16 low dose hydrocortisone therapy attenuates septic.16

346 www.anesthesia-analgesia.org February 2014 • Volume 118 • Number 2

Copyright © 2014 International Anesthesia Research SocietyDOI: 10.1213/ANE.0000000000000050

The incidence of septic shock can be as high as 20% among hospitalized patients.1 Even after the appropriate treat-ment is administered, mortality from septic shock

remains approximately 50%.2,3 Since the first publication of the use of glucocorticoids in severe infection,4 researchers have explored the use of steroids in septic shock. A half-century later, the role of glucocorticoids for decreasing mortality from septic shock remains controversial.5,6 A 1995 meta-analysis found that a short course of high-dose glucocorticoid therapy

provided no advantage for the treatment of septic shock and could have negative effects.7 A 2004 meta-analysis found that steroids did not affect mortality from septic shock.8 However, a subgroup analysis of patients treated with low doses of ste-roids (≤ 300 mg hydrocortisone or equivalent per day) for >5 days found that sustained low-dose steroid therapy reduced 28-day mortality. In 2009, the same group of researchers repeated a meta-analysis on the same issue by integrating recent randomized controlled trials (RCTs).9 The analysis was restricted only to the response to steroid therapy in an adult population with severe sepsis and septic shock. The results demonstrated that long-term, low-dose steroid therapy can increase short-term survival rates.

These studies involved analyses of various corticoste-roid therapies10,11 but did not focus on the effect of a single steroid therapy. Glucocorticoids differ in receptor binding, biological half-life, and glucocorticoid–mineralocorticoid hormone actions.12 Glucocorticoids may differ in their effi-cacy in septic shock. As hydrocortisone is the endogenous glucocorticoid released by the adrenal gland, it might be the best choice of replacement therapy in shock.

An initial literature search found that studies of low-dose corticosteroid therapy for septic shock accounted for most recent studies. Therefore, we investigated the effects of low-dose hydrocortisone on shock reversal and survival in patients with septic shock. We performed a conventional meta-analysis of published trials and a cumulative meta-analysis to evaluate the effects of each study on the final, generalized results.13

BACKGROUND: The role of low-dose hydrocortisone in attenuating septic shock and reducing short-term mortality in adult patients with septic shock is unclear. We conducted a meta-analysis of previous studies to determine whether hydrocortisone could ameliorate the effects of septic shock at 7 and 28 days and reduce 28-day morality.METHODS: Randomized controlled trials (RCTs) of corticosteroids versus placebo (or support-ive treatment alone) were retrieved from electronic searches (Medline, Embase, and Cochrane Library databases; LILACS; and Web of Knowledge) and manual searches (up to May 2012). From a pool of 1949 potentially relevant articles, duplicate independent review identified 10 rel-evant, RCTs of low-dose hydrocortisone therapy in septic shock. Four pairs of reviewers agreed on the criteria for trial eligibility. One reviewer entered the data into the computer, and 3 review-ers checked the data. Missing data were obtained from the authors of the relevant trials. The primary outcome analyzed was an estimate of 28-day mortality.RESULTS: Eight publications were included in the meta-analysis. Low-dose hydrocortisone ther-apy did not reduce 28-day mortality (N = 1063; odds ratio (OR) = 0.891, 95% confidence inter-val (CI), 0.69–1.15). Low-dose hydrocortisone therapy ameliorated shock at 7 days (6 RCTs, N = 964, OR = 2.078, 95% CI, 1.58–2.73, P < 0.0001, and I2 = 26.9%) and 28 days (6 RCTs, N = 947, OR = 1.495, 95% CI, 1.12–1.99, P = 0.006, and I2 = 0.0%).CONCLUSIONS: Although low-dose hydrocortisone therapy ameliorates septic shock at 7 and 28 days, it does not reduce 28-day mortality. (Anesth Analg 2014;118:346–57)

Low-Dose Hydrocortisone Therapy Attenuates Septic Shock in Adult Patients but Does Not Reduce 28-Day Mortality: A Meta-Analysis of Randomized Controlled TrialsChangsong Wang, MD,* Jiaxiao Sun, MSc,* Juanjuan Zheng, MSc,† Lei Guo, MD,* Hongyan Ma, MD,* Yang Zhang,* Fengmin Zhang, PhD,‡§ and Enyou Li, MD*

From the *Department of Anesthesiology, The First Affiliated Hospital of Harbin Medical University; †Department of Medical Records, the First Hospital of Quanzhou, Quanzhou, China; ‡Department of Microbiology, The Heilongjiang Key Laboratory of Immunity and Infection, Pathogenic Biology, Harbin Medical University; and §Key Laboratory of Bio-Pharmaceutical, Harbin Medical University, Ministry of Education, Harbin, China.Jiaxiao Sun, MSc, is currently affiliated with Department of Anesthesiology, the First Hospital of Quanzhou, Quanzhou, China. Juanjuan Zheng, is currently affiliated with Department of Medical Records, the First Hospital of Quanzhou, Quanzhou, China.Accepted for publication October 18, 2013.Funding: Financial support by grants from the National Natural Science Foundation of China (No.30972839), China Postdoctoral Science Foundation (No. 2013M531069), Foundation of Heilongjiang Educational Committee (No.12531245) and Doctoral Fund of the First Affiliated Hospital of Harbin Medical University (No.2012B006) are gratefully acknowledged.Dr. Changsong Wang and Jiaxiao Sun contributed equally to this work. The authors declare no conflicts of interest.Reprints will not be available from the authors.Address correspondence to Enyou Li, MD, Department of Anesthesiology, The First Affiliated Hospital of Harbin Medical University, No 23 Youzheng St., Nangang District, Harbin, Heilongjiang 150001, China. Address e-mail to [email protected].

Section Editor: Avery Tung

Society of Critical Care Anesthesiologists

mailto:[email protected]

February 2014 • Volume 118 • Number 2 www.anesthesia-analgesia.org 347

METHODSWe conducted a systematic review and several meta-analyses of the literature according to the methods recommended in the PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions.

GroupsOur investigators were divided into 4 groups. CW was pri-marily responsible for the literature search group (CW and LG). JS and HM were responsible for the 2 literature review groups (JS, FZ, HM, and YZ). JZ was responsible for the data analysis group (JZ and EL). After the 2 separate litera-ture review groups conducted the literature exclusion and inclusion and the data extraction, the data were verified. If there was an inconsistency, the data extraction was repeated until a consensus was reached.

Literature SearchThe trials were identified by electronic and manual searches. The electronic searches were performed by 2 authors who independently searched the Medline, Embase, and Cochrane Library databases; the Cochrane Controlled Trials

Register; LILACS (http://www.bireme.br; assessed May 2012); and Web of Knowledge (Conference Proceedings Citation Index-Science, Conference Proceedings Citation Index-Social Sciences & Humanities). We did not restrict our search based on language or year of publication. The last search update was May 2012. The Medline database was searched using the PubMed interface. The follow-ing search terms (in all fields) were used: sepsis, septic shock, steroids, corticosteroids, adrenal cortex hormones, and glucocorticoids. Embase was searched using the fol-lowing search terms: sepsis, septic shock, steroids, and corticosteroids. The search terms sepsis and septic shock were searched in the Cochrane infectious diseases group’s trial register. We searched the Cochrane central register using the following search terms: sepsis, septic shock, ste-roids, and corticosteroids. LILACS was searched using the search terms sepsis, steroids, and corticosteroids, and we searched the proceedings of the annual meetings by using the search terms sepsis, septic shock, steroids, and cortico-steroids in the Web of Knowledge (Conference Proceedings Citation Index-Science, Conference Proceedings Citation Index-Social Sciences & Humanities) database. We reviewed the reference lists of published meta-analyses. In

1949 abstracts identified by electronic database search1239 MEDLINE410 COCHRANE LIBRARY56 LILACS

223 EMBASE21 WEB OF KNOWLEDGE Summary of the meeting

120 abstracts of interest identified and further reviewed69 MEDLINE23 COCHRANE LIBRARY

8 LILACS 5 EMBASE

15 WEB OF KNOWLEDGE

MEDLINE: excluded 33 review articles, 2 repetitive articles, and4 repetitive experiments; 23 were excluded for other reasons.

COCHRANE LIBRARY: excluded 8 review articles and 13 repetitive articles; 2 were excluded for other reasons.

LILACS: excluded 7 review articles; 1 was excluded for other reasons.

EMBASE: excluded 4 review articles; 1 was excluded for other reasons.

WEB OF KNOWLEDGE: excluded 6 review articles and 2 repetitive experiments; 6 were excluded for other reasons.

Reference 3, excluded 3.

8 included in meta-analysis

Figure 1. Flow diagram of the literature search.

http://www.bireme.br;

Low-Dose Hydrocortisone Therapy in Septic Shock

348 www.anesthesia-analgesia.org aNesthesia & aNalgesia

addition, we manually searched the Index Medicus of RCTs, meta-analyses, and systematic reviews for studies that were missed in the initial electronic search.

The search strategy identified 1949 studies. Two litera-ture review groups conducted the literature exclusion; 120 studies were included for potential interest. The studies with one or more of the following terms mentioned were consid-ered for inclusion: steroid, any class of glucocorticoid, septic shock, and human study. The selected studies were repeat-edly reviewed for exclusion by the literature search groups. The exclusion and inclusion criteria were independently applied to each study by the 2 study review groups (Fig. 1).

Strategy DesignThe literature search and data extraction strategy were dis-cussed and designed by 2 authors. After all the authors had discussed and reviewed the strategy, the corresponding author approved the final version of the study strategy design.

Definition of Septic ShockSeptic shock was defined according to the standard estab-lished by the 1992 American College of Chest Physicians/Society of Critical Care Medicine (ACCP/SCCM) Consensus

Conference.14 “Shock reversal” was defined as a stable state of systolic blood pressure (> 90 mm Hg) for a period of 24 hours or more without vasopressor support or transfusion.9 Low-dose hydrocortisone was defined as a daily dose of hydrocortisone ≤300 mg.9

Inclusion and Exclusion CriteriaThe literature inclusion and exclusion procedures were per-formed independently by 2 literature review groups. We first excluded retrospective analyses, repeated literature reports, and repeated experiments (the same experiment analyzed and evaluated in different literature reports); purely physi-ological studies (e.g., the effects of steroids on neutrophils in patients with septic shock);15 imaging studies; pediatric studies; studies with high-dose medications; studies on medications other than hydrocortisone (the initial design was to conduct a separate analysis on a single steroid of other glucocorticoid types; however, the analysis could not be performed separately due to the lack of relevant studies on other medications); nonrandomized controlled studies; and studies without a control group (Table 1). If data were missing, the literature search group contacted the authors for the relevant data.

Table 1. Excluded LiteratureAuthors Publication year Exclusion reasonCooperative study group4 1963 No information is provided on the number of patients who may have had shock in this

study. No clear definition of mortality; low methodological quality; unable to get accurate information for adults.

Schumer38 1976 Study medications did not match, dexamethasone, methylprednisoloneLucas and Ledgerwood39 1984 High-doseSprung et al.40 1984 High-doseHughes41 1984 Methylprednisolone, no relevant resultsLederer42 1984 Betamethasone, high-doseHellman and Alestig43 1985 High-doseBone et al.44 1987 High-doseLuce et al.45 1988 High-doseMarks et al.46 1990 Nonrandomized controlled study, physiological study, no relevant resultsOppert et al.47 2000 Nonrandomized controlled study, no control groupBriegel et al.48 2001 Physiological study, no relevant dataAnnane and Bellissant49 2002 Physiological study, no relevant data, study objective did not match analysisKeh et al.50 2003 Study method did not match, steroid application overlapped in 2 groupsLaterre et al.51 2003 Study objective did not match, no relevant dataGuzman et al.52 2005 Study objective did not match, no relevant dataLevy et al.35 2005 Study objective did not match, no relevant dataTandan et al.53 2005 No hydrocortisone dose and durationRinaldi et al.54 2006 Physiological study, effects of steroid on albumin, no relevant dataCicarelli et al.31 2007 Study medication did not match, dexamethasoneLoisa et al.55 2007 No control groupMcgee et al.56 2007 Study method did not match, no control group, comparison of steroid effects with different

dosage.Cicarelli57 2008 Study medication did not match, dexamethasoneKaufmann et al.15 2008 Physiological study, no relevant dataKurugundla et al.58 2008 No relevant dataBüchele et al.59 2009 Nonrandomized controlled studyHayashi60 2009 No relevant dataHu et al.61 2009 Study objective did not match, no relevant dataRussell et al.62 2009 Study objective did not match, comparison of 2 hypertension medication, no control groupYu et al.34 2009 Comparison with effects of methylprednisolone, no control groupValoor et al.63 2009 Pediatric populationBeale et al.64 2010 Report, nonrandomized controlled studyCOIITSS study investigators30 2010 No control groupJung et al.65 2011 Imaging report, not relevant to study resultsSchelling et al.66 2001 Physiological study, no relevant resultsHuh et al.67 2011 Comparison between IV administration of steroids at day 3 and day 7, no control group


Subsequently, the 2 study review groups performed the initial verification. A disagreement occurred only in 1 study, which was eventually excluded after a discussion among all of the authors.

The process yielded 8 published studies, one of which was only published as a meeting abstract. Two groups of researchers independently conducted a second round of data extraction from the literature. Key data were 28-day mortality and shock reversal at 7 and 28 days. If the relevant

data were missing or ambiguous, we contacted the authors for clarification.

Quality AssessmentQuality assessments were performed separately by the 2 literature review groups. Studies that received inconsistent scores were scored again by all of the authors. The quality of the study was assessed using a modified Jadad scale16 in which the generation of random sequences, blinding

Table 2. General Information of Included Studies

ArticleNo. of

samples

No. of research centers

Publication type Design Intervention Outcome

Bollaert et al.,27 41 2 Article Double-blind In the patient group, 100 mg hydrocortisone was IV administered 3 times daily for a total of 5 d with a matching placebo of physiologic saline for the control group.

Mortality (28 d), shock reversal (7 d and 28 d)

Briegel et al.,26 40 1 Article Double-blind In the patient group, 100 mg hydrocortisone was IV administered as a loading dose; then, hydrocortisone was continuously administered at 0.18 mg/kg/h. If shock reversal was present, the dose was reduced to 0.08 mg/kg/h for 6 d. Physiologic saline was used as a matching placebo for the control group.


Chawla et al.,21 44 1 Report Double-blind In the patient group, 100 mg hydrocortisone was IV administered every 8 h for a total of 3 d; the dosage was then tapered over time for 4 d. Physiologic saline was used as a matching placebo for the control group.


Annane et al.,22 299 19 Article Double-blind In the patient group, 50 mg hydrocortisone was IV administered every 6 h, and 50 µg fludrocortisone was taken orally once daily for a total of 7 d. Physiologic saline was used as a matching placebo for the control group.


Oppert et al.,28 41 1 Article Double-blind In the patient group, 100 mg hydrocortisone as a loading dose was administered IV; hydrocortisone was then continuously administered at 0.18 mg/kg/h. If shock reversal was present, the dose was reduced to 0.06 mg/kg/h and subsequently slowly tapered. Physiologic saline was used as a matching placebo for the control group.

Mortality (28 d), shock reversal (7 d)

Mussack et al.,23 (2005)

24 1 Article Double-blind In the patient group, 100 mg hydrocortisone as a loading dose was IV injected; hydrocortisone was then continuously administered at 0.18 mg/kg/h for a total of 6 d. Physiologic saline was used as a matching placebo for the control group.

Mortality (28 d), shock reversal (28 d)

Sprung et al.,24 499 13 Article Double-blind In the patient group, 50 mg hydrocortisone was IV ly administered every 6 h for a total of 5 d. Afterwards, hydrocortisone 50 mg was IV administered every 12 h from day 6 to day 8, and 50 mg every 24 h from day 9 to day 11. Physiologic saline was used as a matching placebo for the control group.


Arabi et al.,25 75 1 Article Double-blind In the patient group, 50 mg hydrocortisone was IV administered every 6 h until hemodynamic stability was achieved and was than tapered over time for 8 d. Physiologic saline was used as a matching placebo for the control group.

Mortality (28 d)



method, reasons for withdrawal, and dropout at the time of follow-up were evaluated. A 7-point scale was used, with 1 to 3 indicating a low-quality study and 4 to 7 indicating a high-quality study. No studies were excluded from the analysis because of the quality assessment.

Statistical AnalysesThe outcomes of interest were 28-day mortality and the shock reversal at 7 and 28 days. The adverse events superinfection, gastrointestinal (GI) bleeding, and hyper-glycemia were also evaluated. Statistical analysis was per-formed using Stata (version 11.1, StataCorp LP, College Station, TX). Because the mortality rate was calculated in most RCTs at different time points, we used the hazard ratio as the parameter for calculating the mortality rate.17 Intratrial variability among the RCTs may have intro-duced bias in the hazard ratio calculation. Considering that the hazard ratio is very similar to the odds ratio (OR), we calculated the OR value and 95% confidence intervals (CIs) as the approximate parameters for evaluating the effects of hydrocortisone therapy on mortality and shock reversal.18,19

The statistical variable I2 was used to compare hetero-geneity among studies (25% indicated low heterogeneity, 50% indicated moderate heterogeneity, and 75% indicated high heterogeneity; I2 > 50% indicated significant hetero-geneity).20 The fixed-effects model was applied if there was significant heterogeneity. The DerSimonian–Laird test was applied for the pooled OR value. The fixed-effects model was applied if there was low significant heterogeneity. The Mantel–Haenszel test was applied for the pooled OR value. The Z-test was applied for the significance test for pooled OR values.

To find the source of heterogeneity and ensure the sta-bility of results, we performed a sensitivity analysis for 28-day mortality and 7-day shock reversal. We performed

subgroup analyses for sample size (< 100 or > 100) and qual-ity score (6 or 7) for 28-day mortality. To explain the relation-ship between the log value of the 28-day mortality and the patients’ average age and gender, we performed a second-ary analysis by meta-regression method. The variables time and sample size were used in a cumulative meta-analysis to investigate the dynamic changes among 3 indictors: 28-day mortality, 7 day shock reversal, and 28-day shock reversal.

To assess publication bias and test for small sample size bias, we used Egger’s test in continuous data analyses. However, the response variable of this study was a binary variable. Therefore, Harbord test was performed for quanti-tative assessment, and Begg’s funnel plot was used to quali-tatively demonstrate the bias.

RESULTSEight publications (Table 2) were incorporated in the meta-analysis, which included 1 meeting abstract.21 All 8 studies were included in the analysis of 28-day mortality. Among these studies, the raw data were provided in 6 stud-ies.20–25 The raw data for the remaining 2 studies21,26 were acquired by writing to the authors. Six studies21,22,24,26–28 were included for shock reversal analysis on day 7. Among these studies, the raw data were originally provided in 4 stud-ies.22,24,27,28 The raw data for the remaining 2 studies21,26 were acquired by writing to the authors. Six studies21–24,26,27 were included in the analysis of shock reversal on day 28. Among these studies, the raw data were provided in 4 studies.22–24,27 The raw data for the remaining 2 studies21,26 were acquired by writing to the authors.

28-Day MortalityEight RCTs with a total of 1063 participants were included in the analysis (535 subjects in the patient group and 528 in the control group). The 28-day mortality values were 227 (42.43%) and 237 (44.89%) in the patient and control groups,

Figure 2. Effects of low-dose hydrocortisone therapy on mortality at 28 days in patients with septic shock.


respectively. The analysis results were OR = 0.891, 95% CI, 0.69–1.15, P = 0.371, and I2 = 29.2%. There were no significant differences in the 28-day mortality analysis (Fig. 2; Table 3).

Sensitivity Analysis of 28-Day MortalityA sensitivity analysis was performed for the included 8 studies to investigate the source of heterogeneity (Fig. 3). The result for 28-day mortality remained stable after the exclusion of any 1 study. No significant differences were found in the 28-day mortality rates (Table 3).

Subgroup Analysis of 28-Day MortalityWe performed subgroup analyses of the 8 studies to inves-tigate the effects of sample size (categorized to >100 and <100) and quality score (divided into a 6-score group and a 7-score group) on heterogeneity (Table 3). Two studies with

sample sizes of >100 were included in this subgroup with OR = 0.972, 95% CI, 0.73–1.30, P = 0.850, and I2 = 40.0%. Six studies were included in the subgroup with sample sizes of <100, with OR = 0.665, 95% CI, 0.39–1.13, P = 0.131, and I2 = 25.2%. The subgroup with quality assessment scores of 6 included 3 studies, with OR = 1.052, 95% CI, 0.74–1.49, P = 0.775, and I2 = 0.0%. Four studies were included in the subgroup with quality assessment scores of 7, with OR = 0.786, 95% CI, 0.53–1.16, P = 0.224, and I2 = 45.7%. One study was only published as a meeting abstract and therefore could not be included for quality analysis.

The results of the subgroup analysis showed no sig-nificant differences in the 28-day mortality rates among subgroups. Heterogeneity decreased in some subgroups (such as the subgroup with sample sizes of <100 and the subgroup with quality assessment scores of 6), whereas

Figure 3. Forest plot of estimates of sensitivity for 28-day mortality.

Table 3. Outcome Effect EstimatesOutcome N OR (95% CI) POR I2 (P) Pharbord

MortalityTotal 8 0.891 (0.691–1.148) 0.371 29.2% (0.195) 0.225Sample size <100 6 0.665 (0.392–1.129) 0.131 25.2% (0.245) 0.543Sample size >100 2 0.972 (0.728–1.299) 0.850 40.0% (0.197) —Quality assessment score = 6 3 1.052 (0.742–1.493) 0.775 0.0% (0.490) 0.051Quality assessment score = 7 4 0.786 (0.534–1.159) 0.224 45.7% (0.137) 0.970

Shock reversal (7 days)Total 6 2.078 (1.582–2.729) <0.0001 26.9% (0.233) 0.553Sample size <100 4 2.848 (1.459–5.559) 0.002 47.6% (0.126) 0.328Sample size >100 2 1.949 (1.445–2.628) <0.0001 0.0% (0.721) —Quality assessment score = 6 2 1.896 (1.316–2.731) 0.001 46.8% (0.170) —Quality assessment score = 7 3 3.118 (1.235–7.873) 0.016 53.1% (0.118) 0.287Shock reversal (28 d) 6 1.495 (1.124–1.988) 0.006 0.0% (0.604) 0.019

Meta-regression (log odds mortality)Sex 7 0.842 (0.325–2.184) 0.662 — —Age 7 1.033 (0.928–1.150) 0.471 — —



heterogeneity increased in the subgroup with sample sizes of >100 and the subgroup with quality assessment scores of 7, compared with overall heterogeneity. These results suggest that sample size and quality assessment were not sources of heterogeneity.

Secondary Analysis of 28-Day MortalityFor exploratory purposes, a secondary analysis was per-formed by meta-regression method between the log value of the 28-day mortality and the patients’ average age and gender (Table 3). Among the 8 studies, 7 studies provided gender information and 7 studies provided age information (P = 0.471). Therefore, gender and age were not associated with the heterogeneity of 28-day mortality.

Shock ReversalThe 7-day shock reversal analysis included 6 RCTs with a total of 964 participants (484 subjects in the patient group and 480 in the control group). The number of patients with 7-day shock reversal was 307 (63.43%) in the patient group and 228 (47.50%) in the control group. The increase in shock reversal at 7 days with hydrocortisone was statistically significant: OR = 2.078, 95% CI, 1.58–2.73, P < 0.0001, and I2 = 26.9% (Fig. 4, Table 3). The source of heterogeneity was not found by a subgroup analysis of sample size or quality assessment score (Table 3).

The 28-day shock reversal analysis included 6 RCTs with a total of 947 participants (478 subjects in the hydrocorti-sone group and 469 in the placebo group). The number of patients with 28-day shock reversal was 328 (68.62%) in the patient group and 283 (60.34%) in the control group. The increase in shock reversal at 28 days with hydrocortisone was statistically significant: OR = 1.495, 95% CI, 1.12–1.99, P = 0.006, and I2 = 0.0% (Fig. 4; Table 3).

A sensitivity analysis was performed to investigate the source of heterogeneity of 7-day shock reversal (Fig. 5). The results remained stable after the exclusion of any 1 study. Because no heterogeneity (I2 = 0.0%) was observed in the 28-day shock reversals, a sensitivity analysis was not per-formed for this variable.

Publication Bias AnalysisWe analyzed publication bias for the studies included in the analyses of 28-day mortality and 7-day/28-day shock reversal. Because the dependent variable was a binary variable, we conducted the Harbord test for quantitative assessment of 3 indicators to determine the possibility of publication bias. Begg’s funnel plot was performed for qualitative analysis. The P value was 0.225 for 28-day mor-tality, 0.553 for 7-day shock reversal, and 0.019 for 28-day shock reversal. Begg’s funnel plot for 28-day mortality is shown in Figure 6.

Figure 4. Effects of low-dose hydrocortisone therapy on 7 and 28-day shock reversal in patients with septic shock.


Cumulative AnalysisUsing the variables publication year and sample size, a cumulative meta-analysis was performed for 28-day mor-tality and 7-day/28-day shock reversal. The cumulative analysis of 28-day mortality showed that the OR value gradually increased from 0.27 to 0.89 and that the 95% CI increased from (0.07–0.99) to (0.69–1.15) as a function of publication date. The tendency of the OR value to approach 1 was significant (Fig. 7).

The cumulative analysis of 7-day shock reversal showed that the OR value gradually decreased from 8.04 to 2.08 and the 95% CI decreased from (1.94–33.30) to (1.58–2.73) as a function of publication date. However, the OR value and 95% CI were still significantly >1 (Fig. 7).

The cumulative analysis of 28-day shock reversal showed that the OR value gradually decreased from 3.67 to 1.49 and

the 95% CI decreased from (1.01–13.40) to (1.12–1.99) as a function of publication date. However, the OR value and 95% CI were still significantly >1 (Fig. 7).

The 3 indicators did not show any trend with increases in sample size.

ComplicationsMeta-analysis of superinfection showed an OR = 1.103, 95% CI, 0.83–1.18, P = 0.507, and I2 = 3.1%. The results were not significant, indicating that low-dose hydrocortisone therapy did not increase the likelihood of superinfection in patients with septic shock. No significant trend was found in the cumulative analysis (Table 4).

Meta-analysis of GI bleeding showed an OR = 1.601, 95% CI, 0.99–2.60, P = 0.057, and I2 = 26.9%. Although the result did not reach statistical significance, the OR (1.6) and the nearly significant results (P = 0.057) do not comfortably exclude an increase in GI bleeding. The cumu lative analysis showed that the negative result became more stabilized in studies reported in recent years (Table 4).

Meta-analysis of hyperglycemia showed an OR = 2.143, 95% CI, 1.41–3.26, P < 0.0001, and I2 = 0.0%. The results were significant, indicating that low-dose hydrocortisone increases the incidence of hyperglycemia in patients with septic shock. Because only 3 studies were included, a cumu-lative analysis was not performed (Table 4).

DISCUSSIONThis meta-analysis demonstrated that low-dose hydro-cortisone therapy attenuated septic shock in adult patients at 7 and 28 days but did not reduce 28-day

Figure 6. Funnel graph for the assessment of potential publication bias for 28-day mortality.

Figure 5. Forest plot of estimates of sensitivity for 7-day mortality.



mortality. Hydrocortisone increased the blood glucose lev-els in patients with septic shock and was associated with increased GI bleeding, although this last finding did not

reach statistical significance. The available evidence does not support the use of low-dose hydrocortisone as a routine treatment for adult patients with septic shock.

Figure 7. Cumulative analysis of 28-day mortality and 7/28-day shock reversal.

Table 4. Complications AnalysisOutcome N OR (95% CI) POR I2 (P) Pharbord

Superinfection 6 1.103(0.825–1.475) 0.507 3.1%(0.397) 0.389GI bleed 6 1.601(0.987–2.598) 0.057 26.9%(0.233) 0.590Hyperglycemia 3 2.143(1.410–3.257) <0.0001 0.0%(0.489) 0.705

GI = gastrointestinal.


Our results are similar to those of Sligl et al.,29 who dem-onstrated that corticosteroid therapy does not reduce mor-tality rates but does appear to consistently reduce the time to shock reversal. The role of hydrocortisone therapy, in addi-tion to fludrocortisone, was also evaluated in the COIITSS Trial.30 The authors failed to demonstrate a survival benefit associated with fludrocortisone treatment, but there may be a higher risk for increased infection.

Our results differ from the results of Annane et al.9 in 2009 because of different inclusion criteria. We limited our analysis to hydrocortisone therapy. Therefore, we excluded the 3 studies that Annane et al.9 included, Cicarelli et al.,31 Yildiz et al.,32 and Meduri et al.,33 because these investiga-tors studied prednisolone, dexamethasone, and methyl-prednisolone, respectively. In addition, we included a 2010 report by Arabi et al.25

Yu et al.34 compared the effects of hydrocortisone and methylprednisolone on septic shock. They found that the survival rates for patients who received hydrocortisone were higher than for patients who received methylpred-nisolone, although the difference was not significant. These results suggest that different types of glucocorticoids may have different effects on septic shock treatment.

The study by Levy et al.35 was not included in our analy-sis because it was a retrospective cohort study; the steroid type, dose, and duration were also unspecified. The study by Raurich et al.36 was also excluded in our analysis because it was a case-control study. The study by Annane et al.22 was excluded because it evaluated hydrocortisone and fludrocortisone.

Cumulative meta-analysis showed that the OR value of 7-day shock reversal gradually decreased from 8.04 to 2.08, whereas the 95% CI decreased from (1.94–33.30) to (1.58–2.73) as a function of publication year. However, the OR values and 95% CI were both significantly higher than 1, indicating that although the positive results of 7-day shock reversal gradually weakened over the years, the results were still significantly positive and became stable in recent years. The cumulative analysis of 28-day shock reversal showed that the OR value decreased gradually from 3.67 to 1.49 and that the 95% CI decreased from (1.01–13.40) to (1.12–1.99) as a function of publication year. However, the OR values and 95% CI were both significantly higher than 1, indicating that although the positive results of 28-day shock reversal gradually weakened over the years, the results were still significantly positive and became stable in recent years.

It is not clear why mortality at 28 days did not decrease, since the data demonstrate that shock was ameliorated at 7 and 28 days in septic patients. This lack of an effect on 28-day mortality rate might be attributed to adverse events such as superinfection, GI bleeding, and hyperglycemia. In this study, we found that low-dose hydrocortisone increased blood glucose levels in patients, had a trend toward increased GI bleeding that was not statistically significant, and did not increase the risk of superinfection. Because of the small sample size and few adverse events in these studies, additional studies with increased sample sizes are warranted to explain the lack of improvement in mortality.

Our study demonstrates that although low-dose hydro-cortisone therapy can improve shock reversal in patients

with sepsis, the therapy has no significant impact on 28-day mortality rate. The new International Guidelines for Management of Severe Sepsis and Septic Shock suggest that it is not advisable to use IV hydrocortisone as a treat-ment for adult septic shock patients if adequate fluid resus-citation and vasopressor therapy can restore hemodynamic stability. If hemodynamic stability cannot be maintained, the guidelines suggest IV hydrocortisone alone at a dose of 200 mg per day.37 Our results are consistent with these new guidelines. The available evidence does not support the argument that low-dose hydrocortisone should be used as a routine treatment in adult patients with septic shock. E

DISCLOSURESName: Changsong Wang, MD.Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.Attestation: Changsong Wang has seen the original study data, reviewed the analysis of the data, approved the final manu-script, and is the author responsible for archiving the study files.Name: Jiaxiao Sun, MSc.Contribution: This author helped conduct the study, analyze the data, and write the manuscript.Attestation: Jiaxiao Sun has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.Name: Juanjuan Zheng, MSc.Contribution: This author helped analyze the data.Attestation: Juanjuan Zheng has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.Name: Lei Guo, MD.Contribution: This author helped conduct the study and write the manuscript.Attestation: Lei Guo has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.Name: Hongyan Ma, MD.Contribution: This author helped conduct the study.Attestation: Hongyan Ma has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.Name: Yang Zhang.Contribution: This author helped conduct the study and write the manuscript.Attestation: Yang Zhang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.Name: Fengmin Zhang, PhD.Contribution: This author helped design the study, analyze the data, and write the manuscript.Attestation: Fengmin Zhang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.Name: Enyou Li, MD.Contribution: This author helped design and conduct the study and write the manuscript.Attestation: Enyou Li has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.This manuscript was handled by: Steven L. Shafer, MD.

REFERENCES 1. Matot I, Sprung CL. Definition of sepsis. Intensive Care Med

2001;27 Suppl 1:S3–9



2. Abraham E, Reinhart K, Opal S, Demeyer I, Doig C, Rodriguez AL, Beale R, Svoboda P, Laterre PF, Simon S, Light B, Spapen H, Stone J, Seibert A, Peckelsen C, De Deyne C, Postier R, Pettilä V, Artigas A, Percell SR, Shu V, Zwingelstein C, Tobias J, Poole L, Stolzenbach JC, Creasey AA; OPTIMIST Trial Study Group. Efficacy and safety of tifacogin (recombinant tissue factor path-way inhibitor) in severe sepsis: a randomized controlled trial. JAMA 2003;290:238–47

3. Dellinger RP. Cardiovascular management of septic shock. Crit Care Med 2003;31:946–55

4. Cooperative Study Group. The effectiveness of hydrocortisone in the management of patients with severe infections. JAMA 1963;183:462–5

5. Russell JA. Management of sepsis. N Engl J Med 2006;355:1 699–713

6. Bone RC, Fisher CJ Jr, Clemmer TP, Slotman GJ, Metz CA, Balk RA. A controlled clinical trial of high-dose methylprednisolone in the treatment of severe sepsis and septic shock. N Engl J Med 1987;317:653–8

7. Cronin L, Cook DJ, Carlet J, Heyland DK, King D, Lansang MA, Fisher CJ Jr. Corticosteroid treatment for sepsis: a critical appraisal and meta-analysis of the literature. Crit Care Med 1995;23:1430–9

8. Annane D, Bellissant E, Bollaert PE, Briegel J, Keh D, Kupfer Y. Corticosteroids for severe sepsis and septic shock: a systematic review and meta-analysis. BMJ 2004;329:480

9. Annane D, Bellissant E, Bollaert PE, Briegel J, Confalonieri M, De Gaudio R, Keh D, Kupfer Y, Oppert M, Meduri GU. Corticosteroids in the treatment of severe sepsis and septic shock in adults: a systematic review. JAMA 2009;301: 2362–75

10. Moran JL, Graham PL, Rockliff S, Bersten AD. Updating the evidence for the role of corticosteroids in severe sepsis and septic shock: a Bayesian meta-analytic perspective. Crit Care 2010;14:R134

11. Kalil AC, Sun J. Low-dose steroids for septic shock and severe sepsis: the use of Bayesian statistics to resolve clinical trial con-troversies. Intensive Care Med 2011;37:420–9

12. Czock D, Keller F, Rasche FM, Häussler U. Pharmacokinetics and pharmacodynamics of systemically administered gluco-corticoids. Clin Pharmacokinet 2005;44:61–98

13. Lau J, Schmid CH, Chalmers TC. Cumulative meta-analysis of clinical trials builds evidence for exemplary medical care. J Clin Epidemiol 1995;48:45–57

14. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RM, Sibbald WJ. 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;101:1644–55

15. Kaufmann I, Briegel J, Schliephake F, Hoelzl A, Chouker A, Hummel T, Schelling G, Thiel M. Stress doses of hydrocortisone in septic shock: beneficial effects on opsonization-dependent neutrophil functions. Intensive Care Med 2008;34:344–9

16. Bañares R, Albillos A, Rincón D, Alonso S, González M, Ruiz-del-Arbol L, Salcedo M, Molinero LM. Endoscopic treat-ment versus endoscopic plus pharmacologic treatment for acute variceal bleeding: a meta-analysis. Hepatology 2002;35:609–15

17. Peter JV, John P, Graham PL, Moran JL, George IA, Bersten A. Corticosteroids in the prevention and treatment of acute respi-ratory distress syndrome (ARDS) in adults: meta-analysis. BMJ 2008;336:1006–9

18. Spruance SL, Reid JE, Grace M, Samore M. Hazard ratio in clini-cal trials. Antimicrob Agents Chemother 2004;48:2787–92

19. Moran J, Solomon P, Warn D. Methodology in meta-analysis: a study from Critical Care meta-analytic practice. Health Serv Outcomes Res Method 2004;5:207–26

20. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557–60

21. Chawla K, Kupfer Y, Tessler S. Hydrocortisone reverses refrac-tory septic shock (abstract). Crit Care Med 1999;27:A33

22. Annane D, Sébille V, Charpentier C, Bollaert PE, François B, Korach JM, Capellier G, Cohen Y, Azoulay E, Troché G, Chaumet-Riffaud P, Chaumet-Riffaut P, Bellissant E. Effect of

treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA 2002;288:862–71

23. Mussack T, Briegel J, Schelling G, Biberthaler P, Jochum M. 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;43:259–68

24. Sprung CL, Annane D, Keh D, Moreno R, Singer M, Freivogel K, Weiss YG, Benbenishty J, Kalenka A, Forst H, Laterre PF, Reinhart K, Cuthbertson BH, Payen D, Briegel J; CORTICUS Study Group. Hydrocortisone therapy for patients with septic shock. N Engl J Med 2008;358:111–24

25. Arabi YM, Aljumah A, Dabbagh O, Tamim HM, Rishu AH, Al-Abdulkareem A, Knawy BA, Hajeer AH, Tamimi W, Cherfan A. Low-dose hydrocortisone in patients with cirrhosis and sep-tic shock: a randomized controlled trial. CMAJ 2010;182:1971–7

26. Briegel J, Forst H, Haller M, Schelling G, Kilger E, Kuprat G, Hemmer B, Hummel T, Lenhart A, Heyduck M, Stoll C, Peter K. Stress doses of hydrocortisone reverse hyperdynamic septic shock: a prospective, randomized, double-blind, single-center study. Crit Care Med 1999;27:723–32

27. Bollaert PE, Charpentier C, Levy B, Debouverie M, Audibert G, Larcan A. Reversal of late septic shock with supraphysiologic doses of hydrocortisone. Crit Care Med 1998;26:645–50

28. Oppert M, Schindler R, Husung C, Offermann K, Gräf KJ, Boenisch O, Barckow D, Frei U, Eckardt KU. Low-dose hydro-cortisone improves shock reversal and reduces cytokine levels in early hyperdynamic septic shock. Crit Care Med 2005;33:2457–64

29. Sligl WI, Milner DA Jr, Sundar S, Mphatswe W, Majumdar SR. Safety and efficacy of corticosteroids for the treatment of septic shock: a systematic review and meta-analysis. Clin Infect Dis 2009;49:93–101

30. Annane D, Cariou A, Maxime V, Azoulay E, D’honneur G, Timsit JF, Cohen Y, Wolf M, Fartoukh M, Adrie C, Santré C, Bollaert PE, Mathonet A, Amathieu R, Tabah A, Clec’h C, Mayaux J, Lejeune J, Chevret S. Corticosteroid treatment and intensive insulin therapy for septic shock in adults: a random-ized controlled trial. JAMA 2010;303:341–8

31. Cicarelli DD, Vieira JE, Benseñor FE. Early dexamethasone treatment for septic shock patients: a prospective randomized clinical trial. Sao Paulo Med J 2007;125:237–41

32. Yildiz O, Doganay M, Aygen B, Güven M, Keleştimur F, Tutuû A. Physiological-dose steroid therapy in sepsis [ISRCTN36253388]. Crit Care 2002;6:251–9

33. Meduri GU, Golden E, Freire AX, Taylor E, Zaman M, Carson SJ, Gibson M, Umberger R. Methylprednisolone infusion in early severe ARDS: results of a randomized controlled trial. Chest 2007;131:954–63

34. Yu TJ, Liu YC, Yu CC, Tseng JC, Hua CC, Wu HP. Comparing hydrocortisone and methylprednisolone in patients with septic shock. Adv Ther 2009;26:728–35

35. Levy H, Laterre PF, Bates B, Qualy RL. Steroid use in PROWESS severe sepsis patients treated with drotrecogin alfa (activated). Crit Care 2005;9:R502–7

36. Raurich JM, Llompart-Pou JA, Ibáñez J, Frontera G, Pérez O, García L, Ayestarán JI. Low-dose steroid therapy does not affect hemodynamic response in septic shock patients. J Crit Care 2007;22:324–9

37. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, Sevransky JE, Sprung CL, Douglas IS, Jaeschke R, Osborn TM, Nunnally ME, Townsend SR, Reinhart K, Kleinpell RM, Angus DC, Deutschman CS, Machado FR, Rubenfeld GD, Webb SA, Beale RJ, Vincent JL, Moreno R; Surviving Sepsis Campaign Guidelines Committee including the Pediatric Subgroup. Surviving sepsis campaign: international guidelines for management of severe sep-sis and septic shock: 2012. Crit Care Med 2013;41:580–637

38. Schumer W. Steroids in the treatment of clinical septic shock. Ann Surg 1976;184:333–41

39. Lucas CE, Ledgerwood AM. The cardiopulmonary response to massive doses of steroids in patients with septic shock. Arch Surg 1984;119:537–41

40. Sprung CL, Caralis PV, Marcial EH, Pierce M, Gelbard MA, Long WM, Duncan RC, Tendler MD, Karpf M. The effects of high-dose corticosteroids in patients with septic shock. A pro-spective, controlled study. N Engl J Med 1984;311:1137–43


41. Hughes GS Jr. Naloxone and methylprednisolone sodium suc-cinate enhance sympathomedullary discharge in patients with septic shock. Life Sci 1984;35:2319–26

42. Lederer V. Betamethasone sodium phosphate injection: high-dose regimen in septic shock. Clin Ther 1984;6:719–26

43. Hellman A, Alestig K. High doses of corticosteroids in the treat-ment of septic shock. Acta Chir Scand Suppl 1985;526:124–8

44. Bone RC, Fisher CJ Jr, Clemmer TP, Slotman GJ, Metz CA, Balk RA. A controlled clinical trial of high-dose methylprednisolone in the treatment of severe sepsis and septic shock. N Engl J Med 1987;317:653–8

45. Luce JM, Montgomery AB, Marks JD, Turner J, Metz CA, Murray JF. Ineffectiveness of high-dose methylprednisolone in preventing parenchymal lung injury and improving mortality in patients with septic shock. Am Rev Respir Dis 1988;138:62–8

46. Marks JD, Marks CB, Luce JM, Montgomery AB, Turner J, Metz CA, Murray JF. Plasma tumor necrosis factor in patients with septic shock. Mortality rate, incidence of adult respiratory dis-tress syndrome, and effects of methylprednisolone administra-tion. Am Rev Respir Dis 1990;141:94–7

47. Oppert M, Reinicke A, Gräf KJ, Barckow D, Frei U, Eckardt KU. Plasma cortisol levels before and during “low-dose” hydro-cortisone therapy and their relationship to hemodynamic improvement in patients with septic shock. Intensive Care Med 2000;26:1747–55

48. Briegel J, Jochum M, Gippner-Steppert C, Thiel M. Immuno-modulation in septic shock: hydrocortisone differentially regulates cytokine responses. J Am Soc Nephrol 2001;12 Suppl 17:S70–4

49. Annane D, Bellissant E. Impact of corticosteroids on the vas-cular response to catecholamines in septic shock. Réanimation 2002;11:111–6

50. Keh D, Boehnke T, Weber-Cartens S, Schulz C, Ahlers O, Bercker S, Volk HD, Doecke WD, Falke KJ, Gerlach H. Immunologic and hemodynamic effects of “low-dose” hydrocortisone in septic shock: a double-blind, randomized, placebo-controlled, crossover study. Am J Respir Crit Care Med 2003;167:512–20

51. Laterre PF, Levy H, Bates BM. Steroid use in prowess patients with septic shock (abstract). Intensive Care Medicine 2003;29:S89

52. Guzman JA; Bander JJ. Guzman CB. Steroids in septic shock; Limited value of cortisol testing to guide replacement therapy. Critical Care Medicine 2005;33:A169

53. Tandan SM, Guleria R, Gupta N. Low dose steroids and adre-nocortical insufficiency in septic shock: a double-blind ran-domised controlled trial from India. In: Proceedings of the American Thoracic Society Meeting. New York, NY: American Thoracic Society 2005:A24

54. Rinaldi S, Adembri C, Grechi S, De Gaudio AR. Low-dose hydrocortisone during severe sepsis: effects on microalbumin-uria. Crit Care Med 2006;34:2334–9

55. Loisa P, Parviainen I, Tenhunen J, Hovilehto S, Ruokonen E. Effect of mode of hydrocortisone administration on glycemic

control in patients with septic shock: a prospective randomized trial. Crit Care 2007;11:R21

56. McGee WT, Higgins TL, Jodka P. Optimal steroid therapy may improve outcome in septic shock (abstract). Crit Care Med 2007;12:A117

57. Cicarelli DD, Vieira JE, Benseñor FE. Comparison of C-reactive protein and serum amyloid a protein in septic shock patients. Mediators Inflamm 2008;2008:631414.

58. Kurugundla N, Irugulapati L, Kilari D. Effect of steroids in septic shock patients without relative adrenal insufficiency-a pilot study [abstract]. American Thoracic Society: International Conference 2008:A116

59. Büchele GL, Silva E, Ospina-Tascón GA, Vincent JL, De Backer D. Effects of hydrocortisone on microcirculatory alterations in patients with septic shock. Crit Care Med 2009;37:1341–7

60. Hayashi M. Activated protein c and corticosteroids decrease the rate of albumin transudation in septic shock (abstract). Crit Care Med 2009;34:A407

61. Hu B, Li JG, Liang H, Zhou Q, Yu Z, Li L, Luo Y, Liu C, Gan Q. [The effect of low-dose hydrocortisone on requirement of norepinephrine and lactate clearance in patients with refrac-tory septic shock]. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2009;21:529–31

62. Russell JA, Walley KR, Gordon AC, Cooper DJ, Hébert PC, Singer J, Holmes CL, Mehta S, Granton JT, Storms MM, Cook DJ, Presneill JJ; Dieter Ayers for the Vasopressin and Septic Shock Trial Investigators. Interaction of vasopressin infusion, corticosteroid treatment, and mortality of septic shock. Crit Care Med 2009;37:811–8

63. Valoor HT, Singhi S, Jayashree M. Low-dose hydrocortisone in pediatric septic shock: an exploratory study in a third world setting. Pediatr Crit Care Med 2009;10:121–5

64. Beale R, Janes JM, Brunkhorst FM, Dobb G, Levy MM, Martin GS, Ramsay G, Silva E, Sprung CL, Vallet B, Vincent JL, Costigan TM, Leishman AG, Williams MD, Reinhart K. Global utilization of low-dose corticosteroids in severe sepsis and septic shock: a report from the PROGRESS registry. Crit Care 2010;14:R102

65. Jung B, Nougaret S, Chanques G, Mercier G, Cisse M, Aufort S, Gallix B, Annane D, Jaber S. The absence of adrenal gland enlargement during septic shock predicts mortality: a com-puted tomography study of 239 patients. Anesthesiology 2011;115:334–43

66. Schelling G, Briegel J, Roozendaal B, Stoll C, Rothenhäusler HB, Kapfhammer HP. The effect of stress doses of hydrocortisone during septic shock on posttraumatic stress disorder in survi-vors. Biol Psychiatry 2001;50:978–85

67. Huh JW, Choi HS, Lim CM, Koh Y, Oh YM, Shim TS, Lee SD, Kim WS, Kim DS, Hong SB. Low-dose hydrocortisone treat-ment for patients with septic shock: a pilot study comparing 3 days with 7 days. Respirology 2011;16:1088–95

16 low dose hydrocortisone therapy attenuates septic.16

Documents

effects of septic shock

treatment of septic

attenuating septic shock

incidence of septic

shock reversal

dose hydrocortisone

lowdose steroid therapy

dose glucocorticoid