- Email: [email protected]
Survival benefits of dexmedetomidine used for sedating septic patients in intensive care setting: A systematic review
Survival benefits of dexmedetomidine used for sedating septic patients in intensive care setting: A systematic review Mohammad Mahdi Zamani, Mahsa Keshavarz-Fathi, Maryam Sadat Fakhri-Bafghi, Armin Hirbod-Mobarakeh, Nima Rezaei, Amir Bahrami, Nader D. Nader PII: DOI: Reference:
S0883-9441(15)00564-X doi: 10.1016/j.jcrc.2015.11.013 YJCRC 52006
To appear in:
Journal of Critical Care
Please cite this article as: Zamani Mohammad Mahdi, Keshavarz-Fathi Mahsa, FakhriBafghi Maryam Sadat, Hirbod-Mobarakeh Armin, Rezaei Nima, Bahrami Amir, Nader Nader D., Survival benefits of dexmedetomidine used for sedating septic patients in intensive care setting: A systematic review, Journal of Critical Care (2015), doi: 10.1016/j.jcrc.2015.11.013
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
T
Survival benefits of dexmedetomidine used for sedating septic patients in intensive care setting: A systematic review
RI P
Authors:
Mohammad Mahdi Zamani1,2,3, Mahsa Keshavarz-Fathi1, Maryam Sadat Fakhri-Bafghi1, Armin Hirbod-
SC
Mobarakeh1,3,4, Nima Rezaei1,3,4,5, Amir Bahrami1,2, Nader D. Nader6
NU
Affiliations: 1
Border of Immune Tolerance Education and Research Network (BITERN), Universal Scientific Education and
MA
Research Network (USERN), Tehran, Iran 2
Department of Anesthesiology and Critical Care, School of Medicine, Tehran University of Medical Sciences,
Tehran, Iran
ED
3
Research Center for Immunodeficiencies, Children's Medical Center, Tehran University of Medical Sciences,
PT
Tehran, Iran 4
Molecular Immunology Research Center; and Department of Immunology, School of Medicine, Tehran University
Department of Infection and Immunity, School of Medicine and Biomedical Sciences, University of Sheffield,
Sheffield, UK 6
AC
5
CE
of Medical Sciences, Tehran, Iran
University of Buffalo, Buffalo, United States of America
Corresponding author: Nader D. Nader MD, PhD, FACC, FCCP Professor and Senior Vice Chair, Dept. of Anesthesiology, 252 Farber Hall, Buffalo, NY 14214 Tel: +1 (716) 345-7909 E-mail: [email protected]
1
ACCEPTED MANUSCRIPT 2
ABSTRACT
T
Purpose: The aim of this systematic review was to evaluate the effectiveness and safety of dexmedetomidine used
RI P
for sedation of patients with sepsis.
Methods: We searched Medline, Scopus, TRIP and CENTRAL, DART, OpenGrey and ProQuest without applying
SC
any language filter up to 15 July 2015. Two of the authors independently reviewed search results for irrelevant and duplicate studies and extracted data and assessed methodological quality of the studies. We used tabulation to
treatment effect across studies using Review Manager.
NU
synthesize the findings of the studies and transformed data into a common rubric and calculated a weighted
MA
Results: We found 124 references in 7 databases and after exclusion of irrelevant and duplicate studies, 6 studies with total number of 242 patients with sepsis were included. The risk ratio for 28-day mortality was 0.49 (95% CI:
ED
0.24, 0.99; P value = 0.05) for dexmedetomidine group versus control group. The weighted mean difference for the length of stay in the ICU was 1.54 (95% CI: -1.73, 4.81; P value = 0.36). No side effect including hypertensive,
PT
hypotensive, or bradycardia response was reported in any studies.
CE
Conclusion: In a small group of studies of patients with sepsis, dexmedetomidine improved short-term mortality compared to other sedatives without affecting the ICU length of stay. Further studies are warranted to confirm
agents.
AC
whether using this particular agent improves sepsis outcomes in comparison to other commonly used sedating
Keywords: alpha-2 agonists, sedation, sepsis, critical care, mortality
2
ACCEPTED MANUSCRIPT 3
INTRODUCTION Septic shock is a life threatening condition that results from catastrophic effects of an exaggerated immune
T
response to infectious agents.1-3 This systemic immune response leads to activation of inflammatory, coagulation
RI P
and fibrinolysis cascades and consequently leads to collateral tissue damage and multiple organ failure.2, 4, 5 On the other hand, the compensatory anti-inflammatory response paves the way for secondary infections.4 In United States,
SC
2% of patients admitted to the hospital has sever sepsis.4 Worldwide, there are up to 19 million cases of severe sepsis each year and this incidence is increasing by 8.7% per year.1,
4, 6-8
Recent advances and breakthroughs in
NU
bundle care significantly decreased risk of immediate death associated with severe sepsis and septic shock and as a result, imminent death rate declined from 80% to 20-30%.1, 4
MA
In view of decreased mortality, support of organ function has gained higher significance in the intensive care units (ICU).1, 2 Sepsis and septic shock commonly results in cardiovascular and respiratory compromise and 1, 2, 4
Clinical picture of severe sepsis is often complicated with acute
ED
central nervous system (CNS) dysfunction
respiratory distress syndrome (ARDS) that necessitate mechanical support of ventilation.1, 6 Proper sedation is often
PT
a necessity in care of patients with sepsis in need of mechanical ventilation to reduce the stress and anxiety associated with tracheal intubation and other invasive interventions.1,
2, 4
The choice of such sedative agent is
CE
critically important as patients with sepsis are generally in shock and extremely vulnerable for cardiovascular compromise.6 Traditionally, gamma amino-butyric acid (GABA) receptor agonists such as benzodiazepines and
AC
propofol are commonly administered sedative agents in the ICU 6, 9. Recently, studies have highlighted sedative and analgesic properties of selective α2 receptor agonists such as dexmedetomidine
10, 11
. Following its binding to
transmembrane G protein adrenoreceptors, dexmedetomidine inhibits of protein kinase A and leads in phosphorylation of downstream enzymes such as adenylate cyclase.12 Hyperpolarization of noradrenergic neurons in the locus ceruleus mediates the sedative effects of dexmedetomidine while its analgesic effect is a result of modulation of pain impulses in the noradrenergic pathways in the posterior horns of the spinal cord. 7, 8, 12 Limited but increasing evidence suggests that dexmedetomidine has a promising future as a sedative agent in the intensive care setting considering its excellent sedative and analgesic properties with wider safety margin due to the lack of suppressive effects on respiration.12,
13
Dexmedetomidine seems to have effects on apoptosis and
modulation of the immune system which might be particularly important and play a critical role in the pathogenesis
3
ACCEPTED MANUSCRIPT 4
of sepsis.7,
14
On the other hand, hypotension and bradycardia, the most common adverse effects of
dexmedetomidine could influence hemodynamic stability in patients with septic shock.15,
16
Therefore, despite
T
extensive research, its potential benefits and risks in patients with sepsis remains a controversy. The aim of
RI P
systematic review was to evaluate the effect of dexmedetomidine on the duration of ICU stay and 28-day mortality of patients diagnosed with sepsis, severe sepsis or septic shock according to the guidelines of Border of Immune
METHODS
NU
SC
Tolerance Education and Research Network (BITERN) and Cochrane collaboration.
The methods described in this systematic review were in accordance with BITERN guidelines and general
Criteria for considering studies for this review
MA
methods recommended by Cochrane collaboration.
ED
In this review, we included all clinical trials (irrespective of randomization and blinding) investigating
PT
dexmedetomidine (irrespective of modes of administration and all variations of dosage, frequency and duration) in patients with sepsis, severe sepsis or septic shock (irrespective of age, gender or race) according to inclusion criteria
CE
stated in the protocol (appendix 1).
Search methods for identification of studies
AC
One author conducted the primary search process on 15 July 2014 in Medline, Scopus, TRIP and CENTRAL databases (as databases for journal article) and DART, OpenGrey and ProQuest (as databases for grey literature) according to the search strategies stated in the protocol (appendix 1). We did not apply any language filter or date restriction and we updated the search process in Medline, Scopus, CENTRAL, ProQuest, DART and OpenGrey databases up to 15 July 2015. In addition, the reference lists of articles identified were searched for relevant trials. Citations from all databases were imported into an Endnote library (version X6, Thomson Reuters, USA). In the Endnote library, we used the „„Find Duplicates‟‟ feature of Endnote software to identify the duplicates among citations. Then two of the authors independently reviewed title and abstract of the remainders of search results for irrelevant studies and obvious irrelevant studies were excluded. We retrieved the full-text of the remaining citations for further screening and data collection process.
4
ACCEPTED MANUSCRIPT 5
Data collection and quality assessment Two reviewers independently examined the full-text of the articles for eligibility according to the inclusion
T
criteria (appendix 1). Reviewers resolved ambiguity or any disagreement regarding the eligibility of studies through
RI P
either discussion or consultation with a third author. They excluded ineligible studies along with documenting reasons for exclusion. Two reviewers independently extracted data from articles to pre-designed and pre-tested data
SC
extraction forms in Microsoft Excel spreadsheets (version 2010, Microsoft Corporation, USA). They also assessed methodological quality of the studies independently by using a modified version of Cochrane “Risk of bias” tool on
NU
following domains (developed by scientific committee of BITERN) (Appendix 2): Random sequence generation (selection bias for controlled trials)
2.
Allocation concealment (selection bias for controlled trials)
3.
Generalizability of the findings to the target population (selection bias for trials without control)
4.
Blinding of participants and personnel (performance bias)
5.
Blinding of outcome assessment (detection bias)
6.
Incomplete outcome data (attrition bias)
7.
Selective reporting (reporting bias)
PT
ED
MA
1.
CE
Each domain was judged to be “low risk” of bias, “high risk” of bias, or “unclear risk” of bias. Any disagreements between data collectors were resolved through either discussion or consultation with a third author.
AC
Furthermore, the quality of the studies was assessed using Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology as described elsewhere.17
This methodology is widely
accepted and used by the guideline-writing committees, which includes assessment of the evidence whether it addresses same population, intervention, comparison and outcome. In brief, the quality of evidence was graded as “Low”, “Moderate” or “High” and the results were interpreted accordingly. Mortality rate with 28-days was the main primary outcome for this study which was invariably addressed within all studies. However, later days of mechanical ventilation and ICU length of stay were added to the existing protocol, retrospectively.
5
ACCEPTED MANUSCRIPT 6
Evidence synthesis We undertook systematic approaches to synthesize the findings of the studies since there was clinical
T
heterogeneity in the included studies. We used tabulation and textual description to synthesize the findings of the
RI P
studies. In addition, for all studies we transformed data into a common rubric and presented dichotomous outcomes as risk ratio (RR) and 95% confidence interval (CI), and continuous outcomes as mean difference (MD) and 95%
SC
CI. All conversions were done using Review Manager (Version 5.3. Copenhagen: The Nordic Cochrane Centre, the Cochrane Collaboration, 2014). For missing data, we considered statistics that allow calculation of missing data in
NU
article or in some cases we contacted the corresponding author.
When between-study heterogeneity allowed, we calculated a weighted treatment effect across studies using
MA
Review Manager. We assessed between-study heterogeneity by the chi-square statistic and its P value, and the extent of inconsistency using the I2 statistic. We considered a P value < 0.1 and I2 > 40% as indicating significant
ED
between-study heterogeneity. In case of significant heterogeneity, we did meta-analysis using a random-effects meta-analysis; otherwise a fixed-effect model was used. We expressed the results as RRs with 95% CI for
RESULTS
AC
Description of studies
CE
PT
dichotomous outcomes, and mean differences (MD, 95% CI) for continuous outcomes.
We found 124 references by recruiting the search strategy in 7 databases (Figure 1). We did not retrieve any studies in reference lists of the main articles. After discarding duplicates, we identified 100 publications. In primary screening of titles and abstracts, 75 articles were excluded due to obvious irrelevancy of the topics. In secondary screening of full-text articles, 6 studies with total number of 242 patients with sepsis or septic shock were identified and included in this systematic review (Table 1). Four studies included patients with either severe sepsis or septic shock and two trials included only patients with septic shock.6, 18-22 Three studies compared dexmedetomidine vs. propofol; two studies compared dexmedetomidine vs. lorazepam and one study compared dexmedetomidine vs. midazolam.6, 18-22 A loading dose of 0.01 to 1 µg/kg/h was used in all studies. The maximum maintenance doses of dexmedetomidine ranged between 0.15 µg/kg/h and 0.5
6
ACCEPTED MANUSCRIPT 7
mg/kg/h. Target sedation levels were Ramsay Score 2 in 3 studies and Ramsay Score 4 in one study by adjustment to the sedative regimen.19-22 In 5 studies, intermittent doses of fentanyl or alfentanil was used to treat apparent pain
RI P
T
based on physiological parameters in addition to facial expressions, limb movement, and ventilator synchrony.6, 18-21 Clinical outcomes investigated in these studies included 28-day mortality, days of ICU care and mechanical
SC
ventilation (MV) in 28-day period, delirium and coma-free days, development of secondary infections and organdysfunction free days. Pathophysiologic outcomes reported were inflammatory responses, intra-abdominal pressure
NU
(IAP) (the steady-state pressure within the abdominal cavity), gastric intramucosal pH, absolute and relative CO2 reactivity (as a measure of cerebral blood flow and cerebral autoregulation). Safety outcomes and adverse events
MA
such as hypertensive, hypotensive, or bradycardic response to infusion of dexmedetomidine were reported in 5 studies. The overall quality of the studies was good and there was not any high risk of bias for any domain (Table 2).
a.
ED
Effects of interventions Clinical outcomes
PT
Data for 28-day mortality and the length of ICU stay in 28-day period were available for 3 trials and for
CE
143 patients.6, 20, 21 There were 9 out of 71 patients in the experimental group who died by day 28 comparing to 19 out of 72 participants in the control group. There was not significant heterogeneity in the results (χ² = 0.57, P = 0.75;
AC
I² = 0%) and therefore we employed fixed effect model for meta-analysis of results. The risk ratio of dying within 28 days was 0.49 (95% CI: 0.24, 0.99; P = 0.05; fixed-effects model) for patients in experimental group vs. control group (Figure 2). In addition to 28-day mortality, only one study reported the first day mortality which was not significant (P =0.64) (Table 1).19 Three studies reported numbers of days in the ICU during 28 days of treatment for a total number of 143 patients.6, 20, 21 The weighted mean difference for the length of stay in the ICU was 1.54 (95% CI: -1.73, 4.81; P = 0.36; fixed-effects model) with no heterogeneity across the studies (χ² = 0.07, P = 0.96; I² = 0%; Figure 3). Two studies with 103 patients reported that there was not any significant association between sedation with dexmedetomidine and duration of mechanical ventilation (Table 1).6,
7
21
The weighted mean difference for the
ACCEPTED MANUSCRIPT 8
mechanical ventilation free days during 28-day period was 1.40 (95% CI: -4.44, 7.24; P = 0.64; random-effects
T
model) with significant heterogeneity across the studies (χ² = 3.25, P = 0.07; I² = 69%)
RI P
Sanders et al evaluated delirium in 63 patients with sepsis using confusion assessment method for the ICU (CAM-ICU) and showed that patients treated with dexmedetomidine had significantly more delirium/coma-free
SC
days in 12 days compared to those patients treated with lorazepam (Table 1).6 In this study, the beneficial effects of dexmedetomidine vs. lorazepam were more prominent on development of attention loss (CAM-ICU Feature 2) and
NU
disorganized thinking (CAM-ICU Feature 3).6 This study reported that this beneficiary effect of dexmedetomidine on delirium/ coma-free days was seen only in patients with sepsis but not in those without sepsis.
MA
Hemodynamic and biochemical parameters and kidney, liver, lung, cardiac and coagulation functions were analyzed in all studies. King et al reported that patients treated with dexmedetomidine had significantly more liver-
ED
dysfunction-free days comparing to patients treated with lorazepam.18 However, no statistically significant differences in hemodynamic and biochemical parameters were reported between different groups in other 5 studies
PT
(exact values were not reported). Four studies compared and reported level of sedation and need for analgesics between patients in experimental group and control group.6, 19-21 Three of them reported that both level of sedation
CE
and need for analgesics were same across experimental and control groups
19-21
. Sanders et al reported that sedation
with dexmedetomidine was significantly more efficacious than sedation with lorazepam. However, the need for
b.
AC
analgesic was greater in patients treated with dexmedetomidine.6 Pathophysiologic outcomes Two studies with 40 patients reported serum levels of tumor necrosis factor alpha (TNFα), interleukin 1 beta (IL-1β) and IL-6 as inflammatory markers after 24 hours.19,
20
Since we found no heterogeneity, we meta-
analyzed results of these studies using fixed-effects model. Levels of TNFα, IL-1β and IL-6 after 24 hours of treatment were significantly lower in patients treated with dexmedetomidine. The weighted mean difference for levels of TNFα after 24 hours was -5.31 (95% CI: -8.00, -2.63; P = 0.0001; fixed-effects model) and for IL-1β and IL-6 after 24 hours were respectively -1.24 (95% CI: -1.88, -0.60; P = 0.0001; fixed-effects model) and -250.77 (95% CI: -371.78, -129.75; P < 0.0001; fixed-effects model). Memis et al also reported significantly lower levels of TNF-α and IL-1β after 48 hours (Table 1).19
8
ACCEPTED MANUSCRIPT 9
One study evaluated IAP in 40 patients with severe sepsis and compared effects of dexmedetomidine with propofol on IAP in these patients.20 Both IAP after 24 hours or 48 hours were significantly lower in 20 patients
T
treated with dexmedetomidine comparing to those treated with propofol (Table 1). Kadoi et al reported that absolute
RI P
and relative CO2 reactivity as a measure of cerebral blood flow and cerebral autoregulation in 20 patients with septic shock was significantly lower under dexmedetomidine than propofol sedation in patients with septic shock.22 Such
Drug Safety
NU
c.
SC
association was not present in patients without sepsis.
No side effect including hypertensive, hypotensive, or bradycardic response to either loading dose or
MA
maintenance dose was reported in patients in any studies. Four studies on 183 patients reported that there were no significant differences in use of inotropic agents, bradycardia or tachycardia, and mean arterial pressure during study
ED
hours between experimental and control groups.6, 19-21 D ISCUSSION
PT
In this review, we performed a comprehensive search of the literature and included 6 clinical trials that evaluated the effect of dexmedetomidine in patients with sepsis or septic shock comparing to other sedatives
CE
commonly used in ICU. The systematic review of six studies showed that dexmedetomidine had beneficial effect on 28-day mortality (moderate quality of evidence based on GRADE approach) and delirium/coma (moderate quality
AC
of evidence based on GRADE approach) in patients with sepsis but no significant effect on length of ICU stay or duration of mechanical ventilation (moderate quality of evidence based on GRADE approach). The survival benefit was prominent in comparison with lorazepam not with propofol. Previous studies have shown that dexmedetomidine has protective effects on several organs including heart, lungs, nervous system and kidneys.12 Dexmedetomidine has a potential to stabilize the hemodynamic profile by attenuating sympathetically-mediated hyperdynamic responses, which may further translate to myocardial protection.12 In addition, dexmedetomidine may offer neuroprotective effects probably through modulating neurotransmitter release in the central and peripheral sympathetic nervous system and preventing apoptosis of the neurons.12 Both neuroprotective and anti-inflammatory effects of dexmedetomidine may play an important role in protective effects of dexmedetomidine on delirium as reported by Sanders et al.6, 23, 24 GABA receptors play a critical
9
ACCEPTED MANUSCRIPT 10
role in modulation of several neurotransmitters responsible for development of delirium in clinical settings.9,
24
Sedative effects of dexmedetomidine are unique in a sense that they are mediated through pathways independent of
T
the GABA receptors and hence has intrinsic delirium-sparing effect.25 Unlike GABA-mediated sedation,
RI P
dexmedetomidine enhances non-rapid eye movement sleep and induces a more natural sleep cycle.9, 12, 26, 27 Sleep deprivation has been associated with higher levels of inflammatory responses, increased insulin resistance and
clinical outcomes such as delirium-free days and mortality.6, 30
SC
activation of the hypothalamic-pituitary axis.28, 29 Therefore, dexmedetomidine effect on sleep quality may improve
NU
Dexmedetomidine further enhances mucosal immunity and bacterial clearance by promoting macrophage phagocytosis and bactericidal killing, properties that are critically important in sepsis and septic shock. 11, 31, 32 In
MA
Safety and Efficacy of Dexmedetomidine Compared with Midazolam (SEDCOM) trial, critically ill patients sedated with dexmedetomidine had reduced number of secondary infections, though such effect was not detected in our
ED
review.11 Results of this systematic review suggest that dexmedetomidine can decrease levels of TNF-α, IL-1β and IL-6 in patients with sepsis after 24 hours and 48 hours of treatment. This anti-inflammatory effect of
PT
dexmedetomidine can suppress the exaggerated production of inflammatory cytokines in septic shock.1, 3, 5, 33-37 Sepsis often impacts CNS as the first organ. Impaired cerebral autoregulation is one of the proposed 39
Kadoi et al reported that absolute and relative CO2 reactivity as a measure of
CE
mechanisms for such effect.38,
cerebral blood flow and cerebral autoregulation was significantly lower with dexmedetomidine than propofol
AC
sedation in patients with septic shock.22 This can be justified by cerebrovascular constrictor effect of dexmedetomidine compared to propofol.40 Both IAP after 24 hours or 48 hours were significantly lower in 20 patients treated with dexmedetomidine comparing to those treated with propofol. Increase in IAP not only has negative effects on splanchnic, respiratory, cardiovascular, renal, and neurological function but is also associated with high mortality.20, 41-43 This systematic review yielded no adverse events (with the exception of bradycardia) in using of dexmedetomidine and there were no differences in liver, renal, cardiac, or endocrine safety outcomes between septic patients treated with dexmedetomidine and other sedative agents. Dexmedetomidine by inhibiting central norepinephrine release can cause hypotension and bradycardia.12 Such an effect would be concerning in septic patients who are at risk for the development of cardiovascular shock.6 However, it seems that a reduction in
10
ACCEPTED MANUSCRIPT 11
proinflammatory cytokines would outweigh any direct hypotensive effect of dexmedetomidine.6 Bradycardia induced by dexmedetomidine is primarily dose-dependent and can be prevented by using slow administration of the 12
T
bolus loading or omitting bolus loading all together.
RI P
Major limitations of this meta-analysis included the small sample sizes of included trials and incomplete reporting of certain outcome measures, which limited our meta-analysis for most of the outcome variables. The
SC
results of this systematic review were based on the available literature including 6 studies with total number of 242 patients. Despite limited sample sizes and data, their overall quality of included studies was good and there were no
NU
high risk of bias in their design. However, it is of note that the randomization method was not described in some papers. Additionally, publication bias was not assessed using a funnel plot due to the small number of studies (< 10)
MA
that included in the meta-analyses of the outcome measures.
In conclusion, results of this systematic review based on the available literature suggest that
ED
dexmedetomidine in patients with sepsis or septic shock may improve short-term clinical outcomes as well as some pathophysiologic outcomes. However, considering limited number of studies with limited sample sizes in the
PT
literature, the beneficial effects found in this systematic review, only justify further studies to evaluate long-term as well as short-term outcomes of dexmedetomidine sedation in patients with sepsis and to compare its effects with
AC
CE
different types and dosages of sedative agents.
11
ACCEPTED MANUSCRIPT 12
REFERENCES 1.
Angus DC, van der Poll T: Severe sepsis and septic shock. The New England journal of medicine
2.
RI P
T
369:2063, 2013
Cawcutt KA, Peters SG: Severe sepsis and septic shock: clinical overview and update on management.
SC
Mayo Clinic proceedings 89:1572-1578, 2014
Pravda J: Metabolic theory of septic shock. World journal of critical care medicine 3:45-54, 2014
4.
Angus DC, van der Poll T: Severe sepsis and septic shock. The New England journal of medicine 369:840-
NU
3.
5.
MA
851, 2013
De Backer D, Orbegozo Cortes D, Donadello K, et al.: Pathophysiology of microcirculatory dysfunction
Pandharipande PP, Sanders RD, Girard TD, et al.: Effect of dexmedetomidine versus lorazepam on
PT
6.
ED
and the pathogenesis of septic shock. Virulence 5:73-79, 2014
outcome in patients with sepsis: an a priori-designed analysis of the MENDS randomized controlled trial.
O'Mara K, Gal P, Wimmer J, et al.: Dexmedetomidine versus standard therapy with fentanyl for sedation in
AC
7.
CE
Critical care 14:R38, 2010
mechanically ventilated premature neonates. The journal of pediatric pharmacology and therapeutics : JPPT : the official journal of PPAG 17:252-262, 2012
8.
Wujtewicz M, Maciejewski D, Misiolek H, et al.: Use of dexmedetomidine in the adult intensive care unit. Anaesthesiology intensive therapy 45:235-240, 2013
9.
Pandharipande PP, Pun BT, Herr DL, et al.: Effect of sedation with dexmedetomidine vs lorazepam on acute brain dysfunction in mechanically ventilated patients: the MENDS randomized controlled trial. Jama 298:2644-2653, 2007
12
ACCEPTED MANUSCRIPT 13
10.
Adams R, Brown GT, Davidson M, et al.: Efficacy of dexmedetomidine compared with midazolam for sedation in adult intensive care patients: a systematic review. British journal of anaesthesia 111:703-710,
11.
RI P
T
2013
Riker RR, Shehabi Y, Bokesch PM, et al.: Dexmedetomidine vs midazolam for sedation of critically ill
12.
SC
patients: a randomized trial. Jama 301:489-499, 2009
Afonso J, Reis F: Dexmedetomidine: current role in anesthesia and intensive care. Revista brasileira de
13.
NU
anestesiologia 62:118-133, 2012
Farag E, Argalious M, Abd-Elsayed A, et al.: The use of dexmedetomidine in anesthesia and intensive care:
Cui J, Zhao H, Wang C, et al.: Dexmedetomidine attenuates oxidative stress induced lung alveolar
ED
14.
MA
a review. Current pharmaceutical design 18:6257-6265, 2012
15.
PT
epithelial cell apoptosis in vitro. Oxidative medicine and cellular longevity 2015:358396, 2015
Lin YY, He B, Chen J, et al.: Can dexmedetomidine be a safe and efficacious sedative agent in post-cardiac
16.
CE
surgery patients? a meta-analysis. Critical care 16:R169, 2012
Piao G, Wu J: Systematic assessment of dexmedetomidine as an anesthetic agent: a meta-analysis of
17.
AC
randomized controlled trials. Archives of medical science : AMS 10:19-24, 2014
Guyatt GH, Oxman AD, Vist GE, et al.: GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. Bmj 336:924-926, 2008
18.
King MS, Ely EW, Girard TD, et al.: Sepsis modifies the beneficial effects of dexmedetomidine on nonbrain organ dysfunction: an a priori sub group analysis of the MENDS randomized controlled trial [Abstract]: American Thoracic Society International Conference, May 15-20, 2009, San Diego, 2009, p A4698 [Poster #K4677]
13
ACCEPTED MANUSCRIPT 14
19.
Memis D, Hekimoglu S, Vatan I, et al.: Effects of midazolam and dexmedetomidine on inflammatory responses and gastric intramucosal pH to sepsis, in critically ill patients. British journal of anaesthesia
20.
RI P
T
98:550-552, 2007
Tasdogan M, Memis D, Sut N, et al.: Results of a pilot study on the effects of propofol and dexmedetomidine on inflammatory responses and intraabdominal pressure in severe sepsis. Journal of
Memis D, Kargi M, Sut N: Effects of propofol and dexmedetomidine on indocyanine green elimination
NU
21.
SC
clinical anesthesia 21:394-400, 2009
assessed with LIMON to patients with early septic shock: a pilot study. Journal of critical care 24:603-608,
22.
MA
2009
Kadoi Y, Saito S, Kawauchi C, et al.: Comparative effects of propofol vs dexmedetomidine on cerebrovascular carbon dioxide reactivity in patients with septic shock. British journal of anaesthesia
de Rooij SE, van Munster BC, Korevaar JC, et al.: Cytokines and acute phase response in delirium. Journal
PT
23.
ED
100:224-229, 2008
24.
CE
of psychosomatic research 62:521-525, 2007
Pandharipande P, Ely EW: Sedative and analgesic medications: risk factors for delirium and sleep
25.
AC
disturbances in the critically ill. Critical care clinics 22:313-327, vii, 2006
Yapici N, Coruh T, Kehlibar T, et al.: Dexmedetomidine in cardiac surgery patients who fail extubation and present with a delirium state. The heart surgery forum 14:E93-98, 2011
26.
Nelson LE, Guo TZ, Lu J, et al.: The sedative component of anesthesia is mediated by GABA(A) receptors in an endogenous sleep pathway. Nature neuroscience 5:979-984, 2002
27.
Nelson LE, Lu J, Guo T, et al.: The alpha2-adrenoceptor agonist dexmedetomidine converges on an endogenous sleep-promoting pathway to exert its sedative effects. Anesthesiology 98:428-436, 2003
28.
Hardin KA: Sleep in the ICU: potential mechanisms and clinical implications. Chest 136:284-294, 2009
14
ACCEPTED MANUSCRIPT 15
29.
Weinhouse GL, Schwab RJ, Watson PL, et al.: Bench-to-bedside review: delirium in ICU patients -
Sanders RD, Xu J, Shu Y, et al.: Dexmedetomidine attenuates isoflurane-induced neurocognitive
RI P
30.
T
importance of sleep deprivation. Critical care 13:234, 2009
impairment in neonatal rats. Anesthesiology 110:1077-1085, 2009
Weatherby KE, Zwilling BS, Lafuse WP: Resistance of macrophages to Mycobacterium avium is induced
SC
31.
32.
NU
by alpha2-adrenergic stimulation. Infection and immunity 71:22-29, 2003
Gets J, Monroy FP: Effects of alpha- and beta-adrenergic agonists on Toxoplasma gondii infection in
33.
MA
murine macrophages. The Journal of parasitology 91:193-195, 2005
King EG, Bauza GJ, Mella JR, et al.: Pathophysiologic mechanisms in septic shock. Laboratory
Maes M, Lin A, Kenis G, et al.: The effects of noradrenaline and alpha-2 adrenoceptor agents on the
PT
34.
ED
investigation; a journal of technical methods and pathology 94:4-12, 2014
CE
production of monocytic products. Psychiatry research 96:245-253, 2000
MacLaren R: Immunosedation: a consideration for sepsis. Critical care 13:191, 2009
36.
Taniguchi T, Kidani Y, Kanakura H, et al.: Effects of dexmedetomidine on mortality rate and inflammatory
AC
35.
responses to endotoxin-induced shock in rats. Critical care medicine 32:1322-1326, 2004
37.
Xiang H, Hu B, Li Z, et al.: Dexmedetomidine controls systemic cytokine levels through the cholinergic anti-inflammatory pathway. Inflammation 37:1763-1770, 2014
38.
Terborg C, Schummer W, Albrecht M, et al.: Dysfunction of vasomotor reactivity in severe sepsis and septic shock. Intensive care medicine 27:1231-1234, 2001
39.
Bowie RA, O'Connor PJ, Mahajan RP: Cerebrovascular reactivity to carbon dioxide in sepsis syndrome. Anaesthesia 58:261-265, 2003
15
ACCEPTED MANUSCRIPT 16
40.
Hinohara H, Kadoi Y, Takahashi K, et al.: Differential effects of propofol on cerebrovascular carbon
Malbrain ML, Chiumello D, Pelosi P, et al.: Incidence and prognosis of intraabdominal hypertension in a
RI P
41.
T
dioxide reactivity in elderly versus young subjects. Journal of clinical anesthesia 17:85-90, 2005
mixed population of critically ill patients: a multiple-center epidemiological study. Critical care medicine
42.
SC
33:315-322, 2005
Vidal MG, Ruiz Weisser J, Gonzalez F, et al.: Incidence and clinical effects of intra-abdominal
43.
NU
hypertension in critically ill patients. Critical care medicine 36:1823-1831, 2008
Regueira T, Bruhn A, Hasbun P, et al.: Intra-abdominal hypertension: incidence and association with organ
Figure Legends:
CE
Figure 1. Study flow diagram
PT
ED
MA
dysfunction during early septic shock. Journal of critical care 23:461-467, 2008
AC
Figure 2. Comparison of 28-day mortality between patients in Dexmedetomidine group and control group
Figure 3. Comparison of ICU stay during 28-day period between patients in Dexmedetomidine group and control group
16
ACCEPTED MANUSCRIPT
PT
ED
MA
NU
SC
RI P
T
17
AC
CE
Figure 1
17
ACCEPTED MANUSCRIPT
NU
SC
RI P
T
18
AC
CE
PT
ED
MA
Figure 2
18
ACCEPTED MANUSCRIPT
NU
SC
RI P
T
19
AC
CE
PT
ED
MA
Figure 3
19
ACCEPTED MANUSCRIPT 20
Table 1. Table of included studies and their main characteristics
Study patients and main groups
Loading dose: 0.2 mg/kg over 10 min
Maintenance : 0.1–0.5 mg/ kg/h
12-
20 patients in control group 20 patients in experimental group
NU
10 patients in control group a.
Kadoi et al 2008 (23)
b.
2-
Age (mean±SD): 65 ±8 APACHE II (mean± SD): 38 ± 4
1
2-
10 patients in experimental group a.
Age (mean± SD): 66 ±7
b.
APACHE II (mean±SD): 37 ±
Inflammatory markers after 24 hours (Mean difference and 95% CI and p value):
Loading dose: 1 µg/kg/h over 10 min
PT AC
CE
1-
Experimental group: Dexmedetomidine
20 patients with septic shock*
Hemodynamic and biochemical parameters: no statistically significant differences between groups in Mean arterial pressure, heart rate, Lactate, platelets, leucocytes, bilirubin, alanine aminotransferase, creatinine, and pH after 24 hours (values were not reported)
ED
Memis et al 2007 (20)
2-
1 day mortality (risk ratio and 95%CI and p value): 0.7 (0.1, 3.6); p value=0.6359
SC
Control group: Midazolam
1-
40 Critically ill patients with sepsis (demographic and clinical data were not reported)
Outcomes, results and relationships
RI P
Interventions
MA
Study (author, year, reference)
T
Articles reporting mortality outcomes
Maintenance: 0.2–2.5 µg/kg/h
1.
TNF-α (pg/mL): value=0.0035
2.
IL-1β (pg/mL): -1.29 (-2.2,-0.4), p value= 0.007
3.
IL-6 (pg/mL): value=0.0086
-4.9
-243.2
(-8,
-1.8),
P
(-415.1,-71.3),
p
Patients were maintained at a Ramsay sedation score <2 by adjustment to the sedative regimen
Gatric intramucosal pH (Mean difference and 95%CI and p value): -0.03 (-0.29,0.23), p value=0.8193
side-effects: no side effect was seen in both groups
Control group: Propofol
Hemodynamic parameters: no statistically significant differences between groups in Mean arterial pressure, heart rate and central venous pressure
Cerebral blood flow and cerebral autoregulation (Mean difference and 95% CI and p value)
Loading dose: 1 mg/kg over 10 min Maintenance : 1–3 mg/ kg/h infusion Experimental group: Dexmedetomidine
Loading dose: 0.01 µg/kg/h over 10 min
Maintenance:
20
1-
Absolute CO2 reactivity (cm/s/ mm Hg): -0.6 (0.9, -0.3), p value: 0.0003
2-
Relative CO2 reactivity: -0.7 (-1, -0.4), p value: 0.0003
Relationship: Sedation with Dexmedetomidine had a greater impact on absolute or relative CO2 reactivity in patients with sepsis than those without sepsis
ACCEPTED MANUSCRIPT 21
5
0.3–0.5 mg/kg/h
Loading dose: 1 mg/kg over 15 min
Days of ICU care in 28-day period (Mean difference and 95%CI and p value): 1.0 (-5.6, 7.6); p value: 0.7688
Maintenance : 1–3 mg/ kg/h
Days of MV in 28-day period (Mean difference and 95%CI and p value): 1 (-1.6, 3.6) p value= 0.4639
Gender (male/female): 11/9
SOFA score (mean±SD): 4.0 ± 2.5
Age (mean, range): 50 (19-74)
b.
Gender (male/female): 14/6
d.
Inflammatory markers after 24 and 48 hours (Mean difference and 95% CI and p value):
1-
Loading dose: 1 µg/kg/h over 10 min Maintenance: 0.2–2.5 µg/kg/h
After 24 hours: -6.55 (-11.9, -1.2), P value= 0.0213
o
After 48 hours: -25.3 (-38.6, -12), P value= 0.0006
o
After 24 hours: 0.0145
o
After 48 hours: -1.41 (-2.6,-0.2), p value= 0.0293
CE
SOFA score (mean±SD): 4.2 ± 1.8
40 Critically ill patients with septic shock 1-
20 patients in control group
Control group: Propofol Loading dose: 1 mg/kg over 15
21
IL-6 (pg/mL): After 24 hours: -258.2 (-428.6,-87.8), p value= 0.0051
o
After 48 hours: - 160.2 (-319,-1.4), p value= 0.0553
1-
-1.2 (-2.1,-0.3), p value=
o
Intra-abdominal pressure (mm Hg) difference and 95%CI and p value): 1-
Memis et al 2009 b (22)
IL-1β (pg/mL)
3-
APACHE II (mean±SD): 19 ± 5
TNF-α (pg/mL)
o
2-
PT
a.
Experimental group: Dexmedetomidine
ED
APACHE II (mean±SD): 18 ± 4
20 patients in experimental group
c.
2-
RI P
b.
NU
Age (mean, range): 50 (19-74)
T
MA
a.
d.
Memis et al 2009 a (21)
28-day mortality (risk ratio and 95%CI and p value): 0.5 (0.1-5.1); p value=0.5580
20 patients in control group
c.
2-
Control Propofol
AC
1-
group:
1-
SC
40 Critically ill patients with sepsis after ileus surgery
(Mean
After 24 hours: -5.8 (-8.6,-2.9), p value= 0.0003
After 48 hours: - 4.8 (-7.9,-1.7), p value= 0.0042
Haemodynamic and biochemical parameters: no statistically significant differences between groups in Mean arterial pressure, heart rate, temperature, and urine output, Lactate, platelets, leucocytes, bilirubin, alanine aminotransferase, creatinine, and pH
28-day mortality (risk ratio and 95%CI and p value): 0.8 (0.2-2.9) p value=0.6790
Days of ICU care in 28-day period (Mean difference and 95%CI and p value): 2 (-2.7, 6.7),
ACCEPTED MANUSCRIPT 22
Gender (male/female): 13/7
c.
APACHE II (mean±SD): 20 ± 8
d.
SOFA score (mean±SD): 4.0 ± 2.9
2-
Maintenance : 1–3 mg/ kg/h infusion Experimental group: Dexmedetomidine
Loading dose: 1 µg/kg/h over 10 min
Maintenance: 0.2–2.5 µg/kg/h
b.
Gender (male/female): 14/6
c.
APACHE II (mean±SD): 22 ± 5
d.
SOFA score (mean±SD): 4.5 ± 2.8
AC
CE
1-
39 Critically ill patients with sepsis king et al 2009 (19)
Sanders et al 2010 (6)
1-
20 patients in control group
2-
19 patients in experimental group
63 Critically ill patients ǂ with sepsis 1-
32 patients in control group
Hepatic blood flow (Mean difference and 95%CI and p value)
Indocyanine green plasma disappearance rate reactivity: -1.8 (-10.5, 6.9), p value: 0.6877
side-effects: no side effect was seen in both groups
MA
Age (mean, range): 60 (31-80)
Hemodynamic and biochemical parameters: no statistically significant differences between groups in Mean arterial pressure, heart rate, Lactate, platelets, bilirubin, alanine aminotransferase, creatinine
ED
a.
NU
20 patients in experimental group
PT
2-
p value: 0.4054
T
b.
min
RI P
Age (mean, range): 54 (25-78)
SC
a.
2-
Control group: Lorazepam Loading dose: 1 mg/h
1.
Experimental group:: 12 (12.12)
2.
Control group: 8.5 (2.12)
kidney dysfunction free days: no statistically significant differences
Experimental group: Dexmedetomidine
Lung dysfunction free days: no statistically significant differences
Hemodynamic and cardiac dysfunction free days: no statistically significant differences
coagulation dysfunction free days: no statistically significant differences
Control group: Lorazepam
28-day mortality (risk ratio and 95%CI and p value): 0.4 (0.2-0.9), p value=0.0247
Loading dose: 1 mg/h
28-day mortality (hazard ratio and 95%CI and p value): 0.3 (0.1, 0.9), p value= 0.0317
Loading dose: 0.15 µg/kg/h
Maintenance: 0.15-1.5 µg/kg/h
liver dysfunction free days (median, IQR) p=0.05:
Maintenance : 1–10 mg/h
1-
22
ACCEPTED MANUSCRIPT 23
Gender (male/female): 13/19
c.
APACHE II (mean, range): 29 (24 to 32)
d.
SOFA score (mean, range): 9 (8 to 12)
2-
Experimental group: Dexmedetomidin e
Loading dose: 0.15 µg/kg/h
Maintenance: 0.15-1.5 µg/kg/h
MA
b.
Age (mean, range): 60 (46 -65) Gender (male/female): 18/13
c.
APACHE II (mean, range): 22 ±5
d.
SOFA score (mean, range): 10 (9 to13)
MV-free days in 28 days (Mean difference and 95%CI and p value): 5.1 (-0.1, 10.3), p value= 0.0574
Days of ICU care in 28-day period (Mean difference and 95%CI and p value): 1.2 (-5.1, 7.5), p value= 0.7087
Delirium/coma free days in 12 days (Mean difference and 95%CI and p value): 3.2 (1.3, 5.1), p value= 0.0014
Delirium-free days in 12 days (Mean difference and 95%CI and p value): 1.4 (-0.1, 2.9), p value=0.0689
Coma-free days in 12 days (Mean difference and 95%CI and p value): 3.5 (1.7, 5.3), p value= 0.0003
Efficacy of sedation accurately Sedated days: 1.3 (1.2, 1.4), p value= 0.0001
Hemodynamic parameters: no statistically significant differences in Mean arterial pressure, heart rate, vasopressor use and cardiac arrhythmias. secondary infections: no statistically significant differences in Development of new secondary infections beyond the first 48 hour side-effects: no side effect was seen in both groups
ED
a.
NU
31 patients in experimental group
PT
2-
Maintenance : 1–10 mg/h
T
b.
RI P
Age (mean, range): 58 (44 to 66)
SC
a.
CE
Relationship: Sedation with Dexmedetomidine had a greater impact on delirium/ coma-free days in patients with sepsis than those without sepsis
*20 patients without septic shock were also included
ǂ40 patients without septic shock were also included
AC
23
Sanders et al 2010 (6) Kadoi et al 2008 (23) NA
king et al 2009 (19)
ED
Memis et al 2009 a (10)
PT
Memis et al 2009 b (22)
CE
AC
[Study ID]
Memis et al 2007 (20)
NA
Hemodynamic and biochemical parameters
Low risk of bias
Unclear risk of bias
High risk of bias
24 NA
T *
NA
Selective reporting for pathophysiologic outcomes
Selective reporting for clinical outcomes
Incomplete outcome data for pathophysiologic outcomes
RI P
Incomplete outcome data for clinical outcomes
SC
NU
Blinding of outcome assessment for pathophysiologic outcomes
MA
Blinding of outcome assessment for clinical outcomes
Blinding of participants and personnel for pathophysiologic outcomes
Blinding of participants and personnel for clinical outcomes
Allocation concealment bias (selection bias)
Random sequence generation (selection bias)
ACCEPTED MANUSCRIPT 24
Table 2
ACCEPTED MANUSCRIPT 25
Highlights:
T
RI P
SC
NU
MA ED
PT
CE
Following sepsis dexmedetomidine was associated with lower 28-mortality compared to other modalities of sedation. There was biochemical evidence for decreased state of systemic inflammation with dexmedetomidine is sepsis. Dexmedetomidine sedation did not change the length of ICU stay in patients with septic shock. Similarly, dexmedetomidine sedation did not affect weaning from mechanical ventilation in sepsis patients
AC
25
S0883-9441(15)00564-X doi: 10.1016/j.jcrc.2015.11.013 YJCRC 52006
To appear in:
Journal of Critical Care
Please cite this article as: Zamani Mohammad Mahdi, Keshavarz-Fathi Mahsa, FakhriBafghi Maryam Sadat, Hirbod-Mobarakeh Armin, Rezaei Nima, Bahrami Amir, Nader Nader D., Survival benefits of dexmedetomidine used for sedating septic patients in intensive care setting: A systematic review, Journal of Critical Care (2015), doi: 10.1016/j.jcrc.2015.11.013
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
T
Survival benefits of dexmedetomidine used for sedating septic patients in intensive care setting: A systematic review
RI P
Authors:
Mohammad Mahdi Zamani1,2,3, Mahsa Keshavarz-Fathi1, Maryam Sadat Fakhri-Bafghi1, Armin Hirbod-
SC
Mobarakeh1,3,4, Nima Rezaei1,3,4,5, Amir Bahrami1,2, Nader D. Nader6
NU
Affiliations: 1
Border of Immune Tolerance Education and Research Network (BITERN), Universal Scientific Education and
MA
Research Network (USERN), Tehran, Iran 2
Department of Anesthesiology and Critical Care, School of Medicine, Tehran University of Medical Sciences,
Tehran, Iran
ED
3
Research Center for Immunodeficiencies, Children's Medical Center, Tehran University of Medical Sciences,
PT
Tehran, Iran 4
Molecular Immunology Research Center; and Department of Immunology, School of Medicine, Tehran University
Department of Infection and Immunity, School of Medicine and Biomedical Sciences, University of Sheffield,
Sheffield, UK 6
AC
5
CE
of Medical Sciences, Tehran, Iran
University of Buffalo, Buffalo, United States of America
Corresponding author: Nader D. Nader MD, PhD, FACC, FCCP Professor and Senior Vice Chair, Dept. of Anesthesiology, 252 Farber Hall, Buffalo, NY 14214 Tel: +1 (716) 345-7909 E-mail: [email protected]
1
ACCEPTED MANUSCRIPT 2
ABSTRACT
T
Purpose: The aim of this systematic review was to evaluate the effectiveness and safety of dexmedetomidine used
RI P
for sedation of patients with sepsis.
Methods: We searched Medline, Scopus, TRIP and CENTRAL, DART, OpenGrey and ProQuest without applying
SC
any language filter up to 15 July 2015. Two of the authors independently reviewed search results for irrelevant and duplicate studies and extracted data and assessed methodological quality of the studies. We used tabulation to
treatment effect across studies using Review Manager.
NU
synthesize the findings of the studies and transformed data into a common rubric and calculated a weighted
MA
Results: We found 124 references in 7 databases and after exclusion of irrelevant and duplicate studies, 6 studies with total number of 242 patients with sepsis were included. The risk ratio for 28-day mortality was 0.49 (95% CI:
ED
0.24, 0.99; P value = 0.05) for dexmedetomidine group versus control group. The weighted mean difference for the length of stay in the ICU was 1.54 (95% CI: -1.73, 4.81; P value = 0.36). No side effect including hypertensive,
PT
hypotensive, or bradycardia response was reported in any studies.
CE
Conclusion: In a small group of studies of patients with sepsis, dexmedetomidine improved short-term mortality compared to other sedatives without affecting the ICU length of stay. Further studies are warranted to confirm
agents.
AC
whether using this particular agent improves sepsis outcomes in comparison to other commonly used sedating
Keywords: alpha-2 agonists, sedation, sepsis, critical care, mortality
2
ACCEPTED MANUSCRIPT 3
INTRODUCTION Septic shock is a life threatening condition that results from catastrophic effects of an exaggerated immune
T
response to infectious agents.1-3 This systemic immune response leads to activation of inflammatory, coagulation
RI P
and fibrinolysis cascades and consequently leads to collateral tissue damage and multiple organ failure.2, 4, 5 On the other hand, the compensatory anti-inflammatory response paves the way for secondary infections.4 In United States,
SC
2% of patients admitted to the hospital has sever sepsis.4 Worldwide, there are up to 19 million cases of severe sepsis each year and this incidence is increasing by 8.7% per year.1,
4, 6-8
Recent advances and breakthroughs in
NU
bundle care significantly decreased risk of immediate death associated with severe sepsis and septic shock and as a result, imminent death rate declined from 80% to 20-30%.1, 4
MA
In view of decreased mortality, support of organ function has gained higher significance in the intensive care units (ICU).1, 2 Sepsis and septic shock commonly results in cardiovascular and respiratory compromise and 1, 2, 4
Clinical picture of severe sepsis is often complicated with acute
ED
central nervous system (CNS) dysfunction
respiratory distress syndrome (ARDS) that necessitate mechanical support of ventilation.1, 6 Proper sedation is often
PT
a necessity in care of patients with sepsis in need of mechanical ventilation to reduce the stress and anxiety associated with tracheal intubation and other invasive interventions.1,
2, 4
The choice of such sedative agent is
CE
critically important as patients with sepsis are generally in shock and extremely vulnerable for cardiovascular compromise.6 Traditionally, gamma amino-butyric acid (GABA) receptor agonists such as benzodiazepines and
AC
propofol are commonly administered sedative agents in the ICU 6, 9. Recently, studies have highlighted sedative and analgesic properties of selective α2 receptor agonists such as dexmedetomidine
10, 11
. Following its binding to
transmembrane G protein adrenoreceptors, dexmedetomidine inhibits of protein kinase A and leads in phosphorylation of downstream enzymes such as adenylate cyclase.12 Hyperpolarization of noradrenergic neurons in the locus ceruleus mediates the sedative effects of dexmedetomidine while its analgesic effect is a result of modulation of pain impulses in the noradrenergic pathways in the posterior horns of the spinal cord. 7, 8, 12 Limited but increasing evidence suggests that dexmedetomidine has a promising future as a sedative agent in the intensive care setting considering its excellent sedative and analgesic properties with wider safety margin due to the lack of suppressive effects on respiration.12,
13
Dexmedetomidine seems to have effects on apoptosis and
modulation of the immune system which might be particularly important and play a critical role in the pathogenesis
3
ACCEPTED MANUSCRIPT 4
of sepsis.7,
14
On the other hand, hypotension and bradycardia, the most common adverse effects of
dexmedetomidine could influence hemodynamic stability in patients with septic shock.15,
16
Therefore, despite
T
extensive research, its potential benefits and risks in patients with sepsis remains a controversy. The aim of
RI P
systematic review was to evaluate the effect of dexmedetomidine on the duration of ICU stay and 28-day mortality of patients diagnosed with sepsis, severe sepsis or septic shock according to the guidelines of Border of Immune
METHODS
NU
SC
Tolerance Education and Research Network (BITERN) and Cochrane collaboration.
The methods described in this systematic review were in accordance with BITERN guidelines and general
Criteria for considering studies for this review
MA
methods recommended by Cochrane collaboration.
ED
In this review, we included all clinical trials (irrespective of randomization and blinding) investigating
PT
dexmedetomidine (irrespective of modes of administration and all variations of dosage, frequency and duration) in patients with sepsis, severe sepsis or septic shock (irrespective of age, gender or race) according to inclusion criteria
CE
stated in the protocol (appendix 1).
Search methods for identification of studies
AC
One author conducted the primary search process on 15 July 2014 in Medline, Scopus, TRIP and CENTRAL databases (as databases for journal article) and DART, OpenGrey and ProQuest (as databases for grey literature) according to the search strategies stated in the protocol (appendix 1). We did not apply any language filter or date restriction and we updated the search process in Medline, Scopus, CENTRAL, ProQuest, DART and OpenGrey databases up to 15 July 2015. In addition, the reference lists of articles identified were searched for relevant trials. Citations from all databases were imported into an Endnote library (version X6, Thomson Reuters, USA). In the Endnote library, we used the „„Find Duplicates‟‟ feature of Endnote software to identify the duplicates among citations. Then two of the authors independently reviewed title and abstract of the remainders of search results for irrelevant studies and obvious irrelevant studies were excluded. We retrieved the full-text of the remaining citations for further screening and data collection process.
4
ACCEPTED MANUSCRIPT 5
Data collection and quality assessment Two reviewers independently examined the full-text of the articles for eligibility according to the inclusion
T
criteria (appendix 1). Reviewers resolved ambiguity or any disagreement regarding the eligibility of studies through
RI P
either discussion or consultation with a third author. They excluded ineligible studies along with documenting reasons for exclusion. Two reviewers independently extracted data from articles to pre-designed and pre-tested data
SC
extraction forms in Microsoft Excel spreadsheets (version 2010, Microsoft Corporation, USA). They also assessed methodological quality of the studies independently by using a modified version of Cochrane “Risk of bias” tool on
NU
following domains (developed by scientific committee of BITERN) (Appendix 2): Random sequence generation (selection bias for controlled trials)
2.
Allocation concealment (selection bias for controlled trials)
3.
Generalizability of the findings to the target population (selection bias for trials without control)
4.
Blinding of participants and personnel (performance bias)
5.
Blinding of outcome assessment (detection bias)
6.
Incomplete outcome data (attrition bias)
7.
Selective reporting (reporting bias)
PT
ED
MA
1.
CE
Each domain was judged to be “low risk” of bias, “high risk” of bias, or “unclear risk” of bias. Any disagreements between data collectors were resolved through either discussion or consultation with a third author.
AC
Furthermore, the quality of the studies was assessed using Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology as described elsewhere.17
This methodology is widely
accepted and used by the guideline-writing committees, which includes assessment of the evidence whether it addresses same population, intervention, comparison and outcome. In brief, the quality of evidence was graded as “Low”, “Moderate” or “High” and the results were interpreted accordingly. Mortality rate with 28-days was the main primary outcome for this study which was invariably addressed within all studies. However, later days of mechanical ventilation and ICU length of stay were added to the existing protocol, retrospectively.
5
ACCEPTED MANUSCRIPT 6
Evidence synthesis We undertook systematic approaches to synthesize the findings of the studies since there was clinical
T
heterogeneity in the included studies. We used tabulation and textual description to synthesize the findings of the
RI P
studies. In addition, for all studies we transformed data into a common rubric and presented dichotomous outcomes as risk ratio (RR) and 95% confidence interval (CI), and continuous outcomes as mean difference (MD) and 95%
SC
CI. All conversions were done using Review Manager (Version 5.3. Copenhagen: The Nordic Cochrane Centre, the Cochrane Collaboration, 2014). For missing data, we considered statistics that allow calculation of missing data in
NU
article or in some cases we contacted the corresponding author.
When between-study heterogeneity allowed, we calculated a weighted treatment effect across studies using
MA
Review Manager. We assessed between-study heterogeneity by the chi-square statistic and its P value, and the extent of inconsistency using the I2 statistic. We considered a P value < 0.1 and I2 > 40% as indicating significant
ED
between-study heterogeneity. In case of significant heterogeneity, we did meta-analysis using a random-effects meta-analysis; otherwise a fixed-effect model was used. We expressed the results as RRs with 95% CI for
RESULTS
AC
Description of studies
CE
PT
dichotomous outcomes, and mean differences (MD, 95% CI) for continuous outcomes.
We found 124 references by recruiting the search strategy in 7 databases (Figure 1). We did not retrieve any studies in reference lists of the main articles. After discarding duplicates, we identified 100 publications. In primary screening of titles and abstracts, 75 articles were excluded due to obvious irrelevancy of the topics. In secondary screening of full-text articles, 6 studies with total number of 242 patients with sepsis or septic shock were identified and included in this systematic review (Table 1). Four studies included patients with either severe sepsis or septic shock and two trials included only patients with septic shock.6, 18-22 Three studies compared dexmedetomidine vs. propofol; two studies compared dexmedetomidine vs. lorazepam and one study compared dexmedetomidine vs. midazolam.6, 18-22 A loading dose of 0.01 to 1 µg/kg/h was used in all studies. The maximum maintenance doses of dexmedetomidine ranged between 0.15 µg/kg/h and 0.5
6
ACCEPTED MANUSCRIPT 7
mg/kg/h. Target sedation levels were Ramsay Score 2 in 3 studies and Ramsay Score 4 in one study by adjustment to the sedative regimen.19-22 In 5 studies, intermittent doses of fentanyl or alfentanil was used to treat apparent pain
RI P
T
based on physiological parameters in addition to facial expressions, limb movement, and ventilator synchrony.6, 18-21 Clinical outcomes investigated in these studies included 28-day mortality, days of ICU care and mechanical
SC
ventilation (MV) in 28-day period, delirium and coma-free days, development of secondary infections and organdysfunction free days. Pathophysiologic outcomes reported were inflammatory responses, intra-abdominal pressure
NU
(IAP) (the steady-state pressure within the abdominal cavity), gastric intramucosal pH, absolute and relative CO2 reactivity (as a measure of cerebral blood flow and cerebral autoregulation). Safety outcomes and adverse events
MA
such as hypertensive, hypotensive, or bradycardic response to infusion of dexmedetomidine were reported in 5 studies. The overall quality of the studies was good and there was not any high risk of bias for any domain (Table 2).
a.
ED
Effects of interventions Clinical outcomes
PT
Data for 28-day mortality and the length of ICU stay in 28-day period were available for 3 trials and for
CE
143 patients.6, 20, 21 There were 9 out of 71 patients in the experimental group who died by day 28 comparing to 19 out of 72 participants in the control group. There was not significant heterogeneity in the results (χ² = 0.57, P = 0.75;
AC
I² = 0%) and therefore we employed fixed effect model for meta-analysis of results. The risk ratio of dying within 28 days was 0.49 (95% CI: 0.24, 0.99; P = 0.05; fixed-effects model) for patients in experimental group vs. control group (Figure 2). In addition to 28-day mortality, only one study reported the first day mortality which was not significant (P =0.64) (Table 1).19 Three studies reported numbers of days in the ICU during 28 days of treatment for a total number of 143 patients.6, 20, 21 The weighted mean difference for the length of stay in the ICU was 1.54 (95% CI: -1.73, 4.81; P = 0.36; fixed-effects model) with no heterogeneity across the studies (χ² = 0.07, P = 0.96; I² = 0%; Figure 3). Two studies with 103 patients reported that there was not any significant association between sedation with dexmedetomidine and duration of mechanical ventilation (Table 1).6,
7
21
The weighted mean difference for the
ACCEPTED MANUSCRIPT 8
mechanical ventilation free days during 28-day period was 1.40 (95% CI: -4.44, 7.24; P = 0.64; random-effects
T
model) with significant heterogeneity across the studies (χ² = 3.25, P = 0.07; I² = 69%)
RI P
Sanders et al evaluated delirium in 63 patients with sepsis using confusion assessment method for the ICU (CAM-ICU) and showed that patients treated with dexmedetomidine had significantly more delirium/coma-free
SC
days in 12 days compared to those patients treated with lorazepam (Table 1).6 In this study, the beneficial effects of dexmedetomidine vs. lorazepam were more prominent on development of attention loss (CAM-ICU Feature 2) and
NU
disorganized thinking (CAM-ICU Feature 3).6 This study reported that this beneficiary effect of dexmedetomidine on delirium/ coma-free days was seen only in patients with sepsis but not in those without sepsis.
MA
Hemodynamic and biochemical parameters and kidney, liver, lung, cardiac and coagulation functions were analyzed in all studies. King et al reported that patients treated with dexmedetomidine had significantly more liver-
ED
dysfunction-free days comparing to patients treated with lorazepam.18 However, no statistically significant differences in hemodynamic and biochemical parameters were reported between different groups in other 5 studies
PT
(exact values were not reported). Four studies compared and reported level of sedation and need for analgesics between patients in experimental group and control group.6, 19-21 Three of them reported that both level of sedation
CE
and need for analgesics were same across experimental and control groups
19-21
. Sanders et al reported that sedation
with dexmedetomidine was significantly more efficacious than sedation with lorazepam. However, the need for
b.
AC
analgesic was greater in patients treated with dexmedetomidine.6 Pathophysiologic outcomes Two studies with 40 patients reported serum levels of tumor necrosis factor alpha (TNFα), interleukin 1 beta (IL-1β) and IL-6 as inflammatory markers after 24 hours.19,
20
Since we found no heterogeneity, we meta-
analyzed results of these studies using fixed-effects model. Levels of TNFα, IL-1β and IL-6 after 24 hours of treatment were significantly lower in patients treated with dexmedetomidine. The weighted mean difference for levels of TNFα after 24 hours was -5.31 (95% CI: -8.00, -2.63; P = 0.0001; fixed-effects model) and for IL-1β and IL-6 after 24 hours were respectively -1.24 (95% CI: -1.88, -0.60; P = 0.0001; fixed-effects model) and -250.77 (95% CI: -371.78, -129.75; P < 0.0001; fixed-effects model). Memis et al also reported significantly lower levels of TNF-α and IL-1β after 48 hours (Table 1).19
8
ACCEPTED MANUSCRIPT 9
One study evaluated IAP in 40 patients with severe sepsis and compared effects of dexmedetomidine with propofol on IAP in these patients.20 Both IAP after 24 hours or 48 hours were significantly lower in 20 patients
T
treated with dexmedetomidine comparing to those treated with propofol (Table 1). Kadoi et al reported that absolute
RI P
and relative CO2 reactivity as a measure of cerebral blood flow and cerebral autoregulation in 20 patients with septic shock was significantly lower under dexmedetomidine than propofol sedation in patients with septic shock.22 Such
Drug Safety
NU
c.
SC
association was not present in patients without sepsis.
No side effect including hypertensive, hypotensive, or bradycardic response to either loading dose or
MA
maintenance dose was reported in patients in any studies. Four studies on 183 patients reported that there were no significant differences in use of inotropic agents, bradycardia or tachycardia, and mean arterial pressure during study
ED
hours between experimental and control groups.6, 19-21 D ISCUSSION
PT
In this review, we performed a comprehensive search of the literature and included 6 clinical trials that evaluated the effect of dexmedetomidine in patients with sepsis or septic shock comparing to other sedatives
CE
commonly used in ICU. The systematic review of six studies showed that dexmedetomidine had beneficial effect on 28-day mortality (moderate quality of evidence based on GRADE approach) and delirium/coma (moderate quality
AC
of evidence based on GRADE approach) in patients with sepsis but no significant effect on length of ICU stay or duration of mechanical ventilation (moderate quality of evidence based on GRADE approach). The survival benefit was prominent in comparison with lorazepam not with propofol. Previous studies have shown that dexmedetomidine has protective effects on several organs including heart, lungs, nervous system and kidneys.12 Dexmedetomidine has a potential to stabilize the hemodynamic profile by attenuating sympathetically-mediated hyperdynamic responses, which may further translate to myocardial protection.12 In addition, dexmedetomidine may offer neuroprotective effects probably through modulating neurotransmitter release in the central and peripheral sympathetic nervous system and preventing apoptosis of the neurons.12 Both neuroprotective and anti-inflammatory effects of dexmedetomidine may play an important role in protective effects of dexmedetomidine on delirium as reported by Sanders et al.6, 23, 24 GABA receptors play a critical
9
ACCEPTED MANUSCRIPT 10
role in modulation of several neurotransmitters responsible for development of delirium in clinical settings.9,
24
Sedative effects of dexmedetomidine are unique in a sense that they are mediated through pathways independent of
T
the GABA receptors and hence has intrinsic delirium-sparing effect.25 Unlike GABA-mediated sedation,
RI P
dexmedetomidine enhances non-rapid eye movement sleep and induces a more natural sleep cycle.9, 12, 26, 27 Sleep deprivation has been associated with higher levels of inflammatory responses, increased insulin resistance and
clinical outcomes such as delirium-free days and mortality.6, 30
SC
activation of the hypothalamic-pituitary axis.28, 29 Therefore, dexmedetomidine effect on sleep quality may improve
NU
Dexmedetomidine further enhances mucosal immunity and bacterial clearance by promoting macrophage phagocytosis and bactericidal killing, properties that are critically important in sepsis and septic shock. 11, 31, 32 In
MA
Safety and Efficacy of Dexmedetomidine Compared with Midazolam (SEDCOM) trial, critically ill patients sedated with dexmedetomidine had reduced number of secondary infections, though such effect was not detected in our
ED
review.11 Results of this systematic review suggest that dexmedetomidine can decrease levels of TNF-α, IL-1β and IL-6 in patients with sepsis after 24 hours and 48 hours of treatment. This anti-inflammatory effect of
PT
dexmedetomidine can suppress the exaggerated production of inflammatory cytokines in septic shock.1, 3, 5, 33-37 Sepsis often impacts CNS as the first organ. Impaired cerebral autoregulation is one of the proposed 39
Kadoi et al reported that absolute and relative CO2 reactivity as a measure of
CE
mechanisms for such effect.38,
cerebral blood flow and cerebral autoregulation was significantly lower with dexmedetomidine than propofol
AC
sedation in patients with septic shock.22 This can be justified by cerebrovascular constrictor effect of dexmedetomidine compared to propofol.40 Both IAP after 24 hours or 48 hours were significantly lower in 20 patients treated with dexmedetomidine comparing to those treated with propofol. Increase in IAP not only has negative effects on splanchnic, respiratory, cardiovascular, renal, and neurological function but is also associated with high mortality.20, 41-43 This systematic review yielded no adverse events (with the exception of bradycardia) in using of dexmedetomidine and there were no differences in liver, renal, cardiac, or endocrine safety outcomes between septic patients treated with dexmedetomidine and other sedative agents. Dexmedetomidine by inhibiting central norepinephrine release can cause hypotension and bradycardia.12 Such an effect would be concerning in septic patients who are at risk for the development of cardiovascular shock.6 However, it seems that a reduction in
10
ACCEPTED MANUSCRIPT 11
proinflammatory cytokines would outweigh any direct hypotensive effect of dexmedetomidine.6 Bradycardia induced by dexmedetomidine is primarily dose-dependent and can be prevented by using slow administration of the 12
T
bolus loading or omitting bolus loading all together.
RI P
Major limitations of this meta-analysis included the small sample sizes of included trials and incomplete reporting of certain outcome measures, which limited our meta-analysis for most of the outcome variables. The
SC
results of this systematic review were based on the available literature including 6 studies with total number of 242 patients. Despite limited sample sizes and data, their overall quality of included studies was good and there were no
NU
high risk of bias in their design. However, it is of note that the randomization method was not described in some papers. Additionally, publication bias was not assessed using a funnel plot due to the small number of studies (< 10)
MA
that included in the meta-analyses of the outcome measures.
In conclusion, results of this systematic review based on the available literature suggest that
ED
dexmedetomidine in patients with sepsis or septic shock may improve short-term clinical outcomes as well as some pathophysiologic outcomes. However, considering limited number of studies with limited sample sizes in the
PT
literature, the beneficial effects found in this systematic review, only justify further studies to evaluate long-term as well as short-term outcomes of dexmedetomidine sedation in patients with sepsis and to compare its effects with
AC
CE
different types and dosages of sedative agents.
11
ACCEPTED MANUSCRIPT 12
REFERENCES 1.
Angus DC, van der Poll T: Severe sepsis and septic shock. The New England journal of medicine
2.
RI P
T
369:2063, 2013
Cawcutt KA, Peters SG: Severe sepsis and septic shock: clinical overview and update on management.
SC
Mayo Clinic proceedings 89:1572-1578, 2014
Pravda J: Metabolic theory of septic shock. World journal of critical care medicine 3:45-54, 2014
4.
Angus DC, van der Poll T: Severe sepsis and septic shock. The New England journal of medicine 369:840-
NU
3.
5.
MA
851, 2013
De Backer D, Orbegozo Cortes D, Donadello K, et al.: Pathophysiology of microcirculatory dysfunction
Pandharipande PP, Sanders RD, Girard TD, et al.: Effect of dexmedetomidine versus lorazepam on
PT
6.
ED
and the pathogenesis of septic shock. Virulence 5:73-79, 2014
outcome in patients with sepsis: an a priori-designed analysis of the MENDS randomized controlled trial.
O'Mara K, Gal P, Wimmer J, et al.: Dexmedetomidine versus standard therapy with fentanyl for sedation in
AC
7.
CE
Critical care 14:R38, 2010
mechanically ventilated premature neonates. The journal of pediatric pharmacology and therapeutics : JPPT : the official journal of PPAG 17:252-262, 2012
8.
Wujtewicz M, Maciejewski D, Misiolek H, et al.: Use of dexmedetomidine in the adult intensive care unit. Anaesthesiology intensive therapy 45:235-240, 2013
9.
Pandharipande PP, Pun BT, Herr DL, et al.: Effect of sedation with dexmedetomidine vs lorazepam on acute brain dysfunction in mechanically ventilated patients: the MENDS randomized controlled trial. Jama 298:2644-2653, 2007
12
ACCEPTED MANUSCRIPT 13
10.
Adams R, Brown GT, Davidson M, et al.: Efficacy of dexmedetomidine compared with midazolam for sedation in adult intensive care patients: a systematic review. British journal of anaesthesia 111:703-710,
11.
RI P
T
2013
Riker RR, Shehabi Y, Bokesch PM, et al.: Dexmedetomidine vs midazolam for sedation of critically ill
12.
SC
patients: a randomized trial. Jama 301:489-499, 2009
Afonso J, Reis F: Dexmedetomidine: current role in anesthesia and intensive care. Revista brasileira de
13.
NU
anestesiologia 62:118-133, 2012
Farag E, Argalious M, Abd-Elsayed A, et al.: The use of dexmedetomidine in anesthesia and intensive care:
Cui J, Zhao H, Wang C, et al.: Dexmedetomidine attenuates oxidative stress induced lung alveolar
ED
14.
MA
a review. Current pharmaceutical design 18:6257-6265, 2012
15.
PT
epithelial cell apoptosis in vitro. Oxidative medicine and cellular longevity 2015:358396, 2015
Lin YY, He B, Chen J, et al.: Can dexmedetomidine be a safe and efficacious sedative agent in post-cardiac
16.
CE
surgery patients? a meta-analysis. Critical care 16:R169, 2012
Piao G, Wu J: Systematic assessment of dexmedetomidine as an anesthetic agent: a meta-analysis of
17.
AC
randomized controlled trials. Archives of medical science : AMS 10:19-24, 2014
Guyatt GH, Oxman AD, Vist GE, et al.: GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. Bmj 336:924-926, 2008
18.
King MS, Ely EW, Girard TD, et al.: Sepsis modifies the beneficial effects of dexmedetomidine on nonbrain organ dysfunction: an a priori sub group analysis of the MENDS randomized controlled trial [Abstract]: American Thoracic Society International Conference, May 15-20, 2009, San Diego, 2009, p A4698 [Poster #K4677]
13
ACCEPTED MANUSCRIPT 14
19.
Memis D, Hekimoglu S, Vatan I, et al.: Effects of midazolam and dexmedetomidine on inflammatory responses and gastric intramucosal pH to sepsis, in critically ill patients. British journal of anaesthesia
20.
RI P
T
98:550-552, 2007
Tasdogan M, Memis D, Sut N, et al.: Results of a pilot study on the effects of propofol and dexmedetomidine on inflammatory responses and intraabdominal pressure in severe sepsis. Journal of
Memis D, Kargi M, Sut N: Effects of propofol and dexmedetomidine on indocyanine green elimination
NU
21.
SC
clinical anesthesia 21:394-400, 2009
assessed with LIMON to patients with early septic shock: a pilot study. Journal of critical care 24:603-608,
22.
MA
2009
Kadoi Y, Saito S, Kawauchi C, et al.: Comparative effects of propofol vs dexmedetomidine on cerebrovascular carbon dioxide reactivity in patients with septic shock. British journal of anaesthesia
de Rooij SE, van Munster BC, Korevaar JC, et al.: Cytokines and acute phase response in delirium. Journal
PT
23.
ED
100:224-229, 2008
24.
CE
of psychosomatic research 62:521-525, 2007
Pandharipande P, Ely EW: Sedative and analgesic medications: risk factors for delirium and sleep
25.
AC
disturbances in the critically ill. Critical care clinics 22:313-327, vii, 2006
Yapici N, Coruh T, Kehlibar T, et al.: Dexmedetomidine in cardiac surgery patients who fail extubation and present with a delirium state. The heart surgery forum 14:E93-98, 2011
26.
Nelson LE, Guo TZ, Lu J, et al.: The sedative component of anesthesia is mediated by GABA(A) receptors in an endogenous sleep pathway. Nature neuroscience 5:979-984, 2002
27.
Nelson LE, Lu J, Guo T, et al.: The alpha2-adrenoceptor agonist dexmedetomidine converges on an endogenous sleep-promoting pathway to exert its sedative effects. Anesthesiology 98:428-436, 2003
28.
Hardin KA: Sleep in the ICU: potential mechanisms and clinical implications. Chest 136:284-294, 2009
14
ACCEPTED MANUSCRIPT 15
29.
Weinhouse GL, Schwab RJ, Watson PL, et al.: Bench-to-bedside review: delirium in ICU patients -
Sanders RD, Xu J, Shu Y, et al.: Dexmedetomidine attenuates isoflurane-induced neurocognitive
RI P
30.
T
importance of sleep deprivation. Critical care 13:234, 2009
impairment in neonatal rats. Anesthesiology 110:1077-1085, 2009
Weatherby KE, Zwilling BS, Lafuse WP: Resistance of macrophages to Mycobacterium avium is induced
SC
31.
32.
NU
by alpha2-adrenergic stimulation. Infection and immunity 71:22-29, 2003
Gets J, Monroy FP: Effects of alpha- and beta-adrenergic agonists on Toxoplasma gondii infection in
33.
MA
murine macrophages. The Journal of parasitology 91:193-195, 2005
King EG, Bauza GJ, Mella JR, et al.: Pathophysiologic mechanisms in septic shock. Laboratory
Maes M, Lin A, Kenis G, et al.: The effects of noradrenaline and alpha-2 adrenoceptor agents on the
PT
34.
ED
investigation; a journal of technical methods and pathology 94:4-12, 2014
CE
production of monocytic products. Psychiatry research 96:245-253, 2000
MacLaren R: Immunosedation: a consideration for sepsis. Critical care 13:191, 2009
36.
Taniguchi T, Kidani Y, Kanakura H, et al.: Effects of dexmedetomidine on mortality rate and inflammatory
AC
35.
responses to endotoxin-induced shock in rats. Critical care medicine 32:1322-1326, 2004
37.
Xiang H, Hu B, Li Z, et al.: Dexmedetomidine controls systemic cytokine levels through the cholinergic anti-inflammatory pathway. Inflammation 37:1763-1770, 2014
38.
Terborg C, Schummer W, Albrecht M, et al.: Dysfunction of vasomotor reactivity in severe sepsis and septic shock. Intensive care medicine 27:1231-1234, 2001
39.
Bowie RA, O'Connor PJ, Mahajan RP: Cerebrovascular reactivity to carbon dioxide in sepsis syndrome. Anaesthesia 58:261-265, 2003
15
ACCEPTED MANUSCRIPT 16
40.
Hinohara H, Kadoi Y, Takahashi K, et al.: Differential effects of propofol on cerebrovascular carbon
Malbrain ML, Chiumello D, Pelosi P, et al.: Incidence and prognosis of intraabdominal hypertension in a
RI P
41.
T
dioxide reactivity in elderly versus young subjects. Journal of clinical anesthesia 17:85-90, 2005
mixed population of critically ill patients: a multiple-center epidemiological study. Critical care medicine
42.
SC
33:315-322, 2005
Vidal MG, Ruiz Weisser J, Gonzalez F, et al.: Incidence and clinical effects of intra-abdominal
43.
NU
hypertension in critically ill patients. Critical care medicine 36:1823-1831, 2008
Regueira T, Bruhn A, Hasbun P, et al.: Intra-abdominal hypertension: incidence and association with organ
Figure Legends:
CE
Figure 1. Study flow diagram
PT
ED
MA
dysfunction during early septic shock. Journal of critical care 23:461-467, 2008
AC
Figure 2. Comparison of 28-day mortality between patients in Dexmedetomidine group and control group
Figure 3. Comparison of ICU stay during 28-day period between patients in Dexmedetomidine group and control group
16
ACCEPTED MANUSCRIPT
PT
ED
MA
NU
SC
RI P
T
17
AC
CE
Figure 1
17
ACCEPTED MANUSCRIPT
NU
SC
RI P
T
18
AC
CE
PT
ED
MA
Figure 2
18
ACCEPTED MANUSCRIPT
NU
SC
RI P
T
19
AC
CE
PT
ED
MA
Figure 3
19
ACCEPTED MANUSCRIPT 20
Table 1. Table of included studies and their main characteristics
Study patients and main groups
Loading dose: 0.2 mg/kg over 10 min
Maintenance : 0.1–0.5 mg/ kg/h
12-
20 patients in control group 20 patients in experimental group
NU
10 patients in control group a.
Kadoi et al 2008 (23)
b.
2-
Age (mean±SD): 65 ±8 APACHE II (mean± SD): 38 ± 4
1
2-
10 patients in experimental group a.
Age (mean± SD): 66 ±7
b.
APACHE II (mean±SD): 37 ±
Inflammatory markers after 24 hours (Mean difference and 95% CI and p value):
Loading dose: 1 µg/kg/h over 10 min
PT AC
CE
1-
Experimental group: Dexmedetomidine
20 patients with septic shock*
Hemodynamic and biochemical parameters: no statistically significant differences between groups in Mean arterial pressure, heart rate, Lactate, platelets, leucocytes, bilirubin, alanine aminotransferase, creatinine, and pH after 24 hours (values were not reported)
ED
Memis et al 2007 (20)
2-
1 day mortality (risk ratio and 95%CI and p value): 0.7 (0.1, 3.6); p value=0.6359
SC
Control group: Midazolam
1-
40 Critically ill patients with sepsis (demographic and clinical data were not reported)
Outcomes, results and relationships
RI P
Interventions
MA
Study (author, year, reference)
T
Articles reporting mortality outcomes
Maintenance: 0.2–2.5 µg/kg/h
1.
TNF-α (pg/mL): value=0.0035
2.
IL-1β (pg/mL): -1.29 (-2.2,-0.4), p value= 0.007
3.
IL-6 (pg/mL): value=0.0086
-4.9
-243.2
(-8,
-1.8),
P
(-415.1,-71.3),
p
Patients were maintained at a Ramsay sedation score <2 by adjustment to the sedative regimen
Gatric intramucosal pH (Mean difference and 95%CI and p value): -0.03 (-0.29,0.23), p value=0.8193
side-effects: no side effect was seen in both groups
Control group: Propofol
Hemodynamic parameters: no statistically significant differences between groups in Mean arterial pressure, heart rate and central venous pressure
Cerebral blood flow and cerebral autoregulation (Mean difference and 95% CI and p value)
Loading dose: 1 mg/kg over 10 min Maintenance : 1–3 mg/ kg/h infusion Experimental group: Dexmedetomidine
Loading dose: 0.01 µg/kg/h over 10 min
Maintenance:
20
1-
Absolute CO2 reactivity (cm/s/ mm Hg): -0.6 (0.9, -0.3), p value: 0.0003
2-
Relative CO2 reactivity: -0.7 (-1, -0.4), p value: 0.0003
Relationship: Sedation with Dexmedetomidine had a greater impact on absolute or relative CO2 reactivity in patients with sepsis than those without sepsis
ACCEPTED MANUSCRIPT 21
5
0.3–0.5 mg/kg/h
Loading dose: 1 mg/kg over 15 min
Days of ICU care in 28-day period (Mean difference and 95%CI and p value): 1.0 (-5.6, 7.6); p value: 0.7688
Maintenance : 1–3 mg/ kg/h
Days of MV in 28-day period (Mean difference and 95%CI and p value): 1 (-1.6, 3.6) p value= 0.4639
Gender (male/female): 11/9
SOFA score (mean±SD): 4.0 ± 2.5
Age (mean, range): 50 (19-74)
b.
Gender (male/female): 14/6
d.
Inflammatory markers after 24 and 48 hours (Mean difference and 95% CI and p value):
1-
Loading dose: 1 µg/kg/h over 10 min Maintenance: 0.2–2.5 µg/kg/h
After 24 hours: -6.55 (-11.9, -1.2), P value= 0.0213
o
After 48 hours: -25.3 (-38.6, -12), P value= 0.0006
o
After 24 hours: 0.0145
o
After 48 hours: -1.41 (-2.6,-0.2), p value= 0.0293
CE
SOFA score (mean±SD): 4.2 ± 1.8
40 Critically ill patients with septic shock 1-
20 patients in control group
Control group: Propofol Loading dose: 1 mg/kg over 15
21
IL-6 (pg/mL): After 24 hours: -258.2 (-428.6,-87.8), p value= 0.0051
o
After 48 hours: - 160.2 (-319,-1.4), p value= 0.0553
1-
-1.2 (-2.1,-0.3), p value=
o
Intra-abdominal pressure (mm Hg) difference and 95%CI and p value): 1-
Memis et al 2009 b (22)
IL-1β (pg/mL)
3-
APACHE II (mean±SD): 19 ± 5
TNF-α (pg/mL)
o
2-
PT
a.
Experimental group: Dexmedetomidine
ED
APACHE II (mean±SD): 18 ± 4
20 patients in experimental group
c.
2-
RI P
b.
NU
Age (mean, range): 50 (19-74)
T
MA
a.
d.
Memis et al 2009 a (21)
28-day mortality (risk ratio and 95%CI and p value): 0.5 (0.1-5.1); p value=0.5580
20 patients in control group
c.
2-
Control Propofol
AC
1-
group:
1-
SC
40 Critically ill patients with sepsis after ileus surgery
(Mean
After 24 hours: -5.8 (-8.6,-2.9), p value= 0.0003
After 48 hours: - 4.8 (-7.9,-1.7), p value= 0.0042
Haemodynamic and biochemical parameters: no statistically significant differences between groups in Mean arterial pressure, heart rate, temperature, and urine output, Lactate, platelets, leucocytes, bilirubin, alanine aminotransferase, creatinine, and pH
28-day mortality (risk ratio and 95%CI and p value): 0.8 (0.2-2.9) p value=0.6790
Days of ICU care in 28-day period (Mean difference and 95%CI and p value): 2 (-2.7, 6.7),
ACCEPTED MANUSCRIPT 22
Gender (male/female): 13/7
c.
APACHE II (mean±SD): 20 ± 8
d.
SOFA score (mean±SD): 4.0 ± 2.9
2-
Maintenance : 1–3 mg/ kg/h infusion Experimental group: Dexmedetomidine
Loading dose: 1 µg/kg/h over 10 min
Maintenance: 0.2–2.5 µg/kg/h
b.
Gender (male/female): 14/6
c.
APACHE II (mean±SD): 22 ± 5
d.
SOFA score (mean±SD): 4.5 ± 2.8
AC
CE
1-
39 Critically ill patients with sepsis king et al 2009 (19)
Sanders et al 2010 (6)
1-
20 patients in control group
2-
19 patients in experimental group
63 Critically ill patients ǂ with sepsis 1-
32 patients in control group
Hepatic blood flow (Mean difference and 95%CI and p value)
Indocyanine green plasma disappearance rate reactivity: -1.8 (-10.5, 6.9), p value: 0.6877
side-effects: no side effect was seen in both groups
MA
Age (mean, range): 60 (31-80)
Hemodynamic and biochemical parameters: no statistically significant differences between groups in Mean arterial pressure, heart rate, Lactate, platelets, bilirubin, alanine aminotransferase, creatinine
ED
a.
NU
20 patients in experimental group
PT
2-
p value: 0.4054
T
b.
min
RI P
Age (mean, range): 54 (25-78)
SC
a.
2-
Control group: Lorazepam Loading dose: 1 mg/h
1.
Experimental group:: 12 (12.12)
2.
Control group: 8.5 (2.12)
kidney dysfunction free days: no statistically significant differences
Experimental group: Dexmedetomidine
Lung dysfunction free days: no statistically significant differences
Hemodynamic and cardiac dysfunction free days: no statistically significant differences
coagulation dysfunction free days: no statistically significant differences
Control group: Lorazepam
28-day mortality (risk ratio and 95%CI and p value): 0.4 (0.2-0.9), p value=0.0247
Loading dose: 1 mg/h
28-day mortality (hazard ratio and 95%CI and p value): 0.3 (0.1, 0.9), p value= 0.0317
Loading dose: 0.15 µg/kg/h
Maintenance: 0.15-1.5 µg/kg/h
liver dysfunction free days (median, IQR) p=0.05:
Maintenance : 1–10 mg/h
1-
22
ACCEPTED MANUSCRIPT 23
Gender (male/female): 13/19
c.
APACHE II (mean, range): 29 (24 to 32)
d.
SOFA score (mean, range): 9 (8 to 12)
2-
Experimental group: Dexmedetomidin e
Loading dose: 0.15 µg/kg/h
Maintenance: 0.15-1.5 µg/kg/h
MA
b.
Age (mean, range): 60 (46 -65) Gender (male/female): 18/13
c.
APACHE II (mean, range): 22 ±5
d.
SOFA score (mean, range): 10 (9 to13)
MV-free days in 28 days (Mean difference and 95%CI and p value): 5.1 (-0.1, 10.3), p value= 0.0574
Days of ICU care in 28-day period (Mean difference and 95%CI and p value): 1.2 (-5.1, 7.5), p value= 0.7087
Delirium/coma free days in 12 days (Mean difference and 95%CI and p value): 3.2 (1.3, 5.1), p value= 0.0014
Delirium-free days in 12 days (Mean difference and 95%CI and p value): 1.4 (-0.1, 2.9), p value=0.0689
Coma-free days in 12 days (Mean difference and 95%CI and p value): 3.5 (1.7, 5.3), p value= 0.0003
Efficacy of sedation accurately Sedated days: 1.3 (1.2, 1.4), p value= 0.0001
Hemodynamic parameters: no statistically significant differences in Mean arterial pressure, heart rate, vasopressor use and cardiac arrhythmias. secondary infections: no statistically significant differences in Development of new secondary infections beyond the first 48 hour side-effects: no side effect was seen in both groups
ED
a.
NU
31 patients in experimental group
PT
2-
Maintenance : 1–10 mg/h
T
b.
RI P
Age (mean, range): 58 (44 to 66)
SC
a.
CE
Relationship: Sedation with Dexmedetomidine had a greater impact on delirium/ coma-free days in patients with sepsis than those without sepsis
*20 patients without septic shock were also included
ǂ40 patients without septic shock were also included
AC
23
Sanders et al 2010 (6) Kadoi et al 2008 (23) NA
king et al 2009 (19)
ED
Memis et al 2009 a (10)
PT
Memis et al 2009 b (22)
CE
AC
[Study ID]
Memis et al 2007 (20)
NA
Hemodynamic and biochemical parameters
Low risk of bias
Unclear risk of bias
High risk of bias
24 NA
T *
NA
Selective reporting for pathophysiologic outcomes
Selective reporting for clinical outcomes
Incomplete outcome data for pathophysiologic outcomes
RI P
Incomplete outcome data for clinical outcomes
SC
NU
Blinding of outcome assessment for pathophysiologic outcomes
MA
Blinding of outcome assessment for clinical outcomes
Blinding of participants and personnel for pathophysiologic outcomes
Blinding of participants and personnel for clinical outcomes
Allocation concealment bias (selection bias)
Random sequence generation (selection bias)
ACCEPTED MANUSCRIPT 24
Table 2
ACCEPTED MANUSCRIPT 25
Highlights:
T
RI P
SC
NU
MA ED
PT
CE
Following sepsis dexmedetomidine was associated with lower 28-mortality compared to other modalities of sedation. There was biochemical evidence for decreased state of systemic inflammation with dexmedetomidine is sepsis. Dexmedetomidine sedation did not change the length of ICU stay in patients with septic shock. Similarly, dexmedetomidine sedation did not affect weaning from mechanical ventilation in sepsis patients
AC
25