We hypothesized that lactate levels even within the normal range are prognostic and that low lactate levels predict a beneficial response to vasopressin infusion in septic shock. We conducted a retrospective analysis using the Vasopressin in Septic Shock Trial (VASST) as a derivation cohort (n = 665), then validated using another single-center septic shock cohort, St Paul's Hospital (SPH) cohort (n = 469). Lactate levels were divided into quartiles. The primary outcome variable was 28-day mortality in both cohorts. We used receiver operating characteristic (ROC) curve analysis to compare the prognostic value of lactate concentrations versus Acute Physiology and Chronic Health Evaluation II scores. We then explored whether lactate concentrations might predict beneficial response to vasopressin compared with noradrenaline in VASST. Normal lactate range is less than 2.3 mmol/L. At enrollment, patients in the second quartile (1.4 < lactate < 2.3 mmol/L) had significantly increased mortality and organ dysfunction compared with patients who had lactate ≤ 1.4 mmol/L (quartile 1) (P < 0.0001). Quartile 2 outcomes were as severe as quartile 3 (2.3 ≤ lactate < 4.4 mmol/L) outcomes. Baseline lactate values (ar ea under the ROC curve = 0.63, 0.66; VASST, SPH) were as good as Acute Physiology and Chronic Health Evaluation II scores (area under the ROC curve = 0.66, 0.73; VASST, SPH) as prognostic indicators of 28-day mortality. Lactate concentrations of 1.4 mmol/L or less predicted a beneficial response in those randomized to vasopressin compared with noradrenaline in VASST (P < 0.05). Lactate concentrations within the "normal" range can be a useful prognostic indicator in septic shock. Furthermore, patients whose lactate level is less than or equal to 1.4 mmol/L may benefit from vasopressin infusion.