors, especially when subgroup categorization is defined by response to therapy and not by intrinsic patient or disease characteristics. In this issue of the British Journal of Pharmacology, Bracht and colleagues present their interpretation of data from trials comparing vasopressin with noradrenaline in patients with septic shock. Here, we present an alternative interpretation.