From Wiki Journal Club
Jump to: navigation, search
Fujii T, et al. "Effect of Vitamin C, Hydrocortisone, and Thiamine vs Hydrocortisone Alone on Time Alive and Free of Vasopressor Support Among Patients With Septic Shock: The VITAMINS Randomized Clinical Trial". JAMA. 2020. 323(5):423-431.

Clinical Question

In patients with septic shock, does intravenous administration of vitamin C, thiamine, and hydrocortisone lead to faster resolution of septic shock than hydrocortisone alone?

Bottom Line

Intravenous vitamin C, thiamine, and hydrocortisone do not lead to faster resolution of septic shock compared to hydrocortisone alone.

Major Points

Prior studies shown that vitamin C has a dose-dependent attenuation of organ failure. Thiamine deficiency has been incidentally found in up to 20% of sepsis patients and supplementation helped cleared lactate. A retrospective, single-center, retrospective study of vitamin C, thiamine, and hydrocortisone treatment showed a shorted time on vasopressors and lower hospital mortality. This study examined hydrocortisone alone versus the combination therapy in a prospective, randomized trial.


No guidelines have been published that reflect the results of this trial.


  • Multicenter, open-label, parallel-group, randomized, controlled trial
  • N=216
    • Hydrocortisone, vitamin C, thiamine (n=107)
    • Hydrocortisone alone (n=104)
  • Setting: 10 ICUs in Australia, New Zealand, and Brazil
  • Enrollment: May 2018 to July 2019
  • Mean follow-up: 90 days
  • Analysis: Intention-to-treat
  • Primary outcomes: duration of time alive, free of vasopressor administration
  • Secondary outcomes: 28-day mortality, 90-day mortality, ICU-mortality, hospital-mortality, 28-day cumulative vasopressor-free days, 28-day cumulative mechanical ventilation-free days, 28-day renal replacement therapy-free days, 3 day SOFA score change, 28-day ICU-free days, hospital length of stay


Inclusion Criteria

  • ICU patients
  • Septic shock as primary diagnosis
  • Sepsis-3 guidelines fulfilled within 24 hours of enrollment (i.e., SOFA score increase >= 2, lactate > 2 mmol/l, vasopressor dependent for 2 hours)

Exclusion Criteria

  • Age < 18 years
  • DNR order in place
  • Imminent death
  • Diagnosis of septic shock more than 24 hours prior to enrollment
  • Known or strong contraindication to any of the study drugs
  • Indication for study drugs other than septic shock

Baseline Characteristics

  • Mean age: 61.7 years (63% male)
  • Weight (kg): 82.0
  • ICU admission source: ED 46%, OR planned 16%, OR elective surgery 5%, Ward 17%, Outside hospital: 11%, Another ICU 4%
  • Chronic Health Conditions: Type 2 Diabetes 24%, Chronic Renal Failure 7%

At time of Randomization

  • Mechanically Ventilated: 62%
  • Pressors
    • Norepinephrine: 93%
    • Vasopressin: 21%
    • Epinephrine: 10%
    • Metaraminol: 9%
  • Inotropes (Milrinone): 4%
  • Primary Site of Infection
    • Pulmonary: 30%
    • Gastrointestinal: 29%
    • Urinary: 15%
    • Skin or Soft Tissue: 14%
    • Blood: 5%
    • Other: 6%


  • Randomized to treatement with intravenous hydrocortizone (50 mg q6h) or IV vitamin C (1.5 g q6h), IV thiamine (200 mg q12h), and IV hydrocortizone (50 mg q6h) until resolution of shock or for 10 days, whichever came first.
  • Follow up at 28 days post-randomization and 90 days post-randomization.


Comparisons are three agent therapy vs. hydrocortisone alone.

Primary Outcomes

Time Alive and free of vasopressors (hours)
122.1 vs. 124.6 (HR -0.6; 95% CI -8.3-7.2; P=0.83)

Secondary Outcomes

28-d Mortality (No.)
24 vs. 21 (HR 2.3; 95% CI -8.9-13.4; P=0.69)
90-d Mortality (No.)
30 vs. 25 (HR 4.1; 95% CI -8.0-16.1; P=0.51)
ICU Mortality (No.)
21 vs. 19 (HR 1.4; 95% CI -9.2-11.9; P=0.80)
Hospital Mortality (No.)
25 vs. 21 (HR 3.0; 95% CI -8.2-14.1; P=0.60)
28-d Cumulative vasopressor free days
25.6 vs. 25.8 (HR -0.2; 95% CI -1.7-1.2; P=0.66)
28-d Cumulative mechanical ventilation free days
25.3 vs. 24.8 (HR 0.4; 95% CI -2.6-3.4; P=0.73)
28-d Renal replacement therapy-free days
28.0 vs. 28.0 (HR 0.0; 95% CI -0.6-0.6; P=0.71)
Change in SOFA score at day 3
-2 vs. -1 (HR -1.0; 95% CI -1.9 to -0.1; P=0.02)
28-d ICU-free days
21.9 vs. 22.1 (HR -0.2; 95% CI -4.1-3.7; P=0.66)
Hospital length of stay (days)
12.3 vs. 12.3 (HR 0.0; -4.9-4.9; P=0.75)

Subgroup Analysis

Study was underpowered for subgroup analysis.

Adverse Events

Two patients with two separate events were noted to have fluid overload and hyperglycemia, respectively. One patient in the control group had gastrointestinal bleeding.


  • The trial was not blinded, which may have affected biases. However, given the number of physicians involved in multiple ICUs with more than 100 attendning specialists and intensive care fellows, it is thought that these biases were mitigated.
  • The adjunct effects of vitamin C or thiamine were not assessed with hydrocortisone alone. The priority was given to the three-agent combination due to the need for better sepsis treatment options. Further analysis will be done in meta-analyses.
  • Thiamine levels were not directly measured which means the degree of hypovitaminosis was not captured, nor was the knowledge of correction of the deficit.
  • Target mean arterial pressures (MAP) were not collected for every patient by every attending intensivist, as such uniformity in vasopressor free days is lost.
  • Duration of antibiotic treatment for each patient is not known. Since every patient had received antibiotics prior to randomization, the randomization act was felt to achieve balance.
  • The trial was underpowered to detect differences in mortality and any subgroup analysis.
  • Patients were not specifically examined for adverse events, but rather only when they were treated for these events


  • Alfred Research Trusts Small Project Grant, the Intensive Care Foundation, and the Institutional Development Support Program of the Unified Health System.
  • Multiple authors with disclosures

Further Reading

[1] [2] [3] [4] [5] [6]