Benefits of combination therapy of hydrocortisone,ascorbic acid and thiamine in sepsis and septic shock: A systematic review

Abstract

Background:

Sepsis and septic shock are severe medical conditions that can damage multiple organs with a higher risk of mortality. Recently, the combination of hydrocortisone, ascorbic acid and thiamine (HAT) was hypothesized to work synergistically to reverse septic shock and reduce mortality.

Aim:

To ascertain the efficacy of HAT therapy and compare whether HAT therapy is more beneficial compared to the standard of care in sepsis and septic shock patients.

Methods:

PubMed, Clinicaltrials.gov, Scopus, Web of Science, Cochrane and Embase are databases that were used to identify trials that conducted a study of the combination of HAT in sepsis or septic shock.

Results:

There were 134 articles identified through a database search and 16 from other sources, which were subsequently reduced to 11 trials (six randomized trials and five non-randomized trials) that were deemed appropriate for inclusion in this review. Most of the outcomes from these studies focused on mortality, the need for renal replacement therapy, duration of vasopressor use, changes in Sequential Organ Failure Assessment score, procalcitonin clearance and lengths of intensive care unit stay.

Conclusion:

Due to inconsistent results from clinical studies, the benefits of HAT therapy cannot be confirmed at this point in sepsis and septic shock. Currently, there are at least 20 randomized controlled trials testing HAT in various combinations and dosages in patients with severe sepsis and septic shock. The results of these studies are required before definitive conclusions can be made regarding the impact of this novel treatment strategy on the morbidity and mortality of patients with sepsis.

Keywords

Sepsis septic shock vitamin C (ascorbic acid)hydrocortisone thiamine

Key points

Eleven clinical studies (six randomized and five non-randomized control trials) have evaluated the clinical benefits of hydrocortisone, ascorbic acid and thiamine (HAT) combo therapy in sepsis or septic shock.

The systematic review of available clinical studies could not confirm the clinical benefits of HAT combo therapy for sepsis or septic shock.

Based on the studies analyzed, the combination treatment protocol did not lead to patient side effects or increased mortality.

Introduction

Sepsis is a dysregulated systemic inflammatory response to an infection that can damage multiple organs, while septic shock is a critical condition of sepsis with severe hypotension and a higher risk of mortality (Pendleton, 2018). In the USA, the incidence of severe sepsis is estimated to be 3 cases per 1000 population and 2.26 cases per 100 hospital discharges (Angus et al., 2001). The average costs per case were US$22,100, with annual total costs of US$16.7 billion nationally (Angus et al., 2001; Mayr et al., 2014). According to the World Health Organization (2018), the global epidemiological burden of sepsis is difficult to ascertain, and difficulties related to sepsis is most likely highest in low- and middle-income countries. It is estimated to affect more than 30 million people worldwide every year (World Health Organization, 2018). In order to reduce the burden of sepsis worldwide, sepsis treatment interventions should be sufficient, inexpensive, safe and readily available (Marik et al., 2017). Recently, the combination of HAT has been used as a new therapeutic approach to sepsis in several trials. These three agents are thought to work synergistically with one another to reduce mortality, duration of vasopressor use and incidence of renal replacement in sepsis and septic shock patients. Plasma levels of vitamin C are significantly wasted in critically ailing patients, and this can contribute to hypotension, exaggerated inflammation, microcirculatory compromise, oxidative organ injury and immune defense (Litwak et al., 2019). Besides that, vitamin C also demonstrates its role in restoring glucocorticoid receptors function once they get oxidized (Marik et al., 2017). Hydrocortisone is an anti-inflammatory agent that can help to mitigate the inflammatory response during septic shock. Glucocorticoids can amplify the effects of catecholamines via glucocorticoid receptors on vascular smooth muscle cells to suppress the production of prostacyclin and nitric oxide, which are the essential factors of vasodilator (Litwak et al., 2019). In addition, glucocorticoids can also increase the expression of the sodium-vitamin C transporter-2, which mediated the transport of vitamin C into the cells (Marik et al., 2017). Accordingly, vitamin C can aid in glucocorticoid receptor renewal, while hydrocortisone can enhance the cellular absorption of vitamin C. Another study also showed that a high dose of intravenous vitamin C could cause calcium oxalate nephropathy; thus, thiamine can be used to prevent this reaction by decreasing the conversion of glyoxylate to oxalate (Kim et al., 2018). Based on the mechanism of action on the three agents, they theoretically work synergistically together to reduce endothelial barrier changes caused by septic shock and prevent nephropathy caused by overproduction of oxalate (Litwak et al., 2019). This review aims to assess those studies to ascertain the potential benefits of the combination therapy of HAT on mortality, the need for renal replacement therapy, duration of vasopressor use, changes in Sequential Organ Failure Assessment (SOFA) score, procalcitonin clearance and length of stay for patients with sepsis/septic shock.

Methods

Database search

These databases were referenced for the research in question: PubMed (1996–April 2021), Clinicaltrials (2000–April 2021), Scopus (2004–April 2021), Web of Science (1900–April 2021), Cochrane (1996–April 2021) and Embase (1947–April 2021). The following terms were used for potential study identification: “vitamin C, hydrocortisone, thiamine, sepsis” and “vitamin C AND hydrocortisone AND thiamine AND sepsis.”

Study selection

Articles reviewed were not restricted to those written in English; however, there were no foreign language studies included. Articles were not included if the following applied: labeled review article; did not combine the three medications; did not reference sepsis or septic shock as the primary diagnosis; did not use human participants. If the study was controlled, at least one standard therapy should be a combination of HAT. Three independent researchers reviewed articles to identify relevant information referencing mortality rates, vasopressor-free days and length of intensive care unit (ICU) stay in clinically diagnosed sepsis or septic shock patients.

The following information was collected for the systematic review: study purpose; study design; whether the setting was hospital versus ICU; time frame of the study performance; drug duration/route/dose; predominant type of infection; score of severity (utilizing SOFA [Acute Physiology and Chronic Health Evaluation] APACHE II, PaO2/FiO2 and EUROSCORE II); inclusion and exclusion criteria; primary and secondary results; population size; hospital mortality index; vasopressor-free days; ventilator-free days; ICU length of stay; changes in lactate level; change in radiologic score at day 7; requirement for renal replacement therapy (with acute kidney injury [AKI] diagnosis); and procalcitonin level.

Risk of bias was evaluated by examining the randomization method, blind method, adverse effects, treatment withdrawals, missing data or any “as-treated” analysis. The risk of bias table was made by one author utilizing the Cochrane Handbook for Systematic Reviews of Interventions and evaluated by two other authors for accuracy (The Cochrane Collaboration, 2011).

Results

Initially, 134 articles were identified, with 16 others from different sources (which includes six articles provided by Nutrition and Health editors), but following a review utilizing the Preferred Reporting Item for Systematic Reviews and Meta-Analyses (PRISMA) statement, that number was reduced to 11 studies (six randomized and five non-randomized trials) (see Figure 1) (Moher et al., 2009). All 11 trials were published between the years of 2017 and 2021. A detailed accounting of the outcomes in each of those trials can be found in Table 1. The following is a discourse on the cumulative results of the major outcomes assessed by this review, such as mortality, renal replacement therapy, duration of vasopressor use, changes in SOFA scores, procalcitonin clearance, lengths of stay and a risk of bias assessment. Common sites of infection included throughout the trials were respiratory, intra-abdominal, urinary tract, pulmonary, gastrointestinal, skin/soft tissue, blood, central nervous system (CNS) and bone/joint (Coloretti et al., 2020; Fujii et al., 2020; Hwang et al., 2020; Kim et al., 2018; Litwak et al., 2019; Marik et al., 2017; Moskowitz et al., 2020; Sevransky et al., 2020; Vail et al., 2020).

Figure 1.

Preferred Reporting Item for Systematic Reviews and Meta-Analyses 2009 flow diagram.

Table 1.

Summary of all studies (Balakrishnan et al., 2018; Coloretti et al., 2020; Fujii et al., 2020; Hwang et al., 2020; Iglesias et al., 2020; Kim et al., 2018; Litwak et al., 2019; Marik et al., 2017; Moskowitz et al., 2020; Sevransky et al., 2020; Vail et al., 2020).

Study, Year/N	Study drugs	Baseline sepsis severity (mean ± SD)	Common infections	Inclusion	Exclusion	Primary outcomes	Secondary outcomes
Randomized controlled trial
Balakrishnan et al., 2018/24	Vitamin C: 1.5 g q6 h for 4 days Hydrocortisone: 50 mg q6 h for 4 days Thiamine: 200 mg q12 h for 4 days	APACHE IV: (TG) 60.91 ± 17.52 vs. (CG) 51.81 ± 10.76, p = 0.1469	Patients with sepsis or septic shock following cardiac surgery	Septic shock with [PCT] level > 7 ng/ml	Septic shock with [PCT] level < 7 ng/ml within the first 24 h; Less than 18 years of age; Pregnancy; Chronic kidney disease; Immunocompromised patients	Vasopressor dose over 4 days	Inpatient mortality
Moskowitz et al., 2020/205	Vitamin C: 1.5 g q6 h for 4 days Thiamine: 100 mg q6 h for 4 days Hydrocortisone: 50 mg q6 h for 4 days	30-d predicted survival, no. (%) High: (TG) 34 (33.7%) vs. (CG) 38 (38.4%) Uncertain: (TG) 61 (60.4%) vs. (CG) 54 (54.6%) Low: (TG) 6 (5.9%) vs. (CG) 7 (7.1%)	Pneumonia: (TG) 33% vs. (CG) 30.1% Intra-abdominal: (TG) 31.9% vs. (CG) 24.7% UTI: (TG) 21.3% vs. (CG) 23.7% Other: (TG) 13.8% vs (CG) 21.5%	18 and older had suspected or confirmed infection and were receiving a vasopressor due to sepsis	Hypersensitivity with drugs; Had clinical indication for any of the study drugs; Symptomatic kidney stone within the last year; Had glucose-6 phosphate dehydrogenase deficiency; Receiving kidney replacement therapy; Pregnancy; Prisoner	Change in SOFA score from enrolment and 72-h follow-up	Kidney failure: defined as KDIGO grade 3 or higher during the index ICU stay and all-cause mortality after the first 30 days
Fujii et al., 2020/211	Vitamin C: 1.5 g q6 h for 3.4 days Hydrocortisone: 50 mg q6 h for 3.4 days Thiamine: 200 mg q12 h for 3.4 days	APACHE III: 77.4 ± 29.7 (TG) vs 83.3 ± 28.8 (CG)	Pulmonary: 29% (TG) vs. 31.7% (CG) GI: 29% (TG) vs. 29.8% (CG) Urinary: 16.8% (TG) vs 13.5% (CG) Skin/soft tissue: 13.1% (TG) vs. 14.4% (CG) Blood: 8.4% (TG) vs. 1.9% (CG) Other: 3.7% (TG) vs. 8.7% (CG)	Admitted to ICU and diagnosed with septic shock fulfilled max 24 h prior to enrolment (Sepsis 3-consensus: sepsis with persistent hypotension requiring vasopressors for at least 2 h, acute increase in SOFA score of 2 or more, and lactate >2 mmol/L)	Age < 18 years; Pregnancy DNR/DNI orders; Death is deemed to be imminent or inevitable during this admission, and either the attending physician, patient or substitute decision-maker is not committed to active treatment; Patients with known HIV infection, G6PD deficiency, hx of oxalate nephropathy, short bowel syndrome/severe, fat-malabsorption, acute beriberi disease, acute Wernicke’s, encephalopathy, malaria, scurvy, Addison’s disease, Cushing’s disease, hyperoxaluria; Patients transferred from another ICU or hospital with a diagnosis of a septic shock for >24 hours; Patients with a diagnosis of a septic shock for >24 hours; Clinician expects to prescribe systemic glucocorticoids for an indication other than septic shock; Patient is receiving treatment for systemic fungal infection or documented Strongyloides infection; Known chronic iron overload due to iron storage and other diseases; Patient previously enrolled in this study; Prescribe high-dose vitamin C for another indication	Duration of time alive and free of vasopressor administration up to day 7	28-day, 90-day, ICU, and hospital mortality, 28-day cumulative vasopressor-free days, 28-day cumulative mechanical ventilation-free days, 28-day renal replacement therapy-free days, change in SOFA score at day 3, 28-day ICU free-days, and hospital length of stay
Hwang et al., 2020/116	Vitamin C: 50 mg/kg (max single dose 3 g, daily dose 6 g) q12 h for 48 h Hydrocortisone: 200 mg q24 h or 50 mg q6 h continuous infusion for 48 h Thiamine: 200 mg q12 h for 48 h	APACHE II:22 (14–32) (TG) vs. 22 (17–32) (CG), p = 0.82	Respiratory: 20.8% (TG) vs. 27.6% (CG) Intra-abdominal: 50.9% (TG) vs. 41.4 (CG) UTI: 18.9% (TG) vs. 17.2% (CG) Other/unknown: 9.4% (TG) vs. 13.8% (CG)	Adults 19–89 years old) Presented to an ED and diagnosed with septic shock (sepsis with persistent hypotension requiring vasopressors, acute increase in SOFA score of 2 or more, and lactate >2 mmol/L)	Transferred with vasopressor administration or mechanical ventilator support Had limitations on treatment; Underlying terminal-stage disease without active treatment plan, and not expected to survive to discharge; Taking at least 1 g/day of vitamin C or receiving intravenous thiamine prior to enrolment; Experiencing cardiac arrest prior to enrolment or who were expected to die within 24 h despite best possible treatment; Pregnant; With G6PDH deficiency, thalassemia, cystinuria, hyperoxaluria, ongoing gout attacks, or oxalate renal stones; Hypersensitivity hx reactions to trial drugs; Patients who met the inclusion criteria more than 24 h after ED arrival or who were screened more than 24 h after diagnosis of septic shock; Declined to participate in the trial	Change in SOFA score (SOFA score at enrolment–SOFA score after 72 h)	Mortality (7-day, 28-day, 90-day, in-hospital, ICU), shock reversal, vasopressor-free days, vasopressor dose (at 24, 48 and 72 h from enrolment and maximum dose during 72 h), duration of mechanical ventilation, ventilator-free days, new-onset or worsening acute kidney injury (AKI) after enrolment, new use of renal replacement therapy (RRT), RRT-free days, ICU length of stay (LOS), ICU-free days, hospital LOS, reduction of CRP for 72 h, and reduction of PCT for 72 h
Iglesias et al., 2020/137	Vitamin C: 1500 mg q6 h for 4 days Thiamine: 200 mg every 12 hours for 4 days Hydrocortisone: 50 mg q6 h for 4 days	APACHE II score TG: 24 ± 7.6 CG: 24.9 ± 8.7 APACHE IV score TG: 88 ± 28.3 CG: 87.5 ± 29.7	Pulmonary: (TG) 43%, (CG) 44% UTI: (TG) 26.5%, (CG) 36% Blood: (TG) 13%, (CG) 16% GI/other: (TG) 13%, (CG) 12%	≥18 years of age with a primary diagnosis of sepsis or septic shock according to the 2016 Surviving Sepsis Campaign, within 12 h of admission to the ICU and compliance to the 3-h sepsis bundle	Pregnant; DNR or DNI order on admission; Terminal end-stage disease (e.g., stage IV cancer, end-stage heart failure); No primary admitting diagnosis of sepsis or septic shock; Required immediate surgery Had HIV and a CD4 <50 mm², had known G6PDH deficiency; Transferred from another hospital; Presented with sepsis/septic shock more than 24 h from admission	Resolution of shock and change in SOFA score	ICU mortality, hospital mortality, procalcitonin clearance (PCT-c), hospital length of stay (LOS), ICU LOS and ventilator-free days
Sevransky et al., 2020/501	Vitamin C: 1.5 g q6 h for 4 days or until death/ICU discharge Thiamine hydrochloride: 100 mg q6 h for 4 days or until death/ICU discharge Hydrocortisone sodium succinate: 50 mg q6 h for 4 days or until death/ICU discharge	APACHE II score, median (IQR) TG: 27 (22–33) CG: 27 (19–33)	Lung: (TG) 37.1% vs. (CG) 39.7% UTI: (TG) 13.2% vs. (CG) 21.1% Intra-abdominal: (TG) 18% vs. (CG) 11.8% Blood/vascular: (TG) 18% vs. (CG) 10.8% Skin/soft tissue: (TG) 5.4% vs. (CG) 8.8% CNS: (TG) 2.4% vs. (CG) 0.5% Bone/joint: (TG) 1% vs. (CG) 1% Other: (TG) 4.4% vs. (CG) 5.4% Unknown: (TG) 0.5% vs. (CG) 1%	18 and older with acute respiratory and/or cardiovascular dysfunction caused by suspected infection with planned ICU admission	Weight < 40 kilograms (kg); Prior enrollment in this study; Qualifying organ dysfunction no longer present at the time subject would be randomized; Cardiovascular or respiratory organ failure caused by an illness other than sepsis; First episode of qualifying organ dysfunction during the current emergency department (ED) or ICU admission occurred > 24 hours before the subject could be randomized; Limitations of care; Current hospitalization > 30 days at time of randomization; Chronic hypoxemia requiring supplemental non-invasive oxygen or home mechanical ventilation; Chronic cardiovascular failure requiring home mechanical hemodynamic support/ home chemical hemodynamic support; Known allergy or contraindication to study meds or known/suspected ethylene glycol ingestion, or G6PD deficiency; Use of vitamin C at a dose of > 1 gram daily within the 24 hours preceding first episode of qualifying organ dysfunction during a given ED/ICU admission; Chronic disease/illness that have an expected lifespan of < 30 days unrelated to current sepsis diagnosis; Pregnancy/active breastfeeding; Prisoner/Incarceration; Current participation in another interventional research study; Inability or unwillingness of subject or legal surrogate/representative to give written informed consent	Ventilator-vasopressor free days (VVFDs) in the first 30 days	Mortality within 30days of randomization
Non-randomized trials
Study,Year/N	Study drugs	Baseline sepsis severity (mean ± SD)	Common infections	Inclusion	Exclusion	Primary outcomes	Secondary outcomes
Marik et al., 2017/94	Vitamin C: 1.5 g q6 h for 4 days or until ICU discharge Hydrocortisone: 50 mg q6 h for 7 days or until ICU discharge followed by a taper over 3 days Thiamine: 200 mg q12 h for 4 days or until ICU discharge	APACHE IV: (TG) 79.5 ± 16.4 vs. (CG) 82.0 ± 27.4; (F value, 13.29; p < 0.001) APACHE II: (TG) 22.1 ± 6.3 vs. (CG) 22.6 ± 5.7	Pneumonia: (TG) 38% vs. (CG) 40% Urosepsis: (TG) 23% vs. (CG) 21% Primary bacteremia: (TG) 15% vs. (CG) 15% GI/biliary: (TG) 13% vs. (CG) 13% Other: (TG) 11% vs. (CG) 11%	Severe sepsis or septic shock with [PCT] level ≥2 ng/ml	Septic shock with [PCT] level <2 ng/ml within the first 24 h; Less than 18 years of age; Pregnancy	Hospital mortality	Duration of vasopressor therapy, requirement for renal replacement therapy in patients with AKI, ICU LOS, Δ[PCT], and Δ SOFA score over the first 72 h
Kim et al., 2018/99	Vitamin C: 1.5 g q6 h for 4 days Hydrocortisone: 50 mg q6 h for 7 days or until ICU discharge followed by a taper over 3 days Thiamine: 200 mg q12 h for 4 days	APACHE II: (TG) 27 (16–31) vs. (CG) 29 (23–34)	Severe pneumonia: (TG) 100% vs. (CG) 100% Concurrent bacteremia: (TG) 8% vs. (CG) 7% ARDS at ICU: (TG) 23% vs. (CG) 22%	Critically ill adult patients with severe pneumonia admitted to ICU	Not admitted to ICU and/or required conventional oxygen therapy only; Had acute pneumonia; Had nosocomial pneumonia, occurred >48 h after hospitalization; The vitamin C protocol infusion occurred >48 h after hospitalization	Hospital mortality	The number of vasopressor-free day, ICU LOS
Vail et al., 2020/338597	Evaluated patients with at least one charge for high-dose intravenous vitamin C (one or more vials totaling ≥500 mg in one calendar day) and at least one charge for both IV hydrocortisone and IV thiamine (of any dose) on or within 1 day of ascorbic acid administration	The study data set does not contain a specific severity of illness measure	N/A	Patients discharged from 1 October 2015 to 30 September 2018 with septic shock defined as the presence of ICD-10 codes for infection and organ dysfunction on or before hospital discharge	<18 years old; Never admitted to an ICU, or who had diagnoses that might result in intravenous ascorbic acid administration for other reasons (vitamin C deficiency and burn injury)	Receipt of HAT at least once during hospitalization	Hospital mortality
Litwak et al., 2019/94	Vitamin C: 1.5 g q6 h Hydrocortisone: 50 mg q6 h or 100 mg q8 h Thiamine: 200 mg q12	APACHE II: (TG) 21.5 ± 8.0 vs. (CG) 20.0 ± 7.4; p = 0.739 APACHE IV: (TG) 88.6 ± 29.1 vs. (CG) 84.1 ± 25.4; p = 0.455 APACHE IV Predicted mortality: (TG) 38.9 ± 27.2 vs. (CG) 39.0 ± 23.6; p = 0.991	Pneumonia: (TG) 46.8% vs. (CG) 38.3%; p = 0.404 Urosepsis: (TG) 8.5% vs. (CG) 17%; p = 0.216 Primary bacteremia: (TG) 6.4% vs. (CG) 6.4%; p = 1 GI/biliary: (TG) 21.3% vs. (CG) 31.9%; p = 0.243 Other: (TG) 14.9% vs. (CG) 8.5%; p = 0.336	Sepsis or septic shock with associated respiratory or cardiovascular dysfunction or both tients receiving vasopressor therapy	Pregnancy	Hospital mortality	ICU mortality, duration of vasopressor therapy, requirement for renal replacement therapy in patients with AKI, ICU LOS, Δ[PCT], and Δ SOFA score over the first 72 h
Coloretti et al., 2020/116	Vitamin C: 1.5 g q6 h up to discontinuation of vasoactive drugs or steroids Hydrocortisone: 240 mg/day continuous infusion in patients requiring noradrenaline Dosages >0.3–0.4 μg/kg/min Thiamine: 200 mg q12 h up to discontinuation of vasoactive drugs or steroid Overall 3 days of therapy (IQR 2–5)	SAPS II score: 61 (47–72) (TG) vs. 56 (47–71) (CG), p = 0.798	Pulmonary: 48.2% (TG) vs. 41.1% (CG) Abdominal: 32.1% (TG) vs. 35.7% (CG) Bloodstream: 12.5% (TG) vs. 17.9% (CG) Other: 25% (TG) vs. 25% (CG)	Admitted polyvalent ICU of the Modena University Hospital with septic shock and receiving low-dose steroids from January 2015 to November 2019	Patients with DNR orders or end of-life decisions during the ICU stay	Modified SOFA scores at 96 h and plasma biomarker levels at 168 h	28-day mortality
Detailed results
Study, Year/N	Use of steroid TG vs CG	Mortality	Renal replacement therapy	Duration of vasopressor	ΔSOFA score	[PCT] clearance	ICU LOS
Marik et al., 2017/94	100% vs. 59.6%	Hospital mortality: (TG) 8.5% (4 of 47 pts) vs. (CG) 40.4% (19 of 47 pts); p < 0.001 Propensity adjusted odds of mortality in patients treated with vitamin C protocol was 0.13 (95% CI, 0.04–0.48; p = 0.002) Note: none of the patients in TG died of complications related to sepsis. All patients survived their ICU stay, received “comfort care” on the hospital floor and died of complications of their underlying diseases (adv. dementia, severe HF, adv. sarcoidosis and severe COPD)	Renal replacement therapy (RRT) (TG) 31 patients (66%) met criteria for AKI vs. (CG) 30 patients (64%) met criteria for AKI (TG) 3 of 31 patients (10%) required RRT vs. (CG) 11 of 30 patients (37%) required RRT; p = 0.02	Duration of vasopressors, (mean ± SD), h: (TG) 18.3 ± 9.8 (CG) 54.9 ± 28.4 P < 0.001	ΔSOFA score within 72 h: (TG) 4.8 ± 2.4 vs. (CG) 0.9 ± 2.7; p < 0.001	Procalcitonin clearance within 72 h, median % and IQR (TG) 86.4% (80.1–90.8%) vs. (CG) 33.9% (–62.4% to 64.3%); p < 0.001	N/A
Balakrishnan et al., 2018/24	N/A	Inpatient mortality: no statistically significant difference, yet did not provide any detailed statistic on mortality Note: small sample size, short period intervention and no assessment of patients’ serum ascorbate baseline level are the major limitations of this study	N/A	Vasopressor dose over 4 days The different requirement for vasopressin from Day 1 to Day 4: (TG) .0025 p = 0.001 (CG) –.0042 p = 0.001	No difference in the SOFA score; p = 0.089 on Day 4. Note: did not provide specific scores	PCT levels on Day 4: (CG) 70.03 ± 29.74 ng/ml vs. (TG) 26.3 ± 23.08 ng/ml; p = 0.0009	N/A
Kim et al., 2018/99	100% vs. 90% (36/40)	Hospital mortality without PC: (TG) died 11 of 53 patients (21%) vs. (CG): died 17 of 37 patients (37%); p = 0.07 Hospitality mortality with PC: (TG) died 6 of 36 patients (17%) (CG) died 14 of 36 patients (39%); p = 0.04	N/A	No. vasopressor-free days at Day 28 (mean ± SD) without PC: (TG) 18.8 ± 11.2 (CG) 19.6 ± 11.4 P = 0.76 No. vasopressor-free days at Day 28 (mean ± SD) with PC: (TG) 19.8 ± 10.8 (CG) 20.5 ± 11.1 p = 0.45	N/A	N/A	Length of ICU stay (days): (TG) 9 (4–14) (CG) 12 (6–17) p = 0.19 (clean copy)
Litwak et al., 2019/94	100% vs. 40.4%	Hospital mortality: (TG) 40.4% (CG) 40.4%; p = 1.000 ICU mortality: (TG) 36.2% vs. (CG) 38.3%; p = 0.831	Requirement for renal replacement therapy for AKI: (TG) 28.9% vs. (CG) 34.4%; p = 0.626	Duration of vasopressors, median (IQR), hour: (TG) 84.2 h (37.0–169.3) (CG) 62.5 h (32.6–105.9) p = 0.324	SOFA scores, (mean ± SD) within the first 72 h: (TG) 1.3 ± 4.2 vs. (CG) 0.1 ± 4.7; p = 0.390	median 72-h change in PCT: (TG) 0.1 (CG) 2.5 p=0.268	ICU LOS: (TG) 11 days (CG) 10 days P = 0.491 Hospital LOS (TG) 19 days (CG)14 days p = 0.346
Moskowitz et al., 2020/205	NA	All-cause mortality over 30 days: (TG) 34.7% vs. (CG) 29.3%; p = 0.26	Kidney failure: (TG) 31.7% vs. (CG) 27.3%; p = 0.58	NA	ΔSOFA score within 72 h: (TG) 4.1% vs. (CG) 4.3%; p = 0.12	NA	NA
Coloretti et al., 2020/116	100% vs .100% (56/56)	Hospital mortality: (TG) 50% (CG) 60.7% p = 0.254	Renal replacement therapy: (TG) 39.2 (CG) 37.5 p = 0.846	Vasopressors length of days (median-IQR) (TG) 3 (2–5) (CG) 4 (2–7) p = 0.533	N/A	Procalcitonin Level (ng/dl, median- IQR) (TG) 49 (8–101) (CG) 34 (12–104)	ICU length of stay (days, median- IQR) (TG) 6 (4–10) (CG) 9 (5–15) p = 0.092
Fujii et al., 2020/211	99.1% vs. 98.1% (106/102)	Hospital mortality: (TG) 23.4% vs. (CG) 20.4% with n = 103 ICU mortality: (TG) 19.6% vs. (CG) 18.3%	28 day renal replacement therapy-free days: (TG) 28 (23.5–28) (CG) 28 (21–28), p = .71	28 day cumulative vasopressor-free days: (TG) 25.6 (17.8–26.8) vs. (CG) 25.8 (19.6–26.8) p = .66	Change in SOFA score at day 3: (TG) –2 (–4 to 0) (CG) –1 (–3 to 0) p = .02	NA	Hospital LOS: (TG) 12.3 (6.2–26) (CG) 12.3 (6.2–26.1) p = 0.75 28 day ICU-free days: (TG) 21.9 (0–25.8) (CG) 22.1 (3.9–25.8) p = .66 (not sig)
Hwang et al., 2020/116	100% vs. 100% (107/107)	Hospital mortality: (TG) 24.5% (CG) 19% p = 0.48 ICU mortality: (TG) 15.2% (CG) 13.5% p = 0.8	New use of RRT: (TG) 17.3% (CG) 19 p = 0.82 RRT-free day: (TG) 14 (14–14) (CG) 14 (14–14) p = 0.7	Vasopressor-free days: (TG) 11 (5–12) (CG) 11 (10–12) p = 0.16	Change in SOFA score at day 4: (TG) 3 (−1 to 5) (CG) 3 (0–4) p = 0.96	Reduction of PCT for 72 h: (TG) 49.2% (−31.9 to 74.7) (CG): 40.3% (−3.4 to 82.8) p = 0.27	ICU LOS (day): (TG) 5 (3–11)(CG) 5.5 (4–12.5) p = 0.22
Iglesias et al., 2020/137	100% vs. 100% (68/69)	Hospital mortality: (TG) 16.4% (CG) 19% p = 0.65 ICU mortality: (TG) 9% (CG) 14% p = 0.37	RRT: (TG) 3% (CG) 11.5% p = 0.1	Duration of vasopressors (days): (TG) 27 ± 22 (CG) 53 ± 38 p ≤ .001	ΔSOFA score at 72 h (TG) 2.9 ± 3.3 (CG) 1.93 ± 3.5 p = 0.1	PCT clearance (%) (TG) 63 ± 170 (CG) 58 ± 66 p = 0.44	ICU LOS (days): (TG) 4.76 ± 4.3 (CG) 4.66 ± 3.45 p =0.88
Sevransky et al., 2020/501	NA	Mortality before ICU discharge: (TG) 20.6% vs. (CG) 19.7; p = 0.79 Mortality at 180 days: (TG) 40.5% vs. (CG) 37.8%; p = 0.79	Kidney replacement therapy-free days, median (IQR): (TG) 30 (0–30) vs. (CG) 30 (0–30); p = 0.58	NA	ΔSOFA score, median (IQR): (TG) 5 (3–7) vs. (CG) 5 (2–7)	NA	ICU LOS, median (IQR): (TG) 4 (2–8) (CG) 4 (2–8) Hospital LOS, median (IQR): (TG) 10 (6–17) (CG) 9 (5–17)
Vail et al., 2020/338,597	N/A	Hospital mortality: (TG) 40.8% vs. (CG) 27.5% Location after hospital discharge Home: (TG) 18.2% vs. (CG) 29% Nursing facility: (TG) 27.6% vs. (CG) 30.9% Other acute care facilities: (TG) 5.4% vs. (CG) 4.4% Hospice: (TG) 5.5% vs. (CG) 7.3% Other: (TG) 0.5% vs. (CG) 1.2%	Renal replacement therapy during admission: (TG) 28% vs. (CG) 15.5%)	Vasopressor use in days, median (IQR): In hospital survivors: (TG) 2(1–4)) vs. (CG) 2(1–3) In hospital non-survivors: (TG) 3(2–6) vs. (CG) 2(1–4)	NA	NA	ICU LOS, median (IQR) Hospital survivors: (TG) 6(3–11) vs. (CG) 4(2–8) Hospital nonsurvivors: (TG) 4(2–9) vs. (CG) 3(1–8)

TG: treatment group; CG: control group; CRP: C-Reactive Protein; SOFA: Sequential Organ Failure Assessment; KDIGO: Kidney Disease: Improving Global Outcomes; UTI: urinary tract infection; ICU: intensive care unit; DNR: do not resuscitate; DNI: do not intubate; G6PD: glucose-6-phosphate-dehydrogenase; ED: emergency department; CNS: central nervous system; PCT-c: procalcitonin clearance; HAT: hydrocortisone, ascorbic acid and thiamine; HF: heart failure; COPD: chronic obstructive pulmonary disease; PCT: procalcitonin; LOS: length of stay; AKI: acute kidney injury; Δ: changes; ARDS: acute respiratory distress syndrome; PC: propensity score; GI: Gastrointestinal; IQR: Interquartile Range.

Mortality

All 11 trials examined in this review evaluated the effects that combination HAT therapy had on mortality in patients with sepsis, with four trials displaying beneficial results. The experiments by Marik et al. (2017), Kim et al. (2018), Coloretti et al. (2020) and Iglesias et al. (2020) showed decreased hospital mortality in sepsis patients who received the combination treatment of HAT. On the other hand, the studies by Litwak et al. (2019), Sevransky et al. (2020), \Moskowitz et al. (2020), Vail et al. (2020), Hwang et al. (2020) and Fujii et al. (2020) displayed no improvement in in-hospital mortality between the treatment group and the control group (Balakrishnan et al., 2018).

Renal replacement therapy

In total, nine trials (Coloretti et al., 2020; Fujii et al., 2020; Hwang et al., 2020; Iglesias et al., 2020; Litwak et al., 2019; Marik et al., 2017; Moskowitz et al., 2020; Sevransky et al., 2020; Vail et al., 2020 ) assessed in this review evaluated the effects of HAT therapy on the need for renal replacement therapy. The studies by Marik et al. (2017), Litwak et al. (2019) and Iglesias et al. (2020) all displayed a reduction in renal replacement therapy for sepsis patients who received HAT therapy compared to the control group, who only received standard care. In contrast, the other studies showed no significant difference in the requirement of renal replacement therapy between the treatment group and the control group (Coloretti et al., 2020; Fujii et al., 2020; Hwang et al., 2020; Moskowitz et al., 2020; Sevransky et al., 2020; Vail et al., 2020).

Duration of vasopressor use

All nine trials assessed in this review evaluated the effects of HAT on the duration of vasopressor(s). Six of the nine trials by Marik et al. (2017), Balakrishnan et al. (2018), Kim et al. (2018), Coloretti et al. (2020), Fujii et al. (2020) and Iglesias et al. (2020) showed beneficial results with less time to vasopressor dependence in the treatment group who received the combination therapy of HAT compared to the control group, who only received the standard care. Nevertheless, the other three studies displayed no significant difference in time to vasopressor dependence in both groups (Hwang et al., 2020; Litwak et al., 2019; Vail et al., 2020).

Changes in SOFA scores

Eight out of the 11 trials in this review evaluated the benefits of HAT on changes in SOFA scores within the first 72 hours of sepsis admission. The studies by Moskowitz et al. (2020) and Fujii et al. (2020) are the only two that showed decreased SOFA scores in patients receiving the combination therapy of HAT. Still, the other six trials displayed no significant difference in changes in SOFA scores between the treatment group and the control group (Balakrishnan et al., 2018; Hwang et al., 2020; Iglesias et al., 2020; Litwak et al., 2019; Marik et al., 2017; Sevransky et al., 2020).

Procalcitonin clearance

Out of the 11 trials, six trials in this review assessed the effects of HAT on the clearance of procalcitonin within 72 hours of sepsis administration. Studies by Marik et al. (2017), Balakrishnan et al. (2018), Hwang et al. (2020) and Iglesias et al. (2020) displayed an increase in procalcitonin clearance of the patients who received the combination therapy with a statistically significant difference in both groups. The other two trials by Coloretti et al. (2020) and Litwak et al. (2019) showed no statistically significant difference in procalcitonin clearance between the treatment group and the control group.

Lengths of stay

Of the 11 studies, eight studies in this review evaluated the benefits of the combination therapy of HAT on lengths of ICU stay in sepsis patients.

Three studies by Kim et al. (2018), Coloretti et al. (2020) and Hwang et al. (2020) exhibited statistically significant decreases in ICU length of stay between the treatment group and the control group. The other five trials did not have such an impact (Fujii et al., 2020; Iglesias et al., 2020; Litwak et al., 2019; Sevransky et al., 2020; Vail et al., 2020).

Risk of bias in randomized controlled trial

There are six randomized studies, Balakrishnan et al. (2018), Moskowitz et al. (2020), Hwang et al. (2020), Fujii et al. (2020), Iglesias et al. (2020) and Sevransky et al. (2020), while the others are non-randomized. After evaluating random sequence generation, allocation concealment, blinding of outcome assessment, blinding of outcome data addressed, incomplete outcome data addressed and selective reporting, there is no risk for bias with Balakrishnan et al. (2018), Hwang et al. (2020) or Sevransky et al. (2020). However, it is minimal in Fujii et al. (2020), Moskowitz et al. (2020), Iglesias et al. (2020) and Vail et al. (2020). All other studies revealed high selection bias, performance bias and detection bias. Table 2 provides a summary of the risk of bias assessment (Balakrishnan et al., 2018; Coloretti et al., 2020; Fujii et al., 2020; Hwang et al., 2020; Iglesias et al., 2020; Kim et al., 2018; Litwak et al., 2019; Marik et al., 2017; Moskowitz et al., 2020; Sevransky et al., 2020; Vail et al., 2020).

Table 2.

Risk of bias assessment.

Study, year	Random sequence generation (selection bias)	Allocation concealment (selection bias)	Blinding of participants and personal (performance bias)	Blinding of outcome assessment (detection bias)	Incomplete outcome data (attrition bias) – mortality	Selective reporting (reporting bias)
Marik et al. (2017)	High	High	High	High	Low	Low
Balakrishnan et al. (2018)	Low	Low	Low	Low	Low	Low
Kim et al. (2018)	High	High	High	High	Low	Low
Litwak et al. (2019)	High	High	High	High	Low	Low
Coloretti et al. (2020)	High	High	High	Low	Low	Low
Fujii et al. (2020)	Low	Low	High	Low	Low	Low
Hwang et al. (2020)	Low	Low	Low	Low	Low	Low
Iglesias et al. (2020)	Low	High	Low	Low	Low	Low
Moskowitz et al. (2020)	High	Low	Low	Low	Low	Low
Sevransky et al. (2020)	Low	Low	Low	Low	Low	Low
Vail et al. (2020)	Low	High	High	Low	Low	Low

Discussion

This systematic review evaluated all published studies with HAT therapy. Although all 11 studies assessed the clinical outcomes in sepsis or septic shock patients, the findings were inconsistent from these trials due to the variations in study design, the severity of diseases, infection types, triple regimen (consistent dose and duration) and the sample size among the studies.

Of the 11 studies, five studies were not randomized, which could increase the risk of selection and reporting bias. Although experiments by Marik et al. (2017), Kim et al. (2018) and Coloretti et al. (2020) showed the benefits of HAT therapy on mortality, these studies have some limitations, such as retrospectively non-randomized controlled, small sample size within a single-center design and the participation of nonconcurrent control participants. In this review, a total of six randomized control studies were the experiments by Balakrishnan et al. (2018), Hwang et al. (2020), Fujii et al. (2020), Iglesias et al. (2020), Moskowitz et al. (2020) and Sevransky et al. (2020). However, the Balakrishnan et al. (2018) trial only enrolled patients with sepsis and septic shock following cardiac operations, with a limited sample size that was inadequate for evaluating the difference in the mortality parameter.

Based on the baseline sepsis severity, differences arise in APACHE II and APACHE IV scores among the studies, which could be a factor that impacts the consistency in outcomes across the experiments. For instance, between the trials by Balakrishnan et al. (2018) and by Marik et al. (2017), the baseline sepsis severity from the Balakrishnan et al. (2018) trial was less severe compared to the trial by Marik et al. (2017). Due to the differences between the APACHE IV scores, the study by Balakrishnan et al. (2018) detected no statistically significant difference in mortality, whereas the study by Marik et al. (2017) displayed a reduction in mortality in the group of patients who were treated with the HAT therapy. In addition, the mortality data is not presented in the published Balakrishnan et al. (2018) trial. This is a significant limitation.

Another variation among the studies is infection types. The frequent infections that lead to sepsis and septic shock were diverse across all studies. One trial only enrolled septic shock patients from cardiac surgery, yet other trials included patients with pneumonia or other types of infections disproportionately. The Balakrishnan et al. (2018) trial did not report the site(s) of infection. The dissimilarity in types of infection may influence the treatment outcomes that lead to inconsistency about the study findings.

Dosage and duration of the triple regimen is another element that is not consistent among the trials. Some studies used the HAT therapy for four days, whereas other studies decided to extend the use of hydrocortisone for seven days or until ICU discharge. The dosage of the three agents was similar across the studies, except the dosage of hydrocortisone from the study by Litwak et al. (2019). In addition, there are several limitations in the study by Litwak et al. (2019) because the combination of HAT regimen was not protocolized.

Although the study by Litwak et al. (2019) tried to replicate the study by Marik et al. (2017), the outcomes were in conflict with one another. The inclusion criteria in the Litwak et al. (2019) study were not provided, while Marik et al. (2017) included sequential patients meeting the inclusion criteria. It is likely that there was a severe selection bias in which patients received HAT. In the study by Litwak et al. (2019), the decision to initiate and the timing of discontinuation of HAT therapy was solely based upon provider preference. About 57% of the patients in the treatment group did not complete the full four-day treatment duration due to late HAT therapy initiation, therapy discontinuation by the primary team and patient death. HAT therapy was initiated at median 13.8 h (7.5–36.8 h) from vasopressor initiation. Due to the lack of treatment protocol and interruption of therapy, the study by Litwak et al. (2019) did not detect any beneficial outcomes from the HAT therapy in sepsis or septic shock patients.

The population sample size is another limitation among these trials. In this review, the sample size ranged from 24 to 99 in four trials. With a limited sample size, some studies did not have enough power to detect the differences in clinical outcomes. Therefore, it is difficult to draw a concrete conclusion about the benefit of HAT therapy in sepsis and septic shock patients at this point. In addition, another essential clinical question has arisen in that whether the clinical benefit is from the HAT therapy or from the steroid itself.

Corticosteroids have been used as an adjunct therapy in septic shock patients for a while (Rochwerg et al., 2018). Rochwerg et al. (2018) and Fang et al. (2019) performed two systematic reviews and meta-analyses that primarily focus on the effects of corticosteroids in sepsis patients compared to the standard of care therapy. There was a small reduction in 28-day mortality with the relative risk (RR) of 0.93 and 0.90 from the studies by Rochwerg et al. (2018) and Fang et al. (2019), respectively. After examining all 11 trials, seven studies in this systematic review included patients who received hydrocortisone in both the control group and treatment group after a septic shock. There were 59.6%, 90%, 40.4% and 98.1% of patients who received hydrocortisone in the control arm in the studies by Marik et al. (2017), Kim et al. (2018), Litwak et al. (2019) and Fujii et al. (2020), respectively, while Iglesias et al. (2020), Hwang et al. (2020) and Coloretti et al. (2020) all had 100%. Four out of 11 studies (Coloretti et al., 2020; Iglesias et al., 2020; Kim et al., 2018; Marik et al., 2017) exhibited the benefit in mortality reduction significantly. By examining the control group population, the study by Kim et al. (2018) consisted of more patients who were on hydrocortisone in the control group as compared to the study by Marik et al. (2017): 59.6% and 90%, respectively. Although there were more patients who received hydrocortisone in the control group from the study by Kim et al. (2018) than the study by Marik et al. (2017), there was not much of difference in mortality reduction when comparing these two control groups from both studies. The mortality rate in the control group from Marik et al. (2017) study was 40.4%, while it was 39% in the control group from the study by Kim et al. (2018). By comparing these studies, the monotherapy of hydrocortisone may not be beneficial on mortality in septic shock. Besides that, the study by Kim et al. (2018) detected that the control group and the treatment group showed similar declines in vasopressor requirements, and it was thought that the use of hydrocortisone might help control patients with shock reversal. However, when the researchers performed subgroup analysis, for which they divided the control group into who did receive the steroid and who did not receive the steroid, they found that there was no difference in terms of hospital mortality and superinfection rates. Again, this could reinstate that the combination of HAT could work synergistically in reducing mortality rates in sepsis/septic shock patients.

Although treatment is inexpensive and devoid of side effects/complications at the recommended dose, there were neither statistically or clinically significant differences in mortality, renal replacement therapy, duration of vasopressor, changes in SOFA score, procalcitonin clearance and length of stay outcomes. However, in weighing the risks versus benefits of this treatment, there is no harm in using it despite the lack of a strong recommendation for its use. Without significant adverse effects, HAT therapy could be considered by providers when evaluating patient sepsis severity, while the clinical benefits remain controversial.

At this moment, there are only 11 trials that examined the benefits of combination therapy of HAT in sepsis and septic shock patients; however, the study outcomes were inconsistent, which makes it difficult for clinicians to apply the results in clinical practice. Fortunately, more ongoing trials are focusing on the benefits of HAT therapy in sepsis and septic shock patients. Seven ongoing trials were identified, with four dates anticipated to conclude as follows: NCT03649633 in September 2021 (still recruiting); NCT03913468 in January 2022 (active, not recruiting); NCT03592693 in September 2021 (recruiting); NCT03872011 in September 2021 (recruiting) (Kory, 2019; Mentzelopoulos and Kotanidou, 2018; Mentzelopoulos and Zakynthinos, 2020; Qingquan, 2019). There are two ongoing trials with an unknown status: NCT03540628 in February 2020 (unknown, was recruiting) and NCT03380507 in April 2019 (unknown, was recruiting) (Abdelaty and Mohamed, 2019; Haines, 2018). There is only one study, NCT03335124, that was terminated on 1 December 2018, due to insufficient recruitment (Stefanovic, 2017). The study populations are not identical among all clinical trials; however, their inclusion and exclusion criteria are comparatively similar. Most clinical trials have inclusion criteria of 18 years of age or older, diagnosis of sepsis or septic shock within 12 hours of admission in ICU and all sexes. Exclusion criteria among trials mostly include healthy volunteers, less than 18 years of age, pregnancy, underlying fatal conditions, glucose-6-phosphate-dehydrogenase (G6PD) deficiency, immunocompromised patients and patients who received steroids for other indications other than septic shock, more than 24 hours of septic shock, do not resuscitate (DNR) or do not intubate (DNI) order on admission, terminal end-stage disease and transferred from another hospital. In addition, every study has the same dosage of vitamin C, 1.5 g Q6 H, hydrocortisone, 50 mg Q6 H, and thiamine, 200 mg Q12 H, except for study NCT03872011, which used 2 g of vitamin C for their treatment group instead of the commonly used 1.5 g. As more studies arise on the benefits of HAT therapy for septic shock, a further comprehensive systematic review is needed to be undertaken once the above ongoing clinical trials have been finalized.

Conclusion

In conclusion, the clinical benefits from the combination of HAT for severe sepsis/septic shock cannot be confirmed at this time. The 11 main studies highlighted in this review did not consistently show a significant change in mortality, procalcitonin levels, duration of vasopressor use, renal replacement requirement or SOFA score to indicate infection improvement. The results of ongoing randomized controlled studies testing HAT therapy in various combinations and dosages are required to make definitive conclusions regarding the impact of this novel treatment strategy on the morbidity and mortality of patients with sepsis.

Footnotes

Authors’ contributions

Young R Lee, Kandace Vo and Jincy T Varughese contributed to the design and implementation of the research, to the analysis of the results and to the writing of the manuscript. All authors read and approved the final manuscript.

Availability of data and materials

The authors confirm that the data supporting the findings of this systematic review are available within the manuscript.

Declaration of conflicting interests

The authors have no conflicts of interest to declare.

Ethical approval

Institutional ethics approval was not applicable in this type of systematic review.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Young Ran Lee

References

Abdelaty

Mohamed

(2019) Evaluation of hydrocortisone, vitamin C and thiamine for the treatment of septic shock (HYVITS). U.S. National Library of Medicine. Available at: https://clinicaltrials.gov/ct2/show/NCT03380507?term=NCT03380507&rank=1 (accessed 9 July 2019).

Angus

, et al. (2001 ) Epidemiology of severe sepsis in the United States: Analysis of incidence, outcome, and associated costs of care. Critical Care Medicine 29(7): 1303–1310.

Balakrishnan

, et al. (2018) Hydrocortisone, vitamin C and thiamine for the treatment of sepsis and septic shock following cardiac surgery. Indian Journal of Anesthesia 62(12): 934–939.

Coloretti

, et al. (2020) Adjunctive therapy with vitamin c and thiamine in patients treated with steroids for refractory septic shock: A propensity matched before-after, case-control study. Journal of Critical Care 59: 37–41.

Fang

, et al. (2019) Association of corticosteroid treatment with outcomes in adult patients with sepsis: A systematic review and meta-analysis. The Journal of the American Medical Association 179(2): 213–223.

Fujii

, et al. (2020) Effect of vitamin C, hydrocortisone, and thiamine vs hydrocortisone alone on time alive and free of vasopressor support among patients with septic shock: The VITAMINS randomized clinical trial. JAMA 323(5): 423–431.

Haines

(2018) Thiamine, vitamin C and hydrocortisone in the treatment of septic shock. U.S. National Library of Medicine. Available at: https://clinicaltrials.gov/ct2/show/NCT03540628?term=NCT03540628&rank=1 (accessed 8 July 2019).

Hwang

, et al. (2020) Combination therapy of vitamin C and thiamine for septic shock: a multi-centre, double-blinded randomized, controlled study. Intensive Care Medicine 46(11): 2015–2025.

Iglesias

, et al. (2020) Outcomes of metabolic resuscitation using ascorbic acid, thiamine, and glucocorticoids in the early treatment of sepsis: The ORANGES Trial. Chest 158(1): 164–173.

10.

Kim

, et al. (2018) 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. Journal of Critical Care 47: 211–218.

11.

Kory

(2019) Outcomes of septic shock patients treated with a metabolic resuscitation bundle consisting of intravenous hydrocortisone, ascorbic acid and thiamine. U.S. National Library of Medicine. Available at: https://clinicaltrials.gov/ct2/show/NCT03913468?term=NCT03913468&rank=1 (accessed 11 July 2019).

12.

Litwak

, et al. (2019) Vitamin C, hydrocortisone, and thiamine for the treatment of severe sepsis and septic shock: a retrospective analysis of real-world application. Journal of Clinical Medicine 8(4): 478.

13.

Marik

, et al. (2017) Hydrocortisone, vitamin C, and thiamine for the treatment of severe sepsis and septic shock: A retrospective before-after study. Chest 151(6): 1229–1238.

14.

Mayr

Angus

Yende

(2014) Epidemiology of severe sepsis. Virulence 5: 4–11.

15.

Mentzelopoulos

Kotanidou

(2018) Vitamin C, hydrocortisone, and thiamine for septic shock (CORVICTES). U.S. National Library of Medicine. Available at: https://clinicaltrials.gov/ct2/show/NCT03592693?term=NCT03592693&rank=1 (accessed 8 July 2019).

16.

Mentzelopoulos

Zakynthinos

(2020) Vitamin C, steroids, and thiamine, and cerebral autoregulation and functional outcome in septic shock (CORVICTES-Υμ). U.S. National Library of Medicine. Available at: https://clinicaltrials.gov/ct2/show/NCT03649633?term=NCT03649633&rank=1 (accessed 11 July 2019).

17.

Moher

, et al. (2009) Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLOS Medicine 6(7): e1000097.

18.

Moskowitz

, et al. (2020) Effect of ascorbic acid, corticosteroids, and thiamine on organ injury in septic shock: The ACTS randomized clinical trial. JAMA 324: 642–650.

19.

Pendleton

(2018) Sepsis update. American College of Physicians, Available at: https://www.acponline.org/system/files/documents/about_acp/chapters/mn/18mtg/pendleton_1.pdf (accessed 11 July 2019).

20.

Prescott

, et al. (2018) Reporting of sepsis cases for performance measurement versus for reimbursement in New York state. Crit Care Med 46: 666–673.

21.

Quingpan

Xiaohua

(2019) Vitamin C, thiamine and hydrocortisone for the treatment of septic shock. U.S.: National Library of Medicine. Available at: https://clinicaltrials.gov/ct2/show/NCT03872011 (acccessed 8 July 2019).

22.

Rochwerg

, et al. (2018) Corticosteroids in sepsis: an updated systematic review and meta-analysis. Critical Care Medicine 46(9): 1411–1420.

23.

Sevransky

(2020) The vitamin C, thiamine and steroids in sepsis (VICTAS) protocol: A prospective, multi-center, double-blind, adaptive sample size, randomized, placebo-controlled, clinical trial. U.S. National Library of Medicine. Available at: https://clinicaltrials.gov/ct2/show/NCT03509350?term=NCT03509350&rank=1 (accessed 11 July 2019).

24.

Stefanovic

(2017) The effect of vitamin C, thiamine and hydrocortisone on clinical course and outcome in patients with severe sepsis and septic shock. U.S. National Library of Medicine. Available at: https://clinicaltrials.gov/ct2/show/NCT03335124?term=NCT03335124&rank=1 (accessed 7 July 2019).

25.

The Cochrane Collaboration (2011) Cochrane handbook for systematic reviews of interventions. Available at: http://handbook-5-1.cochrane.org (accessed 7 July 2019).

26.

Vail

, et al. (2020) Use of hydrocortisone, ascorbic acid, and thiamine in adults with septic shock. American Journal of Respiratory and Critical Care Medicine 202: 1531–1539.

27.

World Health Organization (2018) Sepsis. Available at: https://www.who.int/news-room/fact-sheets/detail/sepsis (accessed 11 July 2019).