Conservative versus liberal fluid therapy for initial severe sepsis and septic shock

  • Protocol
  • Intervention

Authors


Abstract

This is the protocol for a review and there is no abstract. The objectives are as follows:

To determine whether conservative or liberal fluid therapy improves clinical outcome in people with initial severe sepsis and septic shock.

Background

Severe sepsis and septic shock are potentially life-threatening complications of infection with high morbidity and mortality. Death from severe sepsis and septic shock is one of the most common causes of death in non-coronary intensive care units (ICU). Patient admissions to ICU represent a major cost burden for healthcare providers. The current management of severe sepsis includes initial resuscitation (first six hours), antibiotic therapy, and administration of adrenergic receptor agonists and steroids. According to the guidelines from multiple relevant organizations, clinicians always use one or more fluid boluses to treat people with sepsis and help them survive severe sepsis (Dellinger 2008). However, evidence for such care is scant and recent evidence from African hospitals did not support the routine use of bolus resuscitation in children with severe febrile illness and impaired perfusion when compared with a no bolus control group (Maitland 2011). Much conjecture has occurred regarding treatment guidelines, especially in relation to invasive management and critical care. Healthcare professionals consider that fluid therapy is an essential part of the treatment protocol in people with severe sepsis and septic shock although its use is mainly based on consensus opinion.

Description of the condition

Sepsis is defined as the presence or presumed presence of an infection accompanied by evidence of a systemic response called the systemic inflammatory response syndrome (Nguyen 2006). Pulmonary, gastrointestinal, genitourinary, and primary bloodstream infections account for the majority of infectious sources in people with sepsis (Hodgin 2008). Severe sepsis is defined as sepsis plus sepsis-induced organ dysfunction or tissue hypoperfusion (Dellinger 2013). The exact mechanism by which sepsis produces multiple organ dysfunction remains unknown, but is thought to develop due to a dysregulated response of the immune system, which then leads to systemic inflammation or an inflammatory response that leads to global tissue hypoxia and organ dysfunction (e.g. in the liver, lungs, heart, kidneys). The ensuing severe sepsis can progress to multiple organ failure and septic shock. Septic shock is defined as sepsis-induced hypotension that persists after adequate fluid resuscitation (Dellinger 2008), and typically occurs in elderly people; very young children; and in people with illnesses such as diabetes, cancers, human immunodeficiency virus (HIV) or recovering from recent surgical or medical procedures. It is becoming an increasing health burden as incidence rates increase due to multiresistant strains (Annane 2003).

The epidemiology of severe sepsis and septic shock is poorly understood because of the lack of population-based country-specific prospective cohort studies. Historically, the diagnosis of severe sepsis was based on hospital discharge records and not the current consensus definition that was developed in 2001 (Levy 2003). The incidence of severe sepsis (sepsis-induced organ dysfunction) in the European Union has been estimated at 90.4 cases per 100,000 population (Davies 2001). The incidence in Australia and New Zealand is estimated to be 77 per 100,000 (Finfer 2004), and in the USA it is estimated to be from 81 (Martin 2003) to 300 per 100,000 (Angus 2001). It has been estimated that globally 1400 people die each day due to severe sepsis (Bone 1992). The most vulnerable are children under the age of 12 months (Watson 2003), and also elderly people who have the highest incidence of severe sepsis (Angus 2001).

Description of the intervention

Liberal fluid therapy is regarded as a crucial and effective intervention during the initial treatment of severe sepsis in children and adults. It is believed to improve clinical outcome in people with severe sepsis and septic shock. Fluid bolus therapy is widely adopted and is regarded as a key life-saving intervention in the management of sepsis. Its use is based mainly on expert opinion with limited experimental evidence to guide decision making.

The question of whether conservative or liberal fluid therapy could improve clinical outcomes in people with severe sepsis and septic shock has not yet been solved.

Conservative fluid therapy is defined as:

  1. adults: no fluid bolus, titrated according to monitors of heart rate, urine output, capillary refill, and level of consciousness (or total fluid amount less than that of liberal fluid therapy);

  2. children: no fluid bolus, titrated according to monitors of heart rate, urine output, capillary refill, and level of consciousness (or total fluid amount less than that of liberal fluid therapy).

Liberal fluid therapy is defined as:

  1. adults: fluid challenge of 1000 mL or greater of crystalloids or 300 to 500 mL of colloids over 30 minutes before titration (or total fluid amount more than that of conservative fluid therapy) (Dellinger 2008b);

  2. children: fluid bolus of crystalloids with boluses of 20 mL/kg over 5 to 10 minutes before titration (or total fluid amount more than that of conservative fluid therapy) (Dellinger 2008b).

How the intervention might work

Critical care doctors continue to debate the benefits of conservative or liberal fluid therapy on the clinical outcome in people with severe sepsis and septic shock. Too little fluid may cause tissue hypoperfusion and worsen a person's condition, while over-prescription of fluid can cause or exacerbate oedema in the lungs, heart, gastrointestinal tract, skin, brain, and other tissues, leading to organ failure or cerebral oedema and herniation. No consensus has emerged among clinicians on the amount of fluids that people should receive.

Why it is important to do this review

Healthcare professionals frequently use fluid bolus therapy and they presently regard it as a key life-saving intervention in the management of sepsis. Currently, there is insufficient evidence to inform best practice. It is, therefore, important to perform this systematic review to improve outcomes.

Objectives

To determine whether conservative or liberal fluid therapy improves clinical outcome in people with initial severe sepsis and septic shock.

Methods

Criteria for considering studies for this review

Types of studies

We will include randomized controlled trials (RCTs) and quasi-RCTs comparing conservative versus liberal fluid bolus. We will include cluster RCTs.

Types of participants

We will include people with severe sepsis or septic shock.

We will accept study authors' definitions of severe sepsis and septic shock and perform sensitivity analyses comparing operational-defined criteria with less clearly developed definitions.

We will include both adults and children in this review. We will exclude neonatal sepsis studies (both early and late).

Types of interventions

We will define conservative fluid therapy as:

  1. adults: no fluid bolus, titrated to clinical monitors of cardiac output, including heart rate, pulse pressure, central venous pressure, urine output, capillary refill, and level of consciousness (or total fluid amount less than that of liberal fluid therapy);

  2. children: no fluid bolus, titrated to clinical monitors of cardiac output, including heart rate, pulse pressure, central venous pressure, urine output, capillary refill, and level of consciousness (or total fluid amount less than that of liberal fluid therapy).

We will define liberal fluid therapy as:

  1. adults: fluid bolus of greater than 1000 mL of crystalloids or 300 to 500 mL of colloids over 30 minutes before titration (or total fluid amount more than that of conservative fluid therapy) (Dellinger 2008b);

  2. children: fluid bolus of crystalloids with boluses of 20 mL/kg over 5 to 10 minutes before titration (or total fluid amount more than that of conservative fluid therapy) (Dellinger 2008b).

Types of outcome measures

We will group outcomes according to the following time points: 6, 12, and 24 hours.              

Primary outcomes
  1. Death, all-cause mortality in hospital/ICU (28 days mortality).

Secondary outcomes
  1. Vasoactive agent-free days.

  2. Pulmonary oedema.

  3. Organ dysfunction (e.g. renal failure, respiratory failure, need for mechanical ventilation).

  4. Duration of organ dysfunction.

  5. Length of stay.

  6. Ventilator-free days.

Search methods for identification of studies

Electronic searches

We will search the current issue of the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library), MEDLINE (1946 to date); and EMBASE (1980 to date). We will impose no language restrictions. We will search intensive and critical care conference abstracts.

We will develop a specific strategy for each database. Appendix 1 outlines the search strategy for MEDLINE Ovid.

Searching other resources

We will contact the authors of all included studies to obtain further information on related studies. In addition, we will search for ongoing clinical trials using www.clinicaltrials.gov and contact the authors of any unpublished studies identified from the searches.

Data collection and analysis

Selection of studies

Two review authors (DL, XL) will independently inspect the titles and abstracts of all study citations identified by the searches. We will resolve any disagreements by discussion with a third review author (HZ).

Data extraction and management

Two review authors (DL, XL) will independently extract data on methods, participants, interventions, and outcomes, into data extraction forms (see Appendix 2 for a copy of our form) and will enter the study data into Review Manager 5 (RevMan 2011). We will resolve any disagreements through discussion. Where necessary, we will contact the authors of the original publications to obtain additional information about the design of the study and results.

Assessment of risk of bias in included studies

Two review authors (DL, WC) will independently assess the risk of bias using the tool described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will resolve any disagreements by discussion. The 'Risk of bias' tool encourages consideration of how the sequence was generated, how allocation was concealed, the integrity of blinding, the completeness of outcome data, selective reporting, and other biases.

We will report the 'Risk of bias' table as part of the table 'Characteristics of included studies'. We will first describe what is reported to have happened in the study, and then assign a judgement that relates to the risk of bias for that entry, which we will achieve by categorizing judgements as:

  • 'low risk' of bias;

  • 'high risk' of bias; or

  • 'unclear risk' of bias.

Measures of treatment effect

Binary data

Patients and clinicians find that risk ratios (RR) are easier to understand than odds ratios (OR) (Grimes 2008), while ORs are often misinterpreted as RRs (Deeks 2000). Therefore, for binary outcomes we will calculate the RR with 95% confidence intervals (CI) based on the random-effects model. When meta-analysed data are significant and non-heterogeneous (i.e. I2 statistic < 50%), we will calculate the number needed to treat for an additional beneficial outcome (NNTB) and the number needed to treat for an additional harmful outcome (NNTH).

Continuous data

For continuous data, we will calculate the mean difference (MD) with 95% CI. For final point data (endpoint scale data) we will only meta-analyse normally distributed data (non-skewed) by applying the following rule for positive measurements: the standard deviation (SD), when multiplied by two has to be less than the mean (Altman 1996). Skewed data from studies will not be shown graphically, but rather added to 'Other data' tables and briefly commented on in the text.

Unit of analysis issues

We consider that the individual participants in each trial arm will comprise the unit of analysis. We, therefore, anticipate that all trials will have a parallel group design; however, if cluster RCTs are included we will seek statistical advice to allow for clustering effects.

Dealing with missing data

We will follow the guidelines in the Cochrane Handbook for Systematic Reviews of Interventions for dealing with missing data (Higgins 2011). We will contact the original investigators to request missing data. We will assume that participants lost to follow-up did not improve and those losses will be included as part of an intention-to-treat analysis. We will make explicit the assumptions of any methods used to deal with missing data: for example, that the data are assumed missing at random. We will perform sensitivity analyses to assess how sensitive the results were to reasonable changes in the assumptions that were made. We will address the potential impact of missing data on the findings of the review in the 'Discussion' section (Higgins 2011).

Assessment of heterogeneity

For statistical heterogeneity, we will use the I2 statistic. We will interpret an I2 estimate of 50% or greater as evidence of high levels of heterogeneity (Higgins 2003). We will evaluate clinical heterogeneity in selected studies by assessment of the study population and the type of clinician involvement in fluid strategy and fluid type (e.g. crystalloids versus colloids, saline versus Ringer's lactate), vasopressor type, or the source of infection.

Assessment of reporting biases

Reporting biases occur when the reporting of research findings is influenced by the nature and direction of results. These are described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Funnel plots may be useful in investigating reporting biases but are of limited power to detect small-study effects (Egger 1997). We will not use funnel plots for outcomes where there are 10 or fewer studies, or where all studies were of similar sizes. In other cases, where funnel plots are possible, we will inspect the funnel graphs and report the findings.

Data synthesis

We will use a random-effects model meta-analysis using the assumption that the effects being estimated in the different studies are not identical, but follow some distribution across studies.

Subgroup analysis and investigation of heterogeneity

We have not planned to do subgroup analyses. We will investigate the sources of clinical heterogeneity using sensitivity analyses by removing trials at potentially higher risk of bias (e.g. resource-limited care from developing countries).

Sensitivity analysis

Where sufficient data are available we will conduct sensitivity analyses for country-specific outcomes (i.e. comparing developing countries with developed countries). We will examine the effect of recently conducted trials on primary outcomes (i.e. those conducted after the surviving sepsis campaign).

Summary of findings

We will use the principles of the GRADE system to assess the quality of the body of evidence associated with (Guyatt 2008):

  1. death - ICU or hospital mortality, or both at 28 days

and construct a 'Summary of findings' table using the GRADEpro software. The GRADE approach appraises the quality of a body of evidence based on the extent to which one can be confident that an estimate of effect or association reflects the item being assessed. The quality of a body of evidence takes into consideration within-study risk of bias (methodological quality), the directness of the evidence, heterogeneity of the data, precision of effect estimates, and risk of publication bias.

Acknowledgements

We would like to thank Ronan O'Sullivan (Content Editor); Cathal Walsh (Statistical Editor); and Tomer Avni, David T Huang, and Andreas Schibler (peer reviewers) for their help and editorial advice during the preparation of this protocol for the systematic review. We also acknowledge Jane Cracknell (Managing Editor Cochrane Anaesthesia Review Group) for all the support provided in the process of developing this protocol for a review.

Appendices

Appendix 1. MEDLINE (Ovid SP) search strategy

1. Sepsis/ or Shock, Septic/ or Infection/ or (search sepsis* or septic* or sirs or infection* or shock* or mods or mof).ti,ab. or exp Systemic Inflammatory Response Syndrome/ or Multiple Organ Failure/
2. Fluid Therapy/ or fluid.ti,ab.
3. ((randomized controlled trial or controlled clinical trial).pt. or randomized.ab. or placebo.ab. or clinical trials as topic.sh. or randomly.ab. or trial.ti.) not (animals not (humans and animals)).sh.
4. 1 and 2 and 3

Appendix 2. Data extraction form

Trial ID 
Title 
Author 
Country 
Citation (journal, year, volume, page) 
Study type 
Study quality 
Participants 
Diagnosis 
n = 
Age 
Sex 
History 
Included 
Excluded 
Intervention 
The definition of conservative fluid therapy 
The definition of liberal fluid therapy 
Outcomes 
 INTERVENTIONCONTROL
n =  
Mortality at 28 days  
Vasoactive agent-free days  
Pulmonary oedema  
Duration of organ dysfunction  
Number of organ dysfunctions  
Notes  

Contributions of authors

Conceiving the review: Danyang Li (DL)

Co-ordinating the review: Wei Cui (WC)

Undertaking manual searches: DL

Screening search results: DL, WC

Organizing retrieval of papers: WC, Huahao Shen (HS)

Screening retrieved papers against inclusion criteria: DL, WC

Appraising quality of papers: Hong Zhu (HZ), Yi Xia (YX)

Abstracting data from papers: DL, Xueyang Li (XL)

Writing to authors of papers for additional information: Lixueyang

Providing additional data about papers: XL

Obtaining and screening data on unpublished studies: DL, WC

Data management for the review: DL, Zhuyimin

Entering data into Review Manager: DL

RevMan statistical data: DL, Zhuyimin

Other statistical analysis not using Review Manager: Zhuyimin

Interpretation of data: DL, WC

Statistical inferences: DL

Writing the review: DL

Securing funding for the review: none

Performing previous work that was the foundation of the present study: none

Guarantor for the review: DL

Responsible for reading and checking review before submission: WC, DL, HS

Declarations of interest

None known.

Ancillary