mortality in pre-hospital tracheal intubation to emergency department intubation in trauma patients

ICM7059 ASSIGNMENT

1. What is a systematic review? 2. How does this differ to a literature review?

3. When and why is a meta-analysis performed? (5 marks)

1. A systematic review (SR) is defined by Grant et al as “systematically search for, appraise and synthesis research evidence, often adhering to guidelines on the conduct of a review”. In contrast a literature review (LR) is defined as a “generic term: published materials that provide examination of recent or current literature. Can cover wide range of subjects at various levels of completeness and comprehensiveness. May include research findings”¹

The difference is between these two types of review is not straight forward as there are differences in terminology with some identifying a systematic review as a type of literature review like Smith et al who dividesLR into broadcategories of narrative (descriptive) reviews, scoping reviews, rapid evidence assessments (rapid reviews) and systematic reviews.² Other sources like the State University library give advice on how to distinguish between the two with a table from Kysh (fig 1.) with a systematic review being sharply focused on a specific research question and a literature review giving a summary or overview.³

Fig 1.

https://libguides.sjsu.edu/c.php?g=230370&p=1528399

Finally some researches combine the terms, for example “A guide to systematic literature reviews” which adopts Smiths terminology of a SR as a sub type of LR.⁴

3. Some SR include Meta-analysis which is utilizes statistical technique to synthesize data from different studies into a single estimate or effect size. This differs to traditional null-hypothesis testing which while giving statistical significance may not give clinical significance.⁵ Whether or not a SR uses a meta analysis depends on the research question, the data available and the intent of the researches.

There is a vast and continuously growing amount of research available to answer both specific and general clinical questions in medicine, a valuable tool for the analysis of evidence is a summary of the available and appropriate literatureto negotiate this volume of material. There aredifferent strategies for the presentation of available relevant literature, Grant et al list 14 with a systemic review and a literature review making up 2 of the 14.¹ A literature review is defined as a “Generic term: published materials that provide examination of recent or current literature. Can cover wide range of subjects at various levels of completeness and comprehensiveness. May include research findings” and a systematic review as “Seeks to systematically search for, appraise and synthesis research evidence, often adhering to guidelines on the conduct of a review.”Smith et al attempts to divide these into broader categories of narrative (descriptive) reviews, scoping reviews, rapid evidence assessments (rapid reviews) and systematic reviews.² Some articles combine the terms, for example “A guide to systematic literature reviews.”⁴ Jose State University library gives advice on how to distinguish between the two with a table from Kysh(fig 1.) with a systematic review being sharply focused on a specific research question and a literature review giving a summary or overview.³

Fig 1.

https://libguides.sjsu.edu/c.php?g=230370&p=1528399

Cook et al give the following definition of a systematic review “the application of scientific strategies, in ways that limit bias, to the assembly, critical appraisal, and synthesis of all relevant studies that address a specific clinical question.”⁶ Cook divides the types ofreviews into Systematic and Narrative depending on the rigor of scientific review methods with those adopting strict rules to limit bias being and minimize error being termed systematic while less stringent and more broad reviews being termed narrative, this gives us systematic literature reviews and narrative literature reviews.

Critical appraisal tools exist for the evaluation reviews for example the PRISMA (Preferred Reporting Items of Systematic reviews and Meta-Analyses) taken from the Cochran collaboration and the CASP (Critical Appraisal Skills Program) with both providing checklists to aid in critical analysis.

Part 2:

Critically evaluate the following paper:

A systematic review and meta-analysis comparing mortality in pre-hospital tracheal intubation to emergency department intubation in trauma patients

You should include a critical evaluation of the following:

• Background (10 marks) 220 words

Tracheal intubation (TI) is widely considered to be the most effective means to gain definitive airway control. Most developed countries Emergency Medical Services (EMS) have the capability to deliver this intervention in the pre-hospital intubations (PTI). To facilitate the passing of the tube through the glottis drugs may be used, this process is termed Rapid Sequence Induction (RSI). Globally there have been different approaches to intubation with one of the key decisions is weather the intervention is appropriate prehospitally, some EMS systems make the clinical decision that rapid transportation to the hospital where emergency department intubation (EDI) is performed. Included in the reasoning is a permissive environment, greater resources and a wide range of specialists all of which may favour a successful outcome. Other EMS servic4e4es favour prehospital intubation claiming early definitive advanced airway management along with shorter times to surgery may favour patient outcome. Among the EMS systems that favour intubation there are different providers that perform the intubation, the EU utilizing physician intubation having a full range of RSI drugs and the USA, Australia and South Africa utilising Paramedics who may or may not have access to RSI drugs and some systems utilise a doctor paramedic team.

The question of weather this intervention is beneficial to trauma patients has been a topic of much research and debateover the years along with who should perform intubation and which if any drugs should be used, there has also been discussion over the optimum equipment and which procedure is best. Many guidelines and protocols exist for prehospital intubations with Pre-Hospital Trauma Life Support (PHTLS) providing an international framework. The key question in these discussions is does this procedure improve mortality.

Fevang et al have published a systematic review and meta-analysis comparing mortality in pre-hospital tracheal intubation to emergency department intubation in trauma patients in an attempt to break the clinical equipoise.⁷

• Methodology (30 marks) 660 words

The review does address a clearly focused clinical question (1)in both the interventions and the outcomes.

In consideration of the validity of the study results; this study was prospectively registered with PROSPERO https://www.crd.york.ac.uk/prospero/#aboutpage which helps avoid duplication and reporting bias by comparison of the completed reviews and also adhered to PRISMAhttp://www.prisma-statement.org an evidence-based minimum set of items for reporting in systematic reviews and meta-analyses. This gave the study an evidence based well-recognised and accessible study design protocol and increased the chance of successfully capturing and reporting available evidence.

outcome.

The study and report characteristics are given by the authors and the language used.The authors do not state which search tool was used for example PICO, PICOS or SPIDER. The review excluded material considered to be of a lower quality of evidence based on the assessment table used, individual patient groups that could confound the results were also excluded.

The authors look for right type of papers (2) and list the electronic search strategy including the databases (EMBASEMEDLINE and Cochrane Library) with dates of coverage and the search and date last searched. The words used for the search are listed and relevant the thesaurus alternatives were included. Reference lists of publications and review articles were considered in addition to the data search. The researches attempted to contact the authors to confirm the outcome data where it was vague.

The reviewers state that the process for selecting studies was done via a prearranged data extraction form that listed study design, provider type, patient category then a pair of reviewers independently screening titles and abstracts in an attempt to eliminate bias, discrepancies were resolved by discussion which does expose the selection to bias.

The authors attempted to remain within accordance of the Cochrane principleshttps://www.cochrane.org/about-usThe authors graded the risk of bias is high, low or unclear for allocation concealment, blinding, incomplete outcome dataand selective reporting. For the review of the quality of evidence, the Grading of recommendations assessment, development, and evaluation (GRADE) approach was used for review of Randomised Control Trials (RCT) and the Meta-analyses Of Observational Studies in Epidemiology (MOOSE).⁸ Also covered are observational studies. that contains specifications for reporting of meta-analyses of observational studies listing background, search strategy, methods, results, discussion, and conclusionwhich may improve the usefulness of meta-analysis for reviewers. The authors state the National Institutes of Health (NIH)resources were utilized.

The Authors specify the principle summary measures as odds ratios (OR) and adjusted odds ratios (AOR) for mortality and list other data variables that were also collected. The PHI mortality rates were registered as favourable, unfavourable, inconclusive or no proven difference. The type of statistical analysis is given as Mantel-Haenszel method using the analysis model for random effects with valid justification for how this model would improve validity. The Review Manager program was used as the statistical analysis tool along with Forest plots to distinguish between those that received drugs and those that did not. It is implied that the above tools for the assessment of evidence would help to eliminate bias however any further risk mitigation strategies were not stated.

In an additional analyses strategy to reduce bias from heterogeneity and seek different results from the data the mortality groups were subdivided into 3 groups,physicians intubation and groups with similar GCS and ISS. funnel plots for unadjusted and adjusted mortality were used to identify publication bias.

• Results (20 marks) 440 words

The authors attempt to get the important and relevant studies (3)and selecting the best available evidence for this systematic review starting with3211 search referencesthat were assessed for eligibility of which 64 full tests were reviewed, of those 42 were excluded due to; uncertainty of the intervention,poor methodological quality and overlap of data, this left 21 studies that met inclusion criteria of which 17 that were included in meta-analysis, of those 17 studies 10 did not provide OR or AOR which left 7 studies that were retained for AOR meat-analysis. The results of this review have been combined and this seems reasonable. (5)The conclusions from these 21 included studies were split with 12 showing increase, 7 seven showing no difference, one decrease in PHI and mortality, in one studymortality was not powered to study mortality.The authors followed the

The data fromthe included studiesare listed in tables showing clinical information, outcome data, quality assessment findings, results and main conclusions.

Theoverall results show (6)median mortality rate as 48% (range 8–94%) for PHI and 29% (range 6–67%) for EDI and an OR with 95% CI of 2.56 (2.06, 3.18) in favour of EDI, showing the how precise the results are (7). Forest plots were divided between access to RSI and no or limited access to RSI, both plots showed improved survival with EDI. Correction was attempted using clinical parameters and ISS to reduce heterogeneity. Authors state the adjustment varied among studies to for ISS head injury and BP. It is worth noting that the 4 most homogenous studies showed asignificantly higher mortality rate in the PHI group, also a significantly higher OR for mortality was found in the no RSI or some RSI group (2.40 (1.52, 3.77).

In a comparison between physician PHI and Paramedic PHI without RSI Two studiesshowed no significant differences in mortality rate between the groups.

The researches assessment of the quality of the included studies (4)as the risk of biasbeing high and the quality of evidence, very low with no high-quality evidence due to complexity and number of variables, with only one RCT but was not designed or powered to examine mortality. The researches rated the remaining 20 studies as fair on analysis of funnel plots for publication bias. It is also noted that there was inconsistency in the reporting of mortality across the studies. This assessment of the evidence seems valid and unfortunate, the systematic review could have been considered futile at this point.

• Discussion and conclusions (15 marks) 330 words

In the discussion the authors summarize the aim of the study as being PHI vs EDI and that the data consistently showed higher mortality for PHI and, importantly, no studies identified showed PHI to improve mortality over EDI, even when sub-groups were assessed there was still no benefit to PHI over EDI. The authors again stress the high risk of bias, particularly selection bias in the observational studies and heterogeneity in the meta-analysis. Only the one RCT included showed a nonsignificant trend towards favouring PHI, however mortality was not the primary outcome of this study. To limit bias the authors attempted to select similar patient groups in regards to indications for PHI and utilized ISS, this adjustment did not produce any significant change and the authors state that this could be due to the lack of physiological parameter data, whichmay have been more important than ISS. The researchers list the differences in patient population, procedural complexity, staff, EMS systems as important contributors to heterogeneity, with access to RSI drugs and training as possibly the most important differences, this highlights the difficulty of applying these results to currant local patient population (8).

In analysis between PHI and EDI did not show a significant difference in success rates compared to mortality indicating that there are other variables that must account for this difference. The iatrogenic physiological complications of TI like aspiration and hypoxia which may be experienced by as many as 50% of patients, after TI the patients are Positive Pressure Ventilated (PPV) increasing the chance of tension pneumothorax. For patients in shock the combination of TI and PPV is known precipitate cardiac arrest as is found in the included RCT and for this reason some systems intentionally postpone TI until the patient is resuscitated. The authors identify this detail as lacking in this review and an important consideration in future studies as an additional important outcome (9).

The authors argue that although quality of the evidence is low the study is important as an attempt to elicit the differences in mortality rates and although this study shows increased mortality for PHI the procedure may be valid as it provides the highest standard of airway protecting early in the care pathway and this may be more beneficial than the complications of the procedure. The authors conclude the discussion by stating that despite the low quality of the evidence the findings of this systematic review should raise queries and possibly concerns.

In conclusion the authors summarise the main points of the discussion namely the consistency of higher mortality in PHI and the difficulty of this analysis changing practise given the complexity and heterogeneity involved. The importance of the procedure remains and should be a focus of further research, in an analysis of benefit verses harms and cost (10), this systematic review should raise concern over the procedure of PHI.

Summary of the strengths and limitations of the paper (10 marks) 220 words

The strengths of this paper are the adherence to well known and tested strategies for the structure and protocol of conducting a systematic review namely, an evidence-based minimum set of items for reporting in systematic reviews and meta-analysesand is considered particularly effective in the evaluations of interventions.⁹ In a review of the PRISMA Check List (fig 2.) the authors have carefully followed the structure and content of these recommendations, thereby giving the authors the strongest chance of achieving a valid result.

The weakness of this review is the high risk of bias and the low quality of evidence, this concern is repeatedly voiced by the authors. The strength of a systematic review lies in the quality of the studies is selects for data analysis and meta-analysis, as the quality is poor here even though this review has been maliciously conducted the results by definition must be considered poor. That said, the authors sorted the evidence to find the best studies available and continued due to the important nature of the clinical question and their conclusion stands that this review can be considered as a call to arms for researches to conduct a RCT and gather higher quality data, this is surely especially true for subgroups of patients like those in shock.
 

Assignment structure, grammar and spelling (5 marks)

Accurate citation and supporting evidence/references (5 marks)

1. Grant MJ, Booth AJHI, Journal L. A typology of reviews: an analysis of 14 review types and associated methodologies. 2009;26(2):91-108.
2. Smith J, Noble H. Reviewing the literature. 2016;19(1):2-3.
3. Kysh L. Difference between a systematic review and a literature review [Poster presentation]. Exhibited at Medical Library Group of Southern California & Arizona (MLGSCA) and the Northern California and Nevada Medical Library Group (NCNMLG) Joint Meeting in July 20132013.
4. Nightingale A. A guide to systematic literature reviews. Surgery – Oxford International Edition. 2009;27(9):381-4.
5. Uman LS. Systematic reviews and meta-analyses. Journal of the Canadian Academy of Child and Adolescent Psychiatry = Journal de l’Academie canadienne de psychiatrie de l’enfant et de l’adolescent. 2011;20(1):57-9.
6. Cook DJ, Mulrow CD, Haynes RB. Systematic reviews: synthesis of best evidence for clinical decisions. Annals of internal medicine. 1997;126(5):376-80.
7. Fevang E, Perkins Z, Lockey D, Jeppesen E, Lossius HMJCC. A systematic review and meta-analysis comparing mortality in pre-hospital tracheal intubation to emergency department intubation in trauma patients. 2017;21(1):192.
8. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. Jama. 2000;283(15):2008-12.
9. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS medicine. 2009;6(7):e1000097.