By Vibhu Sharma, MD, MS

Assistant Professor of Medicine, University of Colorado, Denver

Dr. Sharma reports no financial relationships relevant to this field of study.

SYNOPSIS: In this randomized, double-blinded, placebo-controlled trial, intravenous vitamin C infusion did not influence a change in the modified Sequential Organ Failure Assessment score from the time of infusion to four days compared to placebo.

SOURCE: Fowler AA 3rd, Truwit JD, Hite RD, et al. Effect of vitamin C infusion on organ failure and biomarkers of inflammation and vascular injury in patients with sepsis and severe acute respiratory failure: The CITRIS-ALI randomized clinical trial. JAMA 2019;322:1261-1270.

The CITRIS-ALI trial randomized critically ill patients with sepsis and acute respiratory distress syndrome (ARDS) to a vitamin C infusion delivered every six hours for 96 hours or an identically compounded placebo solution. Patients were included if they had ARDS by the Berlin definition, had suspected or proven infection, and had two of four of the systemic inflammatory response syndrome (SIRS) criteria (i.e., temperature > 38°C or < 36°C, tachycardia > 90 beats/minute, tachypnea > 20 breaths per minute, leukocytosis > 12,000, or leukopenia < 4,000 cells/microliter). All criteria had to be met within a 24-hour period to be included, and patients had to be randomized within 48 hours of meeting ARDS criteria. Patients were excluded if they were not expected to survive, had an allergy to vitamin C, were pregnant, or if they had an interstitial lung disease (ILD), alveolar hemorrhage, diabetic ketoacidosis (DKA), or an active kidney stone. The primary outcome was a change in Sequential Organ Failure Assessment (SOFA) scores at 96 hours after the start of infusion and a change in inflammatory biomarker levels (C-reactive protein [CRP] and thrombomodulin) at 168 hours. There were multiple (46) prespecified secondary outcomes, the most relevant of which were mortality, intensive care unit (ICU)-free days, and ventilator-free days.

The SOFA score was modified (“mSOFA” score) to exclude bilirubin because of multiple missing values. The authors contended that the exclusion of a SOFA component does not affect its predictive validity based on previously published data. Overall, 1,262 patients were screened for eligibility, and 170 patients were randomized to each group. The largest proportion (~50%) were excluded because of inability to consent, being outside of the 48 hours since meeting ARDS criteria, or receiving home oxygen. The two groups were well matched for all demographic characteristics, and baseline mSOFA scores were identical (9.8 vs. 10.3). Baseline demographics and proportion of comorbidities were comparable between groups. Baseline oxygenation (PaO2/FiO2 and the oxygenation index) also was comparable. Conservative fluid management was instituted in both groups as described in the FACTT-lite protocol.

The authors concluded that a 96-hour vitamin C infusion did not reduce the severity of illness scores over four days compared with placebo. The authors did find a difference in three of the 46 prespecified secondary outcomes: reduced mortality at 28 days, a decrease in the number of days in the ICU to day 28, and hospital-free days to day 60 favored the vitamin C infusion group. There were no differences in the other 43 prespecified secondary outcomes. Additionally, inflammatory biomarker levels (CRP, procalcitonin, angiopoietin, receptor for advanced glycation end [RAGE] products, tissue factor pathway inhibitor [TFPI] levels, and thrombomodulin) were not attenuated over the course of the four days of observation.

COMMENTARY

The plausibility of a clinical effect of vitamin C in the setting of shock and ARDS is well founded. Vitamin C is an antioxidant, and low levels have been demonstrated in the setting of sepsis.1 Vitamin C also increases the synthesis of norepinephrine and vasopressin, having been defined as a cofactor in the synthesis of these molecules.2 Furthermore, vitamin C has been shown to attenuate increases in cytokine levels that lead to activation and sequestration of neutrophils in the lungs in vitro.3 Marik et al published a retrospective before-after study4 assessing the effect of an infusion of hydrocortisone, vitamin C, and thiamine (“HAT” therapy) in the setting of severe sepsis and septic shock. They found an adjusted odds ratio for mortality of 0.13 in favor of HAT therapy. This study has led to a plethora of studies in various phases of completion and has led, arguably, to a premature adoption of the “Marik Protocol” in some institutions.

This study was designed to assess the effect of vitamin C infusion on relevant molecular and clinical outcomes among patients with ARDS in the setting of sepsis in a medical intensive care unit. Patients who developed sepsis resulting from a surgical intervention or who were in a surgical intensive care unit were excluded. While no randomized trials have yet reported results assessing the benefit of vitamin C in sepsis/septic shock, the authors decided to test their hypothesis that given the downstream salutary effects of vitamin C on the coagulation cascade, there would be a benefit in the setting of sepsis-associated ARDS.

There are a few points worthy of attention with respect to the conclusions of the trial. First, while there was a significant difference between the secondary endpoints of importance (mortality, hospital-free days, and ICU-free days), the authors noted that the study was a proof-of-concept trial not designed to assess a mortality difference. These three endpoints were the only significant differences between the groups in a total of 46 secondary endpoints, and they were “based on analyses that did not account for multiple comparisons and therefore must be considered exploratory.” Patients were recruited up to 48 hours after meeting ARDS criteria; as a result, some randomized patients were in an advanced state of disease development, preventing a salutary effect of vitamin C (if any) from taking hold. Second, a large proportion of patients were excluded because either they were unable to provide consent or they were outside the window of randomization. This may have introduced an element of systematic error. Third, the majority of patients in both groups (more than 70%) had ARDS due to a primary thoracic etiology (presumably pneumonia) with a slightly higher proportion (82% vs. 70%) randomized to the vitamin C group. While low tidal volume ventilation affects both ARDS due to pneumonia or aspiration and ARDS due to sepsis (i.e., extrathoracic causes) with respect to a mortality benefit, there are substantial differences in mortality between these two groups (i.e., higher mortality in patients with non-thoracic sepsis).5 Therefore, a larger trial that includes more patients with ARDS due to sepsis of non-thoracic etiology may reach different conclusions with respect to both primary and secondary endpoints. Fourth, the fluid balance was comparable in both groups at day 1; however, at 96 hours (the primary time endpoint of the study), patients receiving placebo had a statistically significantly different net negative fluid balance of approximately 800 mL compared with the vitamin C group. Fluid balance in the placebo group continued to be negative out to day 7 (net negative approximately 500 mL, not statistically significant). This was despite a conservative fluid strategy instituted in both groups. These differences in fluid balance are comparable to those published in the FACTT trial at day 4 and day 7. Fluid balance clearly has been shown to affect both ventilator-free days and ICU-free days, and the differences in fluid balance here make the differences in the secondary endpoints circumspect. The proportion of patients with acute kidney injury (AKI) was equivalent in both groups at enrollment; however, the proportion needing continuous renal replacement therapy (CRRT) over the ensuing week in either group was not reported, and whether more frequent CRRT was the reason for the greater negative fluid balance in the placebo group cannot be deciphered. Finally, more than half of the patients in each group received corticosteroids, but the dose of steroids administered or the duration was not reported. While this remains controversial, both the dose and duration of corticosteroid therapy may influence outcomes in the setting of ARDS6 and certainly in the setting of severe community-acquired pneumonia,7 introducing yet another confounding factor into the results of this study.

It also is important to keep in mind the potential adverse effects of a high-dose vitamin C infusion. Descriptions have included an oxalate nephropathy and interference with point-of-care blood glucose measurements, which is why patients with DKA were excluded. Given the exploratory nature of this trial, it seems premature to recommend vitamin C to all patients with sepsis and ARDS. Careful fluid management is an established intervention with respect to outcomes of interest in this trial, and it ought to be the focus of management rather than who may benefit from vitamin C. Whether “HAT” therapy makes a difference in the setting of sepsis alone is being studied in a randomized, controlled trial. Whether vitamin C works in a larger trial incorporating more patients with ARDS related to sepsis of non-thoracic origin remains to be seen.

REFERENCES

  1. Nakano K, Suzuki S. Stress-induced change in tissue levels of ascorbic acid and histamine in rats. J Nutr 1984;114:1602-1608.
  2. Carr AC, Shaw GM, Fowler AA, Natarajan R. Ascorbate-dependent vasopressor synthesis: A rationale for vitamin C administration in severe sepsis and septic shock? Crit Care 2015;19:418.
  3. Fisher BJ, Kraskauskas D, Martin EJ, et al. Mechanisms of attenuation of abdominal sepsis induced acute lung injury by ascorbic acid. Am J Physiol Lung Cell Mol Physiol 2012;303:L20-L32.
  4. Marik PE, Khangoora V, Rivera R, et al. Hydrocortisone, vitamin C, and thiamine for the treatment of severe sepsis and septic shock: A retrospective before-after study. Chest 2017;151:1229-1238.
  5. Eisner MD, Thompson T, Hudson LD, et al. Efficacy of low tidal volume ventilation in patients with different clinical risk factors for acute lung injury and the acute respiratory distress syndrome. Am J Respir Crit Care Med 2001;164:231-236.
  6. Tongyoo S, Permpikul C, Mongkolpun W, et al. Hydrocortisone treatment in early sepsis-associated acute respiratory distress syndrome: Results of a randomized controlled trial. Crit Care 2016;20:329.
  7. Stern A, Skalsky K, Avni T, et al. Corticosteroids for pneumonia. Cochrane Database Syst Rev 2017;12:CD007720.