Résumé : Hypertonic sodium lactate (HTL) expands intravascular volume and may provide an alternative substrate for cellular metabolism in sepsis. We compared the effects of HTL, hypertonic saline (HTS), 0.9% ("normal") saline (NS) and Ringer's lactate (RL) on hemodynamics, sublingual and renal microcirculation, renal, mesenteric and brain perfusion, renal and cerebral metabolism, and survival in anesthetized, mechanically ventilated, adult female sheep. Animals (7 in each group) were randomized to receive a bolus (over 15-min) of 3 mL/kg 0.5M HTL, 3 mL/kg 3% HTS, 10.8 mL/kg NS, or 10.8 mL/kg RL at 2, 6, and 10 h after induction of fecal peritonitis, followed by 2-h infusions of 1 mL/kg/h (HTL/HTS groups) or 3.6 mL/kg/h (NS/RL groups). Animals also received RL and hydroxyethyl starch (ratio 1:1) titrated to maintain pulmonary artery occlusion pressure at baseline levels throughout the experiment. Animals were observed until their spontaneous death. Fluid balance was lower in the HTL and HTS groups than in the other groups from 4 h. Hemodynamic variables were similar among groups during the first 12 h, but thereafter the HTL group had more pronounced decreases in blood pressure and cardiac function. Sublingual and renal microcirculatory abnormalities occurred earlier in the HTL group. Kidney and brain perfusion decreased more rapidly in the HTL group. Median survival times were significantly shorter in the HTL (17 h) and NS (16 h) groups than in the HTS (22 h) or RL (20 h) groups (P=0.0029). In conclusion, in an ovine model of septic shock, administration of HTL was associated with earlier onset impaired tissue perfusion and shorter survival time. These observations raise concerns about use of HTL in septic shock.