Introduction: This systematic review aimed to assess the efficacy and safety of hydrocortisone, ascorbic acid, and thiamine (HAT) combination therapy in patients with sepsis and septic shock. Methods: We conducted a database search in MEDLINE, Embase, CENTRAL, Web of Science, and CNKI for randomised controlled trials (RCTs) comparing HAT against placebo/standard of care or against hydrocortisone in sepsis/septic shock patients. Outcomes included mortality, ICU/hospital length of stay (LOS), vasopressor durations, mechanical ventilation durations, change in SOFA at 72 h, and adverse events. RCT results were pooled in random-effects meta-analyses. Quality of evidence was assessed using GRADE. Results: Fifteen RCTs (N = 2,594) were included. At 72 h, HAT reduced SOFA scores from baseline (mean difference [MD] −1.16, 95% confidence interval [CI]: −1.58 to −0.74, I2 = 0%) compared to placebo/SoC, based on moderate quality of evidence. HAT also reduced the duration of vasopressor use (MD −18.80 h, 95% CI: −23.67 to −13.93, I2 = 64%) compared to placebo/SoC, based on moderate quality of evidence. HAT increased hospital LOS (MD 2.05 days, 95% CI: 0.15–3.95, I2 = 57%) compared to placebo/SoC, based on very low quality of evidence. HAT did not increase incidence of adverse events compared to placebo/SoC. Conclusions: HAT appears beneficial in reducing vasopressor use and improving organ function in sepsis/septic shock patients. However, its advantages over hydrocortisone alone remain unclear. Future research should use hydrocortisone comparators and distinguish between sepsis-specific and comorbidity- or care-withdrawal-related mortality.

Sepsis is a life-threatening condition characterised by dysregulated host immune responses to infection [1]. In sepsis patients, the recognition of pathogen-associated molecular patterns and/or host-derived damage-associated molecular patterns trigger excessive upregulation of both pro- and anti-inflammatory pathways, leading to system-wide cytokine release, immune cell recruitment, and coagulation and complement cascade activation [2]. The resultant inflammation eventually leads to multiorgan dysfunctions. A subset of sepsis cases progress to septic shock – a critical condition defined by elevated serum lactate levels and the need for vasopressors to maintain adequate mean arterial pressure [1]. While the global burden of sepsis has declined over the past decade, sepsis mortality rates remain at around 22% and accounts for close to 20% of all global mortalities recorded in 2017 [3]. The mortality rate escalates to nearly 40% in patients with septic shock [4].

Currently, treatment of sepsis and septic shock is guided by the 2021 Surviving Sepsis Campaign. According to these guidelines, physicians typically manage sepsis and septic shock patients with a combination of antimicrobials, fluid resuscitation, vasopressors, and mechanical ventilation when appropriate. Given the pathophysiology of sepsis/septic shock and clinical evidence from previous systematic reviews [5, 6], corticosteroids such as hydrocortisone have also been extensively used in sepsis management to modulate the overactive immune response and provide blood pressure support. Previous meta-analyses showed that the use of hydrocortisone confers a 7% reduction in the risk of mortality and increased rate of shock reversal, but with no changes in ICU and hospital length of stay (LOS) and with higher risks of corticosteroid-related adverse events such as hyperglycaemia, hypernatremia, and neuromuscular weakness, compared to standard of care [7, 8]. Recently, the use of ascorbic acid (vitamin C) and thiamine (vitamin B1) have also been proposed for sepsis management due to the observation that critically ill sepsis patients tend to be deficient in these vitamins [9‒11]. Ascorbic acid, as a potent antioxidant, can scavenge intracellular radicals and protect tight junctions in microcirculations [12]. It also exhibits anti-inflammatory and antithrombotic effects [13], and functions as a cofactor for epinephrine, dopamine, and vasopressin synthesis [14]. Thiamine is a precursor for thiamine pyrophosphate – a coenzyme regulating key metabolic pathways like the Krebs cycle, the citric acid cycle, and the pentose phosphate pathway. Thus, thiamine may protect against energy compromise during physiologic stress [15]. Both ascorbic acid and thiamine monotherapy have demonstrated efficacy in reducing mortality during critical illnesses in clinical trials [16‒18].

When combined, hydrocortisone, ascorbic acid, and thiamine (HAT) may exert synergistic effects in improving sepsis outcomes via multiple overlapping mechanisms [13]. However, while the latest Surviving Sepsis Campaign guideline recommended corticosteroid use in patients with refractory septic shock, they cautioned against the use of ascorbic acid and did not make specific recommendations regarding thiamine and HAT combinations [19]. Several previous meta-analyses have assessed the efficacy and safety of HAT regimen, but they were conducted during similar time periods and included the same set of randomised controlled trials (RCTs) with similar findings of SOFA score improvements at 72 h compared to standard sepsis care/placebo. These previous reviews were not able to identify improvements in other outcomes, such as mortality, due to their low sample sizes [20, 21]. By including recent major RCTs, evaluating additional patient-important outcomes, and adopting the most up-to-date systematic review methodologies, we aim to provide a more comprehensive picture of HAT regimen’s role in sepsis management.

We conducted this systematic review and meta-analysis in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement [22] and the Cochrane Handbook [23]. The completed PRISMA checklist is available as online supplementary Table S1 (for all online suppl. material, see https://doi.org/10.1159/000538959). This review was prospectively registered on PROSPERO (CRD42023459882) [24].

Study Identification

A literature search was performed in the following academic databases from inception till August 19, 2023: (1) Ovid MEDLINE, (2) Ovid Embase, (3) Web of Science (Core Collection), (4) Cochrane Central Register of Controlled Trials, and (5) China National Knowledge Infrastructure (CNKI). We did not apply language restrictions.

The search strategy was initially developed and piloted on Ovid MEDLINE and Ovid Embase using subject headings and key-terms relating to sepsis (e.g., “sepsis,” “septic shock,” “systemic inflammatory response syndrome,” and “bacteremia,” etc.) and corticosteroids (e.g., “steroid,” “glucocorticoid,” “corticosteroid,” and “hydrocortisone,” etc.), as well as a modified version of the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE and Embase [23]. The strategies were then translated and adapted for other databases. The search strategy used for all databases, along with the number of records identified, is tabulated in online supplementary Tables S2–6. In addition to systematic database searches, we hand-searched the reference sections of previous reviews, preprint servers including medRxiv and Research Square, as well as ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform.

Outcome Measures

Our efficacy outcomes included mortality incidence, ICU LOS, hospital LOS, duration on vasopressors, duration on mechanical ventilation, and change in SOFA scores from baseline to 72 h. Our safety outcomes included the incidence of adverse events, incidence of serious adverse events, incidence of hypernatremia, incidence of hyperglycaemia, incidence of gastrointestinal bleeding, and incidence of secondary infections.

Eligibility Criteria

We included RCTs that satisfied the following inclusion criteria: (1) included adult patients (age ≥18) diagnosed with sepsis or septic shock based on author-defined criteria, (2) compared the use of HAT therapy against standard sepsis care with/without placebo, and (3) reported any of our outcomes of interest. The review protocol was amended prior to screening to specifically include hydrocortisone monotherapy as an eligible comparator due to its use in recent RCTs.

Study Selection

Pairs of investigators (J.D., Q.K.Z., K.V., J.H., A.Z., S.B., V.P., and K.H.) performed title and abstract screening in-duplicate using Covidence. Records deemed eligible were subsequently retrieved and entered into an in-duplicate full-text screening process. Disagreements were resolved via discussions to reach a consensus.

Data Extraction

Data extraction was performed in-duplicate by pairs of investigators (J.D., Q.K.Z., S.B., and V.P.) using standardised extraction forms developed a priori. Extracted data items included: (1) study information (author names, publication year, country of origin, trial registration numbers), (2) study design (inclusion and exclusion criteria, definition of sepsis/septic shock, blinding methodologies, randomization, and allocation concealment methods), (3) intervention and comparator descriptions (route, dose, and duration of HAT regimen, descriptions of standard sepsis care and/or placebo), (4) patient baseline characteristics (number of patients randomised, sex distributions, mean/median age, mean/median baseline SOFA scores), and (4) outcome/subgroup data (including a narrative summary of study conclusions). Disagreements in the extracted data were identified and resolved by two senior investigators (J.D. and Q.K.Z).

Outcome data from intention-to-treat analyses or modified intention-to-treat analyses were preferentially extracted over data from per-protocol analyses. Similar to previous systematic reviews [6], we preferentially extracted 28- or 30-day mortality if available, and used ICU mortality, in-hospital mortality, or author-defined mortality if 28- or 30-day mortality was not reported. This approach was used based on the observation that mortality time points do not influence the pooled mortality point estimates in the meta-analysis of critical care RCTs [25].

For studies that reported median and range or interquartile range for continuous outcomes, we first performed tests for skewness using methods proposed by Shi et al. [26]. If the non-parametric data were not significantly skewed, we imputed the mean and standard deviations using methods proposed by Shi et al. [27], Luo et al. [28], and Wan et al. [29] to complete the meta-analyses. Skewed, non-parametric data were excluded from the meta-analyses and only described narratively.

For studies that did not report the change in SOFA score at 72 h, but reported the baseline and 72-h SOFA scores, the change in score and its associated variance was imputed using methods recommended by the Cochrane Handbook [23]. Variance of the imputed change in score was estimated based on correlation coefficients calculated from well-reported studies that reported the baseline score, the 72-h score, and the change in score. In this review, the correlation coefficient used for imputing variance of the HAT intervention arm was 0.29, and the correlation coefficient used for the standard sepsis care/placebo arm was 0.50.

Risk of Bias Assessment

Risk of bias was assessed in-duplicate as a part of the data extraction process using the Revised Cochrane risk of Bias Tool for Assessing Randomised Trials (RoB2) [30]. As we preferentially extracted and analysed intention-to-treat data, we evaluated the risk of bias due to deviations from the intended interventions with the aim to assess the effect of assignment to intervention. Disagreements in the RoB2 ratings were resolved by consultation with two senior investigators (J.D. and Q.K.Z.).

Statistical Analysis

The relative treatment effects and the accompanying variances were first calculated from the extracted outcome data in the form of risk ratios (RRs) and 95% confidence intervals (CIs). The treatment effects were then pooled in random-effects meta-analyses using the meta 6.2-1 package in R 4.1.1 [31]. For meta-analyses involving studies reporting zero events in one or both of their treatment arms, we applied treatment-arm continuity correction to complete the analyses [32]. Results from the meta-analyses were illustrated using forest plots. Studies using standard sepsis care or placebo comparators were analysed separately from studies using hydrocortisone monotherapy as comparators.

Heterogeneity was examined using Cochran’s Q test with a significance level of p < 0.10 and further quantified using I2 statistics. We interpreted an I2 between 30 and 75% as moderate heterogeneity, and an I2 ≥ 75% as serious heterogeneity based on guidance from the Cochrane Handbook [23].

Number needed to treat (NNT) was calculated using the summary effects from meta-analyses of dichotomous outcomes to provide an easily interpretable estimate of the absolute treatment effect [33]. Baseline risk used for the NNT calculation was obtained by averaging the control risk in all included studies for each dichotomous outcome.

Publication Bias

Funnel plots were used to assess the presence of small-study effects as an indication for publication bias. Funnel plot asymmetry was assessed quantitatively using Harbord’s test [34] for dichotomous outcomes and Egger’s test [35] for continuous outcomes. Funnel plots were not drawn for meta-analyses with less than 10 studies. This was done with the consideration that statistical tests for funnel plot asymmetry lack power with a low number of studies [36], and visual inspection of funnel plots alone has been shown to be unreliable for determining the presence of small-study effects [37]. Meta-analyses that exhibited signs of small-study effects were reassessed using a trim-and-fill approach to quantify the number of potentially omitted studies and an estimation of the corrected summary effect [38].

Meta-Regressions and Sensitivity Analyses

Meta-regression analyses were conducted based on mean baseline SOFA scores to assess for correlations between the change in treatment effect and baseline disease severity. This analysis was not done for meta-analyses with less than 10 studies based on guidance from the Cochrane Handbook [39]. Additionally, sensitivity analyses were conducted to assess the impact of including only studies with septic shock patients and the impact of excluding studies with a high risk of bias.

Quality of Evidence

The quality of evidence generated from the review was assessed using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) framework [40]. The presence of imprecision was assessed using a minimally contextualised approach for dichotomous outcomes, using a CI ratio of 3 as the cutoff for downgrading [41]. The optimal information size (OIS) method was used to determine the presence of imprecision in continuous outcomes [42]. OIS was calculated based on equivalence trial methods [43] using a significance level of 0.05 and statistical power of 80%. Pooled standard deviation from the included studies and arbitrary clinically important margins were used for the OIS calculation. A summary of study findings and the associated GRADE ratings are presented in a GRADE summary of findings table generated using GRADEpro GDT [44].

The database search yielded 829 results, of which 648 were screened following deduplication. Thirty-four full-texts were retrieved for further screening, and 15 studies [45‒59] (N = 2,594) were included in the systematic review. A list of studies excluded during the full-text screening process is tabulated in online supplementary Table S7. The study selection process is illustrated as a PRISMA flowchart in Figure 1, and study characteristics are tabulated in Table 1.

Fig. 1.

PRISMA 2020 flowchart for the identification and selection of RCTs.

Fig. 1.

PRISMA 2020 flowchart for the identification and selection of RCTs.

Close modal
Table 1.

Characteristics of included studies and participants

StudyRegistrationDesignSettingInclusion criteriaTreatment armsN randomizedAge, yearsN female (%)Baseline SOFA scoreConclusion
ACTS Trial, [45] (2020) NCT03389555 Quadruple-blinded, randomised, placebo-controlled, parallel-group trial 14 sites in USA Adult patients with septic shock. Septic shock was defined as suspected or confirmed infection and receiving a vasopressor because of sepsis IV hydrocortisone 50 mg q6h for 4 days or until ICU discharge + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 100 mg q6h for 4 days or until ICU discharge + standard sepsis care 103 68.9±15.0 44 (43.6) 9.1±3.5 HAT did not significantly reduce SOFA score during the first 72 h after enrolment. Use of HAT for septic shock management is not recommended 
Saline placebo + standard sepsis care, including antibiotics, fluid resuscitation, vasopressors, and source control 102 67.7±13.9 45 (45.5) 9.2±3.2 
Feng et al. [46] (2023) ChiCTR1900026084 Randomised, parallel-group trial (blinding methods not reported) 1 site in China Adult patients with sepsis or septic shock with PCT >2 ng/mL. Sepsis/septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 7 days + IV ascorbic acid 1,500 mg q6h for 3 days + IM thiamine 200 mg q12h for 3 days + standard sepsis care 68 57.2±14.4 28 (41.2) 9.1±3.2 HAT could shorten ICU LOS, reduce vasopressor durations, and reduce ICU mortality rate. However, it does not reduce the in-hospital mortality rate 
Standard sepsis care, including fluid resuscitation, vasoactive agents, and empiric broad-spectrum antibiotics 68 56.7±15.1 26 (38.2) 8.9±2.6 
Hussein et al. [47] (2021) NCT04508946 Open-label, randomised, parallel-group trial 1 site in Egypt Adult patients with septic shock. Septic shock was defined as requirement for vasopressors to maintain MAP ≥65 mm Hg, serum lactate ≥2 mmol/L, and SOFA organ system score ≥2 IV hydrocortisone 50 mg q6h for 7 days or until ICU discharge, followed by a 3-day taper + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 200 mg q12h for 4 days or until ICU discharge + standard sepsis care 56 65.8±17.0 22 (46.8) 8.8±2.3 HAT significantly reduced shock time and duration on vasopressors but did not significantly reduce 28-day mortality 
IV hydrocortisone 50 mg q6h for 7 days or until ICU discharge, followed by a 3-day taper + standard sepsis care 56 61.6±18.2 21 (44.7) 8.7±2.1 
HYVCTTSSS Trial, [48] (2020) NCT03258684 Single-blinded, randomised, placebo-controlled, parallel-group trial 1 site in China Adult patients with sepsis or septic shock with PCT >2 ng/mL when entering the ICU. Sepsis/septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 7 days or until ICU discharge + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 200 mg q12h for 4 days or until ICU discharge + standard sepsis care 40 59.5±15.0 18 (45.0) 9.6±4.5 HAT did not reduce mortality compared to placebo 
Saline placebo + standard sepsis care, including early initial resuscitation, diagnosis of infection and early antimicrobial therapy, vasopressors, mechanical ventilation, and renal replacement therapy 40 63.7±12.8 19 (47.5) 10.1±4.0 
HYVITS Trial, [51] (2023) NCT03380507 Open-label, randomised, parallel-group trial 2 sites in Qatar Adult patients with septic shock requiring norepinephrine at a dose ≥0.1 μg/kg/min. Septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 7 days or until ICU discharge, followed by a 3-day taper + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 200 mg q12h for 4 days or until ICU discharge + standard sepsis care 53 49.2±15.5 10 (18.9) 10 (8–12) HAT did not reduce in-hospital mortality at 60 days of follow up among patients with septic shock, nor did it reduce vasopressor durations or SOFA scores at 72 h 
Standard sepsis care 53 49.1±16.5 21 (39.6) 10 (8–13) 
Jamshidi et al. [49] (2021) IRCT20150825023760N7 Open-label, randomised, parallel-group trial 1 site in Iran Adult patients with septic shock. Shock was defined as requirement for intubation with MAP <65 mm Hg IV hydrocortisone 50 mg q6h for 3 days + IV ascorbic acid 1,500 mg q6h for 3 days + IV thiamine 200 mg q12h for 3 days + standard sepsis care 29 45.4±19.8 8 (27.6) 4.5±1.4 HAT appeared to be effective at improving clinical outcomes and reducing vasopressor requirements among patients with septic shock, with increased improvement rate of sepsis biomarkers 
Standard sepsis care, including antibiotics, deep vein thrombosis prophylaxis, sedation, mechanical ventilation, and vasopressors 29 45.4±15.8 3 (10.3) 5.3±1.6 
Lyu et al. [50] (2022) NCT03872011 Double-blinded, randomised, placebo-controlled, parallel-group trial 1 site in China Adult patients with septic shock within 12 h of ICU admission. Septic shock was defined according to SEPSIS-3 IV hydrocortisone 200 mg/d for 5 days + IV ascorbic acid 2000 mg q6h for 5 days + IV thiamine 200 mg q12h for 5 days + standard sepsis care 213 69 (60–78) 75 (35.2) 10 (7–12) Early use of HAT did not confer survival benefits to patients with septic shock compared to placebo 
Saline placebo + standard sepsis care, provided according to the Surviving Sepsis Campaign guidelines 213 71 (61–78) 66 (31.0) 9 (7–11) 
ORANGES Trial, [52] (2020) NCT03422159 Double-blinded, randomised, placebo-controlled, parallel-group trial 2 sites in USA Adult patients with sepsis or septic shock within 12 h of ICU admission and are compliant with the 3-h sepsis bundle. Sepsis/septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 4 days or until ICU discharge + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 200 mg q12h for 4 days or until ICU discharge + standard sepsis care 68 70.0±12.0 36 (52.9) 8.3±3.0 HAT significantly reduced time to resolution of shock. More studies are needed to confirm this finding and assess potential mortality benefits from this treatment 
Saline placebo + standard sepsis care 69 67.0±14.0 42 (60.9) 7.9±3.5 
Raghu and Ramalingam, [53] (2021) Open-label, randomised, parallel-group trial India (number of sites not reported) Adult patients with sepsis or septic shock with PCT ≥2 ng/mL (sepsis/septic shock criteria not reported) IV hydrocortisone 200 mg loading dose, followed by IV hydrocortisone 50 mg qid for 7 days + IV ascorbic acid 1,500 mg qid for 4 days + IV thiamine 200 mg bid for 4 days + standard sepsis care 120 52.5±17.4 53 (44.2) 10.0±4.3 Early treatment with HAT may be effective in reducing vasopressor dosage and mortality in patients with sepsis and septic shock 
Standard sepsis care 120 56.3±16.6 37 (30.8) 10.1±4.2 
Sang et al. [54] (2020) Randomised, parallel-group trial (blinding methods not reported) 1 site in China ICU-hospitalized geriatric patients (aged 60–85) with septic shock. Septic shock was defined according to SEPSIS-3 IV hydrocortisone 200 mg/d for 7 days + IV ascorbic acid 1,000 mg tid for 7 days + NG thiamine 50 mg tid for 7 days + standard sepsis care 78 69.0±11.6 38 (48.7) HAT can reduce symptoms of capillary leak syndrome in geriatric septic shock patients 
Standard sepsis care 82 68.0±8.5 42 (51.2) 
VICTAS Trial, [55] (2021) NCT03509350 Double-blinded, randomised, placebo-controlled, parallel-group trial 43 sites in USA Adult patients with sepsis-related acute respiratory and/or cardiovascular dysfunction. Sepsis was defined according to blood cultures and administration of ≥1 antimicrobial agent IV hydrocortisone 50 mg q6h for 4 days or until death/ICU discharge + IV ascorbic acid 1,500 mg q6h for 4 days or until death/ICU discharge + IV thiamine 100 mg q6h for 4 days or until death/ICU discharge + standard sepsis care 252 62 (51–69) 113 (44.8) 9 (7–12) HAT did not increase ventilator- and vasopressor-free days with 30-days of follow up among critically ill patients with sepsis. The trial was terminated early for administrative reasons and thus may have been underpowered to detect a clinically important difference 
Saline placebo + standard sepsis care 249 61 (50–72) 115 (46.2) 9 (6–11) 
ViCTOR Trial, [56] (2020) CTRI/2018/07/014787 Open-label, randomised, parallel-group trial 1 site in India Adult patients with septic shock. Septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 4 days + IV ascorbic acid 1,500 mg q6h for 4 days + IV thiamine 200 mg q12h for 4 days + standard sepsis care 45 58.7±14.9 14 (31.1) 11.2±3.0 HAT did not improve inpatient all-cause mortality among patients with septic shock 
Standard sepsis care 45 59.4±15.0 11 (25.6) 10.9±3.8 
VITAMINS Trial, [57] (2020) NCT03333278 Open-label, randomised, parallel-group trial 10 sites in Australia, New Zealand, and Brazil Adult patients with septic shock. Septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 7 days or until shock resolution/ICU discharge, followed by taper + IV ascorbic acid 1,500 mg q6h for 10 days or until shock resolution/ICU discharge + IV thiamine 200 mg q12h for 10 days or until shock resolution/ICU discharge + standard sepsis care 109 61.9±15.9 39 (36.4) 8.6±2.7 HAT did not significantly improve the duration of time alive and free of vasopressor administration over 7 days compared to hydrocortisone alone in patients with septic shock 
IV hydrocortisone 50 mg q6h for 7 days or until shock resolution/ICU discharge, followed by taper + standard sepsis care 107 61.6±13.9 39 (37.5) 8.4±2.7 
Wang et al. [58] (2023) NCT03821714 Double-blinded, randomised, parallel-group trial 1 site in China Adult patients with septic shock requiring norepinephrine doses ≥15 μg/min. Septic shock was defined as sepsis with persisting hypotension requiring vasopressors to maintain MAP ≥65 mm Hg despite adequate fluid resuscitation IV hydrocortisone 200 mg/d for 1 day + IV ascorbic acid 1,500 mg q6h for 1 day + IV thiamine 200 mg q12h for 1 day + standard sepsis care 15 56.0±20.0 6 (50.0) 8.8±2.9 HAT significantly improved microcirculation in septic shock patients compared to hydrocortisone alone 
IV hydrocortisone 200 mg/d for 1 day + standard sepsis care, including infection source control, fluid resuscitation, vasopressors, and antimicrobials 12 59.0±16.0 1 (10.0) 8.3±2.9 
Wani et al. [59] (2020) CTRI/2018/08/015193 Open-label, randomised, parallel-group trial 1 site in India Adult patients with sepsis/septic shock. Sepsis/septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 7 days or until ICU discharge, followed by a 3-day taper + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 200 mg q12h for 4 days or until ICU discharge + standard sepsis care 50 59 (25–72) 22 (44.0) 9.2±3.5 HAT did not improve 30-day or in-hospital mortality; however, lower vasopressor use and faster lactate clearance was observed with treatment 
Standard sepsis care, including antibiotics, IV fluids, vasopressors, and mechanical ventilation 50 56 (25–72) 19 (38.0) 9.4±3.7 
StudyRegistrationDesignSettingInclusion criteriaTreatment armsN randomizedAge, yearsN female (%)Baseline SOFA scoreConclusion
ACTS Trial, [45] (2020) NCT03389555 Quadruple-blinded, randomised, placebo-controlled, parallel-group trial 14 sites in USA Adult patients with septic shock. Septic shock was defined as suspected or confirmed infection and receiving a vasopressor because of sepsis IV hydrocortisone 50 mg q6h for 4 days or until ICU discharge + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 100 mg q6h for 4 days or until ICU discharge + standard sepsis care 103 68.9±15.0 44 (43.6) 9.1±3.5 HAT did not significantly reduce SOFA score during the first 72 h after enrolment. Use of HAT for septic shock management is not recommended 
Saline placebo + standard sepsis care, including antibiotics, fluid resuscitation, vasopressors, and source control 102 67.7±13.9 45 (45.5) 9.2±3.2 
Feng et al. [46] (2023) ChiCTR1900026084 Randomised, parallel-group trial (blinding methods not reported) 1 site in China Adult patients with sepsis or septic shock with PCT >2 ng/mL. Sepsis/septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 7 days + IV ascorbic acid 1,500 mg q6h for 3 days + IM thiamine 200 mg q12h for 3 days + standard sepsis care 68 57.2±14.4 28 (41.2) 9.1±3.2 HAT could shorten ICU LOS, reduce vasopressor durations, and reduce ICU mortality rate. However, it does not reduce the in-hospital mortality rate 
Standard sepsis care, including fluid resuscitation, vasoactive agents, and empiric broad-spectrum antibiotics 68 56.7±15.1 26 (38.2) 8.9±2.6 
Hussein et al. [47] (2021) NCT04508946 Open-label, randomised, parallel-group trial 1 site in Egypt Adult patients with septic shock. Septic shock was defined as requirement for vasopressors to maintain MAP ≥65 mm Hg, serum lactate ≥2 mmol/L, and SOFA organ system score ≥2 IV hydrocortisone 50 mg q6h for 7 days or until ICU discharge, followed by a 3-day taper + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 200 mg q12h for 4 days or until ICU discharge + standard sepsis care 56 65.8±17.0 22 (46.8) 8.8±2.3 HAT significantly reduced shock time and duration on vasopressors but did not significantly reduce 28-day mortality 
IV hydrocortisone 50 mg q6h for 7 days or until ICU discharge, followed by a 3-day taper + standard sepsis care 56 61.6±18.2 21 (44.7) 8.7±2.1 
HYVCTTSSS Trial, [48] (2020) NCT03258684 Single-blinded, randomised, placebo-controlled, parallel-group trial 1 site in China Adult patients with sepsis or septic shock with PCT >2 ng/mL when entering the ICU. Sepsis/septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 7 days or until ICU discharge + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 200 mg q12h for 4 days or until ICU discharge + standard sepsis care 40 59.5±15.0 18 (45.0) 9.6±4.5 HAT did not reduce mortality compared to placebo 
Saline placebo + standard sepsis care, including early initial resuscitation, diagnosis of infection and early antimicrobial therapy, vasopressors, mechanical ventilation, and renal replacement therapy 40 63.7±12.8 19 (47.5) 10.1±4.0 
HYVITS Trial, [51] (2023) NCT03380507 Open-label, randomised, parallel-group trial 2 sites in Qatar Adult patients with septic shock requiring norepinephrine at a dose ≥0.1 μg/kg/min. Septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 7 days or until ICU discharge, followed by a 3-day taper + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 200 mg q12h for 4 days or until ICU discharge + standard sepsis care 53 49.2±15.5 10 (18.9) 10 (8–12) HAT did not reduce in-hospital mortality at 60 days of follow up among patients with septic shock, nor did it reduce vasopressor durations or SOFA scores at 72 h 
Standard sepsis care 53 49.1±16.5 21 (39.6) 10 (8–13) 
Jamshidi et al. [49] (2021) IRCT20150825023760N7 Open-label, randomised, parallel-group trial 1 site in Iran Adult patients with septic shock. Shock was defined as requirement for intubation with MAP <65 mm Hg IV hydrocortisone 50 mg q6h for 3 days + IV ascorbic acid 1,500 mg q6h for 3 days + IV thiamine 200 mg q12h for 3 days + standard sepsis care 29 45.4±19.8 8 (27.6) 4.5±1.4 HAT appeared to be effective at improving clinical outcomes and reducing vasopressor requirements among patients with septic shock, with increased improvement rate of sepsis biomarkers 
Standard sepsis care, including antibiotics, deep vein thrombosis prophylaxis, sedation, mechanical ventilation, and vasopressors 29 45.4±15.8 3 (10.3) 5.3±1.6 
Lyu et al. [50] (2022) NCT03872011 Double-blinded, randomised, placebo-controlled, parallel-group trial 1 site in China Adult patients with septic shock within 12 h of ICU admission. Septic shock was defined according to SEPSIS-3 IV hydrocortisone 200 mg/d for 5 days + IV ascorbic acid 2000 mg q6h for 5 days + IV thiamine 200 mg q12h for 5 days + standard sepsis care 213 69 (60–78) 75 (35.2) 10 (7–12) Early use of HAT did not confer survival benefits to patients with septic shock compared to placebo 
Saline placebo + standard sepsis care, provided according to the Surviving Sepsis Campaign guidelines 213 71 (61–78) 66 (31.0) 9 (7–11) 
ORANGES Trial, [52] (2020) NCT03422159 Double-blinded, randomised, placebo-controlled, parallel-group trial 2 sites in USA Adult patients with sepsis or septic shock within 12 h of ICU admission and are compliant with the 3-h sepsis bundle. Sepsis/septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 4 days or until ICU discharge + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 200 mg q12h for 4 days or until ICU discharge + standard sepsis care 68 70.0±12.0 36 (52.9) 8.3±3.0 HAT significantly reduced time to resolution of shock. More studies are needed to confirm this finding and assess potential mortality benefits from this treatment 
Saline placebo + standard sepsis care 69 67.0±14.0 42 (60.9) 7.9±3.5 
Raghu and Ramalingam, [53] (2021) Open-label, randomised, parallel-group trial India (number of sites not reported) Adult patients with sepsis or septic shock with PCT ≥2 ng/mL (sepsis/septic shock criteria not reported) IV hydrocortisone 200 mg loading dose, followed by IV hydrocortisone 50 mg qid for 7 days + IV ascorbic acid 1,500 mg qid for 4 days + IV thiamine 200 mg bid for 4 days + standard sepsis care 120 52.5±17.4 53 (44.2) 10.0±4.3 Early treatment with HAT may be effective in reducing vasopressor dosage and mortality in patients with sepsis and septic shock 
Standard sepsis care 120 56.3±16.6 37 (30.8) 10.1±4.2 
Sang et al. [54] (2020) Randomised, parallel-group trial (blinding methods not reported) 1 site in China ICU-hospitalized geriatric patients (aged 60–85) with septic shock. Septic shock was defined according to SEPSIS-3 IV hydrocortisone 200 mg/d for 7 days + IV ascorbic acid 1,000 mg tid for 7 days + NG thiamine 50 mg tid for 7 days + standard sepsis care 78 69.0±11.6 38 (48.7) HAT can reduce symptoms of capillary leak syndrome in geriatric septic shock patients 
Standard sepsis care 82 68.0±8.5 42 (51.2) 
VICTAS Trial, [55] (2021) NCT03509350 Double-blinded, randomised, placebo-controlled, parallel-group trial 43 sites in USA Adult patients with sepsis-related acute respiratory and/or cardiovascular dysfunction. Sepsis was defined according to blood cultures and administration of ≥1 antimicrobial agent IV hydrocortisone 50 mg q6h for 4 days or until death/ICU discharge + IV ascorbic acid 1,500 mg q6h for 4 days or until death/ICU discharge + IV thiamine 100 mg q6h for 4 days or until death/ICU discharge + standard sepsis care 252 62 (51–69) 113 (44.8) 9 (7–12) HAT did not increase ventilator- and vasopressor-free days with 30-days of follow up among critically ill patients with sepsis. The trial was terminated early for administrative reasons and thus may have been underpowered to detect a clinically important difference 
Saline placebo + standard sepsis care 249 61 (50–72) 115 (46.2) 9 (6–11) 
ViCTOR Trial, [56] (2020) CTRI/2018/07/014787 Open-label, randomised, parallel-group trial 1 site in India Adult patients with septic shock. Septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 4 days + IV ascorbic acid 1,500 mg q6h for 4 days + IV thiamine 200 mg q12h for 4 days + standard sepsis care 45 58.7±14.9 14 (31.1) 11.2±3.0 HAT did not improve inpatient all-cause mortality among patients with septic shock 
Standard sepsis care 45 59.4±15.0 11 (25.6) 10.9±3.8 
VITAMINS Trial, [57] (2020) NCT03333278 Open-label, randomised, parallel-group trial 10 sites in Australia, New Zealand, and Brazil Adult patients with septic shock. Septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 7 days or until shock resolution/ICU discharge, followed by taper + IV ascorbic acid 1,500 mg q6h for 10 days or until shock resolution/ICU discharge + IV thiamine 200 mg q12h for 10 days or until shock resolution/ICU discharge + standard sepsis care 109 61.9±15.9 39 (36.4) 8.6±2.7 HAT did not significantly improve the duration of time alive and free of vasopressor administration over 7 days compared to hydrocortisone alone in patients with septic shock 
IV hydrocortisone 50 mg q6h for 7 days or until shock resolution/ICU discharge, followed by taper + standard sepsis care 107 61.6±13.9 39 (37.5) 8.4±2.7 
Wang et al. [58] (2023) NCT03821714 Double-blinded, randomised, parallel-group trial 1 site in China Adult patients with septic shock requiring norepinephrine doses ≥15 μg/min. Septic shock was defined as sepsis with persisting hypotension requiring vasopressors to maintain MAP ≥65 mm Hg despite adequate fluid resuscitation IV hydrocortisone 200 mg/d for 1 day + IV ascorbic acid 1,500 mg q6h for 1 day + IV thiamine 200 mg q12h for 1 day + standard sepsis care 15 56.0±20.0 6 (50.0) 8.8±2.9 HAT significantly improved microcirculation in septic shock patients compared to hydrocortisone alone 
IV hydrocortisone 200 mg/d for 1 day + standard sepsis care, including infection source control, fluid resuscitation, vasopressors, and antimicrobials 12 59.0±16.0 1 (10.0) 8.3±2.9 
Wani et al. [59] (2020) CTRI/2018/08/015193 Open-label, randomised, parallel-group trial 1 site in India Adult patients with sepsis/septic shock. Sepsis/septic shock was defined according to SEPSIS-3 IV hydrocortisone 50 mg q6h for 7 days or until ICU discharge, followed by a 3-day taper + IV ascorbic acid 1,500 mg q6h for 4 days or until ICU discharge + IV thiamine 200 mg q12h for 4 days or until ICU discharge + standard sepsis care 50 59 (25–72) 22 (44.0) 9.2±3.5 HAT did not improve 30-day or in-hospital mortality; however, lower vasopressor use and faster lactate clearance was observed with treatment 
Standard sepsis care, including antibiotics, IV fluids, vasopressors, and mechanical ventilation 50 56 (25–72) 19 (38.0) 9.4±3.7 

Age and baseline SOFA scores are reported as mean ± standard deviation or median (interquartile range) unless otherwise indicated.

PCT, procalcitonin; SOFA, Sequential Organ Failure Assessment; MAP, mean arterial pressure; LOS, length of stay; HAT, hydrocortisone, ascorbic acid, and thiamine combination regimen.

Three RCTs [47, 57, 58] used hydrocortisone as an active comparator group, while the remaining studies used standard sepsis care and/or placebo as the comparator group. Nine RCTs [45, 47, 49‒51, 54, 56‒58] only included patients diagnosed with septic shock, 1 RCT [55] only included sepsis patients without shock, and the remaining RCTs included patients regardless of the presence of shock symptoms. Most included studies defined sepsis and septic shock according to the Third International Consensus Definitions for Sepsis and Septic Shock (SEPSIS-3) or defined according to a similar criteria [1], with the exception of the ACTS trial [45], the study by Wang et al. [58], Jamshidi et al. [49], and Raghu and Ramalingam [53]. Both the ACTS trial and Wang et al. [58] did not use serum lactate measurement as a part of their septic shock definition. Jamshidi et al. [49] defined septic shock as a requirement for intubation and mean arterial pressure <65 mm Hg, while Raghu and Ramalingam did not report their diagnostic criteria for sepsis and septic shock.

The most common HAT regimen assessed was IV hydrocortisone 200 mg/d (divided into 50 mg every 6 h) for 7 days or until ICU discharge + IV ascorbic acid 1,500 mg every 6 h for 4 days or until ICU discharge + IV thiamine 200 mg every 12 h for 4 days or until ICU discharge. Several trials deviated from this regimen, using either lower doses or shorter durations of hydrocortisone, ascorbic acid, and/or thiamine (see Table 1). The most substantial deviation occurred in Wang et al. [58] study, which only used HAT regimen for 24 h.

Risk of Bias

Five studies [47, 51, 53, 54, 58] were rated as having a high risk of bias, with the main concerns being inadequate reporting of allocation concealment methods with imbalanced baseline characteristics, open-label trial designs, and inadequate reporting of deviation from intended interventions. Five studies [46, 49, 52, 56, 59] were rated as having some concerns for risk of bias, mainly due to inadequate reporting of allocation concealment methods without imbalance baseline characteristics. The remaining five studies had a low risk of bias (see Fig. 2 and online suppl. Table S8).

Fig. 2.

Bar plot showing the distribution of risk of bias judgements within each bias domain, as assessed using RoB2.

Fig. 2.

Bar plot showing the distribution of risk of bias judgements within each bias domain, as assessed using RoB2.

Close modal

Incidence of Mortality

The use of HAT regimen was not associated with reduced risk of mortality compared to standard sepsis care and/or placebo (RR 0.81, 95% CI: 0.63–1.06; see Fig. 3a) based on 12 RCTs (N = 2,234) [45, 46, 48‒56, 59]. There was significant and moderate heterogeneity (I2 = 61%, pQ < 0.01). The NNT for an additional beneficial outcome was 17 based on a baseline risk of 0.31.

Fig. 3.

Forest plot illustrating the individual and pooled relative treatment effects from meta-analyses of RCTs for the outcome of mortality. A RR <1 indicates beneficial treatment effects associated with HAT regimen. a Meta-analysis of studies using standard sepsis care/placebo as the comparator arm. b Meta-analysis of studies using hydrocortisone monotherapy as the comparator arm. HAT, hydrocortisone, ascorbic acid, and thiamine; SOC, standard of care; RR, risk ratio; CI, confidence interval; RE, random-effects.

Fig. 3.

Forest plot illustrating the individual and pooled relative treatment effects from meta-analyses of RCTs for the outcome of mortality. A RR <1 indicates beneficial treatment effects associated with HAT regimen. a Meta-analysis of studies using standard sepsis care/placebo as the comparator arm. b Meta-analysis of studies using hydrocortisone monotherapy as the comparator arm. HAT, hydrocortisone, ascorbic acid, and thiamine; SOC, standard of care; RR, risk ratio; CI, confidence interval; RE, random-effects.

Close modal

There was no change in significance in the sensitivity analyses, with reduced summary effect and reduced heterogeneity (see online suppl. Fig. S1, S2). Meta-regression by mean baseline SOFA score did not reveal a significant correlation (p = 0.23; see online suppl. Fig. S3). There was significant asymmetry observed in the funnel plot (p < 0.01; see online suppl. Fig. S4). Trim-and-fill analysis showed 4 potentially omitted studies and a reduced corrected summary effect, but no change in significance of findings (see online suppl. Fig. S5).

HAT regimen was also not associated with a reduced risk of mortality when compared to hydrocortisone monotherapy (RR 0.93, 95% CI: 0.66–1.31; see Fig. 3b) based on 3 RCTs (N = 327) [47, 57, 58] with no heterogeneity (I2 = 0%, pQ = 0.67). All patients included in the analysis were septic shock patients. The NNT for an additional beneficial outcome was 41 based on a baseline risk of 0.35. Only one RCT, the VITAMINS trial [57], had a low or moderate risk of bias; it did not associate HAT with a reduced risk of mortality compared to hydrocortisone monotherapy (RR 1.11, 95% CI: 0.66–1.87).

ICU and Hospital LOS

The use of HAT regimen was associated with increased hospital LOS by 2 days compared to standard sepsis care and/or placebo (mean difference [MD] 2.05 days, 95% CI: 0.15–3.95; see Fig. 4a) based on 4 RCTs (N = 565) [52, 53, 56, 59]. There was significant and moderate heterogeneity (I2 = 57%, pQ = 0.07). Only the ViCTOR trial [56] was restricted to septic shock patients, which found a larger increase in hospital LOS associated with HAT compared to the original meta-analysis (MD 10.68 days, 95% CI: 0.56–20.80). HAT was no longer associated with increased hospital LOS following the exclusion of studies with a high risk of bias (MD 1.02 days, 95% CI: −0.66 to 2.69; see online suppl. Fig. S6). Three RCTs [50, 51, 55] comparing HAT against standard sepsis care/placebo reported non-parametric hospital LOS data that could not be imputed. None of the studies associated use of HAT with changes in hospital LOS (see online suppl. Table S9).

Fig. 4.

Forest plot illustrating the individual and pooled relative treatment effects from meta-analysis of RCTs for the outcome of hospital and ICU LOS. A MD <0 indicates beneficial treatment effects associated with HAT regimen. a Meta-analysis for hospital LOS using standard sepsis care/placebo as the comparator arm. b Meta-analysis for hospital LOS using hydrocortisone monotherapy as the comparator arm. c Meta-analysis for ICU LOS using standard sepsis care/placebo as the comparator arm. d Meta-analysis for ICU LOS using hydrocortisone monotherapy as the comparator arm. LOS, length of stay; HAT, hydrocortisone, ascorbic acid, and thiamine; SOC, standard of care; MD, mean difference; CI, confidence interval; RE, random-effects; SD, standard deviation.

Fig. 4.

Forest plot illustrating the individual and pooled relative treatment effects from meta-analysis of RCTs for the outcome of hospital and ICU LOS. A MD <0 indicates beneficial treatment effects associated with HAT regimen. a Meta-analysis for hospital LOS using standard sepsis care/placebo as the comparator arm. b Meta-analysis for hospital LOS using hydrocortisone monotherapy as the comparator arm. c Meta-analysis for ICU LOS using standard sepsis care/placebo as the comparator arm. d Meta-analysis for ICU LOS using hydrocortisone monotherapy as the comparator arm. LOS, length of stay; HAT, hydrocortisone, ascorbic acid, and thiamine; SOC, standard of care; MD, mean difference; CI, confidence interval; RE, random-effects; SD, standard deviation.

Close modal

HAT regimen was not associated with changes in ICU LOS compared to standard sepsis care and/or placebo (MD −0.75 days, 95% CI: −2.65 to 1.16; see Fig. 4c) based on 7 RCTs (N = 947) [46, 48, 51‒54, 56]. There was significant and serious heterogeneity (I2 = 86%, pQ < 0.01). There was no change in significance in the sensitivity analyses (see online suppl. Fig. S7, S8). Two RCTs [50, 55] comparing HAT against placebo reported non-parametric ICU LOS data that could not be imputed. None of the studies associated the use of HAT with changes in ICU LOS (see online suppl. Table S10).

HAT regimen was not associated with changes in hospital LOS compared to hydrocortisone monotherapy (MD −1.77 days, 95% CI: −3.61 to 0.07; see Fig. 4b) based on 2 RCTs (N = 116) [47, 58] with no heterogeneity (I2 = 0%, pQ = 0.81). Both RCTs had a high risk of bias and only enrolled patients with septic shock. Meta-analysis of the same 2 RCTs also did not associate HAT regimen with changes in ICU LOS compared to hydrocortisone monotherapy (MD −1.55 days, 95% CI: −3.19 to 0.09; see Fig. 4d) with no heterogeneity (I2 = 0%, pQ = 0.62). One RCT [57] compared HAT against hydrocortisone reported non-parametric hospital LOS data that could not be imputed, and it did not associate the use of HAT with changes in LOS (see online suppl. Table S9).

Duration of Vasopressor Use

The use of HAT regimen was associated with decreased duration of vasopressor use by 19 h compared to standard sepsis care and/or placebo (MD −18.80 h, 95% CI: −23.67 to −13.93; see Fig. 5a) based on 8 RCTs (N = 1,025) [46, 49, 51‒54, 56, 59]. There was significant and moderate heterogeneity (I2 = 64%, pQ < 0.01). This association remained in sensitivity analyses (see online suppl. Fig. S9, S10). Two RCTs [48, 51] compared HAT against standard sepsis care/placebo reported non-parametric vasopressor duration data that could not be imputed, and they did not associate the use of HAT with changes in vasopressor durations (see online suppl. Table S11). There were no RCTs that used hydrocortisone monotherapy as the comparator arm and reported the duration of vasopressor therapy.

Fig. 5.

Forest plot illustrating the individual and pooled relative treatment effects from meta-analysis of RCTs for other continuous outcomes. A MD <0 indicates beneficial treatment effects associated with HAT regimen. a Meta-analysis for duration on vasopressors using standard sepsis care/placebo as the comparator arm. b Meta-analysis for duration on mechanical ventilation using standard sepsis care/placebo as the comparator arm. c Meta-analysis for change in 72-h SOFA score from baseline using standard sepsis care/placebo as the comparator arm. MV, mechanical ventilation; HAT, hydrocortisone, ascorbic acid, and thiamine; SOC, standard of care; MD, mean difference; CI, confidence interval; RE, random-effects; SD, standard deviation.

Fig. 5.

Forest plot illustrating the individual and pooled relative treatment effects from meta-analysis of RCTs for other continuous outcomes. A MD <0 indicates beneficial treatment effects associated with HAT regimen. a Meta-analysis for duration on vasopressors using standard sepsis care/placebo as the comparator arm. b Meta-analysis for duration on mechanical ventilation using standard sepsis care/placebo as the comparator arm. c Meta-analysis for change in 72-h SOFA score from baseline using standard sepsis care/placebo as the comparator arm. MV, mechanical ventilation; HAT, hydrocortisone, ascorbic acid, and thiamine; SOC, standard of care; MD, mean difference; CI, confidence interval; RE, random-effects; SD, standard deviation.

Close modal

Duration of Mechanical Ventilation

The use of HAT regimen was not associated with changes in the duration of mechanical ventilation compared to standard sepsis care and/or placebo (MD −1.82 days, 95% CI: −5.06–1.41; see Fig. 5b) based on 4 RCTs (N = 586) [48, 51, 53, 54]. There was significant and serious heterogeneity (I2 = 95%, pQ < 0.01). There was no change in significance in the sensitivity analyses restricted to septic shock patients (see online suppl. Fig. S11). All analysed studies, except the HYVCTTSSS trial [48], were rated as having a high risk of bias. The HYVCTTSSS trial did not associate the use of HAT regimen with decreased duration of mechanical ventilation (MD 1.16 days, 95% CI: −1.28 to 3.59). Only 1 RCT (N = 94) [47] used hydrocortisone monotherapy as the comparator arm, and it did not associate the use of HAT regimen with decreased duration of mechanical ventilation (MD 0.01 days, 95% CI: −1.46 to 1.49).

Change in SOFA Scores at 72 h

The use of HAT regimen was associated with increased reduction of 72-h SOFA score from baseline compared to standard sepsis care and/or placebo (MD −1.16, 95% CI: −1.58 to −0.74; see Fig. 5c) based on 9 RCTs (N = 1,145) [45, 46, 48, 49, 51‒53, 56, 59]. There was no heterogeneity (I2 = 0%, pQ = 0.44). This association remained in sensitivity analyses (see online suppl. Fig. S12, 13). Two RCTs [50, 55] compared HAT against placebo and reported non-imputable SOFA score data. None of these studies associated HAT with change in SOFA score change from baseline (see online suppl. Table S12).

Only one RCT [57] using hydrocortisone as a comparator reported change in SOFA scores at 72 h. The study reported non-parametric data that could not be imputed (see online suppl. Table S12). It associated the use of HAT with increased reduction in 72-h SOFA score from baseline compared to hydrocortisone monotherapy (difference based on the Hodges-Lehmann estimate −1.0, 95% CI: −1.9 to −0.1).

Safety Outcomes

The VITAMINS trial, which compared HAT regimen against hydrocortisone monotherapy, reported 2 incidences of adverse events in the HAT arm compared to 1 in the hydrocortisone arm (RR 1.94, 95% CI: 0.18–21.11, NNT for an additional harmful outcome was 110 based on a baseline risk of 0.01) [57]. There were no serious adverse events in the trial. The VICTAS trial, which compared HAT regimen against placebo, reported no serious adverse events in the trial [55]. None of the remaining RCTs reported information on the overall incidence of adverse events and/or the incidence of serious adverse events.

In terms of HAT-associated adverse events, the use of HAT regimen was not associated with increased risk of hyperglycaemia, hypernatremia, gastrointestinal bleeding, nor secondary infections, compared to standard sepsis care/placebo (see Fig. 6), based on 4 RCTs [45, 48, 50, 51]. These findings remained consistent in sensitivity analyses. The NNT for an additional harmful outcome was 4 for hyperglycaemia (from baseline risk of 0.04), 19 for hypernatremia (from baseline risk of 0.07), 44 for gastrointestinal bleeding (from baseline risk of 0.03), and 202 for secondary infections (from baseline risk of 0.05).

Fig. 6.

Forest plot illustrating the individual and pooled relative treatment effects from meta-analysis of RCTs for safety outcomes. A RR <1 indicates beneficial treatment effects associated with HAT regimen. a Meta-analysis for hyperglycaemia incidence using standard sepsis care/placebo as the comparator arm. b Meta-analysis for hypernatremia incidence using standard sepsis care/placebo as the comparator arm. c Meta-analysis for gastrointestinal bleeding incidence using standard sepsis care/placebo as the comparator arm. d Meta-analysis for secondary infection incidence using standard sepsis care/placebo as the comparator arm. GI, gastrointestinal; HAT, hydrocortisone, ascorbic acid, and thiamine; SOC, standard of care; RR, risk ratio; CI, confidence interval; RE, random-effects.

Fig. 6.

Forest plot illustrating the individual and pooled relative treatment effects from meta-analysis of RCTs for safety outcomes. A RR <1 indicates beneficial treatment effects associated with HAT regimen. a Meta-analysis for hyperglycaemia incidence using standard sepsis care/placebo as the comparator arm. b Meta-analysis for hypernatremia incidence using standard sepsis care/placebo as the comparator arm. c Meta-analysis for gastrointestinal bleeding incidence using standard sepsis care/placebo as the comparator arm. d Meta-analysis for secondary infection incidence using standard sepsis care/placebo as the comparator arm. GI, gastrointestinal; HAT, hydrocortisone, ascorbic acid, and thiamine; SOC, standard of care; RR, risk ratio; CI, confidence interval; RE, random-effects.

Close modal

For comparison with hydrocortisone monotherapy, the VITAMINS trial [57] reported 1 incidence of hyperglycaemia in the HAT group compared to 0 in the hydrocortisone group (RR 2.97, 95% CI: 0.12–73.25, NNT not calculated as baseline risk was 0), and 1 incidence of gastrointestinal bleeding in the hydrocortisone group compared to 0 in the HAT group (RR 0.33, 95% CI: 0.01–7.89, NNT for an additional beneficial outcome was 155 based on a baseline risk of 0.01).

Quality of Evidence

A summary of findings from this review, along with the associated GRADE quality of evidence ratings, is tabulated in Table 2.

Table 2.

GRADE summary of findings table

OutcomesRelative effect (95% CI)Anticipated absolute effects (95% CI)aPatients, n (N studies)Quality of evidence (GRADE)
risk without HATrisk with HATrisk difference (95% CI)
Incidence of mortality 
 Standard sepsis care/placebo comparator RR 0.81 (0.63–1.06) 281 per 1,000 228 per 1,000 (177–298) 53 fewer per 1,000 (104 fewer to 17 more) 2,234 (12 RCTs) ⨁⨁◯◯ Lowb,c 
 Hydrocortisone comparator RR 0.93 (0.66–1.31) 286 per 1,000 266 per 1,000 (189–375) 20 fewer per 1,000 (97 fewer to 89 more) 327 (3 RCTs) ⨁⨁◯◯ Lowd 
Hospital LOS 
 Standard sepsis care/placebo comparator The mean hospital LOS was 12.15 days The anticipated mean hospital LOS is 14.20 days (12.30–16.10) MD 2.05 more days (0.15 more to 3.95 more) 565 (4 RCTs) ⨁◯◯◯ Very lowb,e,f 
 Hydrocortisone comparator The mean hospital LOS was 12.06 days The anticipated mean hospital LOS is 10.29 days (8.45–12.13) MD 1.77 fewer days (3.61 fewer to 0.07 more) 116 (2 RCTs) ⨁◯◯◯ Very lowd,e,f 
ICU LOS 
 Standard sepsis care/placebo comparator The mean ICU LOS was 9.60 days The anticipated mean ICU LOS is 8.85 days (6.95–10.76) MD 0.75 fewer days (2.65 fewer to 1.16 more) 947 (7 RCTs) ⨁◯◯◯ Very lowf,g,h 
 Hydrocortisone comparator The mean ICU LOS was 10.07 days The anticipated mean ICU LOS is 8.52 days (6.88–10.16) MD 1.55 fewer days (3.19 fewer to 0.09 more) 116 (2 RCTs) ⨁◯◯◯ Very lowd,f,g 
Duration on vasopressors 
 Standard sepsis care/placebo comparator The mean duration on vasopressors was 63.87 h The anticipated mean duration on vasopressors is 45.07 h (40.20–77.80) MD 18.80 fewer hours (23.67 fewer to 13.93 more) 1,025 (8 RCTs) ⨁⨁⨁◯ Moderateb,i 
Duration on mechanical ventilation 
 Standard sepsis care/placebo comparator The mean duration on mechanical ventilation was 8.97 days The anticipated mean duration on mechanical ventilation is 7.15 days (3.91–10.38) MD 1.82 fewer days (5.06 fewer to 1.41 more) 586 (4 RCTs) ⨁◯◯◯ Very lowd,f,h,j 
 Hydrocortisone comparator The mean duration on mechanical ventilation was 5.38 days The anticipated mean duration on mechanical ventilation is 5.39 days (3.92–6.87) MD 0.01 more days (1.46 fewer to 1.49 more) 94 (1 RCT) ⨁◯◯◯ Very lowd,f,j 
Change in 72-h SOFA score from baseline 
 Standard sepsis care/placebo comparator The mean change in 72-h SOFA score from baseline was −1.89 The anticipated mean change in 72-h SOFA score from baseline is –3.05 (–3.47 to −2.63) MD 1.16 fewer (1.58 fewer to 0.74 fewer) 1,145 (9 RCTs) ⨁⨁⨁◯ Moderatek,l 
Incidence of adverse events 
 Hydrocortisone comparator RR 1.94 (0.18–21.11) 10 per 1,000 19 per 1,000 (2–203) 9 more per 1,000 (8 fewer to 193 more) 211 (1 RCT) ⨁⨁◯◯ Lowm 
Incidence of hyperglycaemia 
 Standard sepsis care/placebo comparator RR 7.41 (0.25–217.50) 22 per 1,000 166 per 1,000 (5–1,000) 144 more per 1,000 (17 fewer to 1,000 more) 626 (2 RCTs) ⨁◯◯◯ Very lowh,m 
 Hydrocortisone comparator RR 2.97 (0.12–73.25) 0 per 1,000 211 (1 RCT) ⨁⨁◯◯ Lowm 
Incidence of hypernatremia 
 Standard sepsis care/placebo comparator RR 1.70 (0.78–3.70) 52 per 1,000 88 per 1,000 (41–192) 36 more per 1,000 (11 fewer to 140 more) 812 (4 RCTs) ⨁◯◯◯ Very lowb,m 
Incidence of gastrointestinal bleeding 
 Standard sepsis care/placebo comparator RR 1.66 (0.41–6.72) 32 per 1,000 53 per 1,000 (13–217) 21 more per 1,000 (19 fewer to 185 more) 186 (2 RCTs) ⨁◯◯◯ Very lown,m 
 Hydrocortisone comparator RR 0.33 (0.01–7.89) 10 per 1,000 4 per 1,000 (0–76) 6 fewer per 1,000 (10 fewer to 66 more) 211 (1 RCT) ⨁⨁◯◯ Lowm 
Incidence of secondary infections 
 Standard sepsis care/placebo comparator RR 1.09 (0.55–2.17) 73 per 1,000 80 per 1,000 (40–158) 7 more per 1,000 (33 fewer to 85 more) 386 (3 RCTs) ⨁⨁◯◯ Lowm 
OutcomesRelative effect (95% CI)Anticipated absolute effects (95% CI)aPatients, n (N studies)Quality of evidence (GRADE)
risk without HATrisk with HATrisk difference (95% CI)
Incidence of mortality 
 Standard sepsis care/placebo comparator RR 0.81 (0.63–1.06) 281 per 1,000 228 per 1,000 (177–298) 53 fewer per 1,000 (104 fewer to 17 more) 2,234 (12 RCTs) ⨁⨁◯◯ Lowb,c 
 Hydrocortisone comparator RR 0.93 (0.66–1.31) 286 per 1,000 266 per 1,000 (189–375) 20 fewer per 1,000 (97 fewer to 89 more) 327 (3 RCTs) ⨁⨁◯◯ Lowd 
Hospital LOS 
 Standard sepsis care/placebo comparator The mean hospital LOS was 12.15 days The anticipated mean hospital LOS is 14.20 days (12.30–16.10) MD 2.05 more days (0.15 more to 3.95 more) 565 (4 RCTs) ⨁◯◯◯ Very lowb,e,f 
 Hydrocortisone comparator The mean hospital LOS was 12.06 days The anticipated mean hospital LOS is 10.29 days (8.45–12.13) MD 1.77 fewer days (3.61 fewer to 0.07 more) 116 (2 RCTs) ⨁◯◯◯ Very lowd,e,f 
ICU LOS 
 Standard sepsis care/placebo comparator The mean ICU LOS was 9.60 days The anticipated mean ICU LOS is 8.85 days (6.95–10.76) MD 0.75 fewer days (2.65 fewer to 1.16 more) 947 (7 RCTs) ⨁◯◯◯ Very lowf,g,h 
 Hydrocortisone comparator The mean ICU LOS was 10.07 days The anticipated mean ICU LOS is 8.52 days (6.88–10.16) MD 1.55 fewer days (3.19 fewer to 0.09 more) 116 (2 RCTs) ⨁◯◯◯ Very lowd,f,g 
Duration on vasopressors 
 Standard sepsis care/placebo comparator The mean duration on vasopressors was 63.87 h The anticipated mean duration on vasopressors is 45.07 h (40.20–77.80) MD 18.80 fewer hours (23.67 fewer to 13.93 more) 1,025 (8 RCTs) ⨁⨁⨁◯ Moderateb,i 
Duration on mechanical ventilation 
 Standard sepsis care/placebo comparator The mean duration on mechanical ventilation was 8.97 days The anticipated mean duration on mechanical ventilation is 7.15 days (3.91–10.38) MD 1.82 fewer days (5.06 fewer to 1.41 more) 586 (4 RCTs) ⨁◯◯◯ Very lowd,f,h,j 
 Hydrocortisone comparator The mean duration on mechanical ventilation was 5.38 days The anticipated mean duration on mechanical ventilation is 5.39 days (3.92–6.87) MD 0.01 more days (1.46 fewer to 1.49 more) 94 (1 RCT) ⨁◯◯◯ Very lowd,f,j 
Change in 72-h SOFA score from baseline 
 Standard sepsis care/placebo comparator The mean change in 72-h SOFA score from baseline was −1.89 The anticipated mean change in 72-h SOFA score from baseline is –3.05 (–3.47 to −2.63) MD 1.16 fewer (1.58 fewer to 0.74 fewer) 1,145 (9 RCTs) ⨁⨁⨁◯ Moderatek,l 
Incidence of adverse events 
 Hydrocortisone comparator RR 1.94 (0.18–21.11) 10 per 1,000 19 per 1,000 (2–203) 9 more per 1,000 (8 fewer to 193 more) 211 (1 RCT) ⨁⨁◯◯ Lowm 
Incidence of hyperglycaemia 
 Standard sepsis care/placebo comparator RR 7.41 (0.25–217.50) 22 per 1,000 166 per 1,000 (5–1,000) 144 more per 1,000 (17 fewer to 1,000 more) 626 (2 RCTs) ⨁◯◯◯ Very lowh,m 
 Hydrocortisone comparator RR 2.97 (0.12–73.25) 0 per 1,000 211 (1 RCT) ⨁⨁◯◯ Lowm 
Incidence of hypernatremia 
 Standard sepsis care/placebo comparator RR 1.70 (0.78–3.70) 52 per 1,000 88 per 1,000 (41–192) 36 more per 1,000 (11 fewer to 140 more) 812 (4 RCTs) ⨁◯◯◯ Very lowb,m 
Incidence of gastrointestinal bleeding 
 Standard sepsis care/placebo comparator RR 1.66 (0.41–6.72) 32 per 1,000 53 per 1,000 (13–217) 21 more per 1,000 (19 fewer to 185 more) 186 (2 RCTs) ⨁◯◯◯ Very lown,m 
 Hydrocortisone comparator RR 0.33 (0.01–7.89) 10 per 1,000 4 per 1,000 (0–76) 6 fewer per 1,000 (10 fewer to 66 more) 211 (1 RCT) ⨁⨁◯◯ Lowm 
Incidence of secondary infections 
 Standard sepsis care/placebo comparator RR 1.09 (0.55–2.17) 73 per 1,000 80 per 1,000 (40–158) 7 more per 1,000 (33 fewer to 85 more) 386 (3 RCTs) ⨁⨁◯◯ Lowm 

GRADE Working Group quality of evidence rating [40].

High quality: We are very confident that the true effect lies close to that of the estimate of the effect.

Moderate quality: We are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

Low quality: Our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect.

Very low quality: We have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

CI, confidence interval; RR, risk ratio; RCT, randomized controlled trial; LOS, length of stay; MD, mean difference; OIS, optimal information size.

aThe absolute risk in the intervention group (and its 95% confidence interval) is based on the baseline risk in the control group and the relative effect of the intervention (and its 95% CI). The baseline risk in the control group is calculated using GRADEpro GDT, by dividing the total control group sample size by the total incidence in the control group. This is different from the methods used for NNT calculations, which is the average control risk in the included studies for each outcome.

bDowngraded by one level due to inconsistency of results; I2 statistics indicate the possible presence of moderate heterogeneity.

cDowngraded by one level due to suspected publication bias; Harbord’s test was significant, indicating the presence of small-study effects as an indication for publication bias.

dDowngraded by two levels due to limitations in study design or execution (risk of bias); high risk of bias studies account for ≥75% of included studies.

eAn arbitrary clinically important margin of 48 h (2 days) was used for the OIS calculation for hospital LOS. The final OIS for imprecision assessment was 538 patients per arm for comparison against standard sepsis care/placebo, and 154 patients per arm for comparison against hydrocortisone monotherapy.

fDowngraded by two levels due to imprecision; the total sample size of the included studies was lower than the calculated OIS.

gAn arbitrary clinically important margin of 24 h (1 day) was used for the OIS calculation for ICU LOS. The final OIS for imprecision assessment was 629 patients per arm for comparison against standard sepsis care/placebo, and 471 patients per arm for comparison against hydrocortisone monotherapy.

hDowngraded by two levels due to inconsistency of results; I2 statistics indicate the possible presence of serious heterogeneity.

iAn arbitrary clinically important margin of 12 h was used for the OIS calculation for duration on vasopressors. The final OIS for imprecision assessment was 89 patients per arm.

jAn arbitrary clinically important margin of 12 h was used for the OIS calculation for duration on mechanical ventilation. The final OIS for imprecision assessment was 1,897 patients per arm for comparison against standard sepsis care/placebo, and 834 patients per arm for comparison against hydrocortisone monotherapy.

kAn arbitrary was used for the OIS calculation for change in 72-h SOFA score from baseline. The final OIS for imprecision assessment was 228 patients per arm.

lDowngraded by one level due to indirectness; SOFA score serves as a surrogate outcome for predicting the risk of in-hospital mortality in patients with sepsis, and is not itself a patient-important outcome.

mDowngraded by two levels due to imprecision; the 95% CI of the relative effect crosses null, and the ratio of the upper and lower bounds of the 95% CI is greater than 3.

nDowngraded by one level due to limitations in study design or execution (risk of bias); high risk of bias studies account for ≥50% of included studies.

The current systematic review and meta-analysis associated the use of HAT regimen with reduced SOFA score at 72 h from baseline compared to standard sepsis care/placebo, based on moderate quality of evidence. This is similar to findings from previous reviews. However, using a larger set of RCT evidence, we were able to associate the use of HAT with a lower duration of vasopressor use and a 2 day increase in hospital LOS compared to standard sepsis care/placebo, based on moderate and very low quality of evidence, respectively. These are novel observations that were not identified in previous systematic reviews.

Interpretation of Findings

Haemodynamic optimisation remains a cornerstone in the management of sepsis and septic shock [60]. Findings from the current review indicate that HAT regimen may help attain haemodynamic stability, allowing patients to be weaned off of vasopressors earlier. This reduces the risk of adverse events associated with prolonged vasopressor use [61, 62] as well as costs associated with prolonged intensive care management. However, it is notable that corticosteroids have been previously associated with reduced vasopressor durations without the addition of ascorbic acid and thiamine [6]. Attributing additional benefits to the adjuvant use of ascorbic acid and thiamine would require RCT investigations using hydrocortisone as a comparator arm.

Unfortunately, a lack of comparison to hydrocortisone monotherapy is a major limitation in the current evidence base, as a majority of our included studies used placebo and/or standard sepsis care as the control arm. The current review only included 3 RCTs using hydrocortisone controls, and while meta-analyses of these RCTs showed potentially beneficial signals in reducing hospital and ICU LOS, the data are too imprecise to draw any meaningful conclusions. Thus, it is currently unknown whether the adjuvant vitamin therapies can lead to additional patient benefits compared to hydrocortisone alone. This critical gap in evidence would need to be addressed in additional trials using corticosteroid comparators.

An interesting observation in the current review is the increased hospital LOS associated with HAT use, which is counterintuitive to the beneficial findings of SOFA improvements and reduced vasopressor durations. There are several potential explanations for this finding. Firstly, corticosteroid use is known to increase the risk of adverse events such as hyperglycaemia, hypernatremia, gastrointestinal bleeding, and secondary infections, which have been associated with worsened patient prognosis and increased hospital LOS [63, 64]. While our meta-analyses had limited precision to definitively assess the incidence of HAT-associated adverse events, previous reviews have associated corticosteroid use with these adverse events in critically ill patients [5, 6, 65]. The conduct of the included RCTs could have also played a major role. Three of the four studies included in the analysis used open-label trial designs, which could have influenced the investigators to delay patient discharge to monitor for adverse events or to reduce the risk of undesirable outcomes post-discharge. In the current review, the association between hospital LOS and HAT use was no longer significant after excluding an open-label study with a high risk of bias.

Lastly, the current review reaffirms previous findings that while HAT regimen can significantly improve SOFA scores at 72 h, it does not reduce the risk of mortality. The use of all-cause mortality as endpoints in sepsis trials has been controversial as most sepsis-associated mortality results from exacerbation of underlying comorbidities and/or pre-specified, conservative goals of care leading to treatment withdrawal [66, 67]. As such, the improvements on SOFA and reduction of vasopressor durations observed did not translate to reduced mortality risks. Future investigations in both HAT regimen and sepsis settings in general should report information relating to withdrawal of care and cause of mortality [68, 69]. Nevertheless, refractory shock remains a major cause of sepsis-associated mortality in addition to comorbidity and withdrawal of care. It is possible that mortality benefits associated with improved organ functions and haemodynamic stability would become observable with larger sample sizes to overcome the weight of patients with comorbidities and/or conservative care plans [66].

Review Limitations

The current review is limited by the small number of published RCTs comparing HAT against hydrocortisone, making it difficult to assess whether beneficial effects observed in the review is attributable to the addition of ascorbic acid and thiamine or only to hydrocortisone. Additionally, the reporting of adverse events and serious adverse events was poor in the current literature; as such, the safety and tolerability of HAT regimen could not be accurately assessed due to imprecision. Lastly, a majority of the outcomes in this review were assessed based on very low quality of evidence according to GRADE. This is mainly due to poor RCT reporting and/or conduct (resulting in high risk of bias), contradictory treatment effects (resulting in high heterogeneity), low sample sizes, and the presence of suspected publication bias (in the outcome of mortality). These low quality findings are not suitable for guiding clinical practice and should be confirmed in future investigations.

In this systematic review, we found that HAT regimen may be useful for reducing the duration of vasopressor use and improve organ functions in patients with sepsis and septic shock, although it was difficult to ascertain whether the addition of ascorbic acid and thiamine lead to more benefits compared to corticosteroids alone. HAT regimen was also associated with increased LOS, which could be attributable to corticosteroid-associated side effects and/or the open-label designs of the included RCTs. Future trials should continue to assess HAT regimen compared to corticosteroid comparators, and to differentiate between sepsis-specific mortality versus mortality from comorbidities and withdrawal of care.

An ethics statement is not applicable because only aggregate patient data from published literature is used in this study.

The authors declare no conflicts of interest.

This review and its authors received no funding support from any funding agency in the public, commercial, or not-for-profit sectors.

J.D. conceived the study, designed the review methodology, supervised, and participated in article screening and data extraction, performed statistical analyses, produced all figures, tables, and supplementary materials, drafted the original manuscript, and conducted critical revisions to the manuscript. Q.K.Z. conceived the study, supervised, and participated in article screening and data extraction, and critically revised the manuscript. K.V., J.H., A.Z., S.B., V.P., and K.H. participated in article screening and critically revised the manuscript. M.M. performed statistical analyses and critically revised the manuscript. All authors gave approval for the manuscript to be submitted for publication and agree to be held accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Additional Information

Preliminary results from this review were presented at the 2024 American Thoracic Society (ATS) International Conference in San Diego, California.Jiawen Deng and Qi Kang Zuo contributed equally to the completion of this review and both serve as co-first authors.Edited by: H.-U. Simon, Bern.

All relevant data are disclosed in the manuscript, its associated figures, and the supplementary materials. No additional data are available. Further enquiries can be directed to the corresponding author.

1.
Singer
M
,
Deutschman
CS
,
Seymour
CW
,
Shankar-Hari
M
,
Annane
D
,
Bauer
M
, et al
.
The Third international consensus definitions for sepsis and septic shock (Sepsis-3)
.
JAMA
.
2016
;
315
(
8
):
801
10
.
2.
Chousterman
BG
,
Swirski
FK
,
Weber
GF
.
Cytokine storm and sepsis disease pathogenesis
.
Semin Immunopathol
.
2017
;
39
(
5
):
517
28
.
3.
Rudd
KE
,
Johnson
SC
,
Agesa
KM
,
Shackelford
KA
,
Tsoi
D
,
Kievlan
DR
, et al
.
Global, regional, and national sepsis incidence and mortality, 1990-2017: analysis for the Global Burden of Disease Study
.
Lancet
.
2020
;
395
(
10219
):
200
11
.
4.
Bauer
M
,
Gerlach
H
,
Vogelmann
T
,
Preissing
F
,
Stiefel
J
,
Adam
D
.
Mortality in sepsis and septic shock in Europe, North America and Australia between 2009 and 2019- results from a systematic review and meta-analysis
.
Crit Care
.
2020
;
24
(
1
):
239
.
5.
Annane
D
,
Bellissant
E
,
Bollaert
P-E
,
Briegel
J
,
Confalonieri
M
,
De Gaudio
R
, et al
.
Corticosteroids in the treatment of severe sepsis and septic shock in adults: a systematic review
.
JAMA
.
2009
;
301
(
22
):
2362
75
.
6.
Fang
F
,
Zhang
Y
,
Tang
J
,
Lunsford
LD
,
Li
T
,
Tang
R
, et al
.
Association of corticosteroid treatment with outcomes in adult patients with sepsis: a systematic review and meta-analysis
.
JAMA Intern Med
.
2019
;
179
(
2
):
213
23
.
7.
Pitre
T
,
Drover
K
,
Chaudhuri
D
,
Zeraaktkar
D
,
Menon
K
,
Gershengorn
HB
, et al
.
Corticosteroids in sepsis and septic shock: a systematic review, pairwise, and dose-response meta-analysis
.
Crit Care Explor
.
2024
;
6
(
1
):
e1000
.
8.
Rochwerg
B
,
Oczkowski
SJ
,
Siemieniuk
RAC
,
Agoritsas
T
,
Belley-Cote
E
,
D’Aragon
F
, et al
.
Corticosteroids in sepsis: an updated systematic review and meta-analysis
.
Crit Care Med
.
2018
;
46
(
9
):
1411
20
.
9.
Evans-Olders
R
,
Eintracht
S
,
Hoffer
LJ
.
Metabolic origin of hypovitaminosis C in acutely hospitalized patients
.
Nutrition
.
2010
;
26
(
11–12
):
1070
4
.
10.
Borrelli
E
,
Roux-Lombard
P
,
Grau
GE
,
Girardin
E
,
Ricou
B
,
Dayer
J
, et al
.
Plasma concentrations of cytokines, their soluble receptors, and antioxidant vitamins can predict the development of multiple organ failure in patients at risk
.
Crit Care Med
.
1996
;
24
(
3
):
392
7
.
11.
Galley
HF
,
Davies
MJ
,
Webster
NR
.
Ascorbyl radical formation in patients with sepsis: effect of ascorbate loading
.
Free Radic Biol Med
.
1996
;
20
(
1
):
139
43
.
12.
Wilson
JX
.
Evaluation of vitamin C for adjuvant sepsis therapy
.
Antioxid Redox Signal
.
2013
;
19
(
17
):
2129
40
.
13.
Marik
PE
.
Hydrocortisone, ascorbic acid and thiamine (HAT therapy) for the treatment of sepsis. Focus on ascorbic acid
.
Nutrients
.
2018
;
10
(
11
):
1762
.
14.
Carr
AC
,
Shaw
GM
,
Fowler
AA
,
Natarajan
R
.
Ascorbate-dependent vasopressor synthesis: a rationale for vitamin C administration in severe sepsis and septic shock
.
Crit Care
.
2015
;
19
:
418
.
15.
Collie
JTB
,
Greaves
RF
,
Jones
OAH
,
Lam
Q
,
Eastwood
GM
,
Bellomo
R
.
Vitamin B1 in critically ill patients: needs and challenges
.
Clin Chem Lab Med
.
2017
;
55
(
11
):
1652
68
.
16.
Donnino
MW
,
Andersen
LW
,
Chase
M
,
Berg
KM
,
Tidswell
M
,
Giberson
T
, et al
.
Randomized, double-blind, placebo-controlled trial of thiamine as a metabolic resuscitator in septic shock: a pilot study
.
Crit Care Med
.
2016
;
44
(
2
):
360
7
.
17.
Woolum
JA
,
Abner
EL
,
Kelly
A
,
Thompson Bastin
ML
,
Morris
PE
,
Flannery
AH
.
Effect of thiamine administration on lactate clearance and mortality in patients with septic shock
.
Crit Care Med
.
2018
;
46
(
11
):
1747
52
.
18.
Shrestha
DB
,
Budhathoki
P
,
Sedhai
YR
,
Mandal
SK
,
Shikhrakar
S
,
Karki
S
, et al
.
Vitamin C in critically ill patients: an updated systematic review and meta-analysis
.
Nutrients
.
2021
;
13
(
10
):
3564
.
19.
Evans
L
,
Rhodes
A
,
Alhazzani
W
,
Antonelli
M
,
Coopersmith
CM
,
French
C
, et al
.
Surviving sepsis Campaign: international guidelines for management of sepsis and septic shock 2021
.
Crit Care Med
.
2021
;
49
(
11
):
e1063
143
.
20.
Assouline
B
,
Faivre
A
,
Verissimo
T
,
Sangla
F
,
Berchtold
L
,
Giraud
R
, et al
.
Thiamine, ascorbic acid, and hydrocortisone as a metabolic resuscitation cocktail in sepsis: a meta-analysis of randomized controlled trials with trial sequential analysis
.
Crit Care Med
.
2021
;
49
(
12
):
2112
20
.
21.
Na
W
,
Shen
H
,
Li
Y
,
Qu
D
.
Hydrocortisone, ascorbic acid, and thiamine (HAT) for sepsis and septic shock: a meta-analysis with sequential trial analysis
.
J Intensive Care
.
2021
;
9
(
1
):
75
.
22.
Page
MJ
,
McKenzie
JE
,
Bossuyt
PM
,
Boutron
I
,
Hoffmann
TC
,
Mulrow
CD
, et al
.
The PRISMA 2020 statement: an updated guideline for reporting systematic reviews
.
BMJ
.
2021
;
372
:
n71
.
23.
Higgins
JPT
,
Thomas
J
,
Chandler
J
,
Cumpston
M
,
Li
T
,
Page
MJ
, et al
, editors.
Cochrane Handbook for systematic reviews of interventions (v6.2)
.
The Cochrane Collaboration
;
2021
.
24.
Schiavo
JH
.
PROSPERO: an international register of systematic review protocols
.
Med Ref Serv Q
.
2019
;
38
(
2
):
171
80
.
25.
Roth
D
,
Heidinger
B
,
Havel
C
,
Herkner
H
.
Different mortality time points in critical care trials: current practice and influence on effect estimates in meta-analyses
.
Crit Care Med
.
2016
;
44
(
8
):
e737
41
.
26.
Shi
J
,
Luo
D
,
Wan
X
,
Liu
Y
,
Liu
J
,
Bian
Z
, et al
.
Detecting the skewness of data from the five-number summary and its application in meta-analysis
.
Stat Methods Med Res
.
2023
;
32
(
7
):
1338
60
.
27.
Shi
J
,
Luo
D
,
Weng
H
,
Zeng
X-T
,
Lin
L
,
Chu
H
, et al
.
Optimally estimating the sample standard deviation from the five-number summary
.
Res Synth Methods
.
2020
;
11
(
5
):
641
54
.
28.
Luo
D
,
Wan
X
,
Liu
J
,
Tong
T
.
Optimally estimating the sample mean from the sample size, median, mid-range, and/or mid-quartile range
.
Stat Methods Med Res
.
2018
;
27
(
6
):
1785
805
.
29.
Wan
X
,
Wang
W
,
Liu
J
,
Tong
T
.
Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range
.
BMC Med Res Methodol
.
2014
;
14
:
135
.
30.
Sterne
JAC
,
Savović
J
,
Page
MJ
,
Elbers
RG
,
Blencowe
NS
,
Boutron
I
, et al
.
RoB 2: a revised tool for assessing risk of bias in randomised trials
.
BMJ
.
2019
;
366
:
l4898
.
31.
Schwarzer
G
,
Carpenter
JR
,
Rücker
G
.
Meta-analysis with R
.
Basel, Switzerland
:
Springer International Publishing
;
2015
.
32.
Sweeting
MJ
,
Sutton
AJ
,
Lambert
PC
.
What to add to nothing? Use and avoidance of continuity corrections in meta-analysis of sparse data
.
Stat Med
.
2004
;
23
(
9
):
1351
75
.
33.
Cook
RJ
,
Sackett
DL
.
The number needed to treat: a clinically useful measure of treatment effect
.
BMJ
.
1995
;
310
(
6977
):
452
4
.
34.
Harbord
RM
,
Egger
M
,
Sterne
JAC
.
A modified test for small-study effects in meta-analyses of controlled trials with binary endpoints
.
Stat Med
.
2006
;
25
(
20
):
3443
57
.
35.
Egger
M
,
Davey Smith
G
,
Schneider
M
,
Minder
C
.
Bias in meta-analysis detected by a simple, graphical test
.
BMJ
.
1997
;
315
(
7109
):
629
34
.
36.
Sterne
JAC
,
Sutton
AJ
,
Ioannidis
JPA
,
Terrin
N
,
Jones
DR
,
Lau
J
, et al
.
Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials
.
BMJ
.
2011
;
343
:
d4002
.
37.
Terrin
N
,
Schmid
CH
,
Lau
J
.
In an empirical evaluation of the funnel plot, researchers could not visually identify publication bias
.
J Clin Epidemiol
.
2005
;
58
(
9
):
894
901
.
38.
Duval
S
,
Tweedie
R
.
Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis
.
Biometrics
.
2000
;
56
(
2
):
455
63
.
39.
Geissbühler
M
,
Hincapié
CA
,
Aghlmandi
S
,
Zwahlen
M
,
Jüni
P
,
da Costa
BR
.
Most published meta-regression analyses based on aggregate data suffer from methodological pitfalls: a meta-epidemiological study
.
BMC Med Res Methodol
.
2021
;
21
(
1
):
123
.
40.
Guyatt
GH
,
Oxman
AD
,
Vist
GE
,
Kunz
R
,
Falck-Ytter
Y
,
Alonso-Coello
P
, et al
.
GRADE: an emerging consensus on rating quality of evidence and strength of recommendations
.
BMJ
.
2008
;
336
(
7650
):
924
6
.
41.
Zeng
L
,
Brignardello-Petersen
R
,
Hultcrantz
M
,
Mustafa
RA
,
Murad
MH
,
Iorio
A
, et al
.
GRADE Guidance 34: update on rating imprecision using a minimally contextualized approach
.
J Clin Epidemiol
.
2022
;
150
:
216
24
.
42.
Guyatt
GH
,
Oxman
AD
,
Kunz
R
,
Brozek
J
,
Alonso-Coello
P
,
Rind
D
, et al
.
GRADE guidelines 6. Rating the quality of evidence: imprecision
.
J Clin Epidemiol
.
2011
;
64
(
12
):
1283
93
.
43.
Zhong
B
.
How to calculate sample size in randomized controlled trial
.
J Thorac Dis
.
2009
;
1
(
1
):
51
4
.
44.
Guyatt
G
,
Oxman
AD
,
Akl
EA
,
Kunz
R
,
Vist
G
,
Brozek
J
, et al
.
GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables
.
J Clin Epidemiol
.
2011
;
64
(
4
):
383
94
.
45.
Moskowitz
A
,
Huang
DT
,
Hou
PC
,
Gong
J
,
Doshi
PB
,
Grossestreuer
AV
, et al
.
Effect of ascorbic acid, corticosteroids, and thiamine on organ injury in septic shock: the ACTS randomized clinical trial
.
JAMA
.
2020
;
324
(
7
):
642
50
.
46.
Feng
F
,
Yang
H
,
Yang
W
,
Chen
Y
.
Metabolic resuscitation therapy in critically ill patients with sepsis and septic shock: a pilot prospective randomized controlled trial
.
Open Med
.
2023
;
18
(
1
):
20230637
.
47.
Hussein
AA
,
Sabry
NA
,
Abdalla
MS
,
Farid
SF
.
A prospective, randomised clinical study comparing triple therapy regimen to hydrocortisone monotherapy in reducing mortality in septic shock patients
.
Int J Clin Pract
.
2021
;
75
(
9
):
e14376
.
48.
Chang
P
,
Liao
Y
,
Guan
J
,
Guo
Y
,
Zhao
M
,
Hu
J
, et al
.
Combined treatment with hydrocortisone, vitamin C, and thiamine for sepsis and septic shock: a randomized controlled trial
.
Chest
.
2020
;
158
(
1
):
174
82
.
49.
Jamshidi
MR
,
Zeraati
MR
,
Forouzanfar
B
,
Tahrekhani
M
,
Motamed
N
.
Effects of triple combination of hydrocortisone, thiamine, and Vitamin C on clinical outcome in patients with septic shock: a single-center randomized controlled trial
.
J Res Med Sci
.
2021
;
26
:
47
.
50.
Lyu
Q-Q
,
Zheng
R-Q
,
Chen
Q-H
,
Yu
J-Q
,
Shao
J
,
Gu
X-H
.
Early administration of hydrocortisone, vitamin C, and thiamine in adult patients with septic shock: a randomized controlled clinical trial
.
Crit Care
.
2022
;
26
(
1
):
295
.
51.
Mohamed
A
,
Abdelaty
M
,
Saad
MO
,
Shible
A
,
Mitwally
H
,
Akkari
A-R
, et al
.
Evaluation of hydrocortisone, vitamin C, and thiamine for the treatment of septic SHOCK: a randomized controlled trial (the hyvits trial)
.
Shock
.
2023
;
59
(
5
):
697
701
.
52.
Iglesias
J
,
Vassallo
AV
,
Patel
VV
,
Sullivan
JB
,
Cavanaugh
J
,
Elbaga
Y
.
Outcomes of metabolic resuscitation using ascorbic acid, thiamine, and glucocorticoids in the early treatment of sepsis: the ORANGES trial
.
Chest
.
2020
;
158
(
1
):
164
73
.
53.
Raghu
K
,
Ramalingam
K
.
Safety and Efficacy of Vitamin C, Vitamin B1, and Hydrocortisone in clinical outcome of septic shock receiving standard care: a quasi experimental randomized open label two arm parallel group study
.
Eur J Mol Clin Med
.
2021
;
8
(
2
):
873
91
.
54.
Sang
ZZ
,
Wang
SJ
,
Wang
W
,
Feng
SY
,
Wang
SL
,
Li
Y
.
Clinical efficacy of hydrocortisone combined with vitamin C and vitamin B1 in the treatment of septic shock in the elderly
.
Chin J Gerontol
.
2020
;
40
(
20
):
4351
4
.
55.
Sevransky
JE
,
Rothman
RE
,
Hager
DN
,
Bernard
GR
,
Brown
SM
,
Buchman
TG
, et al
.
Effect of vitamin C, thiamine, and hydrocortisone on ventilator- and vasopressor-free days in patients with sepsis: the VICTAS randomized clinical trial
.
JAMA
.
2021
;
325
(
8
):
742
50
.
56.
Mohamed
ZU
,
Prasannan
P
,
Moni
M
,
Edathadathil
F
,
Prasanna
P
,
Menon
A
, et al
.
Vitamin C therapy for routine care in septic shock (ViCTOR) trial: effect of intravenous vitamin C, thiamine, and hydrocortisone administration on inpatient mortality among patients with septic shock
.
Indian J Crit Care Med
.
2020
;
24
(
8
):
653
61
.
57.
Fujii
T
,
Luethi
N
,
Young
PJ
,
Frei
DR
,
Eastwood
GM
,
French
CJ
, et al
.
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
.
2020
;
323
(
5
):
423
31
.
58.
Wang
J
,
Song
Q
,
Yang
S
,
Wang
H
,
Meng
S
,
Huang
L
, et al
.
Effects of hydrocortisone combined with vitamin C and vitamin B1 versus hydrocortisone alone on microcirculation in septic shock patients: a pilot study
.
Clin Hemorheol Microcirc
.
2023
;
84
(
2
):
111
23
.
59.
Wani
SJ
,
Mufti
SA
,
Jan
RA
,
Shah
SU
,
Qadri
SM
,
Khan
UH
, et al
.
Combination of vitamin C, thiamine and hydrocortisone added to standard treatment in the management of sepsis: results from an open label randomised controlled clinical trial and a review of the literature
.
Infect Dis
.
2020
;
52
(
4
):
271
8
.
60.
Virág
M
,
Leiner
T
,
Rottler
M
,
Ocskay
K
,
Molnar
Z
.
Individualized hemodynamic management in sepsis
.
J Pers Med
.
2021
;
11
(
2
):
157
.
61.
Lamontagne
F
,
Day
AG
,
Meade
MO
,
Cook
DJ
,
Guyatt
GH
,
Hylands
M
, et al
.
Pooled analysis of higher versus lower blood pressure targets for vasopressor therapy septic and vasodilatory shock
.
Intensive Care Med
.
2018
;
44
(
1
):
12
21
.
62.
Guinot
P-G
,
Abou-Arab
O
,
Guilbart
M
,
Bar
S
,
Zogheib
E
,
Daher
M
, et al
.
Monitoring dynamic arterial elastance as a means of decreasing the duration of norepinephrine treatment in vasoplegic syndrome following cardiac surgery: a prospective, randomized trial
.
Intensive Care Med
.
2017
;
43
(
5
):
643
51
.
63.
Leite
SA
,
Locatelli
SB
,
Niece
SP
,
Oliveira
AR
,
Tockus
D
,
Tosin
T
.
Impact of hyperglycemia on morbidity and mortality, length of hospitalization and rates of re-hospitalization in a general hospital setting in Brazil
.
Diabetol Metab Syndr
.
2010
;
2
:
49
.
64.
Arzhan
S
,
Roumelioti
M-E
,
Litvinovich
I
,
Bologa
CG
,
Myers
OB
,
Unruh
ML
.
Hypernatremia in hospitalized patients: a large population-based study
.
Kidney360
.
2022
;
3
(
7
):
1144
57
.
65.
Zhou
F
,
Deng
J
,
Heybati
K
,
Zuo
QK
,
Ali
S
,
Hou
W
, et al
.
Efficacy and safety of corticosteroid regimens for the treatment of hospitalized COVID-19 patients: a meta-analysis
.
Future Virol
.
2022
;
17
(
7
):
463
89
.
66.
Moskowitz
A
,
Omar
Y
,
Chase
M
,
Lokhandwala
S
,
Patel
P
,
Andersen
LW
, et al
.
Reasons for death in patients with sepsis and septic shock
.
J Crit Care
.
2017
;
38
:
284
8
.
67.
Rhee
C
,
Jones
TM
,
Hamad
Y
,
Pande
A
,
Varon
J
,
O’Brien
C
, et al
.
Prevalence, underlying causes, and preventability of sepsis-associated mortality in US acute care hospitals
.
JAMA Netw Open
.
2019
;
2
(
2
):
e187571
.
68.
Mebazaa
A
,
Laterre
PF
,
Russell
JA
,
Bergmann
A
,
Gattinoni
L
,
Gayat
E
, et al
.
Designing phase 3 sepsis trials: application of learned experiences from critical care trials in acute heart failure
.
J Intensive Care
.
2016
;
4
:
24
.
69.
Zannad
F
,
Garcia
AA
,
Anker
SD
,
Armstrong
PW
,
Calvo
G
,
Cleland
JGF
, et al
.
Clinical outcome endpoints in heart failure trials: a European Society of Cardiology Heart Failure Association consensus document
.
Eur J Heart Fail
.
2013
;
15
(
10
):
1082
94
.