A preclinical randomised controlled dose optimization of megadose sodium ascorbate for reversal of gram-negative sepsis-induced cardiovascular, brain and kidney dysfunction.

BACKGROUND: Megadose sodium ascorbate has shown promise as a treatment to reverse the pathophysiological effects of ovine Gram-negative sepsis. In human septic shock, lower doses of sodium ascorbate improved urine output and reduced vasopressor requirements compared with placebo. We sought to determine the minimum therapeutic dose of sodium ascorbate required to reverse sepsis-induced cardiovascular and renal dysfunction in sheep. METHODS: Healthy young adult sheep were instrumented with renal artery flow probes, and oxygen-sensing and laser Doppler probes in the kidneys. Non-anaesthetised animals were infused with live Escherichia coli for 31-h. At 23.5-h of sepsis, four groups (n = 7-8/group) received fluid resuscitation (30 mL/kg Hartmann's solution) and were randomized to intravenous sodium ascorbate (1.0, 2.0, or 3.0 g/kg) or vehicle, delivered as a bolus followed by 7-h infusion. Norepinephrine was titrated to maintain mean arterial pressure (MAP) at ~ 70 mmHg. RESULTS: At 23-h of sepsis, animals developed hypotension, hyperlactatemia, acute kidney injury, and renal medullary hypoxia. Vehicle-treated sheep required escalating doses of norepinephrine (from 0.4 to 0.8 ± 0.2 µg/kg/min) to restore MAP. Sodium ascorbate at 3.0 g/kg (achieving plasma ascorbate levels of ~ 10 mmol/L) rapidly restored MAP, allowing withdrawal of norepinephrine in half the animals (P = 0.007). Lower doses of sodium ascorbate (1.0 and 2.0 g/kg) had no significant effect on vasopressor requirements. The improvements in renal medullary oxygenation (25.2 ± 3.3 to 43.4 ± 4.5 mmHg, P = 0.04) and urine flow (from 0.5 ± 0.2 to 6.9 ± 2.4 ml/kg/h, P < 0.0001) were dose-dependent. Renal medullary tissue protein expression of nuclear factor kappa-light chain-enhancer B was significantly reduced with 3.0 g/kg of sodium ascorbate (to -52.9 ± 13.3%, P = 0.0005) and phosphorylated endothelial nitric oxide synthase at Ser-1177 was upregulated (to +219.5 ± 51.4%, P = 0.04) compared with vehicle-treated sheep. CONCLUSIONS: In established ovine Gram-negative sepsis, only 3.0 g/kg sodium ascorbate effectively restored cardiovascular and renal dysfunction, which was associated with suppression of renal inflammatory signalling and restoration of endothelial nitric oxide activity. These findings demonstrate a clear dose-dependent therapeutic threshold, where achieving plasma ascorbate concentrations of ~ 10 mmol/L is essential to elicit multi-organ protection.