Now that VITAMINS is published, it’s worth trying to look at the big picture of sepsis treatment with hydrocortisone, ascorbate, and thiamine (HAT).
Marik et al. 2017: The beginning
HAT therapy began with a single-center, before/after study.1 There was a stark mortality reduction following the routine adoption of HAT therapy in septic shock (figure below, p=0.0006).
This study has numerous limitations (single-center, before/after, retrospective design). It seems like HAT therapy is doing something, but the study cannot prove it.
subsequent before/after studies
HAT therapy was adopted in several centers, leading to additional before/after studies (comparing outcomes before versus institution of HAT protocols versus afterwards). One of these studies was neutral,2 but most showed benefit from HAT.3–5
These studies have the same limitations as Marik’s original study (single-center, before/after, retrospective design). Nonetheless, it is notable that several institutions did find that using HAT correlated with improved outcomes.
why might HAT therapy improve outcomes?
In the spirit of full disclosure, I drank some of the cool aid and used HAT in my septic patients. My impression is that this does improve outcomes (consistent with the above studies). Exactly how or why it might work, however, is complicated.
Applying HAT could lead to a host of downstream consequences. This creates numerous potential mechanisms whereby HAT might improve patient outcomes, perhaps most notably…
(#1) Ascorbate and thiamine might cause direct benefit.
- This is the most obvious potential mechanism of benefit – HAT works as intended.
- There are multiple theoretical mechanisms whereby ascorbate and thiamine could improve the pathophysiology of septic shock. Additionally, this is supported by a fair body of basic science and clinical evidence.5–9
(#2) HAT leads to early administration of hydrocortisone.
- Hydrocortisone has been proven in multi-center RCTs to improve organ function (including pulmonary function and cardiovascular stability).10,11
- Whether hydrocortisone affects mortality is debatable and probably unknowable. APPROCHSS did find mortality benefit, possibly related to early administration of hydrocortisone (more on this here).11
- Early steroid therapy is beneficial for the treatment of bacterial meningitis (whereas delayed steroid therapy is ineffective).12 It’s conceivable that early steroid treatment could have augmented benefits in other forms of infection as well.
(#3) Avoiding over-resuscitation and unnecessary procedures
- If you believe that HAT is going to save the patient’s life, then you may be more willing to tolerate mild instability while waiting for the patient to improve (rather than, for example, bolusing additional volume in efforts to immediately improve the hemodynamics).
- Likewise, if you feel that HAT will turn things around, you might be less likely to immediately intubate the patient or place a central line.
- In many cases, willingness to tolerate moderate abnormalities with a watch-and-wait mentality may avoid unnecessary procedures and iatrogenic harm.
(#4) Self-fulfilling prophecy
- Using HAT may promote a belief that patients shouldn’t die from septic shock. If clinicians believe that the patient will survive, this belief itself may improve survival. The team will be more willing to deploy extraordinary measures, and less willing to withdraw life-sustaining therapies.
- For example, the use of HAT has indirectly led me to deploy super-human doses of vasopressors in some scenarios which appeared hopeless. These patients have generally survived.
randomized controlled trials
Retrospective studies show that at some hospitals, the institution of HAT has correlated with improvements in outcomes. This doesn’t prove that HAT has caused improvements in outcome. Nor does it clarify whether improvements might result from ascorbate/thiamine, or result from other downstream consequences of HAT treatment. To clarify causality and mechanism, we need RCTs.
Unfortunately, RCTs have some inherent limitations. Enrollment, consent, and randomization all take time. This will inevitably delay the initiation of study therapies. This may render traditional RCTs a limited strategy for studying immediate, up-front initiation of HAT.
CITRIS-ALI
CITRIS-ALI isn’t actually a study of HAT, but it is the first large multi-center RCT investigating IV vitamin C. It investigated relatively high-dose IV vitamin C in patients with sepsis and ARDS.13
Interpreting this study is controversial, to say the least. Patients treated with IV vitamin C had lower mortality, but this wasn’t the primary endpoint. Mortality differences compromised the validity of the primary endpoint (changes in SOFA score), because SOFA score was recorded only in survivors. Thus, IV vitamin C may have been a victim of its own success (because sicker patients in the Vitamin C group survived – instead of dying – this dragged down SOFA scores in the Vitamin C group!).
It’s controversial how to deal with this statistical conundrum of survivorship bias. Re-analysis of the study is arguably warranted, but it will be difficult to achieve this in a fashion which will be widely accepted. Ultimately, mortality reduction is an important endpoint and a signal of potential clinical benefit (regardless of whether it was designated as a primary or secondary endpoint). This conundrum is discussed further here.
VITAMINS
This is a multi-center RCT of patients in septic shock which compares hydrocortisone alone versus HAT therapy. 216 patients were included as shown here:
No impact was seen on the primary endpoint (vasopressor-free time) or mortality (table below). Ironically, HAT did cause an improvement in the SOFA score (the much-debated primary endpoint of CITRIS-ALI). However, since this is an isolated secondary endpoint, it cannot carry much weight.
The main limitation of VITAMINS is that it’s not a study of early metabolic resuscitation:
- The median delay from ICU admission to randomization was 12 hours. This doesn’t account for time delays prior to ICU arrival (which for some patients included inter-hospital transfer). All told, patients received HAT relatively late in the course of septic shock.
- 40% of patients had already been initiated on stress-dose steroid prior to study enrollment. This illustrates that VITAMINS wasn’t a study of the initial resuscitative strategy.
Timing of therapies in septic shock matters. A retrospective study of HAT suggests that early therapy is important (ideally within <6 hours):5
Overall, VITAMINS is an outstanding RCT which argues against benefit from delayed HAT therapy in septic shock. This suggests that progressive depletion of vitamin C levels over time isn’t clinically relevant (as such deficiencies often arise days after the initiation of critical illness). However, this study doesn’t investigate whether early HAT therapy could protect the vascular endothelium during the initial phase of sepsis (which is purported to be a mechanism of HAT treatment).14
CITRIS-ALI vs. VITAMINS:
Administration of IV vitamin C was delayed in both CITRIS-ALI and also VITAMINS. Why, then, might Vitamin C appear to have benefit in CITRIS-ALI, but not VITAMINS? There are a few potential reasons:
- CITRIS-ALI involved higher doses of IV vitamin C. A small pilot RCT on sepsis previously suggested that higher doses may be more effective.15
- CITRIS-ALI involved intubated patients with ARDS, who were likely experiencing ongoing stress on alveolar capillaries and endothelium. Vitamin C could be beneficial in this context, given the ongoing and progressive endothelial injury.
- Of course, another explanation is that both studies were actually negative (with mortality differences in CITRIS-ALI being a statistical fluke).
where do we go from here?
Since the publication of Marik’s 2017 paper, about a hundred studies have been published on HAT therapy. Despite this, equipoise persists. Pro and Con arguments regarding HAT might be summarized roughly as follows:
Pro arguments
- Before/after studies suggest that HAT therapy was beneficial in multiple hospitals.
- CITRIS-ALI found a mortality reduction with high-dose IV vitamin C therapy in ARDS due to sepsis.
- Studies disagree about whether HAT is beneficial, but there is no persuasive evidence of harm. The worst-case scenario seems to be that vitamin C and thiamine are a harmless waste of resources.
- VITAMINS excludes delayed benefit from HAT, but doesn’t evaluate early HAT therapy.
Con arguments
- VITAMINS indicates that ascorbic acid and thiamine add nothing to hydrocortisone.
- CITRIS-ALI was a negative study (based on its primary endpoint of SOFA scores).
- Before/after studies are vulnerable to numerous biases and confounders.
There are currently several additional RCTs underway on vitamin C. Hopefully these studies will continue to completion (since equipoise on the topic persists). We may need to be patient and wait a couple more years for a plurality of additional data from these studies.
It is increasingly likely that when the dust finally settles, we may recognize that the benefit of early HAT therapy was solely due to early steroid paired with our conviction that sepsis is a survivable illness (with the vitamin C and thiamine functioning as placebos). That would be disappointing, but not a bad outcome. It’s possible that vitamin C and thiamine served as training wheels, allowing us to think about sepsis resuscitation and mortality in a fresh light. Perhaps the cure for sepsis was with us all along – early source control, antibiotics, steroid, gentle resuscitation, patience… and hope.
- Several retrospective before/after studies have suggested benefit from HAT therapy. However, these studies don’t clarify the mechanism of benefit (e.g. early hydrocortisone, ascorbic acid, or merely clinicians’ behavior when they are more confident that the patient will recover).
- VITAMINS is a multi-center RCT comparing hydrocortisone alone vs. HAT therapy. The study found no significant differences, suggesting that the effectiveness of HAT may be due largely to early hydrocortisone administration. However, HAT therapy was administered well after the initial resuscitation, potentially compromising its efficacy.
- Numerous additional RCTs are currently underway to investigate HAT therapy. Hopefully these will shed additional light on the topic.
related
- Accelerated therapeutic strategy for sepsis (2015)
- HAT therapy: evidentiary background (2017)
- ADRENAL (2018)
- APPROCHSS (2018)
- CITRIS-ALI (2019)
references
- 1.Marik P, Khangoora V, Rivera R, Hooper M, Catravas J. Hydrocortisone, Vitamin C, and Thiamine for the Treatment of Severe Sepsis and Septic Shock: A Retrospective Before-After Study. Chest. 2017;151(6):1229-1238. doi:10.1016/j.chest.2016.11.036
- 2.Litwak J, Cho N, Nguyen H, Moussavi K, Bushell T. Vitamin C, Hydrocortisone, and Thiamine for the Treatment of Severe Sepsis and Septic Shock: A Retrospective Analysis of Real-World Application. J Clin Med. 2019;8(4). doi:10.3390/jcm8040478
- 3.Sadaka F, Grady J, Organti N, et al. Ascorbic Acid, Thiamine, and Steroids in Septic Shock: Propensity Matched Analysis. J Intensive Care Med. July 2019:885066619864541. doi:10.1177/0885066619864541
- 4.Kim W, Jo E, Eom J, et al. Combined vitamin C, hydrocortisone, and thiamine therapy for patients with severe pneumonia who were admitted to the intensive care unit: Propensity score-based analysis of a before-after cohort study. J Crit Care. 2018;47:211-218. doi:10.1016/j.jcrc.2018.07.004
- 5.Long M, Frommelt M, Ries M, et al. Early Hydrocortisone, Ascorbate and Thiamine Therapy for Severe Septic Shock: A Retrospective Cohort Analysis. Critical Care and Shock. 2020:Pending.
- 6.Moskowitz A, Andersen L, Huang D, et al. Ascorbic acid, corticosteroids, and thiamine in sepsis: a review of the biologic rationale and the present state of clinical evaluation. Crit Care. 2018;22(1):283. doi:10.1186/s13054-018-2217-4
- 7.Marik P. Vitamin C for the treatment of sepsis: The scientific rationale. Pharmacol Ther. 2018;189:63-70. doi:10.1016/j.pharmthera.2018.04.007
- 8.Marik P. Hydrocortisone, Ascorbic Acid and Thiamine (HAT Therapy) for the Treatment of Sepsis. Focus on Ascorbic Acid. Nutrients. 2018;10(11). doi:10.3390/nu10111762
- 9.Badeaux J, Martin J. Emerging Adjunctive Approach for the Treatment of Sepsis: Vitamin C and Thiamine. Crit Care Nurs Clin North Am. 2018;30(3):343-351. doi:10.1016/j.cnc.2018.05.002
- 10.Venkatesh B, Finfer S, Cohen J, et al. Adjunctive Glucocorticoid Therapy in Patients with Septic Shock. N Engl J Med. 2018;378(9):797-808. doi:10.1056/NEJMoa1705835
- 11.Annane D, Renault A, Brun-Buisson C, et al. Hydrocortisone plus Fludrocortisone for Adults with Septic Shock. N Engl J Med. 2018;378(9):809-818. doi:10.1056/NEJMoa1705716
- 12.Tunkel A, Hartman B, Kaplan S, et al. Practice guidelines for the management of bacterial meningitis. Clin Infect Dis. 2004;39(9):1267-1284. doi:10.1086/425368
- 13.Fowler A, Truwit J, Hite R, 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(13):1261-1270. doi:10.1001/jama.2019.11825
- 14.Barabutis N, Khangoora V, Marik P, Catravas J. Hydrocortisone and Ascorbic Acid Synergistically Prevent and Repair Lipopolysaccharide-Induced Pulmonary Endothelial Barrier Dysfunction. Chest. 2017;152(5):954-962. doi:10.1016/j.chest.2017.07.014
- 15.Fowler A, Syed A, Knowlson S, et al. Phase I safety trial of intravenous ascorbic acid in patients with severe sepsis. J Transl Med. 2014;12:32. doi:10.1186/1479-5876-12-32
- PulmCrit Wee: Rational selection of infusion rate based on loading dose - June 25, 2024
- PulmCrit: PPIs are safe and effective for GI prophylaxis… the end. - June 18, 2024
- PulmCrit: Bilevel Sequence Intubation (BSI) – The new standard - June 17, 2024
The fact that they gave the treatment 32 hours after presenting symptoms invalidates the entire study. That last graph you show proves it … It works only when given less than 6-12 hours of presenting symptoms (actually most benefit is lost if waiting more than 8 hours). A huge waste of money time and effort … TIMING IS THE MOST CRITICAL FACTOR when giving HAT (which should NOT be a surprise to anyone in critical care today). see https://is.gd/lEdFOG
Marik’s original before after study treated within 24 hrs after ICU admission. Very similar to VITAMINS. See my comment below.
Josh, does VITAMINS change your personal practice?
I agree that the jury is still out and there is equipoise for more studies, but it seems like overall the evidence we have is starting to swing in a different direction. I will personally find it more difficult to advocate for HAT against the masses, and worry that it will (hopefully temporarily) be hung up next to methylene blue.., reserved only for deteriorating patients on multiple pressors… used occasionally, much later than it should be used.
Thanks for the great analysis.
Yes the study does change my practice.
At this juncture:
– I will continue to discuss this with colleagues and await more evidence with an open mind.
– I will stop using HAT for most septic patients.
– I will continue using HAT for patients with ARDS, on the basis of mortality benefit seen in CITRIS-ALI.
There is a lot of equipoise here, so it’s really very very hard to know what the right thing to do is. That’s why more studies are being done. I don’t claim to have the answers.
Hi I can’t find the Kory 2020 data published anywhere. Was the time to antibiotics/ vasopressors/ etc, also reported? If not, the reported mortality trend could well be due to delayed administration of co-interventions (e.g. antibiotics) other than HAT. I have been unable to find any published data identifying timing as critical to HAT (only speculation based on the assumption that it is important). Is anyone able to share these studies if they exist? Given that in VITAMINS the “median time from meeting eligibility criteria to the first dose of vitamin C in the intervention group was 12.1 hours (IQR,… Read more »
Kory’s study isn’t published yet, I’m sorry, it has been accepted for publication though so it should be out soon.
Kory’s time frame starts at ED presentation, whereas the time-frame in VITAMINS starts at ICU admission. This would make it impossible to align the two studies, because in VITAMINS there is no data about the delay between ED presentation to ICU admission (such delay would be additive to the delays documented within the VITAMINS manuscript).
I’m not sure we can know when sepsis, or septic shock, starts in any of these studies – it could be well before presentation to hospital or occur while in hospital.
If one is to believe the benefit shown by CITRIS-ALI (which lacked adjustment for multiple secondary endpoints), time did not seem to be a factor. Some patients received Vit C days after onset of sepsis, as stated by Dr Fowler at the CCR 20 meeting.
Chris
To find out why high doses of IV vitamin C works in sepsis, I would recommend doing some background reading and looking at pre-clinical studies long before Marik came up with HAT. How do you think Marik came up with vitamin C in sepsis? It was not a voice from the skies. He just followed what others have shown in pre-clinical animal models.
Hi Ramesh
I have read a lot of the previous literature. The rationale for HAT is very clever. Unfortunately, animal and laboratory studies have many limitations (eg summary here: https://litfl.com/animal-and-laboratory-studies/ – I am a former lab scientist myself). it was this rationale/ biological plausibility that encouraged the VITAMINS trialists to proceed with a phase 2 clinical study..
Cheers
Chris
I agree with you Chris about the difference in results between animal and human studies, particularly in sepsis. But there is so much more evidence for IV Ascorbic acid as compared to “H” and “T”. And the mechanisms by which IV Ascorbic acid could work are numerous and well studied in animals too.
Because it works. The white washing denial cloaking was Rockefeller financed and goes back to Jungblut and Sabine.
I’ve seen a lot of people repeat Paul Marik’s claim (at CCR 20) that the time to treatment was too long and this invalidates VITAMINS results. This is moving the goalposts after the fact. Marik’s original before after study treated within 24 hrs after ICU admission. Not within 6 hrs after ED admission as he is now claiming post hoc to be necessary. Marik Before-After (Chest 2016): “consecutive patients […] were treated with intravenous hydrocortisone, vitamin C, and thiamine (vitamin C protocol) within 24 h of ICU admission (treatment group)” VITAMINS: “the median time from meeting eligibility criteria to the… Read more »
I was just about to mention that myself. The original Marik trial enrolled patients up to 24 hours after admission and I can’t find anywhere that reports the average time to first dose. Certainly would be nice to have that data for some relative comparison. I’m wondering if this criticism about later administration (which even made it to NPR!) will prompt the VITAMINS authors to do a post-hoc analysis of patients receiving first dose in less than 12 hours.
the challenge to all of us at this point is judging HAT on the basis of its scientific and evidentiary merits, rather than its association with various personalities, media hype, or overblown claims. most treatments in medicine don’t wind up being as effective as initially believed/hoped – so the fact that initial hopes didn’t pan out doesn’t mean that the treatment has zero efficacy.
Agreed. I think there was a need to provide a counterpoint to some of the stronger claims made by others, but now further trials are coming we can acknowledge we don’t have the final answer and hope for positive results.
From what I have read I understand Dr. marik is still getting the same beneficial results at his ICU. Why??
If I get Sepsis again I will definitely be asking for adjunctive HAT, even if I have to pay the extra $20 for vitamin C. However, in Australia, I doubt they would do it …….. unless I threatened to sue for malpractice.
Just to increase uncertainty, I’ve found this publication right now in PubMed. Unfortunately the abstract is missing and I was unable to download the article:
Am J Respir Crit Care Med. 2020 Jan 9. doi: 10.1164/rccm.201908-1543LE. [Epub ahead of print]
Hydrocortisone-Ascorbic Acid-Thiamine Use Associated with Lower Mortality in Pediatric Septic Shock.
Has anyone read the full text?
I am the lead author in the AJRCCM study. It was a single-center retrospective, propensity-scored matched analysis of children with septic shock. We saw a dramatic 30 and 90 day mortality benefit but no other changes in secondary outcomes. Clearly it is preliminary data, but in line with Marik and CITRIS-ALI- a large proportion of children had moderate ARDS- leading to the hypothesis that patients with multiple oxidative stress hits (sepsis, high FIO2 conc from ALI/ARDS) will benefit more readily than others. Phenotype appears to be important. More fodder for debate…..
Eric,
Any chance you have numbers comparing controls to HAT restricted just to after May 2017, when HAT became available.
Untreated controls and hydrocortisone only patients were drawn from both the pre-HAT protocol epoch (1/2014 to 4/2017), as well as the post-HAT protocol epoch (after 5/2017). Because HAT therapy was primarily used on the sickest patients, we would not have been able to find sufficiently sick untreated patients to match if we had used a contemporaneous cohort alone, and hence the need to use the current broader cohort of patients. I don’t have the exact # of patients post-5/2017 but it was relatively low. I don’t think our sepsis treatment has changed dramatically since 2014- but the reviewers asked a… Read more »
Eric,
based on your retrospective data, would you now consider ethical to perform a randomized clinical trial with one treatment group with vitamin c and one without?
Equipoise is tricky here because I do feel that vitamin C confers a benefit. But in the end, we don’t know enough…. about phenotype, etc to know, so I think I’d be open to an RCT yes
“For example, the use of HAT has indirectly led me to deploy super-human doses of vasopressors in some scenarios which appeared hopeless. These patients have generally survived. “
What do you think this means? We give up too early? We need to reset what constitutes “futile”? Or was there something, either tangible or not, that made you push the extra bit over others where you did not? Was there something that made you think they were not actually hopeless?
Thanks for the post
It would be interesting to determine the relationship between the blood ascorbate level & primary/secondary outcomes. In the past Vitamin C studies, the researchers have always found a negative relationship between BLOOD ASCORBATE LEVEL & mortality/disease rate.
Germicidal actinic ray is acknowledged as a good suggests that of deactivating bacterium and bound viruses. It is, however, probably harmful to humans. The spectral vary of 200-280 nanometers, conjointly referred to as UV-C, was till recently believed to be unworkable as a steriliser once folks actively occupied a space. because of studies completed by Dr. David Brenner and his team at Columbia University, additional analysis is proving that the ranges of 207nm to 230 nm, or “Far-UV” isn’t harmful to humans.