Abstract
Background: Vasopressin is one of the strong vasopressor agents associated with ischemic events. Responses to the administration of vasopressin differ among patients with septic shock. Although the administration of a high dose of vasopressin needs to be avoided, the effects of bolus loading have not yet been examined. Since the half-life of vasopressin is longer than that of catecholamines, we hypothesized that vasopressin loading may be effective for predicting responses to its continuous administration. Methods: We retrospectively analyzed consecutive cases of septic shock for which vasopressin was introduced with loading under noradrenaline at >0.2 μg/kg/min during the study period. Vasopressin was administered in a 1 U bolus followed by its continuous administration at 1 U/h. The proportion of patients with a negative catecholamine index (CAI) change 6 h after the introduction of vasopressin was set as the primary outcome. We defined non-responders for exploration as those with a mean arterial pressure change <18 mmHg 1 min after vasopressin loading, among whom none had a change in CAI <0. Results: Twenty-one consecutive cases were examined in the present study, and included 14 responders and 7 non-responders. The primary outcome accounted for 71.4% of responders and 0% of non-responders, with a significant difference (p = 0.0039). Median CAI changes 2, 4, and 6 h after the administration of vasopressin were 0, -5, and -10 in responders and +20, +10, and +10 in non-responders, respectively. CAI was not reduced in any non-responder. Outcomes including mortality were not significantly different between responders and non-responders. Digital ischemia (1/21) and mesenteric ischemia (1/21) were observed. Conclusions: Vasopressin loading may predict responses to its continuous administration in septic shock patients. Further investigations involving a safety analysis are needed.