Effects and safety of separate low-dose hydrocortisone use in patients with septic shock: A meta-analysis

Abstract

Objective:

The aim of this study was to explore the effects and safety of low-dose hydrocortisone in patients with septic shock.

Methods:

The PubMed, EMBASE, and Cochrane Central Register of Controlled Trials were searched from database inception until 1 August 2018. Two reviewers performed literature selection, data extraction, and quality evaluation independently.

Results:

Twelve randomized controlled trials were included in this meta-analysis. The combined results showed that low-dose hydrocortisone use had no survival benefit in patients with septic shock (relative risk = 1.09; 95% confidence interval = 0.88–1.05; P = 0.37). But low-dose hydrocortisone use was useful for shock reverse (relative risk = 1.09; 95% confidence interval = 1.00–1.19; P = 0.04) and in shortening the time of vasopressor support (weighted mean difference = −1.79, 95% confidence interval = −2.05 to −1.52; P < 0.00001). In addition, use of low-dose hydrocortisone was associated with a higher risk of hyperglycemia (relative risk = 1.21; 95% confidence interval = 1.04–1.40; P = 0.01) and hypernatremia (relative risk = 6.34; 95% confidence interval = 1.19–33.81; P = 0.03). There was no significant improvement of intensive care unit mortality (relative risk = 1.11; 95% confidence interval = 0.93–1.33; P = 0.23) or hospital mortality (relative risk = 1.08; 95% confidence interval = 0.94–1.24; P = 0.29), length of intensive care unit (weighted mean difference = −1.84; 95% confidence interval = −5.80 to 2.11; P = 0.36) or length of hospital (weighted mean difference = 0.11; 95% confidence interval = −2.06 to 2.29; P = 0.98), and time of mechanical support (weighted mean difference = −0.69; 95% confidence interval = −1.76 to −0.38; P = 0.20) with the use of low-dose hydrocortisone. There was no significant difference in secondary infection (relative risk = 1.04; 95% confidence interval = 0.91–1.18; P = 0.57), recurrence of shock (relative risk = 1.47; 95% confidence interval = 0.64–3.39; P = 0.36), and gastrointestinal bleeding (relative risk = 1.41; 95% confidence interval = 0.89–2.22; P = 0.14) with the use of low-dose hydrocortisone.

Conclusion:

Although there was no effect of low-dose hydrocortisone on survival of patients with septic shock, it is associated with a higher rate of shock reversal and shortening duration of vasopressor support; thus, low-dose hydrocortisone may be an alternative drug in septic shock patients who are refractory to fluid resuscitation and vasopressors.

Keywords

Low-dose hydrocortisone septic shock meta-analysis

Background

Septic shock is the most common diagnosis associated with poor outcome.¹ Nevertheless, besides early fluid resuscitation and antibiotics use, no consensus reached by physicians on other pharmacologic treatments. Glucocorticoid is one of hot topics. High-dose glucocorticoids used in septic shock patients were reported to be associated with worse outcome.² In 2004, Annane et al.³ found that low-dose glucocorticoid could decrease the 28-day mortality rate. Hydrocortisone is the most reported drug. But, the positive effects of low-dose glucocorticoid were not confirmed by following studies.⁴ The aim of this study was to explore the effects and safety of low-dose hydrocortisone alone in patients with septic shock by meta-analysis.

Methods

Registration and protocol

The present meta-analysis was registered in PROSPERO (https://www.crd.york.ac.uk/PROSPERO/), and registration number is CRD42018090792. The conduct and reporting of the study follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

Eligibility criteria

Patients: The patients are septic shock patients. Septic shock was defined according to the 1992 American College of Chest Physicians/Society of Critical Care Medicine (ACCP/SCCM) Consensus (before 2016),⁵ or the Third International Consensus Definitions for Sepsis and Septic Shock.⁶

Intervention: Low-dose hydrocortisone, a daily dose of hydrocortisone ≤300 mg.⁷

Comparisons: Placebo or standard treatment irrespective of dose, duration, intravenous infusion method, or co-intervention.

Outcomes: The primary endpoint was 28-day mortality; The secondary endpoints included long-term mortality, reversal of shock, length of stay, renal replacement therapy, mechanical ventilation, duration of vasopressor support, and related adverse events. Reversal of shock was defined as the maintenance of a systolic blood pressure of at least 90 mm Hg without vasopressor support for at least 24 h or as described in each included study.⁷

Study type: Randomized clinical trials or quasi-randomized controlled trials.

Data source

PubMed, EMBASE, and Cochrane library were searched for eligible studies from its inception until 1 August 2018. The combination of Medical subject headings (MeSH) and text word search was used in this study.^8–10 Detailed literature search strategy was in Appendix 1. Ongoing and unpublished clinical trials were searched via http://www.isrctn.com, ClinicalTrials.gov, and the World Health Organization trial registry. In addition, the references of eligible papers were screened to obtain additional articles.

Literature selection and data extraction

Two reviewers performed literature selection through three steps: (1) excluding duplicates, (2) excluding clearly non-relevant studies by reading titles and abstracts, and (3) selecting eligible studies for meta-analysis by reading the remaining studies in full text. Two reviewers performed data extraction using a pre-defined sheet and the characteristics of each study, the baseline information of included patients, items necessary for methodological evaluation, and the primary and secondary endpoints. Any disagreement was solved by discussion or consulting with the third reviewer.

Quality evaluation

Two reviewers evaluated the quality of included studies independently using the Cochrane Risk Assessment Tool.¹¹ The following domains were evaluated: sequence generation, allocation concealment, blinding of patients and personnel, blinding of outcome assessors, incomplete outcome data, and selective reporting. Sequence generation will be considered adequate if the study explicitly described an appropriate randomization procedure to generate an unpredictable sequence of allocation, including computerized randomization, use of random number tables, and coin tossing. Concealment of allocation will be considered adequate if specific methods to protect allocation were documented and implemented. Performance bias will be considered low if a study reported participant, caregiver, and researcher blinding. Blinding of outcome assessment will be considered adequate if outcome assessors and adjudicators were blinded. And each domain was assessed as adequate, inadequate, or unclear.

Meta-analysis

The risk ratio was used for categorical variables and mean deviation was used for continuous variables. Heterogeneity between studies was explored using the I² test and chi-square test. I² > 50 or P < 0.1 indicates significant heterogeneity and random-effect model would be applied. Otherwise, fix-effect model was applied. Publication bias was evaluated by funnel plot and Egger’s test. In some studies, only median was reported, and we used the approach that proposed by Wan et al.¹² and Luo et al.¹³ to estimate the mean and standard deviation. One pre-specified subgroup analysis was performed based on adrenal responsiveness (corticotropin stimulation test responders vs. non-responders).

Results

A total of 1777 citations were obtained. Finally, 12 studies with 5400 patients met eligibility criteria.^14–25 Figure 1 shows the process of literature search and reasons for exclusion of studies. The reason of excluded literature was in Appendix 2. Among them, five trials were multicenter.^{15,17,18,23,25} In most studies, low-dose hydrocortisone treatment was compared with placebo. And in the study by Hu et al.,¹⁹ low-dose hydrocortisone treatment was compared with standard treatment. Ten trials^{14,15,17,18,20–25} reported 28-day mortality, four trials^16–18,23 reported intensive care unit (ICU) mortality, and five trials^{14,17–19,23} reported hospital mortality. Table 1 shows an overview of the detailed information of the included studies. Table 2 shows the excluded reason of these excluded studies. Table 3 shows the summary of risk of bias assessments for all included studies.

Figure 1.

Flow diagram of the literature search.

Table 1.

Excluded literature.

First author	Publication year	Exclusion reason
Chawla²⁶	1999	Only abstract
Annane²⁷	2002	Not only hydrocortisone
Yildiz²⁸	2002	Not only hydrocortisone, not septic shock patients
Ngaosuwan²⁹	2017	Study method did not match, comparison of steroid effects with different dosage
Mirea³⁰	2014	Only abstract
Bennett Jr³¹	1962	Only abstract
Meduri³²	2009	Only abstract
Annane³³	2010	Not only hydrocortisone
Annane³⁴	2018	Not only hydrocortisone
Schumer³⁵	1976	High dose
Keh³⁶	2016	Not septic shock patients
Cicarelli³⁷	2006	Not hydrocortisone
Yildiz³⁸	2011	Not low dose, not hydrocortisone
Tandan³⁹	2005	Not hydrocortisone
Aboab⁴⁰	2008	No relevant data
Kurugundla⁴¹	2008	No relevant data
Deng⁴²	2011	Not hydrocortisone
Torres⁴³	2015	Not hydrocortisone, not septic shock patients

Table 2.

Characteristics of the included trials.

Author	Year	No. of patients	Intervention	Duration	Control	Primary outcome
Mussack et al.²¹	2005	24	100 mg bolus for 30 min then 0.18 mg/kg/h	6 days	Placebo	Median S-100B serum levels
Sprung et al.²³	2008	500	50 mg q 6 h ivi for 5 days 50 mg q 12 h ivi for 3 days 50 mg daily ivi for 3 days	11 days	Placebo	Mortality at 28 days
Arabi et al.¹⁴	2010	75	50 mg q 6 h ivi	Shock reverse	Placebo	Mortality at 28 days
Briegel et al.¹⁶	1999	40	100 mg bolus Then 0.18 mg/kg/h Then 0.08 mg/kg/h	Duration of vasopressor therapy	Placebo	Time to shock reverse
Bollaert et al.¹⁵	1998	41	100 mg q 8 h ivi	≥5 days	Placebo	Shock reverse
Oppert et al.²²	2005	48	50 mg bolus followed by 0.18 mg/kg/h	Until resolution of shock	Placebo	Time to cessation of vasopressor support
Lv et al.²⁰	2017	118	200 mg ivi	6 days	Placebo	Mortality at 28 days
Tongyoo et al.²⁴	2016	206	50 mg q 6 h ivi	7 days	Placebo	Mortality at 28 days
Gordon et al.¹⁷	2014	61	50 mg ivi q 6 h for 5 days 50 mg ivi q 12 h for 3 days 50 mg ivi daily for 3 days	11 days	Placebo	Vasopressin concentrations
Venkatesh et al.²⁵	2018	3800	200 mg ivi	7 days	Placebo	Mortality at 90 days
Gordon et al.¹⁸	2016	409	50 mg ivi q 6 h for 5 days 50 mg ivi q 12 h for 3 days 50 mg ivi daily for 3 days	11 days	Placebo	Kidney failure free-days to day 28
Hu et al.¹⁹	2009	77	50 mg ivi q 6 h for 7 days 50 mg ivi q 8 h for 3 days 50 mg ivi q 12 h for 2 days 50 mg ivi daily for 2 days	14 days	Standard therapy	Mortality

No: number; min: minutes; kg: kilogram; h: hours; q: quaque; ivi: intravenous.

Table 3.

The methodological qualities of included studies.

Source	Sequence Generation	Allocation Concealment	Blinding of participants and personnel	Blinding of outcome assessment	Incomplete Outcome Data	Selective Reporting	Other Bias
Mussack et al.²¹	Unclear	Unclear	Adequate	Unclear	Adequate	Adequate	Unclear
Sprung et al.²³	Adequate	Adequate	Adequate	Adequate	Adequate	Adequate	Adequate
Arabi et al.¹⁴	Adequate	Adequate	Adequate	Unclear	Unclear	Unclear	Adequate
Briegel et al.¹⁶	Adequate	Adequate	Adequate	Adequate	Adequate	Unclear	Adequate
Bollaert et al.¹⁵	Adequate	Adequate	Adequate	Adequate	Adequate	Adequate	Inadequate
Oppert et al.²²	Unclear	Adequate	Adequate	Adequate	Unclear	Unclear	Adequate
Lv et al.²⁰	Adequate	Inadequate	Adequate	Adequate	Adequate	Adequate	Adequate
Tongyoo et al.²⁴	Adequate	Adequate	Adequate	Adequate	Adequate	Unclear	Adequate
Gordon et al.¹⁷	Adequate	Adequate	Unclear	Unclear	Unclear	Adequate	Adequate
Venkatesh et al.²⁵	Adequate	Adequate	Adequate	Adequate	Adequate	Adequate	Adequate
Gordon et al.¹⁸	Adequate	Adequate	Adequate	Adequate	Adequate	Adequate	Adequate
Hu et al.¹⁹	Unclear	Unclear	Unclear	Unclear	Inadequate	Unclear	Inadequate

Primary outcome

28-day mortality

Ten studies^{14,15,17,18,20–25} and 5283 patients eligible for analysis. During the 28-day follow-up, 662 (25.1%) patients in the low-dose hydrocortisone group died and 689 (38.5%) patients in the control group died. The pooled relative risk (RR) was 0.96 (95% confidence interval (CI) = 0.88–1.05; P = 0.37) with no significant heterogeneity (I² = 21% by fixed-effects model) (Figure 2).

Figure 2.

Forrest plot showing the effect of low-dose hydrocortisone compared to placebo or control on 28-day mortality in adult septic shock patient.

Sensitive analyses

The result of 28-day mortality was driven by the study by Venkatesh et al.,²⁵ which carried 64.9% of the weight. A sensitivity analysis was performed by excluding the study by Venkatesh et al. And it did not significantly change the result (RR = 1.05; 95% CI = 0.91–1.21; P = 0.49; I² = 8%) (Appendix 2).

Subgroup analyses

In two trials,^15,23 the authors explored the different effects of hydrocortisone according to the responsiveness of patients. In the group of patients who were responsive to hydrocortisone, the combined RR was 0.80 (95% CI = 0.43–1.47), and in the non-responders, the combined RR was 0.86 (95% CI = 0.60–1.22) (Appendix 2).

Secondary outcomes

ICU mortality

Four trials^16–18,23 were eligible for this analysis. The ICU mortality in the intervention and control group was 171 of 504 (33.9%) patients versus 154 of 507 (30.4%) patients. The combined RR was 1.11 (95% CI = 0.93–1.33; P = 0.23; I² = 0%) (Table 3).

One-year mortality

Two trials^16,23 (n = 540) reported 1-year mortality. One-year mortality for treated versus control patients was 142 of 271 (52.4%) versus 133 of 269 (49.4%) (RR = 1.06; 95% CI = 0.90–1.25; P = 0.46; I² = 0%) (Table 3).

Shock reversal

Two trials^15,22 (n = 89) reported the shock reversal rate within 7 days, and five trials^{14,16,20,21,23} (n = 757) reported the shock reversal rate in entire treatment period. Shock reversal by day 7 for treated versus control patients was 27 of 45 (60.0%) versus 22 of 44 (50.0%) (RR = 1.46; 95% CI = 0.31–6.96; P = 0.63; I² = 89%). Shock reversal in entire treatment period for treated versus control patients was 291 of 380 (76.6%) versus 265 of 377 (70.3%) (RR = 1.09; 95% CI = 1.00–1.19; P = 0.04; I² = 6%) in the fixed-effects model (Figure 3); and RR = 1.09; 95% CI = 0.99–1.19; P = 0.07; I² = 6% in the randomized-effects model. Shock reversal in the whole seven trials for treated versus control patients was 318 of 425 (74.8%) versus 287 of 421 (68.2%) (RR = 1.11; 95% CI = 0.94–1.31; P = 0.23; I² = 53%) (Appendix 2).

Figure 3.

Forrest plot showing the effect of low-dose hydrocortisone compared to placebo or control on shock reversal in five trials.

Time to resolution of shock

Three trials^16,23,25 (n = 4340) reported the time to resolution of shock. The combined weighted mean difference (WMD) was −0.80 days (95% CI = −1.48 to −0.11; P = 0.02; I² = 91%) (Table 3). The source of heterogeneity might be due to the trial of Sprung et al.²³ A sensitivity analysis was performed by excluding the study by Sprung et al., and the results did not change the overall conclusion (WMD −0.25 days, 95% CI = −0.34 to −0.15; P < 0.001; I² = 0%).

Duration of vasopressor support

Five trials^{15,19,21,22,24} (n = 395) reported the duration of vasopressor support of two groups. Among them, the mean and standard deviation were reported in two studies,^19,24 and mean and quartile were reported in three studies.^15,21,22 The combined WMD was −1.78 days (95% CI = −2.04 to −1.52; P < 0.001; I² = 0%) (Figure 4). Two trials^19,24 reported average time of vasopressor also showed hydrocortisone can reduce the time of vasopressor support (WMD = −1.79, 95% CI = −2.05 to −1.52; P < 0.001; I² = 0%) (Appendix 2).

Figure 4.

Forrest plot showing the effect of low-dose hydrocortisone compared to placebo or control on time of vasopressor support in adult septic shock patient.

Renal replacement therapy

Five trials^{16,18,21,24,25} reported the number of patients requiring renal replacement therapy. The reported renal replacement therapy rates for treated versus control patients were 652 of 2236 (29.2%) patients versus 714 of 2243 (31.8%) patients, and the pooled RR was 0.92 (95% CI = 0.84–1.00; P = 0.05; I² = 0%) (Table 3). And the outcome was driven by the study by Venkatesh et al.,²⁵ which carried 85.9% of the weight. A sensitivity analysis was performed after removing the study by Venkatesh et al. (RR = 0.82; 95% CI = 064–1.04; P = 0.10; I² = 0%) (Appendix 2).

Serious adverse events

Recurrence of shock was reported in two trials^14,25 (n = 3875 patients), and the pooled RR was 1.47 (95% CI = 0.64–3.39; P = 0.36; I² = 72%) (Table 4). Gastrointestinal bleeding was reported in six trials^{14–16,23–25} (n = 4662 patients), and the pooled RR was 1.41 (95% CI = 0.89–2.22; P = 0.14; I² = 5%) (Table 3). Secondary infections were reported in five trials^{14,15,23–25} and the pooled RR was 1.04 (95% CI = 0.91–1.18; P = 0.57; I² = 0%) (Table 4). Hyperglycemia was reported in three trials,^14,24,25 and the pooled RR was 1.21 (95% CI = 1.04–1.40; P = 0.01; I² = 0%) (Table 4). Hypernatremia was reported in two studies,^16,25 and the pooled RR was 6.34 (95% CI = 1.19–33.81; P = 0.03; I² = 0%) (Table 4).

Table 4.

The effect of hydrocortisone on tertiary outcomes in patients with septic shock.

Outcome	No. of trials	No. of patients	Treatment effect	95% CI	P	I² (%)
ICU mortality	4	1011	RR 1.11	0.93 to 1.33	0.23	0
30-day mortality	2	281	RR 0.87	0.67 to 1.12	0.28	0
Hospital mortality	5	1123	RR 1.08	0.94 to 1.24	0.29	0
One-year mortality	2	540	RR 1.06	0.90 to 1.25	0.46	0
Time to resolution of shock	3	4340	MD −0.80	−1.48 to −0.11	0.02	0
Time to cessation of mechanical ventilation	3	4030	MD −0.69	−1.76 to −0.38	0.2	0
Length of ICU stay	4	770	MD −1.84	−5.80 to 2.11	0.36	55
Length of hospital stay	3	693	MD 0.11	−2.06 to 2.29	0.98	0
Recurrence of shock	2	3875	RR 1.47	0.64 to 3.39	0.36	72
Gastrointestinal bleeding	6	4662	RR 1.41	0.89 to 2.22	0.14	5
Secondary infections	5	4622	RR 1.04	0.91 to 1.18	0.57	0
Renal replacement therapy	5	4479	RR 0.92	0.84 to 1.00	0.05	0
Hyperglycemia	3	4081	RR 1.21	1.04 to 1.40	0.01	0
Hypernatremia	2	3840	RR 6.34	1.19 to 33.81	0.03	0

No: number; CI: confidence interval; ICU: intensive care unit; RR: relative risk; MD: mean difference.

Discussion

The results of the present study showed that low-dose hydrocortisone has no significant effects on survival in patients with septic shock. However, use of low-dose hydrocortisone could shorten the duration of vasopressor support and be associated with a higher rate of shock reversal. Meanwhile, we found that low-dose hydrocortisone could reduce the occurrence of renal replacement therapy. This may be related to less use of vasopressor in the intervention group. In addition, hydrocortisone was associated with an increased risk of developing hyperglycemia and hypernatremia. There were no significant effects of low-dose hydrocortisone treatment on ICU/hospital length of stay, gastrointestinal bleeding, and secondary infection.

The pathophysiology of septic shock involves complicated interactions between infectious organisms and host that can result in multi-system organ failure and death.⁴⁴ Corticosteroids may be beneficial in patients with septic shock due to its multi-effect.^27,45–47 However, in view of the current evidence, the effects of corticosteroids on these patients remain unclear.^25,34 In 1951, Hahn et al.⁴⁸ first reported the therapeutic effect of hydrocortisone on severe infection. From the 1950s to 1980s, high-dose corticosteroids like methylprednisolone (30 mg/kg) or dexamethasone (3–6 mg/kg) for 1 to 2 days were accepted in adult septic shock patients.⁴⁹ Although Schumer et al.³⁵ found that high-dose dexamethasone and methylprednisolone (corticosteroid replacement therapy) could significantly reduce the mortality of patients with septic shock in 1976, their conclusion did not corroborate in the following studies. And high-dose corticosteroids even might increase the mortality rate.^2,50,51 Then corticosteroid replacement therapy stopped. In the late 1990s, Bollaert et al.¹⁵ reported low-dose hydrocortisone could significantly improve the hemodynamic status and facilitate shock reversal in septic shock patients. And their results were confirmed in several following studies (usually 200–300 mg of hydrocortisone per day for 5–7 days).^27,49,52,53 So, low-dose hydrocortisone was recommended in septic shock patients by the 2004 Surviving Sepsis Campaign guidelines.⁵⁴ But following studies reported inconsistent results of low-dose hydrocortisone on mortality.^23,25,27,34 Our study showed low-dose hydrocortisone could shorten the duration of vasopressor use and was beneficial for shock reversal. But there was no statistical effect of low-dose hydrocortisone on mortality. Previous meta-analyses focusing on corticosteroids in septic shock patients also reached inconsistent conclusion.^7,52,55,56 Rochwerg et al.⁵⁶ in 2018 and Annane et al.⁷ in 2009 in their meta-analysis reported that corticosteroids were associated with a higher survival rate when compared with patients in the control group. Different inclusion criteria may be responsible for this inconsistent result. In these two studies, the authors did not focus on patients with septic shock. However, in the present study, only septic shock patients were eligible for meta-analysis. Then, in the Rochwerg⁵⁶ study, besides hydrocortisone, other corticosteroids were also analyzed in combination. And in the present study, only the effects of hydrocortisone were analyzed to minimize the heterogeneity.

Our results are similar to those of Rygard et al.⁴⁵ who demonstrated that corticosteroid therapy does not reduce mortality rates but does appear to consistently reduce the time to shock reversal. Eleven of these studies met our inclusion criteria.^{14–18,20–25} It may be the reason that our results are similar to the study of Rygard, there was a little different in length of ICU stay between our study with the study of Rygard et al., it may be associated with the different inclusion criteria, our study only includes the trials of hydrocortisone alone, and specific length of ICU stay for each of these subpopulations was not reported.

The physiologic action of fludrocortisone is similar to that of hydrocortisone. However, the former has considerably heightened and prolonged effects on electrolyte balance and carbohydrate metabolism. Two large randomized controlled trials (RCTs) by Annane et al.^27,34 in 2002 and 2018 demonstrated the mortality beneficial in septic shock adults with the use of hydrocortisone combined with fludrocortisone. However, another large RCT named COIITSS study, also was launched by Annane et al.,³³ included 509 patients with septic shock adults, and found that the use of fludrocortisone in conjunction with hydrocortisone was associated with an increased risk of infection. The different outcomes may associate with the more severe degree of the septic shock in COIITSS study or the time to use corticosteroids. A meta-analysis by Rygard⁴⁵ reported a survival benefit of hydrocortisone plus fludrocortisone therapy in a subgroup of three trials. But their result was driven by the study by Annane,³⁴ and more studies are required.

As for the optimal time to use hydrocortisone, the Surviving Sepsis Campaign guidelines⁵⁷ suggest consideration of corticosteroids therapy for patients with septic shock who were refractory to fluids and vasopressors, based on the results of several studies.^7,25 In the study by Annane et al.²⁷ (50 mg intravenous bolus every 6 h and fludrocortisone 50 µg once daily), they enrolled patients within 8 h from onset of septic shock to randomization and they found mortality beneficial. In the APROCCHSS trial³⁴ (50 mg intravenous bolus every 6 h and fludrocortisone 50 µg once daily for 7 days), 1241 patients were randomized within less than 24 h from onset of septic shock, and they also reported mortality beneficial. In the CORTICUS study²³ (50 mg of intravenous hydrocortisone every 6 h for 5 days) and ADRENAL trial²⁵ (200 mg hydrocortisone per day for 7 days), patients were randomized 24 to 72 h and 20 to 90 h after the onset of septic shock and they did not found mortality beneficial. A retrospective study by Park et al.⁵⁸ (≤300 mg/day of hydrocortisone or equivalent) in 2012 showed that septic shock patients in the early therapy group (within 6 h after the onset of septic shock) had a lower mortality rate than those in the late therapy group. Another prospective study by Katsenos⁵⁹ (daily administration of 50 mg q 6 h for 7 days) in 2014 also reported early therapy with hydrocortisone within 9 h after the onset of shock was associated with a lower mortality rate than those received hydrocortisone more than 9 h after the onset of shock. It seems that there is a trend of decreasing mortality rate of early use of hydrocortisone. But high-quality RCTs focusing on early use versus late use of hydrocortisone are scarce and more evidence are needed. Keh et al.³⁶ explored the effects of hydrocortisone on the development of septic shock in patients with severe sepsis. And they found hydrocortisone therapy in patients with severe sepsis can not reduce the risk of development of septic shock within 14 days. This evidence does not support the preventive use of hydrocortisone in sepsis patients. As for the dosing strategy of hydrocortisone, common regimens for corticosteroids include hydrocortisone 100 mg intravenous bolus every 8 h or 50 mg intravenous bolus every 6 h. Another option for providers is 100 mg intravenous bolus followed by an infusion of 0.18 mg/kg/h. Lots of studies reported there was no mortality beneficial between continuous infusion and bolus infusion, but the glycemic controls well when hydrocortisone is administered as a continuous infusion as compared with intermittent bolus.⁴⁴

The strength and limitation of this meta-analysis

In order to minimize potential bias, only randomized controlled trials evaluating low-dose hydrocortisone in septic shock patients were eligible. And trials focusing on other types of corticosteroids were excluded, and observational studies were also excluded. Several limitations of this meta-analysis should be concerned. First, different septic shock diagnostic criteria, different administration route, and different hydrocortisone dose may bias the results of this meta-analysis. But due to the limited sample size, we cannot perform further subgroup analysis. Second, the interaction between corticosteroids and other drugs like vasopressin is not evaluated in the present study. It has been reported that corticosteroids and vasopressin used together might improve the prognosis of patients with septic shock.⁵³

The implications for clinical practice and further studies

Based on the results of this study, low-dose hydrocortisone is beneficial for hemodynamic status in septic shock patients, although there is no association between use of low-dose hydrocortisone and survival rate. Meanwhile, use of low-dose hydrocortisone did not result in a higher risk of serious adverse events. Thus, if there are no obvious contraindications, including myopathy with neuromuscular blocking agents, diabetic ketoacidosis, human immunodeficiency virus infection, etc,⁵⁹ low-dose hydrocortisone can be considered as an adjunctive treatment in patients with septic shock. In future studies, the septic shock should be confirmed using the newest criteria. Based on the results of this study, more studies are required on economic impact of utilizing hydrocortisone in patients with septic shock. It is essential to standardize the timing, method, discontinuation of hydrocortisone administration, and adjunctive therapy in septic shock patients. In addition, long-term endpoints and functional rehabilitation should be the primary outcome.

Conclusion

The results of this study show no survival benefit of low-dose hydrocortisone in patients with septic shock, but it is associated with a higher rate of shock reversal and shortening duration of vasopressor support. Thus, low-dose hydrocortisone may be an alternative drug in septic shock patients who are refractory to fluid resuscitation and vasopressors. Further studies focusing on precision medicine of hydrocortisone in septic shock patients are needed.

Footnotes

Appendix 1

Appendix 2 Acknowledgements

My deepest gratitude goes first and foremost to Professor Ma, my supervisor, for his constant encouragement and guidance. Second, I would like to express my heartfelt gratitude to my senior fellow, who gave me a lot of help. I am also greatly indebted to my alma mater—Zhejiang University. Finally, our deepest gratitude go to the anonymous reviewers for their careful work and thoughtful suggestions that have helped improve this paper substantially.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Abraham

Reinhart

Opal

, et al. Efficacy and safety of tifacogin (recombinant tissue factor pathway inhibitor) in severe sepsis: a randomized controlled trial. JAMA 2003; 290: 238–247.

Cronin

Cook

Carlet

, et al. Corticosteroid treatment for sepsis: a critical appraisal and meta-analysis of the literature. Crit Care Med 1995; 23(8): 1430–1439.

Annane

Bellissant

Bollaert

, et al. Corticosteroids for severe sepsis and septic shock: a systematic review and meta-analysis. BMJ 2004; 329(7464): 480.

Volbeda

Wetterslev

Gluud

, et al. Glucocorticosteroids for sepsis: systematic review with meta-analysis and trial sequential analysis. Intensive Care Med 2015; 41(7): 1220–1234.

Bone

Balk

Cerra

, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992; 101: 1644–1655.

Singer

Deutschman

Seymour

, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016; 315: 801–810.

Annane

Bellissant

Bollaert

, et al. Corticosteroids in the treatment of severe sepsis and septic shock in adults: a systematic review. JAMA 2009; 301(22): 2362–2375.

Haynes

McKibbon

Wilczynski

, et al. Optimal search strategies for retrieving scientifically strong studies of treatment from Medline: analytical survey. BMJ 2005; 330(7501): 1179.

Wilczynski

McKibbon

Walter

, et al. MEDLINE clinical queries are robust when searching in recent publishing years. J Am Med Inform Assoc 2013; 20(2): 363–368.

10.

Wong

Wilczynski

Haynes

RB.

Developing optimal search strategies for detecting clinically sound treatment studies in EMBASE. J Med Libr Assoc 2006; 94(1): 41–47.

11.

Higgins

Altman

Gøtzsche

, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011; 343: d5928.

12.

Wan

Wang

Liu

, et al. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol 2014; 14: 135.

13.

Luo

Wan

Liu

, et al. Optimally estimating the sample mean from the sample size, median, mid-range, and/or mid-quartile range. Stat Methods Med Res 2018; 27(6): 1785–1805.

14.

Arabi

Aljumah

Dabbagh

, et al. Low-dose hydrocortisone in patients with cirrhosis and septic shock: a randomized controlled trial. CMAJ 2010; 182(18): 1971–1977.

15.

Bollaert

Charpentier

Levy

, et al. Reversal of late septic shock with supraphysiologic doses of hydrocortisone. Crit Care Med 1998; 26(4): 645–650.

16.

Briegel

Forst

Haller

, et al. Stress doses of hydrocortisone reverse hyperdynamic septic shock: a prospective, randomized, double-blind, single-center study. Crit Care Med 1999; 27(4): 723–732.

17.

Gordon

Mason

Perkins

, et al. The interaction of vasopressin and corticosteroids in septic shock: a pilot randomized controlled trial. Crit Care Med 2014; 42(6): 1325–1333.

18.

Gordon

Mason

Thirunavukkarasu

, et al. Effect of early vasopressin vs norepinephrine on kidney failure in patients with septic shock: the VANISH randomized clinical trial. JAMA 2016; 316(5): 509–518.

19.

Liang

, et al. The effect of low-dose hydrocortisone on requirement of norepinephrine and lactate clearance in patients with refractory septic shock. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2009; 21(9): 529–531.

20.

Chen

, et al. Early initiation of low-dose hydrocortisone treatment for septic shock in adults: a randomized clinical trial. Am J Emerg Med 2017; 35(12): 1810–1814.

21.

Mussack

Briegel

Schelling

, et al. Effect of stress doses of hydrocortisone on S-100B vs. interleukin-8 and polymorphonuclear elastase levels in human septic shock. Clin Chem Lab Med 2005; 43(3): 259–268.

22.

Oppert

Schindler

Husung

, et al. Low-dose hydrocortisone improves shock reversal and reduces cytokine levels in early hyperdynamic septic shock. Crit Care Med 2005; 33: 2457–2464.

23.

Sprung

Annane

Keh

, et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med 2008; 358: 111–124.

24.

Tongyoo

Permpikul

Mongkolpun

, et al. Hydrocortisone treatment in early sepsis-associated acute respiratory distress syndrome: results of a randomized controlled trial. Crit Care 2016; 20(1): 329.

25.

Venkatesh

Finfer

Cohen

, et al. Adjunctive glucocorticoid therapy in patients with septic shock. N Engl J Med 2018; 378: 797–808.

26.

Chawla

Tessler

Hydrocortisone reverses refractory septic shock. Crit Care Med. 1999; 27: A33.

27.

Annane

Sébille

Charpentier

, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA 2002; 288: 862–871.

28.

Yildiz

Doganay

Aygen

, et al. Physiological-dose steroid therapy in sepsis [ISRCTN36253388]. Crit Care 2002; 6(3): 251–259.

29.

Ngaosuwan

Ounchokdee

Chalermchai

Clinical outcomes of minimized hydrocortisone dosage of 100 mg/day on lower occurrence of hyperglycemia in septic shock patients. Shock 2018; 50(3): 280–285.

30.

Mirea

Pavelescu

Grintescu

, et al. Continuous administration of corticosteroids in septic shock can reduce risk of hypernatremia. Crit Care 2014; 18: P239.

31.

Bennett

Jr Finland

Hamburger

, et al. A double-blind study of the effectiveness of cortisol in the management of severe infections. T Assoc Am Phys 1962; 75: 198–207.

32.

Meduri

Golden

Umberger

Prospective double-blind randomized clinical trial on the effects of low-dose hydrocortisone infusion in patients with severe sepsis. Chest 2009; 136: 45S.

33.

Annane

Cariou

Maxime

, et al. Corticosteroid treatment and intensive insulin therapy for septic shock in adults: a randomized controlled trial. JAMA 2010; 303(4): 341–348.

34.

Annane

Renault

Brun-Buisson

, et al. Hydrocortisone plus fludrocortisone for adults with septic shock. N Engl J Med 2018; 378: 809–818.

35.

Schumer

Steroids in the treatment of clinical septic shock. Ann Surg 1976; 184(3): 333–341.

36.

Keh

Trips

Marx

, et al. Effect of hydrocortisone on development of shock among patients with severe sepsis: the HYPRESS randomized clinical trial. JAMA 2016; 316(17): 1775–1785.

37.

Cicarelli

Bensenor

Vieira

JE.

Effects of single dose of dexamethasone on patients with systemic inflammatory response. Sao Paulo Med J 2006; 124(2): 90–95.

38.

Yildiz

Tanriverdi

Simsek

, et al. The effects of moderate-dose steroid therapy in sepsis: a placebo-controlled, randomized study. J Res Med Sci 2011; 16(11): 1410–1421.

39.

Tandan

SMGR

Gupta

. Low dose steroids and adrenocortical insufficiency in septic shock: a double-blind randomised controlled trial from India. In: Proceedings of the American Thoracic Society meeting, New York, 20 May–25 May 2005, p. A24. New York: American Thoracic Society.

40.

Aboab

Polito

Orlikowski

, et al. Hydrocortisone effects on cardiovascular variability in septic shock: a spectral analysis approach. Crit Care Med 2008; 36(5): 1481–1486.

41.

Kurugundla

NIL

Kilari

. Effect of steroids in septic shock patients without relative adrenal insufficiency—a pilot study. In: American Thoracic Society international conference, Toronto, ON, Canada, 16–21 May 2008, p. A116. New York: American Thoracic Society.

42.

Deng

Shang

Wan

XY.

Effects of low-dose glucocorticoid on renal function and prognosis in patients with sepsis. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2011; 23(3): 183–184.

43.

Torres

Sibila

Ferrer

, et al. Effect of corticosteroids on treatment failure among hospitalized patients with severe community-acquired pneumonia and high inflammatory response: a randomized clinical trial. JAMA 2015; 313(7): 677–686.

44.

Marik

PE.

The role of glucocorticoids as adjunctive treatment for sepsis in the modern era. Lancet Respir Med 2018; 6(10): 793–800.

45.

Rygard

Butler

Granholm

, et al. Low-dose corticosteroids for adult patients with septic shock: a systematic review with meta-analysis and trial sequential analysis. Intensive Care Med 2018; 44(7): 1003–1016.

46.

Russell

Rush

Boyd

Pathophysiology of septic shock. Crit Care Clin 2018; 34: 43–61.

47.

Boyer

Chadda

Salah

, et al. Glucocorticoid treatment in patients with septic shock: effects on vasopressor use and mortality. Int J Clin Pharmacol Ther 2006; 44(7): 309–318.

48.

Hahn

Houser

Rammelkamp

Jr , et al. Effect of cortisone on acute streptococcal infections and poststreptococcal complications. J Clin Invest 1951; 30(3): 274–281.

49.

Patel

Balk

RA.

Systemic steroids in severe sepsis and septic shock. Am J Respir Crit Care Med 2012; 185(2): 133–139.

50.

Bone

Fisher

Jr Clemmer

, et al. A controlled clinical trial of high-dose methylprednisolone in the treatment of severe sepsis and septic shock. N Engl J Med 1987; 317(11): 653–658.

51.

Batzofin

Sprung

Weiss

YG.

The use of steroids in the treatment of severe sepsis and septic shock. Best Pract Res Clin Endocrinol Metab 2011; 25: 735–743.

52.

Wang

Sun

Zheng

, et al. Low-dose hydrocortisone therapy attenuates septic shock in adult patients but does not reduce 28-day mortality: a meta-analysis of randomized controlled trials. Anesth Analg 2014; 118: 346–357.

53.

Tan

Chen

, et al. Vasopressin and its analog terlipressin versus norepinephrine in the treatment of septic shock: a meta-analysis. Int J Clin Exp Med 2016; 9(7): 14183–14190.

54.

Dellinger

Carlet

Masur

, et al. Surviving Sepsis Campaign guidelines for management of severe sepsis and septic shock. Crit Care Med 2004; 32: 858–873.

55.

Moreno

Sprung

Annane

, et al. Time course of organ failure in patients with septic shock treated with hydrocortisone: results of the Corticus study. Intensive Care Med 2011; 37(11): 1765–1772.

56.

Rochwerg

Oczkowski

Siemieniuk

RAC

, et al. Corticosteroids in sepsis: an updated systematic review and meta-analysis. Crit Care Med 2018; 46: 1411–1420.

57.

Rhodes

Evans

Alhazzani

, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2016. Intensive Care Med 2017; 43: 304–377.

58.

Park

Suh

Song

, et al. Early initiation of low-dose corticosteroid therapy in the management of septic shock: a retrospective observational study. Crit Care 2012; 16(1): R3.

59.

Katsenos

Antonopoulou

Apostolidou

, et al. Early administration of hydrocortisone replacement after the advent of septic shock: impact on survival and immune response. Crit Care Med 2014; 42: 1651–1657.

60.

Long

Koyfman

Controversies in corticosteroid use for sepsis. J Emerg Med 2017; 53: 653–661.