INTRODUCTION: Thiamine (TH) is a co-factor for pyruvate dehydrogenase, an enzyme necessary for pyruvate entry into the Krebs cycle, and without this enzyme, pyruvate would be converted to lactate. Elevated lactate, which is often used as a marker of perfusion, is proportionally associated with increased mortality in septic shock. The few publications on TH in septic shock are inconclusive. This study aims to ascertain if there is benefit to adding TH to standard of care (SOC) in the management of septic shock. METHOD(S): IRB-approved, multicenter, retrospective review from 2016 to 2021. Adult patients admitted to the ICU for septic shock and receiving >= 400 mg a day of IV TH (in divided doses) were included. Patients < 18, pregnant, admitted for SARS-COV-2, or whom received < 400 mg of TH daily were excluded. Two matched cohorts were evaluated, SOC plus TH versus SOC alone. The primary endpoint is time to shock reversal, defined as off vasopressors for at least 12 hrs and alive. Secondary endpoints include time to lactate clearance (< 2 mmol/L), lactate trends at 6, 12, 24, 48 hrs, and end of therapy, hospital and ICU lengths of stay, new end organ dysfunction, and in-hospital mortality. RESULT(S): Data from 50 patients were analyzed: 25 in the SOC plus TH and 25 in the SOC arm. The TH arm had greater number of vasopressors (2 vs. 1, p=.019), and greater utilization of stress-dose steroids (72% vs. 8%, p<.001), however there was no difference in cumulative vasopressor dose in norepinephrine equivalents at baseline (BL) (30.1 vs. 25.8 mcg/min, p=.248). There was no difference in SOFA score at ICU admission (10 vs. 8.5, p=.106) or lactate level at ICU admission (5.9 vs. 3.9 mmol/L, p=.055). There was a longer time to shock reversal from vasopressor initiation time in the TH arm (93 vs. 37.1 hrs, p=.023). Lactate clearance was slower in the TH arm (44.75 vs. 15.8 hrs, p=.027), and there was increased in-hospital mortality in the TH arm (13 vs. 5, p=.018). CONCLUSION(S): Compared to SOC alone, TH treated patients had longer times to shock reversal. However, this outcome may have been confounded by differences at BL with regard to number of vasopressors, and stress-dose steroid utilization, which indicate these patients were sicker at BL. Larger, prospective studies are required to confirm these findings.