Intravenous vitamin C administration to patients with septic shock: a pilot randomised controlled trial

Abstract

Background: Intravenous vitamin C administration in septic shock may have a sparing effect on vasopressor requirements, and vitamin C's enzyme cofactor functions provide a mechanistic rationale. Our study aimed to determine the effect of intravenous vitamin C administration on vasopressor requirements and other outcomes in patients with septic shock.

Methods: This was a double-blind, randomised placebo-controlled trial in 40 patients with septic shock who were randomised (1:1) to receive intravenous vitamin C (at a dose of 25 mg/kg of body weight every 6 h) or placebo (intravenous 5% dextrose) for up to 96 h, or until death or discharge. The primary outcome was intravenous vasopressor requirements (dose and duration), and secondary outcomes included Sequential Organ Failure Assessment (SOFA) scores, intensive care unit (ICU) and hospital length of stay, and mortality. In addition, blood samples were collected to determine vitamin C kinetics and inflammatory marker concentrations.

Results: Median plasma vitamin C concentrations were deficient at baseline (9.2 [4.4, 12] µmol/L) and increased to 408 (227, 560) µmol/L following 72 h of intervention. The mean duration of intravenous vasopressor infusion in the vitamin C group was 48 (95% CI 35-62) hours and in the placebo group was 54 (95% CI 41-62) hours (p = 0.52). The dose of vasopressor delivered over time was comparable between the two groups, as were SOFA scores (p > 0.05). The median ICU length of stay in the intervention group was 3.8 (2.2, 9.8) days versus 7.1 (3.1, 20) days in the placebo group (p = 0.12). The median hospital length of stay for the vitamin C group was 18 (11, 35) days versus 22 (10, 52) days for the placebo group (p = 0.65). Mortality was comparable between the two groups (p > 0.05). Of the inflammatory markers, neutrophil counts were elevated in the vitamin C group relative to placebo by 72 h (p = 0.01). C-reactive protein and myeloperoxidase concentrations were elevated at baseline, however, the two groups were comparable over time (p > 0.05).

Conclusions: Our pilot study indicated that intravenous vitamin C did not provide significant decreases in the mean dose or duration of vasopressor infusion. Further research that takes into account the potential impact of intervention timing, dose and duration, and location of trial, may provide more definitive evidence.

Trial registration: ACTRN12617001184369 (11/8/2017).

Keywords: ICU length of stay; Noradrenaline; Sepsis; Septic shock; Vasopressor; Vitamin C.

Fig. 1

CONSORT flow diagram

Fig. 2

Plasma vitamin C concentrations for patients with septic shock who received intravenous vitamin C relative to placebo. Intravenous vitamin C was administered at a dose of 25 mg/kg six-hourly, and blood samples were collected daily to analyse vitamin C concentrations by HPLC. Box plots show median values with 25th and 75th percentiles as boundaries and whiskers indicate range; circles indicate mean values and 95% CI

Fig. 3

Dose and duration of intravenous vasopressors delivered to patients with septic shock by treatment group. a Dot plot of mean vasopressor dose (units/min) delivered over the four-day study period (96 h) or until death by treatment group. Filled circles represent mean dose for each individual patient, black diamond with vertical line represent group means with 95% CIs. b Kaplan Meier plot of time from randomisation until cessation of vasopressor therapy by treatment groups. There was no significant difference between the patients who received intravenous vitamin C and those who received the placebo in mean dose (p = 0.35) nor duration (p = 0.64) of vasopressor administration. See Additional file 1: Figure S1 for the individual data

Fig. 4

Kaplan–Meier plot for ICU length of stay of the study participants. Lines (bands) indicate the percentage of patients (95% CIs) still in ICU by time in days since study randomisation. Crosses indicate participants censored for death. There was no significant difference between the intervention and placebo groups (p = 0.12, log rank test)