Arginine vasopressin compromises gut mucosal microcirculation in septic rats

OBJECTIVE: Arginine vasopressin (AVP) is increasingly used in the therapy of septic patients with hypotension. However, its effects on the microvascular networks have not been studied in detail. This study was designed to determine the effects of AVP infusion on the villus microcirculation of the septic rat ileum. DESIGN: Prospective, placebo-controlled, randomized, single-blinded trial. SETTING: University research laboratory. SUBJECTS: Fifteen male Sprague-Dawley rats. INTERVENTIONS: Twenty-four hours after cecal ligation and perforation to create sepsis (M1), rats (n = 8) received a continuous AVP infusion to increase mean arterial pressure by 20 mm Hg (M2) and 40 mm Hg (M3) from M1. In the control group (n = 7), an equivalent volume of normal saline was infused. MEASUREMENTS AND MAIN RESULTS: Videomicroscopy was performed on 6-10 villi of ileum mucosa at M1 and was repeated at M2 and M3. Blood was drawn to determine plasma levels of AVP and interleukin-6. At M1, both study groups were hypotensive compared with preseptic data (mean arterial pressure, -25%). The increase in mean arterial pressure was linked to supraphysiologic AVP plasma levels and was accompanied by a decrease in mean mucosal blood flow by 76% at M2 and 81% at M3 (p <.001 vs. control). Red blood cell velocity fell by 45% and 47%, respectively (p <.05 vs. control). Whereas periods of arrested villus blood flow increased from 8.1 +/- 2.6 secs/min to 43.8 +/- 5.2 and 47 +/- 6.2 secs/min at M2 and M3 (p <.001), the diameter of terminal arterioles remained unchanged. In addition, AVP infusion further augmented the sepsis-associated increase in interleukin-6 levels (AVP, 905 +/- 160 vs. control, 638 +/- 55 pg/mL; p =.022). CONCLUSIONS: This study provides evidence for severe abnormalities in gut mucosal blood flow after AVP infusion in septic rats, accompanied by an augmented inflammatory response to the septic injury. The effects of AVP on microvascular blood flow in this model may be related to AVP activities on larger arterioles (>40 microm), a concomitant reduction in cardiac output, or even both.