The debate over vitamin C for sepsis

Mixed trial results have stirred controversy.


A storm has been brewing in critical care. It's about whether to use vitamin C, either alone or in combination with thiamine and hydrocortisone, to treat sepsis.

The excitement started in June 2017, when Paul Marik, MD, FACP, professor of medicine at Eastern Virginia Medical School in Norfolk, and colleagues published a retrospective study in CHEST. Forty-seven septic patients who were treated with a cocktail of vitamin C, thiamine, and hydrocortisone had an inpatient mortality rate of 8.5%, compared to 40.4% among 47 patients who did not receive that treatment.

However, a larger, randomized trial didn't find significant improvements in any of its primary outcomes when patients with sepsis and acute respiratory distress syndrome (ARDS) were given IV vitamin C. The CITRIS-ALI trial, published in JAMA on Oct. 1, 2019, initially found similar modified Sequential Organ Failure Assessment (SOFA) scores and levels of C-reactive protein and thrombomodulin in 167 intervention and control patients. But the groups did differ on the secondary outcome of 28-day mortality: 46.3% with placebo versus 29.8% in the vitamin C group.

Then there was the VITAMINS trial, a randomized investigation of vitamin C, thiamine, and hydrocortisone in 216 septic shock patients. Results published by JAMA on Jan. 20 showed no difference in time alive and free of vasopressor administration up to day seven in those who received the intervention compared to patients receiving hydrocortisone alone.

Image by Getty Images
Image by Getty Images

Given these differing findings, the issue of vitamin C in sepsis is likely to be “contentious for a long time,” according to the lead author of the CITRIS-ALI study, Alpha A. Fowler III, MD, who is the William Taliaferro Thompson professor of medicine in the division of pulmonary disease and critical care medicine at Virginia Commonwealth University in Richmond.

Explaining the findings

One major mystery in the vitamin C debate is why the CITRIS-ALI trial could find an effect on mortality without any change in the outcomes that presumably reflect the chances of survival, such as SOFA scores and time on vasopressors.

Dr. Fowler now believes the trial had a survivorship bias. He and colleagues published a reanalysis of the data as a letter in the Feb. 25 JAMA which shows a significant reduction in the SOFA score with vitamin C when all patients, including those who died, were included.

“Changing the practice of medicine is like turning a supertanker: It doesn't happen except by one inch at a time. And I think that's kind of where we are with vitamin C,” he said. “We're going to hang in there. We have our ideas as to why vitamin C is working from all the basic work that we did in mice.”

Another explanation for the conflicting findings in the research is variation in the interventions. “There are two issues. One is timing and the other is dosing—giving 1.5 grams of vitamin C like the VITAMINS trial doesn't get the plasma level of ascorbate up into these very high ranges,” said Dr. Fowler.

Dr. Marik attributes the lack of effect in the VITAMINS study to timing, noting that the median time in which patients in the intervention arm received the first dose of vitamin C after meeting eligibility criteria was 12.1 hours. “Ideally it should be given in the ER at the same time as the first dose of antibiotics. This is what we do. It seems most effective if given within six hours of presentation,” he said. Dr. Marik also noted that more than a fifth of the trial patients were surgical rather than medical.

Another factor is what patients are given. Dr. Fowler emphasized that his protocol only involves vitamin C. “We have no use for hydrocortisone. We have no use for thiamine. It's high concentration of ascorbic acid in the circulation,” he said. Dr. Marik, however, said his research and other studies have “demonstrated clear synergy” of all three together.

Dr. Fowler and his team are currently enrolling patients in the ASTER trial, which will randomize 900 septic patients to placebo, IV acetaminophen, or vitamin C, 50 mg/kg every six hours for five days. “[CITRIS-ALI] was a proof-of-concept trial. The ASTER trial is going to be the larger trial that may lead to a phase 3 equivocal trial if it turns out that vitamin C is as useful as it was in CITRIS-ALI,” he said. “The final word on vitamin C infusion at this point is not in.”

The Vitamin C, Thiamine And Steroids in Sepsis (VICTAS) study is another investigation of the subject that's currently underway. “VICTAS has a larger sample size and will I hope be able, in conjunction with the other trials, [to provide] more guidance about whether the use of vitamin C, thiamine, and steroids are helpful for patients with sepsis and respiratory or cardiovascular compromise,” said Jonathan Sevransky, MD, MHS, lead investigator and professor of medicine at Emory University in Atlanta.

However, other physicians are already convinced by the negative results of the randomized trials. “We should really be giving therapies to our patients that we think are helpful, and once therapies are proven to not be helpful, I don't think we should use them,” said James Walter, MD, an associate professor of pulmonary and critical care medicine at Northwestern Medicine in Chicago. “I think using the vitamin C cocktail after VITAMINS now falls into the latter category.”

Risks and costs

Although the experts agreed that vitamin C is generally not harmful and costs relatively little, research has shown a potential risk of kidney stones, especially in men. Another issue is that treated patients would need to have their blood glucose levels checked by a central laboratory, not with a bedside device, because high levels of vitamin C can interfere with these measurements.

There's also the concern that a focus on vitamin C might distract from provision of other interventions. “Everything that you do has an opportunity cost, and if you concentrate on something that hasn't been proven, you might forget about things that have been proven—early antibiotics, early identification, early fluids, and then making sure that you don't give too many fluids,” said Dr. Sevransky. “All of these things have been shown to be helpful for patients, and every time you add something, it can cause your focus to shift away from those things that we know work.”

Dr. Walter worries about the risk of raising unrealistic hopes in patients and families. He cowrote a letter to the editor of CHEST critiquing the public claims of vitamin C supporters and said he has had patients and families come to him asking about using vitamin C for sepsis after reading about its curative qualities in a Whole Foods newsletter. “I think that's troubling,” he said. “I think that sets up an unfair expectation for families about holding on for miracle cures and putting off important conversations about prognosis and end-of-life care and realistic discussions about a really difficult-to-treat disease.”

The overall lack of certainty about both the good and bad effects of vitamin C on sepsis patients makes things tough for practicing clinicians. Although Dr. Sevransky is currently studying vitamin C, he doesn't use it in his routine practice. “I think that's something that each clinician needs to look at the evidence and decide whether or not the evidence that's been published to date supports the use in their patient population,” he said.

Dr. Fowler said his institution has long used vitamin C alone for septic patients on extracorporeal membrane oxygenation and also in those who have gone on to develop ARDS from bacterial or fungal sepsis. He expects the data to someday support that treatment choice. “It's going to take years to work out. I just turned 70, so I'm at the end of my career, but I'm expecting when I'm sitting on the beach with teeth falling out and hair falling out that I'll pick up a copy of JAMA and they'll say, ‘Look at what the results with vitamin C found in this 2,000-patient trial.’”

Emily B. Brant, MD, a critical care specialist at the University of Pittsburgh who wrote an editorial accompanying the CITRIS-ALI study, said that she has previously tried vitamin C “as a last-ditch effort” with mixed results. Now, however, with the results of the VITAMINS study, she does not believe “there is equipoise about the effectiveness of vitamin C, thiamine, and hydrocortisone compared to hydrocortisone alone for patients with sepsis and septic shock. Thus, I no longer use it.”

That's actually the opposite of how Dr. Marik would like to see the therapy used. “If you select out patients who are really close to death, which is what some people do . . . they say, ‘See, it doesn't work,’” he said. He said he would prefer additional rigorous randomized trials to help settle the issue. “If you do a randomized control trial, it has to be meticulously done. It has to be really a well-done clinical trial in which patients are enrolled early, their fluid management is appropriate, and you select out the right patients,” he said.