a b s t r a c t

Background: The effects of corticosteroids on clinical outcomes of patients with sepsis remains controversial. We aimed to further determine the effectiveness of corticosteroids in reducing mortality in adult patients with severe sepsis by comparison with placebo.

Methods: Pubmed, Embase, Medline, Cochrane Central Register of Controlled Trials (CENTRAL) as well as the Information Sciences Institute (ISI) Web of Science were searched for all controlled studies that compared corticosteroids and placebo in adult patients with severe sepsis. The primary outcome was the mortality 28-day mortality and the secondary outcomes were mortality at longest follow up, occurrence, and reoccurrence of septic shock.

Results: A total of 19 trials involving 7035 patients were pooled in our final analyses. No significant heterogeneity was found in any of the outcome measures. Compared with placebo, corticosteroids were associated with a lower 28-day mortality (RR 0.91, 95% CI 0.85-0.98, Z = 2.57, P = 0.01) both in patients having sepsis and in those who developed septic shock (RR 0.92, 95% CI 0.85-0.99, Z = 2.19, P = 0.03), while no significant difference was found in mortality with the longest follow up in patients either having sepsis (RR 0.94, 95% CI 0.89-1.00, Z = 1.93, P = 0.05), or occurrence (RR 0.83, 95% CI 0.56-1.24, Z = 0.90, P = 0.37) or reoccurrence

of septic shock (RR 1.08, 95% CI 1.00-1.16, Z = 1.89, P = 0.06).

Conclusions: Corticosteroids were effective in reducing the 28-day mortality in patients with severe sepsis and in those with septic shock.

(C) 2018

Background

Sepsis is a clinical syndrome characterized by systemic inflammation due to infection. Studies showed that, between 2003 and 2015, global incidence of sepsis was about 437 per 100,000 [1]. Vascular dysfunction and systematic inflammatory response contribute to tissue hypoxia, Multiple organ dysfunction syndrome (MODS) and death. Thus, sepsis is a syndrome associated with a mortality of 25% and up to 52% once de- veloped to septic shock [2]. Up to now, no effective therapy for sepsis has been developed other than resuscitative therapy, respiratory sup- port, and antibiotic therapy [3].

The hypothalamic-pituitary-adrenal (HPA) axis is usually impaired in patients with sepsis. [4] In these patients, the hypothalamic- pituitary-adrenal axis affects inflammation through white blood cells,

* Corresponding author at: No.37 Guoxue Alley, Chengdu 610041, China.

?? Corresponding author.

E-mail address: [email protected] (Z.-A. Liang).

¹ These authors contributed equally to this work.

cytokines, and nitric oxide production [5]. The same time, inflammatory cytokines may either suppress Cortisol response to adrenocorticotropin, resulting in insufficient adrenal output, or compete with intracellular corticosteroids receptor function, resulting in peripheral tissue cortico- steroids resistance [6]. Thus, the major theoretical purpose of glucocor- ticoids in sepsis is to restore balance of the altered HPA axis with the goal of improved outcomes such as mortality.

Nevertheless, the role of corticosteroids in reducing mortality in pa- tients with severe sepsis remains unclear despite numerous published Randomized controlled trials and meta-analysis papers. In 2009, a systematic review published in The Journal of American Medical Association reported that low dose corticosteroids might have a benefi- cial effect on short-term mortality of patients with severe sepsis [7]. However, a meta analysis in 2015 found no advantage in the use of cor- ticosteroids in any dose [8]. Recently, two double-blind RCTs have been published which reported conflicting results about the use of corticoste- roids in patients with severe sepsis. In the study of Venkatech et al., hy- drocortisone did not result in lower mortality than placebo [9]. On the contrary, in the study of Annane et al., mortality was lower in patients who received hydrocortisone plus fludrocortisone [10].

https://doi.org/10.1016/j.ajem.2018.11.040

0735-6757/(C) 2018

Considering the controversial findings on corticosteroids in adult pa- tients with severe sepsis, we therefore assumed that corticosteroids may be more effective than placebo in reducing mortality. To identify the role of corticosteroids in improving clinical outcomes of patients with severe sepsis, we conducted a systematic review and meta- analysis of all published trials comparing the mortality of patients with severe sepsis patients who received corticosteroids and those who received placebo.

Methods

Search strategies

From 1946 to March 2018, a comprehensive computer search was conducted in Pubmed, Embase, Medline, Cochrane Central Register of Controlled Trails (CENTRAL), and Information Sciences Institute (ISI) Web of Science using the keywords of “glucocorticoid” or “corticoste- roid” or “steroid” or “cortisone” or “hydrocortisone” or “prednisolon” or “methylprednisolon” or “prednison” or “dexamethason” or “triamcinolon” and “sepsis” or “septic” without limitation in the publication type or language. We also reviewed the references listed in each identified article and manually searched the related articles to identify all eligible studies and minimize any potential publication bias.

Inclusion and exclusion criteria

Eligible clinical trials were identified based on the following criteria:

1) the subjects enrolled in each study included patients with severe sep- sis, which was defined in 1992 [11]; 2) patients were randomly divided into experimental group, in which corticosteroids were applied, and control group, in which patients received placebo; and 3) outcomes in- cluded but were not limited to mortality, occurrence, or reoccurrence of septic shock. We excluded studies if they were performed in animals or in patients under 18 years, or published as non-randomized study, re- views, or case reports.

Study selection

Two independent investigators (YNN and BML) performed the study selection in two phases. In the first phase, they discarded duplicated and non-controlled studies by screening titles and abstracts. In the second phase, eligible studies were extracted by reviewing full texts in accor- dance with the previously designed study inclusion criteria. Any disagreement was resolved by mutual consensus in the presence of a third investigator (ZAL).

Data extraction

Independently, the two data collectors extracted and recorded desir- able information of each enrolled study in a standard form recom- mended by Cochrane [12], which consisted of authors, Publication year, study design, country, NCT No., population, demographic charac- teristics (age, gender, etc.), disease conditions (The Acute Physiologic and Chronic Health Evaluation II (APACHE II) and Simplified Acute Physiologic Score II (SAPS II)), Sequential Organ Failure Assessment , and outcome measures (such as mortality and reoccurrence of septic shock). For any missing data or information, corresponding au- thors were contacted by email to request the full original data. Different opinions between the two collectors were determined by reaching a consensus or consulting a third investigator.

Quality assessment

For the assessment of risk of bias in estimating the study outcomes, we used the Cochrane risk of bias tool [12]. Each study was assessed for:

Fig. 1. Study flow.

1) random sequence generation (selection bias); 2) allocation conceal- ment (selection bias); 3) blinding of participants and personnel (performance bias); 4) blinding of related outcomes assessment (detection bias); 5) incomplete outcome data (attrition bias); 6) selec- tive reporting (reporting bias); and 7) other biases. Two investigators conducted the quality assessment for the study methodology, indepen- dently and separately. Any divergence was resolved by mutual consen- sus in the presence of a third investigator.

Statistical analysis

Statistical analysis of our study was accomplished by an indepen- dent statistician using Cochrane systematic review software Review Manager (RevMan; Version 5.3.5; The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, 2014). We used Mann-Whitney U test to verify hypothesis and rendered statistical significance as a Z- value and P-value b 0.05, and the results were displayed in Forest plots. Continuous variables were reported as mean and standard deriva- tion (SD), while dichotomous variables were shown as frequency and proportion. An initial test for clinical, methodological, and statistical heterogeneities was conducted, and the ?² test with P b 0.1 and I² N 50% were used to indicate significance. We also performed the sensi- tivity analysis to substitute alternative decisions or ranges of values for decisions that were arbitrary or unclear. Random-effects model was ap- plied in the presence of statistical heterogeneity. For continuous data we calculated mean difference (MD) and 95% confidence interval (CI), while for dichotomous data we calculated Risk ratio (RR) and 95% CI.

Results

Initially 5776 records were identified, of which 5772 were extracted from electronic databases and 4 from reference lists review. (Fig. 1) By screening the titles and abstracts, we discarded 5729 studies for dupli- cation (n = 766), Animal experiments (n = 1945), non-adult patients (n = 1774), and non-controlled studies (n = 1244). We searched the

full-text articles for the remaining 47 studies, and eventually 19 trials [9,10,13-28] entered our final analysis after excluding 21 studies for not reporting related outcomes and 7 for not designed as expected.

Study description

All 19 studies compared the outcomes of corticosteroids and placebo. The 28-day mortality was recorded in 18 studies [9,10,13-19,21-28], mortality at longest follow up was reported in 19 studies [9,10,13-29], the occurrence of septic shock was reported in 2 studies [21,26], and the reoccurrence of septic shock was reported in 9 studies [9,10,13- 16,24,26,27]. Details of each enrolled study were presented in Table 1.

A total of 7035 patients were pooled from all the included trials in our final systematic review and meta-analysis, among whom 3517 were treated with corticosteroids and 3518 with placebo. Details of baseline characteristics of patients in each enrolled study were shown in Table 2.

Quality assessment

Quality assessment of the 19 enrolled studies showed no bias in at- tribution, detection, blinding, or reporting in 11 studies, but high bias in 2 studies with regard to performance because blinding of patients and personnel. (Figs. 2 and 3) No studies were excluded for low quality or dubious decisions in the sensitivity analysis. There was no evidence of reporting bias for the on inspection of the funnel plot (Fig. 4).

Table 1

Basic characteristics of 18 enrolled studies

Study ID	Study design	RCT no.	Population	Disease	Dose Outcome
Annane 2002 Annane 2018	Placebo controlled, randomized, double-blind, parallel-group trial Multicenter, double blind, randomized	NR NCT00625209	300 1241	Septic shock Septic shock	50 mg hydrocortisone bolus every 6 h for 7 days (1)(2)(4) and 50ug fludrocortisone orally per day for 7 days 50 mg hydrocortisone bolus for every 6 h and 50 ug (1)(2)(4)
Arabi 2010	trial randomized controlled study	ISRCTN99675218	75	Septic shock	fludrocortisone tablet daily for 7 days 50 mg hydrocortisone every 6 h and dose reduction (1)(2)(4)
Bollaert 1998	Prospective, randomized, double-blind,	NR	41	Septic shock	by 10 mg every 2 days after shock resolution 100 mg hydrocortisone bolus every 8 h for 5 days (1)(2)(4)
Briegel 1999	placebo-controlled study. Prospective, randomized, double-blind,	NR	40	Septic shock	then weaned over 6 days 100 mg hydrocortisone within 30 min and (1)(2)(4)
Cicarelli 2007	single-center study Prospective randomized clinical trial	NR	29	Septic shock	followed by 0.18 mg/Kg/h until shock reversal 0.2 mg/Kg dexamethasone, 3 doses at intervals of (1)(2)
Confalonieri	A preliminary randomized study	NR	46	Sepsis	36 h 200 mg hydrocortisone bolus followed by 10 mg/h (1)(2)
2005					for 7 days, and then weaned over 4 days
Gordon 2014 Hu 2009	Prospective open-label randomized controlled pilot trial Randomized controlled study	ISRCTN66727957 NR	31 77	Septic shock Septic shock	50 mg hydrocortisone bolus 6 h for 5 days, 12 h for (1)(2) 3 days, daily for 3 days 50 mg hydrocortisone bolus every 6 h for 7 days (1)
Keh 2016	Double-blind, randomized clinical trial	NCT00670254	380	sepsis	200 mg/d 50 mg hydrocortisone bolus, 200 mg for 5 days, 100 (1)(2)(3)
Lv 2017	Placebo controlled, Randomized clinical	NCT 02580240	118	Septic shock	mg for 2 days, 50 mg for 2 days, 25 mg for 2 days 200 mg/d hydrocortisone for 6 days, then tapered off (1)(2)(4)
	study
Meduri 2007 Oppert 2005	Randomized clinical trial Prospective, randomized, double-blind,	NR NR	80 41	Severe sepsis Septic shock	1 mg/kg/d methylpredisolone for 14 days, 0.5 (1)(2) mg/kg/d for 7 days, 0.25 mg/kg/d for 4 days and 0.125 mg/kg/d for 3 days. If MV free before day 14, patients was advanced to day 15 of drug therapy. 50 mg hydrocortisone bolus followed by a (1)(2)(4)
	single-center study				continuous infusion of 0.18 mg/Kg until vasopressor cessation N1 h, then weaned by steps of 0.02 mg/kg/h
Rinaldi 2006	Randomized prospective study	NR	40	Severe sepsis	300 mg/d hydrocortisone for 6 days, then tapered off (1)(2)
Tongyoo 2016	Double blind, single center, randomized,	NCT 01284452	197	Sepsis	50 mg hydrocortisone every 6 h for 7 days (1)(2)(3)
Venkatesh	placebo-controlled trial International, pragmatic, double blind,	NCT01448109	3658	Septic shock	200 mg/d hydrocortisone for 7 days (1)(2)(4)
2018 Yildiz2002	parallel group, randomized, controlled trial Placebo-controlled, randomized,	ISRCTN36253388	40	sepsis	5 mg prednisolone at 6:00 and 2.5 mg at 18:00 for (1)(2)
Yildiz 2011	double-blinded, single center study Prospective, randomized, double-blind,	NCT01275638	55	sepsis	10 days 10 mg Prednisolone at 6:00, 5 mg at 14:00, and 5 (1)
	placebo-controlled trial				mg/d at 22.00 for 10 days

⁽¹⁾Mortality at longest follow up; ⁽²⁾28-day mortality; ⁽³⁾Occurrence of septic shock; ⁽⁴⁾Reoccurrence of septic shock. RCT, randomized controlled trial.

Table 2

Basic characteristics of patients

Study ID

Hydrocortisone

Control

APACHE II

Age, year

Male

SOFA

SAPS II

APACHE II

Age, year

Male

SOFA

SAPS II

Mean (SD) or

Mean (SD) or

N (%)

Mean (SD) or

Mean (SD) or

Mean (SD) or

Mean (SD) or

N (%)

Mean (SD) or

Mean (SD) or

Median (IQR)

Median (IQR)

Median (IQR)

Median (IQR)

Median (IQR)

Median (IQR)

Median (IQR)

Median (IQR)

Annane

NR

62 +- 15

64%

NR

60 +- 19

NR

60 +- 17

70%

NR

57 +- 19

2002 Annane

NR

66 +- 14

65.5%

12 +- 3

56 +- 19

NR

66 +- 15

67.7%

11 +- 3

56 +- 19

2018 Arabi 2010

30.0 +- 7.4

60.6 +- 12.6

56%

14.6 +- 3.7

NR

29.3 +- 8.0

59.3 +- 12.2

56%

14.3 +- 3.7

NR

Bollaert 1998 Briegel 1999

NR 26 +- 1

66 (21-83) 47 +- 4

68.2% 45%

NR 12.8 +- 0.6

14 +- 3 55 +- 4

NR 27 +- 1

56 (34-81) 51 +- 4

63.2% 60%

NR 11.8 +- 0.6

14 +- 3 55 +- 5

Cicarelli 2007 Confalonieri

20 +- 5 17.2 4.1

69 +- 11 60.4 +- 17.3

42.9% 74%

9 +- 3 NR

NR NR

19 +- 4 18.2 4.0

61 +- 15 66.6 +- 14.7

46.7% 65%

10 +- 2 NR

NR NR

2005 Gordon

19 (14-22)

61 (54-68)

58%

NR

NR

20 (17,25)

60(48-76)

60%

NR

NR

2014 Hu 2009

18.75 +- 19.54

56.17 +- 33.85

61%

NR

NR

10.47 +- 11.58

54.91 +- 35.36

64%

NR

NR

Keh 2016

19.5 +- 6.9

65.5 +- 14.2

66.7%

6.4 +- 2.6

56.1 +- 13.3

18.5 +- 6.0

64.6 +- 14.6

63.1%

6.2 +- 2.4

52.2 +- 9.9

Lv 2017

25.5 +- 9.5

68.8 +- 12.6

57%

11.9 +- 3.3

NR

21.3 +- 6.9

64.8 +- 16.7

62%

9.9 +- 3.0

NR

Meduri 2007 Oppert 2005

NR 25 (19-30)

NR 59

NR 72%

NR 11(10-13)

NR NR

NR 25.5(19.8-29)

NR 47

NR 83%

NR 10 (8.5-13)

NR NR

Rinaldi 2006

NR

68 (49-74)

NR

NR

54 (41-58)

NR

66 (51-72)

NR

57 (44-60)

NR

Tongyoo 2016 Venkatesh

21.7 +- 5.7 24 (19-29)

64.5 +- 17.3 62.3 +- 14.9

51 60.4%

10.9 +- 3.5 NR

NR NR

21.9 +- 5.7 23 (18-29)

64.3 +- 16 62.7 +- 15.2

51.5 61.3%

10.8 +- 3.6 NR

NR NR

2018 Yildiz2002

15.4 +- 5.5

57.8 +- 17.7

54.2%

7.8 +- 3.9

NR

17.9 +- 8.0

56.5 +- 16.4

45.8%

9.3 +- 4.0

NR

Yildiz 2011

75 (35-90)

9.9 +- 3.4

78%

NR

NR

64 (20-91)

7.9 +- 4.0

55%

NR

NR

APACHE, The Acute Physiologic and Chronic Health Evaluation; IQR, interquartile range; NR, not report; SAPS, Simplified Acute Physiologic Score; SOFA, Sequential Organ Failure Assess- ment; SD, standard derivation.

Heterogeneity

No significant statistical heterogeneity was found in any of the comparisons between corticosteroids and placebo: 28-day mortality (I² = 11%, ?² = 19.13, P = 0.32); mortality at longest follow up (I² = 22%, ?² = 23.15, P = 0.18), occurrence of shock (I² = 0%, ?² = 0.25, P = 0.62), or reoccurrence of septic shock (I² = 0%, ?² = 4.99, P = 0.76).

Mortality

Significant differences in the 28-day mortality (RR 0.91, 95% CI 0.85-0.98, Z = 2.57, P = 0.01) were found in corticosteroids compared with placebo but not in the mortality at longest follow up (RR 0.94, 95% CI 0.89-1.00, Z = 1.93, P = 0.05). (Figs. 5 and 6).

Occurrence of septic shock

No significant difference was identified either in occurrence of septic shock between treatment with corticosteroids and placebo (RR 0.83,

Fig. 2. Risk of bias.

95% CI 0.56-1.24, Z = 0.90, P = 0.37) (Fig. 7) or reoccurrence of septic shock (RR 1.08, 95% CI 1.00-1.16, Z = 1.89, P = 0.06). (Fig. 8).

Post hoc analysis

Subgroup analysis of corticosteroids in patients with septic shock (n = 6209) was performed, and corticosteroids were associated with a significantly lower 28-day mortality (RR 0.92, 95% CI 0.85-0.99, Z= 2.19, P = 0.03) than placebo. (Supplemental File 1).

Discussion

In this meta-analysis, we found that corticosteroids could reduce the 28-day mortality of both patients with severe sepsis and those with septic shock. However, in terms of the occurrence and reoccurrence of septic shock, corticosteroids did not show any advantage.

We found a significantly lower 28-day mortality in patients with se- vere sepsis treated with corticosteroids. This may have resulted from several factors, among which the improvement of hemodynamics was the main mechanism. By binding to mineralocorticoid receptors in the kidney, corticosteroids could improve glomerular function [30], free water clearance, and sodium renal excretion [31]. Thus, patients with sepsis can restore effective blood volume through sodium and water re- tention. Second, corticosteroids are known to regulate the contractility of vascular smooth muscle through controlling alpha 1-adrenoceptor- mediated second-messenger, adenosine triphosphate-sensitive potas- sium channels and depressing pressor sensitivity to noradrenaline [32]. Third, endothelial glucocorticoid receptors is a critical regulator in preventing NF-?B activation and nitric oxide synthesis in sepsis [33]. In this way, corticosteroids play a significant role in increasing the cardiac output [34]. Studies showed that the administration of corticosteroids result in decrease in requirement for vasoactive drugs such as norepinephrine [16,24] and become potentially beneficial for

Fig. 3. Risk of bias summary.

Fig. 4. Funnel plot.

Organ function through a peripheral vascular circulation recovery [35]. Fourth, corticosteroids also improve micro-circulation and tissue perfusion in septic shock [36]. This effect may be mediated by non- transcriptional activation of endothelial nitric oxide synthase [37] and stabilization of capillary permeability. Corticosteroids have been shown to increase renal blood flow, inhibit inducible nitric oxide syn- thase activation in the renal cortex, prevent the appearance of cortical microcirculatory hypoxic areas, improve renal oxygen delivery, and significantly restore oxygen consumption, kidney function and tubular sodium reabsorption to baseline values [38]. Fifth, corticosteroids attenuate inflammation in various organs in patients with sepsis. For example, they have been shown to dramatically inhibit the nuclear factor-kappa B in peripheral blood and pulmonary mononuclear cells [38]. They also stimulate the albumin synthesis and decrease pro- infammatory mediators such as interleukin-1 and interleukin-6 [39].

As for mortality at longest follow up, we did not find any significant

advantage of corticosteroids over placebos. The longest follow up time in the included studies varied from right out of hospital to 1 year after hospital discharge, during which we could not sure the exact reason for death.

However, our meta-analysis also indicated that corticosteroids could not further prevent the occurrence and reoccurrence of septic shock. First of all, only two studies involving 494 patients were eligible for the analysis of occurrence of septic shock; and type II error might exist be- cause of the limited number of patients. Second, the sensitization of the vasculature to vasopressors could be increased by corticosteroids [32]. For this reason, vasopressors were less needed in patients treated with corticosteroids. Because hydrocortisone was tapered in some patients while vasopressors were still necessary, Hemodynamic deterioration may be underestimated if the patients were treated with corticosteroids. Third, the use of corticosteroids might lead to increased risk of sec- ondary infection and sepsis because of inhibiting inflammation, which is commonly known as adverse events [40]. Finally, studies showed that short term use of corticosteroids would lead to a rebound of the systematic inflammatory response [41]. These disadvantages might off- set the advantages of corticosteroids, which may account for the result that no significant difference between corticosteroids and placebos

was found in the rate of occurrence and reoccurrence of septic shock.

Although all enrolled studies used low dose corticosteroids (b300 mg per day), the exact dose and duration of corticosteroids still varied. Nevertheless, we did not find the optimal dose or duration for reducing the mortality of sepsis. Moreover, our study did not explore the exact initiating time of corticosteroids. However, Studies showed that in patients with septic shock, receiving hydrocortisone for, early initiation of corticosteroids was associated with improved survival. [42] The conclusions we drew from our meta analysis differed from those of previous analyses. This may be attributed to the fact that we included

Fig. 5. The 28-day mortality. CI, confidence interval; SD, standard deviation.

Fig. 6. Mortality at longest follow up.

more large-sample RCTs published recently [43]. Moreover, we excluded all the studies conducted before 1992 for two reasons. The first reason was that the earliest definition and diagnosis criteria of sepsis were not confirmed until 1992 [11]. In the clinical studies published before 1992, the definition of “sepsis” was not consistent with the one made in 1992. The other reason was that the methods used widely in clinical settings in recent years such as early-goal- directed therapy, which could reduce the mortality of sepsis, were not

popular N20 years ago [44]. Moreover, compared with the ones pub- lished in 2015 [45] we excluded the studies involving patients with community-acquired pneumonia but not stating clearly whether these patients met the diagnosis criteria of sepsis.

Recently, two large-sample, multicenter RCTs have been published but reported contradictory findings [9,10]. In Annane’s study, hydrocortisone-plus-fludrocortisone could reduce the 90-day mortality, while Venkatesh found no significant difference in the 90-day mortality

Fig. 7. The occurrence of septic shock.

Fig. 8. The reoccurrence of septic shock.

even though a trend of lower mortality in the hydrocortisone group was identified. Meanwhile, hydrocortisone-plus-fludrocortisone might be more powerful than hydrocortisone alone.

Despite our findings, the present study still has some limitations that needed to be addressed. First of all, no significant heterogeneity was identified in any of the comparisons, but clinical heterogeneity still existed. Although low dose corticosteroids (b300 mg per day) were administrated in all the studies, the exact dose and duration of cortico- steroids still varied. We did not find the optimal dose or duration for re- ducing the mortality of sepsis. Second, in spite of the subgroup analysis on patients with septic shock, the severity of sepsis differed by studies, which could be seen from the APACHE II and SOFA score. Third, our analysis showed that, in general, patients with sepsis could benefit from corticosteroids, but not all patients responded to corticosteroids. Thus, more studies should be done to profile the characteristics of sepsis patients who respond well to corticosteroids.

Conclusions

Corticosteroids could reduce the 28-day mortality of both patients with severe sepsis and those with septic shock. However, it did not show any advantage in preventing the occurrence and reoccurrence of septic shock.

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2018.11.040.

Abbreviations

APACHE The Acute Physiologic and Chronic Health Evaluation CENTRAL Cochrane Central Register of Controlled Trails

CI confidence interval

ISI Information Sciences Institute MD mean difference

RR risk ratio

RCT randomized controlled trial

SAPS Simplified Acute Physiologic Score SD standard derivation

SOFA Sequential Organ Failure Assessment

Competing interests

None of all authors have any financial or non-financial competing in- terests in this manuscript.

Declarations

Ethics approval and consent to participate

Each enrolled trial was approved by the corresponding institutional ethical committee, and all participants provided written informed consent.

Consent for publication

Not applicable.

Availability of data and material

Not applicable.

Competing interests

None of all authors have any financial or non-financial competing in- terests in this manuscript.

Funding

This study was partly supported by Ministry of Science and Technol- ogy of the People’s Republic of China (2013BAI09B09).

Author contributions

Y-NN and Y-WW designed the study, drafted the manuscript, con- ducted literature search and data analysis; Y-NN, Y-ML and B-ML re- vised the manuscript critically for important intellectual content; Z-AL made the decision to submit the report for publication. All authors read and approved the final manuscript.

Acknowledgements

We thank Professor Dongtao Lin (College of Foreign Languages and Cultures, Sichuan University), who is specialized in biomedical writing and editing, for copyediting this manuscript.

References

1. Fleischmann C, Scherag A, Adhikari NK, Hartog CS, Tsaganos T, Schlattmann P, et al. International forum of acute care trialists. Assessment of global incidence and mor- tality of hospital-treated sepsis. Current estimates and limitations. Am J Respir Crit Care Med 2016 Feb 1;193(3):259-72.
2. Leligdowicz A, Dodek PM, Norena M, Wong H, Kumar A, Kumar A. Co-operative an- timicrobial therapy of septic shock database research group. Association between source of infection and hospital mortality in patients who have septic shock. Am J Respir Crit Care Med 2014 May 15;189(10):1204-13.
3. Fleischmann C, Scherag A, Adhikari NK, Hartog CS, Tsaganos T, Schlattmann P, et al. International forum of acute care trialists. Assessment of global incidence and mor- tality of hospital-treated Sepsis. Current estimates and limitations. Am J Respir Crit Care Med 2016 Feb 1;193(3):259-72.
4. Cooper MS, Stewart PM. corticosteroid insufficiency in acutely ill patients. N Engl J Med 2003 Feb 20;348(8):727-34.
5. Annane D, Cavaillon JM. Corticosteroids in sepsis: from bench to bedside? Shock 2003 Sep;20(3):197-207.
6. Molijn GJ, Spek JJ, van Uffelen JC, de Jong FH, Brinkmann AO, Bruining HA, et al. Dif- ferential adaptation of glucocorticoid sensitivity of peripheral blood mononuclear leukocytes in patients with sepsis or septic shock. J Clin Endocrinol Metab 1995 Jun;80(6):1799-803.
7. Annane D, Bellissant E, Bollaert PE, Briegel J, Confalonieri M, De Gaudio R, et al. Cor- ticosteroids in the treatment of severe sepsis and septic shock in adults: a systematic review. JAMA 2009 Jun 10;301(22):2362-75.
8. Volbeda M, Wetterslev J, Gluud C, Zijlstra JG, van der Horst IC, Keus F. Glucocorticosteroids for sepsis: systematic review with meta-analysis and trial se- quential analysis. Intensive Care Med 2015 Jul;41(7):1220-34.
9. Venkatesh B, Finfer S, Cohen J, Rajbhandari D, Arabi Y, Bellomo R, et al. Adjunctive glucocorticoid therapy in patients with septic shock. N Engl J Med 2018 Mar 1;378 (9):797-808.
10. Annane D, Renault A, Brun-Buisson C, Megarbane B, Quenot JP, Siami S, et al. Hydro- cortisone plus fludrocortisone for adults with septic shock. N Engl J Med 2018 Mar 1; 378(9):809-18.
11. American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: definitions for sepsis and organ failure and guidelines for the use of in- novative therapies in sepsis, Crit Care Med 1992 Jun;20(6):864-74.
12. Higgins JP, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. Oxford: The Cochrane Collaboration. Updated March 2011 www. cochrane-handbook.org; 2011. Accessed 20 March 2011.
13. Annane D, Sebille V, Charpentier C, Bollaert PE, Francois B, Korach JM, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in pa- tients with septic shock. JAMA 2002 Aug 21;288(7):862-71.
14. Arabi YM, Aljumah A, Dabbagh O, Tamim HM, Rishu AH, Al-Abdulkareem A, et al. Low-dose hydrocortisone in patients with cirrhosis and septic shock: a randomized controlled trial. CMAJ 2010 Dec 14;182(18):1971-7.
15. Bollaert PE, Charpentier C, Levy B, Debouverie M, Audibert G, Larcan A. Reoccurrence of late septic shock with supraphysiologic doses of hydrocortisone. Crit Care Med 1998 Apr;26(4):645-50.
16. Briegel J, Forst H, Haller M, Schelling G, Kilger E, Kuprat G, et al. Stress doses of hy- drocortisone reverse hyperdynamic septic shock: a prospective, randomized, double-blind, single-center study. Crit Care Med 1999 Apr;27(4):723-32.
17. Cicarelli DD, Vieira JE, Bensenor FE. Early dexamethasone treatment for septic shock patients: a prospective randomized clinical trial. Sao Paulo Med J 2007;125(4): 237-41.
18. Confalonieri M, Urbino R, Potena A, Piattella M, Parigi P, Puccio G, et al. Hydrocorti- sone infusion for severe community-acquired pneumonia: a preliminary random- ized study. Am J Respir Crit Care Med 2005 Feb 1;171(3):242-8.
19. Gordon AC, Mason AJ, Perkins GD, Stotz M, Terblanche M, Ashby D, et al. The inter- action of vasopressin and corticosteroids in septic shock: a pilot randomized con- trolled trial. Crit Care Med 2014 Jun;42(6):1325-33.
20. Hu B, Li JG, Liang H, Zhou Q, Yu Z, Li L, et al. The effect of low-dose hydrocortisone on requirement of norepinephrine and Lactate clearance in patients with refractory sep- tic shock. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2009 Sep;21(9):529-31.
21. Keh D, Trips E, Marx G, Wirtz SP, Abduljawwad E, Bercker S, et al. Effect of hydrocor- tisone on development of shock among patients with severe sepsis: the HYPRESS randomized clinical trial. JAMA 2016 Nov 1;316(17):1775-85 (Jul 5;125(4):237-41).
22. Lv QQ, Gu XH, Chen QH, Yu JQ, Zheng RQ. Early initiation of low-dose hydrocortisone treatment for septic shock in adults: a randomized clinical trial. Am J Emerg Med 2017 Dec;35(12):1810-4.
23. Meduri GU, Golden E, Umberger R. Prospective double-blind randomized clinical trial on the effects of low-dose hydrocortisone infusion in patients with severe sep- sis. Chest 2009;136:45S.
24. Oppert M, Schindler R, Husung C, Offermann K, Graf KJ, Boenisch O, et al. Low-dose hydrocortisone improves shock reoccurrence and reduces cytokine levels in early hyperdynamic septic shock. Crit Care Med 2005 Nov;33(11):2457-64.
25. Rinaldi S, Adembri C, Grechi S, De Gaudio AR. Low-dose hydrocortisone during se- vere sepsis: effects on microalbuminuria. Crit Care Med 2006 Sep;34(9):2334-9.
26. Sprung CL, Annane D, Keh D, Moreno R, Singer M, Freivogel K, et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med 2008 Jan 10;358(2):111-24.
27. Tongyoo S, Permpikul C, Mongkolpun W, Vattanavanit V, Udompanturak S, Kocak M, et al. Hydrocortisone treatment in early sepsis-associated acute respiratory distress syndrome: results of a randomized controlled trial. Crit Care 2016 Oct 15;20(1):329.
28. Yildiz O, Doganay M, Aygen B, Guven M, Kelestimur F, Tutuu A. Physiological-dose Steroid therapy in sepsis [ISRCTN36253388]. Crit Care 2002 Jun;6(3):251-9.
29. Yildiz O, Tanriverdi F, Simsek S, Aygen B, Kelestimur F. The effects of moderate-dose steroid therapy in sepsis: a placebo-controlled, randomized study. J Res Med Sci 2011;16:1410-21.
30. Laviolle B, Annane D, Fougerou C, Bellissant E. Gluco- and mineralocorticoid biolog- ical effects of a 7-day treatment with low doses of hydrocortisone and fludrocortisone in septic shock. Intensive Care Med 2012 Aug;38(8):1306-14.
31. Liu C, Zhao Q, Zhen Y, Zhai J, Liu G, Zheng M, et al. Effect of corticosteroid on renal water

    and sodium excretion in symptomatic heart failure: prednisone for renal function im- provement evaluation study. J Cardiovasc Pharmacol 2015 Sep;66(3):316-22.

    Annane D, Bellissant E, Sebille V, Lesieur O, Mathieu B, Raphael JC, et al. Impaired pressor sensitivity to noradrenaline in septic shock patients with and without im- paired adrenal function reserve. Br J Clin Pharmacol 1998 Dec;46(6):589-97.

32. Prager R, Kirsh MM, Dunn E, Nishiyama R, Straker J, Lee R, et al. The benefits of cor- ticosteroids in endotoxic shock. Ann Thorac Surg 1975;19(2):142-52.
33. Patel GP, Balk RA. Systemic steroids in severe sepsis and septic shock. Am J Respir Crit Care Med 2012 Jan 15;185(2):133-9.
34. De Bosscher K, Vanden Berghe W, Haegeman G. Mechanisms of anti-inflammatory action and of immunosuppression by glucocorticoids: negative interference of acti- vated glucocorticoid receptor with transcription factors. J Neuroimmunol 2000 Sep 1;109(1):16-22.
35. Buchele GL, Silva E, Ospina-Tascon GA, Vincent JL, De Backer D. Effects of hydrocor- tisone on microcirculatory alterations in patients with septic shock. Crit Care Med 2009 Apr;37(4):1341-7.
36. Johannes T, Mik EG, Klingel K, Dieterich HJ, Unertl KE, Ince C. Low-dose dexamethasone-supplemented fluid resuscitation reverses endotoxin-induced acute renal failure and prevents cortical microvascular hypoxia. Shock 2009 May; 31(5):521-8.
37. Meduri GU, Muthiah MP, Carratu P, Eltorky M, Chrousos GP. Nuclear factor-kappaB- and glucocorticoid receptor alpha- mediated mechanisms in the regulation of systemic and pulmonary inflammation during sepsis and acute respiratory distress syndrome. Evidence for inflammation-induced target tissue resistance to glucocorti- coids. Neuroimmunomodulation 2005;12(6):321-38.
38. Kaufmann I, Briegel J, Schliephake F, Hoelzl A, Chouker A, Hummel T, et al. Stress doses of hydrocortisone in septic shock: beneficial effects on opsonization- dependent neutrophil functions. Intensive Care Med 2008 Feb;34(2):344-9.
39. Sprung CL, Annane D, Keh D, Moreno R, Singer M, Freivogel K, et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med 2008 Jan 10;358(2):111-24.
40. Klastersky J, Cappel R, Debusscher L. Effectiveness of betamethasone in management of severe infections. A double-blind study. N Engl J Med 1971 Jun 3;284(22): 1248-50.
41. Katsenos CS, Antonopoulou AN, Apostolidou EN, Ioakeimidou A, Kalpakou GT, Papanikolaou MN, et al. Hellenic sepsis study group. Early administration of hydro- cortisone replacement after the advent of septic shock: impact on survival and immuneresponse. Crit Care Med 2014 Jul;42(7):1651-7.
42. Annane D, Bellissant E, Bollaert PE, Briegel J, Confalonieri M, De Gaudio R, et al. Cor- ticosteroids in the treatment of severe sepsis and septic shock in adults: a systematic review. JAMA 2009 Jun 10;301(22):2362-75.
43. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al. Early goal- directed therapy collaborative group. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001 Nov 8;345(19):1368-77.
44. Volbeda M, Wetterslev J, Gluud C, Zijlstra JG, van der Horst IC, Keus F. Glucocorticosteroids for sepsis: systematic review with meta-analysis and trial se- quential analysis. Intensive Care Med 2015 Jul;41(7):1220-34.