Vitamin C in Sepsis: Splashes in the popular press
Hydrocortisone-Vitamin C-Thiamine in Sepsis
When articles pertaining to emergency medicine and critical care make a splash in the popular press, we’ll do our best to provide a timely appraisal of the evidence.
Marik PE, Khangoora V, Rivera R, Hooper MH, Catravas J, Hydrocortisone, Vitamin C and Thiamine for the Treatment of Severe Sepsis and Septic Shock: A Retrospective Before- After Study, CHEST (2017), doi: 10.1016/j.chest.2016.11.036.
Bottom Line Up Front
In a single center, before-after ICU study comparing 47 patients with severe sepsis or septic shock and a procalcitonin ≥ 2ng/ml, a regimen of Hydrocortisone, Vitamin C, and Thiamine was shown to have a dramatic absolute risk reduction of 31.9% (NNT= 3.1) in mortality (40.4% vs. 8.5%, p<0.001) when compared to retrospective controls not receiving the 3 drug combination. The inclusion criteria included procalcitonin measurement, which limits its generalizability. However, if these results are replicated in large-scale RCTs it would be practice changing in the management of severe sepsis and septic shock. Only a relatively small benefit would need to be demonstrated to support wide-spread adoption because this regimen is very low-cost and has a reassuring safety profile. However, implementing this intervention should not distract us from doing the things we do know work in sepsis, such as early recognition, resuscitation, appropriate antibiotics and lung-protective strategies for patients at risk for ARDS.
Does it live up to the hype?
It’s too early to tell. There have been dozens of promising therapies in sepsis that have failed to show benefit in large randomized controlled trials. Until more research is done, it is prudent to maintain focus and improve compliance on interventions that have been more robustly shown to work.
A Case
You are working in the trauma and critical care area of your emergency department when a 60-year-old woman comes in with a three-day history of productive cough, fever, and today developing lethargy. She has diabetes and hypertension. On arrival, her vital signs are significant for fever to 39.2, tachycardia to the 130s, blood pressure of 93/50 and tachypnea to 28. She has coarse crackles on the right lower lung fields & she appears lethargic. You begin fluid resuscitation with a 30 mL/kg bolus and attempt to gather collateral information from her family. Her chest x-ray shows a dense right lower lobe pneumonia, and small bilateral effusions. Her CBC shows a white count of 18 and her lactate is 12. After the fluid bolus, her blood pressure improves to 110/60 but her tachypnea worsens & she looks like she is tiring out. She is intubated for respiratory failure & her blood pressure remains soft. Central access is obtained & she is started on norepinephrine with a MAP goal of > 65. Broad spectrum antibiotics are initiated after blood cultures are obtained, and she is admitted to the ICU. There, her lactate has only cleared to 10.
Vasopressor requirements increase despite the addition of vasopressin and she is started on stress dose steroids with minimal improvement in her vasopressor requirement. You ask the attending about the utility of vitamin C and thiamine in septic shock, remembering an article that has been circulating on your Newsfeed the last few days about a group in Virginia that is described as having “cured sepsis” with this intervention.
Vitamin C: Source: https://en.wikipedia.org/wiki/Vitamin_C#/media/File:L-Ascorbic_acid.svg
Critical Analysis
Study Location: Medical service of the General ICU of a tertiary referral center.
Patient Selection: Electronic chart query for patients with the diagnosis of severe sepsis or septic shock (1992 American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference definition). The “before” group included patients from June – December 2015 and the “after” group included patients from January - July 2016. Forty-seven patients were included in each of the intervention and control group.
Inclusion Criteria: The intervention was implemented in all consecutive patients with a primary diagnosis of severe sepsis or septic shock, and a procalcitonin level greater than or equal to 2 ng/mL. The included control patients were consecutive patients from the preceding six months using the same inclusion criteria.
Exclusion Criteria: Age less than or equal to 18, pregnant patients, and patients who were not full code on presentation to the ICU were excluded from analysis.
The Intervention: The intervention group received IV vitamin C 1500 mg Q 6 hours for 4 days, IV thiamine 200 mg Q 12 hours (both for either 4 days or until transfer from the ICU), and IV hydrocortisone 50 mg Q6 hours. Hydrocortisone was also given to the control group with 59.6% receiving it “per current guidelines and at the discretion of the attending physician.”
Outcomes: The primary end-point of the study was survival to hospital discharge. Secondary end points were ICU length of stay, need for renal replacement therapy, duration of vasopressor use, delta in SOFA score, and procalcitonin clearance.
Results: 47% of patients in each group required vasopressors meeting the definition for septic shock while the remainder were severe sepsis. Hospital mortality in the control group was 40.4% (expected mortality 39.7%) and 8.5% (expected 41.6%) in the intervention group yielding a number needed to treat of 3.1 to avoid 1 death.
NNT for death was 3.1 in this study, rounded up to 4 treated patients to avoid one death
The odds ratio for mortality in patients on the protocol was 0.13 (0.04-0.48 95% CI). The intervention group had a significantly shorter duration of vasopressors (18.3 vs. 54.9 hours, p<0.001); less need for RRT (10% vs. 33%, p=0.02); greater improvement in SOFA scores (Delta SOFA 72hr 4.8 +/- 2.4 vs. 0.9 +/- 2.7, p<0.001); and greater procalcitonin clearance (86.4% vs. 33.9%, p<0.001). There was no significant difference in ICU LOS (4 days for both groups).
Limitations: The main limitations of this study are relatively small sample size as well as the retrospective nature of the study. Due to their inherent limitations, retrospective studies can only be hypothesis generating and limited to associations. These results are impressive but there have been many interventions in sepsis that have shown initial promise but have failed to be validated on large-scale randomized controlled studies.
The other major limitation regarding the generalizability of this study to broader practice is the use of procalcitonin to identify patients that would benefit from the intervention. This is not a routinely ordered lab at our institution & the utility of procalcitonin as a biomarker in sepsis is still a matter of debate.
Discussion
These results need to be replicated at other centers and in a randomized fashion before this intervention should be widely adopted. The harms associated with vitamin C are almost nil (except with renal dysfunction where oxalate accumulation can occur with mega-dose VitC) and the costs are low making it a potentially safe and cost-effective intervention.
This intervention was so practice changing at their institution that they feel they no longer have clinical equipoise and will not be able to complete an RCT themselves. Therefore, an RCT would have to be completed in an institution that does not have this as their standard of care in order to allow randomization to the control arm.
Reversals in medical evidence are common. An analysis of 363 articles testing standard of care in medicine were reversed 40% of the time over a 10-year period. There is a figurative graveyard of proposed therapies in sepsis management that were initially promising including statins, hemofiltration, immune globulin, drotrecogin alfa, deltibant, nitric oxide inhibitors, growth hormone, calcitriol, etc. Sepsis is not a single disease that will have a silver-bullet solution; rather, there are subsets of patients who are likely to benefit from some of the above interventions but we simply do not have the mechanism to identify them yet. Analogous to the admonishment in finance that “past returns do not guarantee future results,” past failures in sepsis research do not necessarily predict current or future failures and this study demonstrates the need for further investigation.
Applying the evidence: In this hypothetical case, the potential benefits of the intervention outweigh the known risk for harm and I would have a low threshold to trial the addition of vitamin C and thiamine to her treatment if her care had otherwise been optimized and she continued to worsen.
Written by: Adam Rieves, MD MS
Faculty reviewers: Chris Holthaus, MD, Brian Fuller, MD, MSCI, Chris Carpenter, MD, MSc, FACEP, FAAEM
References:
A Decade of Reversal: An Analysis of 146 Contradicted Medical Practices. Prasad, Vinay et al. Mayo Clinic Proceedings , Volume 88 , Issue 8 , 790 - 798
Marik PE, Khangoora V, Rivera R, Hooper MH, Catravas J, Hydrocortisone, Vitamin C and Thiamine for the Treatment of Severe Sepsis and Septic Shock: A Retrospective Before- After Study, CHEST (2017), doi: 10.1016/j.chest.2016.11.036.
https://www.uptodate.com/contents/investigational-and-ineffective-therapies-for-sepsis#H2552815
Rhodes A, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2016. Crit Care Med. 2017 Jan 17. doi: 10.1097/CCM.0000000000002255. [Epub ahead of print] PMID: 28098591 OR Intensive Care Med. 2017 Jan 18. doi: 10.1007/s00134-017-4683-6. [Epub ahead of print]