Note from the National Guideline Clearinghouse (NGC): The National Institute for Health and Clinical Excellence (NICE) commissioned an independent academic centre to perform a systematic literature review on the technology considered in this appraisal and prepare an assessment report. The assessment report for this technology appraisal was prepared by the West Midlands Health Technology Assessment Collaboration (see the "Companion Documents" field).
Search Strategy
The primary question addressed by this review is how effective intravenous (IV) fluids are in the resuscitation of hypovolaemic trauma patients with no head injury in a pre-hospital setting. Preliminary scoping searches suggested that high quality randomised controlled trial evidence directly addressing this question was unlikely to be sufficient to provide an unequivocal answer to this question. The guideline developers therefore decided to look at evidence from other settings that may be generalisable to the pre-hospital setting.
Two separate search strategies were used: a highly sensitive search strategy, designed not to miss any relevant studies, was developed to identify studies relating to the use of fluids in a pre-hospital setting (immediate versus delayed fluids, different volumes, or speed of infusion), and a more specific search strategy was used to identify additional randomized controlled trials (RCTs) of fluid administration in other settings (e.g. after admission to hospital), as tens of thousands of studies would have otherwise been identified. Full search strategies are listed in Appendix 3 of the assessment report.
Databases
The following electronic databases were searched: the Cochrane Central Register of Controlled Trials (Issue 1, 2003), MEDLINE (OVID, 1966-2003), EMBASE (OVID, 1980- 2003) and the Science Citation Index (1980-2003).
Strategy
Text and MeSH terms relating to the population (e.g. trauma, hypovolaemia), the intervention (e.g. IV fluid, fluid resuscitation) and the setting where applicable (e.g. pre-hospital, emergency) were combined with filters for randomised controlled trials. There were no language restrictions.
Citation Searching, Handsearching
In addition, citation lists of relevant publications (included studies and reviews) were checked and the Journal of Trauma, Injury, Infection & Critical Care was hand searched for the years 1998 (volume 44) - 2003 (volume 54 (2)) inclusive.
Unpublished Data
Unpublished data were sought by contacting organisations and individual experts, and by checking research registers of ongoing trials and other relevant web sites (list of web sites searched in Appendix 1 of the assessment report). Data from the industry and other submissions were checked for relevant published and unpublished studies.
Additional Questions
- What is the effect of basic life support (BLS) versus advanced life support (ALS) on patient outcome?
- What is the effect of fluid replacement for different types of injuries (e.g. blunt, penetrating) on patient outcome?
- What is the effect of different types of fluid (e.g. different crystalloids or colloids or crystalloids versus colloids) on patient outcome?
- What is the effect of fluid replacement in paediatric trauma patients?
- How accurate are paramedics at diagnosing hypovolaemia in trauma patients at the scene and can this affect patient outcomes?
- Is there evidence on whether naturally occurring physiological shock mechanisms have a protective effect? How does fluid resuscitation interact with these mechanisms?
In order to identify the evidence base concerning additional relevant issues relating to fluid replacement, search strategies were developed to identify systematic reviews relating to these issues. Search filters for reviews were combined with relevant MeSH terms and text words.
The following databases were searched: Cochrane Library (Issue 4, 2002), MEDLINE (OVID, 1966-2003) and EMBASE (OVID, 1980-2003). There were no language restrictions.
Individual randomised controlled trials were not systematically sought.
Observational Studies
A separate systematic review of observational studies was ruled out at the protocol stage as these would not have informed the question adequately due to the intrinsically confounded nature of the study designs. However, some observational studies are frequently cited. Therefore, for the purpose of providing an adequate appraisal of current policy, all observational studies cited in the Consensus Statement or Joint Royal Colleges Ambulance Liaison Committee (JRCALC) guidelines were retrieved and critically appraised.
Inclusion and Exclusion Criteria
Primary research question: Immediate versus delayed fluid replacement or differential volume replacement in a pre-hospital or other setting
Inclusion Criteria
Study design: Randomised controlled trials
Population: Patients of any age with haemorrhagic hypovolaemia resulting from trauma
Intervention: Immediate or early fluid replacement (pre-hospital or other setting)
Comparator: Delayed or no fluid replacement (pre-hospital or other setting); different volume of fluid given (pre-hospital or other setting); fluids given at different speed (pre-hospital or other setting)
Exclusion Criteria
Study design: Observational studies
Population: Randomised controlled trials* with primarily:
- Head injured patients
- Patients with burns
- Patients with septic shock
Intervention/Comparator: Randomised controlled trials comparing different types of fluids; randomised controlled trials comparing blood or blood products to other fluids
* Studies were not excluded if they had mixed populations providing the majority were patients with haemorrhagic hypovolaemia resulting from trauma.
The inclusion and exclusion criteria were applied independently by two reviewers to all identified citations, and any disagreement resolved by a third reviewer. Where a decision on inclusion or exclusion could not be made on the basis of title or abstract, the full study was retrieved.
Inclusion Criteria for Systematic Reviews for Additional Research Questions
Systematic reviews of primary evidence of any study design that addressed the questions outlined above. Two reviewers independently assessed reviews for their relevance.