See Dr. Marik’s response to this article.
A recent small single-center before-and-after trial by Marik et al showed that vitamins in combination with other relatively safe therapies may improve outcomes in sepsis. We asked three critical care physicians to give their thoughts on the debate on vitamin C in sepsis, and our own co-founder and healthcare finance expert Joe Habboushe to weigh in on the cost/price argument.
Jarone Lee, MD, MPH, is Assistant Professor of Emergency Medicine and Surgery at Harvard Medical School, Medical Director, Blake 12 ICU, Massachusetts General Hospital. Previously he was an emergency physician at St. Luke’s-Roosevelt Hospital Center, where he also did his residency training.
Christopher Seymour, MD, is creator of the qSOFA score and Assistant Professor of Critical Care and Emergency Medicine at the University of Pittsburgh School of Medicine. Dr. Seymour works clinically in the Medical Intensive Care Unit at the University of Pittsburgh Medical Center Mercy Hospital.
Gregg Chesney, MD, is an emergency medicine physician and critical care physician at NYU Langone Medical Center and NYU Lutheran Medical Centers. He is also Assistant Professor of Emergency Medicine and an Assistant Professor of Medicine at NYU School of Medicine.
Have questions or comments? Ask the experts.
On the Evidence Base
Jarone Lee: Paul Marik’s recent paper on using a combination of Vitamin C, thiamine and hydrocortisone found an impressive and profound impact on treating septic shock with a decrease in mortality by 32%, which represents a number-needed-to-treat of about 1 in 3! Amazing results that, if true, would alter the way we treat sepsis. The two vitamins and steroids given are generally benign and should have limited side effects, so we should definitely consider using this, especially for patients that require salvage therapy.
On the other hand, his amazing results are a bit too good. When something seems too good, we have to be cautious. His study, while well done for a retrospective, propensity-matched trial, is still a retrospective, single-center, study using historical controls. Without a true randomized control trial, we will not know if there are other factors that affected the mortality decrease. For example, the way sepsis is treated might have changed dramatically between now and his historical controls and as such, this result could be just from improved protocols, nursing care, general ICU care, technology, antibiotic choices, etc… Lots of factors that cannot be controlled for or evaluated in a retrospective, propensity-matched trial.
Christopher Seymour: It is hypothesis-generating, that combinations of vitamins and otherwise relatively safe therapies may improve outcomes in sepsis, but this has not been subjected to rigorous study or has any adequate evidence base to broadly change practice. The National Institutes of Health has invested money into a randomized trial to test vitamin C and other accompanying therapies and we look forward to the results of the more rigorous studies.
Gregg Chesney: While the results are intriguing (31.9% absolute risk reduction in mortality), the quality of evidence provided in this study is undeniably poor. This is a very small (47 patients) nonrandomized retrospective observational open label trial with historical controls, a far cry from the rigorous randomized control trials that are the underpinning of our evidence-based approach to sepsis care as evaluated in the Surviving Sepsis guidelines. With this and other studies of vitamin C being so small, it is certainly possible that the treatment effect noted is simply due to chance or other factors unrelated to the therapy provided. As a result this study generates more questions for me than it answers.
Does it make sense? Yes, there is a reasonable underlying physiologic rationale: Vitamin C is a cofactor in catecholamine synthesis and serves to help maintain the endothelial glycocalyx, the thin proteinaceous layer in capillaries thought to play a key role in vascular integrity and function. Relative deficiencies in Vitamin C, endogenous steroids, and thiamine have all been noted in sepsis. However, there are numerous endogenous substances that are markedly depleted in sepsis, thought to be more a result of the underlying critical illness than a cause of the illness and in many cases replacing these substances has proven to be useless or harmful (such as activated protein C). Why should this particular cocktail be special?
Why does this combination work? It’s been proposed that the steroids potentiate the effect of the vitamin C and that the thiamine helps prevent calcium oxalate formation. However, with three different non-standard therapies included in the treatment arm in this study, it is difficult to tease the treatment effect out, especially when the individual therapies have failed to yield benefits anywhere near those reported by this combination. As a result, any provider choosing to utilize this new therapy is obligated to follow the treatment cocktail exactly as it was used in Marik’s study.
Jarone Lee: Unlike hydrocortisone and thiamine, vitamin C IV is not a regular medication given. As we start giving more, there will be unanticipated side effects despite no current reports. We want to not do harm, so I would recommend considering this protocol mostly in patients that have failed our regular, and maximized sepsis care.
Christopher Seymour: This combination therapy involves hydrocortisone, which is a steroid, and corticosteroids have been tested for decades in sepsis and septic shock, and, through meta analyses and RCTs have been thought to be reserved for the sickest patients, those who are already on two vasopressors.
There’s also a very large phase 3 randomized trial called ADRENAL that’s underway in Australia and New Zealand that will likely give us state-of-the-art evidence across hundreds of center in many thousands of patients, about the role of this potentially dangerous medication that is included as part of this cocktail.
So I do think vitamins are safe, and we can buy them over the counter, but when combined with other medicines that have real and significant side effects, we need to have caution when it’s from a single center before-and-after study that was not randomized.
Gregg Chesney: Is it safe? Probably. The corticosteroid dose used is the dosing regimen currently recommended in the Surviving Sepsis guidelines. The thiamine dose used is less than the dose safely utilized in treatment of Wernicke’s encephalopathy. And the vitamin C dose is less than the dose studied in a small (24 patients) Phase I RCT (Fowler 2014). At higher doses of vitamin C, calcium oxalate nephropathy has been reported (Buehner 2016), and steroids in sepsis have been associated with critical illness polymyoneuropathy. This combination, though, has never been studied for safety before. There’s always the risk that in the setting of a much larger patient population, unforeseen harms may arise. And regular vitamin supplementation in large populations of healthy individuals (including vitamin C) has previously been associated with no measurable benefit, and in certain cases found to be harmful.
Gregg Chesney: Is it cheap? Maybe not as cheap as it seems. According to my critical care pharmacist, the course of treatment costs approximately $1,5501. Of note, all generic IV vitamin C is made by a single manufacturer, so the cost of vitamin C may be subject to increase, and supplies may become more limited in light of this trial. But if it really works and saves lives and ICU days, the cost becomes easily justified.
Jarone Lee: The cost of IV vitamin C is definitely an order or two magnitude above PO vitamin C. In the grand scheme of things with an average cost of an ICU bed of approximately $2,000–3,000 per day in 2000 (Halpern 2010), it is not that big of a deal. As always, I think cost is always a difficult discussion. I think that we have to be efficient, and be clear that we are treating patients appropriately. Unfortunately, guidelines and best data sometimes do not work, as they are only as good as the patients enrolled and examined. So in the ICU, many of our patients do not fit any of the guidelines or studies, and as such, the treatment is still in the grey area with equipoise in terms of different treatment options.
Gregg Chesney: Much like other therapies we utilize in the treatment of sepsis, this Vitamin C, hydrocortisone, thiamine cocktail warrants further more rigorous investigation prior to consideration for wide scale adoption. Until then, the tenets of sepsis care will remain the same. We are not going to be prioritizing giving vitamin C over antibiotics. It may be worth now considering this cocktail as a therapeutic option in refractory shock when traditional sepsis care is faltering, but I won’t be using this in all comers until there is much better data to support its use. And if Paul Marik is in fact right and the benefit is as robust as he suggests, then it won’t take too long to perform a good RCT since we won’t have to enroll very many patients to adequately power a study that demonstrates benefit.
Jarone Lee: I would consider it as much as any other salvage treatment that has potential benefit, especially if it is in line with my family members’ goals of care. We do this regularly in our ICU patients—many patients do not fit criteria and need the consideration of new therapies.
Hospital “costs” can be quite confusing, as they can mean very different things and are often not explained in articles and sometimes even peer-reviewed manuscripts. Three common options: (A) the hospital’s “charge master”, or what they send out on a line item to charge the insurance company (yet insurers often have a negotiated payment that’s an average ~10-25% of this amount, or sometimes “bundle” the payment for the patient visit, such that there is no extra payment). This rarely represents an additionally paid amount, expect for the uncommon situation when a patient doesn’t have insurance and has means to pay and doesn’t negotiate with the hospital; (B) the average additional amount typically collected from the insurance company, which can be quite variable; or (C) the additional variable cost to the hospital, i.e., that would have been avoided if the treatment was not ordered, regardless of what is collected.
We were able to confirm with Dr. Chesney’s source that the quoted “costs” for intravenous Vitamin C represent (C), which we feel is the most relevant figure for this particular discussion.
Joseph Habboushe, MD, MBA
- Could vitamin C save lives in sepsis? These hospitals aren’t waiting for proof.(PulmCCM)
- PulmCrit- Metabolic sepsis resuscitation: the evidence behind Vitamin C (PulmCrit)
- The Marik Protocol: Have We Found a “Cure” for Severe Sepsis and Septic Shock? (R.E.B.E.L. EM)
- Vitamin C and Sepsis. All Sound and Fury? Much Ado About Nothing? (Science-Based Medicine)
- Glycocalyx in Critical Illness (LITFL)
- Insights from Dr. Christopher Seymour, Creator of the qSOFA Score (Paging MDCalc)
- Centers for Disease Control and Prevention, Category: Sepsis (CDC)
- Discover: the qSOFA Score for Sepsis! (Paging MDCalc)