Evaluating Vitamin C in Septic Shock

A Randomized Controlled Trial of Vitamin C Monotherapy

David A. Wacker, MD, PhD; Susan L. Burton, MD; John P. Berger, DO; A. J. Hegg, MD; Jamie Heisdorffer, MPH, RN; Qi Wang, MS; Emily J. Medcraft, PharmD; Ronald A. Reilkoff, MD

Disclosures

Crit Care Med. 2022;50(5):e458-e467.

Abstract and Introduction

Abstract

Objectives: To determine whether IV vitamin C therapy reduces 28-day mortality in patients with septic shock.

Design: Multicenter, double-blinded, randomized controlled trial.

Setting: One academic medical ICU and four community ICUs.

Patients: Of 167 adult patients within 24 hours of vasopressor initiation for septic shock, 126 consented to participation, and 124 received study drug and were included in analysis.

Interventions: IV vitamin C (10 mg/mL in normal saline) administered as a 1,000-mg bolus over 30 minutes followed by continuous infusion of 250 mg/hr for 96 hours or placebo of equal volumes of normal saline.

Measurements And Main Results: Of 124 subjects receiving study drug and included in analysis, 60 received vitamin C and 64 placebo. The primary outcome of all-cause 28-day mortality (vitamin C, 26.7%; placebo, 40.6%; p = 0.10) was lower in the vitamin C arm but did not reach statistical significance. Initiation of renal replacement therapy was higher in the vitamin C arm (vitamin C, 16.7%; placebo, 3.3%; p = 0.015), as was volume of fluid administration within 6 hours of study drug initiation (vitamin C, 1.07 L; placebo, 0.76 L; p = 0.03). There were no statistically significant differences in other secondary outcomes. In post hoc subgroup analysis, there was a decrease in 28-day mortality in the vitamin C arm among patients requiring positive-pressure ventilation at the time of enrollment (vitamin C, 36.3%; placebo, 60.0%; p = 0.05). This trial is registered at clinicaltrials.gov under identifier NCT03338569.

Conclusions: Vitamin C monotherapy failed to significantly reduce mortality in septic shock patients as hypothesized. Our findings do not support its routine clinical use for this purpose.

Introduction

Sepsis and septic shock are common reasons for admission to the ICU with 49 million cases and 11 million sepsis-associated deaths each year worldwide.^[1] A therapeutic role for vitamin C (ascorbic acid) supplementation in the treatment of sepsis and septic shock has been suggested for its anti-inflammatory and antioxidant properties,^[2–4] and in 2017, an observational before-and-after study reported a remarkable survival benefit with the use of a combination of hydrocortisone, vitamin C (ascorbic acid), and thiamine (HAT therapy).^[5] Following this, several randomized controlled trials of HAT therapy^[6–11] or vitamin C and thiamine in combination^[12] for sepsis and septic shock have been conducted, but these studies did not show the same dramatic effects. Although modest improvements in time to reversal of shock and organ dysfunction were occasionally noted, no study demonstrated a statistically significant survival benefit.^[14–16] One study of vitamin C alone for treatment of patients with sepsis-induced acute respiratory distress syndrome (ARDS) did show a reduction in all-cause 28-day mortality in the vitamin C arm with a p value of 0.03; however, this was one of numerous secondary outcomes that obfuscates the statistical significance of the result.^[13]

Despite the existing trials of HAT therapy, questions about vitamin C as adjuvant monotherapy remain. First, prior investigations into vitamin C alone have been limited to small pilot studies and unblinded trials,^[2–4] and although the interactions between constituents in HAT therapy have been assumed to be synergistically positive,^[5] this has never been proven. Unexpected interactions may cancel out or otherwise convolute effects, thus limiting the ability to judge the performance of individual components from trials of a combined regimen. Second, randomized controlled trials of adjuvant corticosteroids^[14–16] for septic shock have recently demonstrated clinical benefit with a consistent signal of decreased duration of vasopressor therapy, which has the potential to confound any positive results from prospective investigations of HAT therapy.

In designing our trial, we noted pharmacokinetic studies of vitamin C, which have shown that trough levels of the drug occurring with bolus dosing may result in periodic hypovitaminosis, whereas continuous vitamin C delivery resulted in more stable serum levels.^[17] We, therefore, designed the Evaluating Vitamin C in SepTic Shock trial to assess the effect of continuously infused vitamin C monotherapy on all-cause ICU and 28-day mortality in patients with septic shock.

1 2 3 4 5

Next Section

Table 1. Baseline Characteristics of Subjects Included in Analysis
Characteristic	Vitamin C Group (n = 60)	Placebo Group (n = 64)	p
Age, yr, median (IQR)	68.9 (60.1–79.9)	73.0 (60.8–80.0)	0.47
Sex male, n (%)	30 (50)	33 (52)	0.71
Race, n (%)
Asian	2 (3.3)	2 (3.1)	0.95
Black or African American	0 (0.0)	2 (3.1)	0.50
White	58 (96.7)	59 (91.8)	0.28
Unknown or declined to answer	0 (0.0)	1 (2)	> 0.99
Ethnicity, n (%)
Latino/Latina	0 (0.0)	0 (0.0)	> 0.99
Baseline Sequential Organ Failure Assessment score, median (IQR)	10 (7–11)	9 (7–12)	0.63
Baseline Acute Physiology and Chronic Health Evaluation II score, median (IQR)	22 (16.5–28)	23 (17–28.5)	0.38
Source of infection, n (%)
Pulmonary	13 (21.6)	16 (25.0)	0.66
Urinary	16 (26.6)	14 (21.9)	0.53
Gastrointestinal/biliary	12 (20.0)	14 (21.9)	0.80
Soft tissue/skin	4 (6.7)	7 (10.9)	0.40
Primary bacteremia	6 (10.0)	4 (6.2)	0.44
Endocarditis	1 (1.7)	2 (3.1)	0.60
Bacterial meningitis	1 (1.7)	0 (0.0)	0.48
Other/unknown	8 (13.3)	7 (10.9)	0.68
Time to antibiotics,^a hr, n (%)
< 1	32 (53.3)	28 (43.8)	0.29
1–3	22 (36.7)	24 (37.5)	0.92
3–6	5 (8.3)	9 (14.1)	0.31
> 6	2 (3.3)	3 (4.7)	0.70
Total fluid administered preenrollment (mL), median (IQR)	4,270 (3,305–5,781)	3,810.5 (3,000–5,800)	0.40
Positive-pressure ventilation (noninvasive or via endotracheal tube) at enrollment, n (%)	34 (56.7)	37 (57.8)	0.90
Received steroids, n (%)	30 (50.0)	42 (65.6)	0.08
Received thiamine, n (%)	4 (6.7)	5 (7.8)	0.81
Time from pressor initiation to study drug initiation (hr), median (IQR)	11.4 (5.4–17.3)	8.8 (5.4–17.4)	0.21

IQR = interquartile range.
^aDefined as the amount of time between the patient meeting criteria for sepsis (either two or more systemic inflammatory response syndrome criteria positive or decrease from baseline Sequential Organ Failure Assessment score of 2 or more) and administration of antibiotics.

Table 2. Primary and Secondary Outcomes
Outcome	Vitamin C Group	Placebo Group	p
Primary outcome
28-d mortality, n (%)	16 (26.7) (n = 60)	26 (40.6) (n = 64)	0.10
Secondary outcomes
ICU mortality, n (%)	14 (23.3) (n = 60)	20 (31.1) (n = 64)	0.32
Organ failure scores
Paired improvement in Sequential Organ Failure Assessment score,^a median (IQR)	3.5 (1–6) (n = 58^b)	4 (1–6) (n = 61^b)	0.68
Paired improvement in Acute Physiology and Chronic Health Evaluation II score,^a median (IQR)	4.5 (2–9) (n = 58^b)	7 (–2 to 11) (n = 61^b)	0.22
Renal function outcomes
Paired improvement in creatinine (mg/dL),^a median (IQR)	0.4 (0–0.7) (n = 49^b–d)	0.3 (–0.1 to 0.7) (n = 56^b–d)	0.55
Renal replacement therapy required during 96-hr study period, n (%)	10 (16.7) (n = 60^d)	2 (3.3) (n = 60^d)	0.02
Lactate clearance,^e n (%)
Within 24 hr	12 (27.9) (n = 43^f)	13 (28.9) (n = 45^f)	0.92
Within 48 hr	16 (37.2) (n = 43^f)	18 (40.0) (n = 45^f)	0.79
Within 72 hr	18 (41.9) (n = 43^f)	18 (40.0) (n = 45^f)	0.86
Within 96 hr	18 (41.9) (n = 43^f)	19 (42.2) (n = 45^f)	0.86
Hospital length of stay following study drug initiation (d), median ( IQR)	8.9 (4.0–20.0) (n = 60)	6.3 (3.8–12.5) (n = 64)	0.15
ICU length of stay following study drug initiation (d), median (IQR)	2.9 (1.8–7.5) (n = 60)	2.6 (1.5–5.3) (n = 64)	0.47
Duration of pressors following initiation of study drug (hr), median (IQR)	27.7 (13.6–47.6) (n = 60)	27.1 (16.4–45.2) (n = 64)	0.79
Duration of mechanical ventilation following initiation of study drug (hr), median (IQR)	0 (0–60) (n = 60)	5 (0–48) (n = 64)	0.45
Total IV fluid administration (L), median (IQR)
6 hr after study drug initiation	1.07 (0.72–1.64) (n = 60)	0.76 (0.36–1.26) (n = 64)	0.03
24 hr after study drug initiation	3.59 (2.52–5.02) (n = 60)	3.37 (1.98–4.70) (n = 64)	0.47
Fluid balance (total intake minus output, L), median (IQR)
24 hr after study drug initiation	2.12 (1.10–3.59) (n = 60)	2.05 (1.24–3.11) (n = 64)	0.93
96 hr after study drug initiation	2.87 (1.70–4.30) (n = 60)	2.69 (0.69–4.84) (n = 64)	0.53

IQR = interquartile range.
^aBaseline-to-final measured value within 96-hr study period.
^bPatients for whom only baseline values were obtained were excluded from this analysis.
^cPatients receiving renal replacement therapy at any point during the 96-hr study period were excluded from this analysis.
^dPatients with dialysis dependence prior to hospital admission were excluded from this analysis.
^eClearance is defined as reaching a serum lactate level of 2.0 mmol/L or less.
^fPatients whose serum lactate level had cleared prior to study drug initiation, and those for whom only a baseline value was obtained were excluded from this analysis.

Table 3. Post Hoc Subgroup Analysis of Mortality Outcomes by Positive-Pressure Ventilation Status and Presence of Acute Hypoxemic Respiratory Failure at Enrollment
Outcome	Stratification by Positive-Pressure Ventilation^a Status at Enrollment						Stratification by Presence of Hypoxemic Respiratory Failure^b at Enrollment
	Positive-Pressure Ventilation Needed at Enrollment			Positive-Pressure Ventilation Not Needed at Enrollment			Hypoxemic Respiratory Failure Present at Enrollment			Hypoxemic Respiratory Failure Not Present at Enrollment
	Vitamin C Group (n = 33)	Placebo Group (n = 35)	p	Vitamin C Group (n = 27)	Placebo Group (n = 29)	p	Vitamin C Group (n = 24)	Placebo Group (n = 25)	p	Vitamin C Group (n = 36)	Placebo Group (n = 39)	p
28-d mortality, n (%)	12 (36.3)	21 (60.0)	0.05	4 (14.8)	5 (19.2)	> 0.99	8 (33.3)	14 (56.0)	0.11	8 (22.2)	12 (30.8)	0.40
ICU mortality, n (%)	10 (30.3)	17 (48.6)	0.12	4 (14.8)	3 (10.3)	0.70	7 (29.2)	11 (44.0)	0.28	7 (19.4)	9 (23.1)	0.70

^aPositive-pressure ventilation provided by either noninvasive positive-pressure ventilation or mechanical ventilation.
^bAcute hypoxemic respiratory failure defined as Pao₂/Fio₂ ratio less than 300 and need for positive-pressure ventilation with positive end-expiratory pressure or expiratory positive airway pressure of at least 5 cm H₂O.

processing....

Evaluating Vitamin C in Septic Shock

A Randomized Controlled Trial of Vitamin C Monotherapy

Abstract and Introduction

Abstract

Introduction

Tables

Tables

Tables

References

Authors and Disclosures

Authors and Disclosures

Most Popular Articles

Email This

Feedback