KC Chan, NKC Tse, SY Lam, LCK Leung, MC Yam, KL Siu, DCW Chan, SM Tai, AWY Yung

Abstract

A child with decreased conscious level is a challenging medical problem to health care professionals. The present guideline was developed on an evidence based approach. It facilitate doctors in dealing the problem systemically. Easy to follow and detailed algorithms (Appendix A1-A7) were included to streamline the investigations and treatments. Latest literatures after the publication of the guideline were reviewed and no new evidence was found in subsequent reviews to justify any amendment of the original recommendations. Discussions on local adoptation are highlighted in the conclusion of the present paper.

Background

Children presenting with decreased conscious level can be due to a variety of reasons. It was found that about 30 children out of every 100,000 children per year would present in coma not caused by trauma. The overall mortality in this group of children could be up to 46%.¹ While some causes are obvious, others could be much more obscure. The present evidence-based guideline was developed in 2005 to help front line doctors approach the problem systematically, recognise clinically important problems, investigate and treat them.²

The scope of the guideline confines to any child <18 years old, with a Glasgow coma score less than 15 or not being Alert (i.e. responding only to Voice, Pain or being Unresponsive) on the AVPU score.

This guideline should not be applied to preterm infants, children with known cause of their decreased conscious level and children with a chronic abnormal conscious level state.

The guideline mainly targets on the problems that can be identified and treated within the first hours of presenting to a hospital. Specific conditions or diseases which require specific treatment protocols are not covered in detail.

The purposes of our working group in reviewing the original guideline are:

I) To review the methodology of the guideline and the appraisal summary prepared by the Royal College of Paediatrics and Child Health, UK.³

II) To review latest literature after the guideline was published in 2005 and to see if any major changes are required.

III) To assess the applicability of the guideline to the local Hospital Authority hospitals.

I) Introduction to Methodology of the Original Guideline

The guideline was produced by the University of Nottingham, Paediatric Accident and Emergency Research Group which included medical and nursing professionals from paediatric emergency medicine, paediatric intensive care, metabolic medicine, neurology, general paediatrics, clinical chemistry, patient and lay representatives plus input from other stakeholder subspecialty societies or associations.

Literature was electronically and hand searched between March 2004 and July 2005 for a list of clinically relevant questions drawn up according to the scope and targeted clinical conditions the guideline aimed to answer. Papers selected were then appraised on methodological quality using critical appraisal checklist developed by the Scottish Intercollegiate Guideline Network⁴ and were given level of evidence according to the criteria developed by the Oxford Centre for Evidence-based Medicine.⁵

Papers which contributed to grade A and B recommendations were appraised by second member of the Guideline Development Group to ensure validity of the appraisal methodology.

When there was no published evidence found, a consensus approach was adopted. The guideline group utilised a large multi-professional Delphi panel for the Delphi Consensus process which enabled members of the panel to have their opinions registered anonymously, analysed, and then fed back to the same panel for further consideration. The whole panel results were reviewed. The group would help members to reconsider their initial position and panel members were at liberty to change their original opinion or their initial position. Consensus was aimed to achieve after one, two or three rounds of the Delphi panel discussion.^6,7 The guideline recommendation and good practice points were thus based on agreement using evidence tables or the Delphi consensus results. Disagreement on wordings was settled by discussion or consultation with stakeholder groups. The draft guideline was then reviewed by all stakeholder groups, followed by an open forum discussion before it was finalised.

Formal appraisal process for the guideline were then performed by the Royal College of Pediatrics and Child Health and the British Association for Accident and Emergency Medicine in November 2005 using the AGREE instrument⁸ to assess the methodology quality followed by independent reviewers to examine the original research papers deriving the grade A and B recommendations in the guideline. Only minor regrading of the recommendation levels was made.³ An algorithm for the patient management was prepared (Appendix A1-A7).⁹

II) Review of the Latest Literature After the Guideline was Published in 2005

The guideline was originally planned to be reviewed 2 years after being published. Our working group thus initiated a simplified but systematic literature search for the latest evidence in the year 2005 and after.

Papers were searched electronically via the Hospital Authority eKG platform using database: Embase, Medline, and all All EBM Reviews - Cochrane DSR, ACP Journal Club, DARE, CCTR, CMR, HTA, and NHSEED and employing keywords: altered consciousness or consciousness disorder and the disease or clinical condition categories of the original guideline.² Search was done around May 2009 for review articles and with patient age less than 18 year old and later search was repeated in early November 2009 to include additional keywords: meta analysis, systematic review, unconsciousness, transient loss of consciousness, coma, stupor, drowsiness, obtundation and delirium. Additional searches were done in 2010 on selected topics, especially on those with no Delphi consensus reached in the original guideline (investigation of borderline hypoglycaemia and management of raised intracranial pressure - fluid regime, use of mannitol or hypertonic saline and indication of invasive monitoring device)

1. Herpes simplex encephalitis - A review article by De Tiege et al was published in 2008.¹⁰ The content of the review was in line with the recommendations made by the Guideline on Herpes simplex encephalitis.
2. Bacterial meningitis - The source articles of the review published in 2009 by Prasad et al¹¹ concerning the use of corticosteroid were already included in the reference list of the original guideline. Another meta-analysis of individual patient data extracted from recent randomised, double-blind, placebo-controlled trials was published in March 2010.¹² Though adjunctive dexamethasone treatment does not seem to significantly reduce death, dexamethasone seemed to reduce hearing loss among survivors.^12,13 This is line with the finding of another meta-analysis¹⁴ and the latest review from the Cochrane collaboration review finding that corticosteroid dexamethasone leads to a reduction in hearing loss.¹⁵ The recent NICE guideline¹⁶ also confirmed the beneficial effect of early (before or with the first dose antibiotics) steroid treatment on long term neurological sequelae. The beneficial effect on reducing severe hearing loss can also be observed when steroid is given shortly after the first dose of antibiotics (5 RCTs involving 501 children, RR 0.29, 95% CI 0.14 to 0.63, p = 0.002). A steroid regime was thus proposed in the NICE guideline for patients older than 3 months old with high likelihood of bacterial meningitis.
3. Use of steroid in tuberculous meningitis - The original guideline does not cover the treatment of the tuberculous meningitis. In a systematic review by Prasad et al¹⁷ involving seven trials, 1140 participants (including 411 deaths) with all having used dexamethasone or prednisolone, corticosteroids reduced the risk of death (RR 0.78, 95% CI 0.67 to 0.91). Data on disabling residual neurological deficit from three trials showed that corticosteroids reduce the risk of death or disabling residual neurological deficit (RR 0.82, 95% CI 0.70 to 0.97). Hence, corticosteroids are recommended in HIV-negative people with tuberculous meningitis to reduce death and disabling residual neurological deficit. The use of steroid with concomitant anti-tuberculous treatment is included in the NICE guideline.¹⁸
4. Use of mannitol for acute traumatic brain injury - In the review by Wakai and Robert,¹⁹ four trials published in 1984 to 2003 were reviewed but there was no articles included after the present guideline was published. Thus there is no new reliable evidence to make recommendations on the use of mannitol in the management of patients with traumatic brain injury.During the literature search, latest articles possibly related to guideline grade A or B recommendations were then appraised on their methodological quality similar to the approach as stipulated in the original guideline and discussed in the Working group meetings. After rounds of discussions, it was finally concluded that there are no new evidence to justify any amendment of the original grade A and B recommendations of the guideline.

III) The Applicability of the Guideline to the Local Hospital Authority Hospitals

Our working group acknowledges that certain recommendations of the original guideline may need local modifications or considerations.

1. Core laboratory investigations for metabolic causes of reduced conscious level and hypoglycaemia in children with no other clear explanation.²⁰
• Local surveys were done covering all Hospital Authority paediatric departments with emergency admissions. All the core investigations are available within their own cluster hospitals. However, some of the investigations for hypoglycaemia are not available locally, like plasma insulin, growth hormone, free fatty acid, plasma beta-hydroxybutyrate, acyl-carnitine profile and urine organic acid. After some follow up arrangements, they can now be accessed through the two university hospitals (Prince of Wales and Queen Mary Hospitals) or through Princess Margaret and Queen Elizabeth Hospitals.
2. Service arrangements like timing of consultation of experienced paediatrician, paediatric subspecialist or specialist of other discipline.
• While recruiting subspecialist or experts in other fields early in the patient care process is the trend, our working group found these service arrangement recommendation being beyond the scope of our working group review. Differences in medical systems and variations in the settings and organisations of individual hospitals may imply some local adaptation being required before adoption of these recommendations.
3. Conditions identified in the guideline requiring management protocol agreed at a local level.
• Specific conditions like the management of raised intracranial pressure are recommended to have local guidelines. On the other hand, though the guideline recommends the NICE guideline for the management of diabetes ketoacidosis, it is envisaged that there could also be minor variation among local departments in fine tuning the management of the diabetes ketoacidosis patients.
4. Peri-arrest arrangement or investigations taken at post mortem.
   • The original guideline has grade D recommendations on certain tests to be performed based on grade 5 evidence (expert opinions). While these are useful references, it is worth noting that Hospital Authority has also issued a Guidance note on Perimortem / Postmortem Specimen Collection for Paediatric Patients suspected of unknown infectious diseases in July 2009.²¹ In the local guidance notes, tests for toxicology and inborn errors of metabolism are also included. Individual hospitals are advised to review the local guidance note and to liaise with the local pathology department accordingly.

Overall Guideline Review Summary

The present guideline together with the appraisal summary from Royal College of Paediatrics and Child Health and British Association for Emergency Medicine provide concise information and an easy to follow algorithm based on available evidence and consensus of a large representative group of different disciplines. Though the guideline was due for revision in 2007, we could not find new evidence (focused systematic search up to 2009) to justify any amendment of the grade A and B recommendation of the guideline. Additional evidences were noted on the use of steroid in tuberculous meningitis and meningitis likely of bacterial origin. Laboratory investigations for unclear causes of altered conscious are now streamlined in all Hospital Authority paediatric departments after local surveys and follow up actions. Recently, some has advocated including the investigation of poisoning, especially on carbon monoxide poisoning in the use of the guideline²² Individual departments are encouraged to explore and consolidate the consultation mechanism of subspecialists and the peri-arrest arrangement of investigations in their local setting if required.

Appendix

(A1-A7) Algorithm. The management of a child (aged 0-18 years) with a decreased conscious level (http://www.nottingham.ac.uk/paediatric-guideline/Guideline algorithm.pdf (With kind permission from Dr Richard Bowker, lead author of the Paediatric Accident and Emergency Research Group which developed the guideline and algorithm).

Appendix A1 Guideline for the management of a child aged 0-18 years with a decreased conscious level.

Appendix A2 Algorithm for the management of a child aged 0-18 years with a decreased conscious level.

Appendix A3 Indentify all problems.

Appendix A4 Indentify all problems (continued from A3).

Appendix A5 Management of all 16 identified problems.

Appendix A6 Management of all 16 indentified problems (continued from A5).

Appendix A7 Management of all 16 indentified problems (continued from A5 & A6).

References

1. Wong CP, Forsyth RJ, Kelly TP, Eyre JA. Incidence, aetiology, and outcome of non-traumatic coma: a population base study. Arch Dis Child 2001;84:193-9.

2. The management of a child with a decreased conscious level - a nationally developed evidence-based guideline for hospital practitioners (The Paediatric Accident and Emergency Research Group), http://www.nottingham.ac.uk/paediatric-guideline/home2.htm; summary of recommendation: http://www.nottingham.ac.uk/paediatric-guideline/recdoc.pdf; full technical report: http://www.nottingham.ac.uk/paediatric-guideline/Tecdoc.pdf

3. The management of a child with a decreased conscious level. An evidence based guideline for health professionals based in the hospital setting. Appraised by Royal College of Paediatrics and Child Health and British Association of Emergency Medicine. http://www.nottingham.ac.uk/paediatric-guideline/Guideline%20algorithm.pdf

4. Scottish Intercollegiate Guideline Network. SIGN 50: A guideline developers' handbook-Annex C. Critical appraisal: Notes and checklists. SIGN 2001. www.sign.ac.uk

5. Philips B, Ball C, Sackett D, et al. Oxford Centre for Evidence-Based Medicine Levels of evidence, 1998. www.cebm.net

6. Murphy M, Black N, Lamping D, et al. Consensus development methods, and their use in clinical guideline development. Health Technol Assessment 1998;2:1-88.

7. The Delphi Consensus process. Guidleine for the management of a chid aged 0-18 years with a decreased conscious level. Appendix B. http://www.nottingham.ac.uk/paediatric-guideline/B.pdf; http://www.nottingham.ac.uk/paediatric-guideline/delphi. htm

8. Appraisal of Guidelines for Research and Evaluation (AGREE) Instrument. The AGREE collaboration, 2001. www.agreecollaboration.org

9. Algorithm. The management of a child (aged 9-18 years) with a decreased conscious level. http://www.nottingham.ac.uk/paediatric-guideline/Guideline%20algorithm.pdf

10. De Tiege X, RozenbergF, Heron B. The spectrum of herpes simplex encephalitis in children. Eur J Paediatr Neurol 2008;12:72-81.

11. Prasad K, Karlupia N, Kumar A. Treatment of bacterial meningitis: an overview of Cochrane systematic reviews. Resp Med 2009;103:945-59.

12. van de Beek D, Farrar JJ, de Gans J et al. Adjunctive dexamethasone in bacterial meningitis: a meta-analysis of individual patient data. Lancet Neurol 2010:9:254-63.

13. McIntyre P. Adjunctive dexamethasone in bacterial meningitis: does value depend on clinical setting? Lancet Neurol 2010:9:229-31.

14. Assiri AM, Alasmari FA, Zimmerman VA, Baddour LM, Erwin PJ, Tleyjeh IM. Corticosteroid administration and outcome of adolescents and adults with acute bacterial meningitis: a meta-analysis. Mayo Clin Proc 2009;84:403-9.

15. Brouwer MC, McIntyre P, de Gans J, Prasad K, van de Beek D. Corticosteroids for acute bacterial meningitis. Cochrane Database Syst Rev 2010;(9):CD004405. DOI:10.1002/14651858.CD004405.pub3

16. NICE clinical guideline 102 - Bacterial meningitis and meningococcal septicaemia Management of bacterial meningitis and meningococcal septicaemia in children and young people younger than 16 years in primary and secondary care. Issue date: June 2010, revised September 2010. http://www.nice.org.uk/nicemedia/live/13027/49339/49339.pdf

17. Prasad K, Singh MB. Corticosteroids for managing tuberculous meningitis. Cochrane Database of Systematic Reviews 2009: Volume (4).

18. NICE Clinical Guideline 33 -Tuberculosis: clinical diagnosis and management of tuberculosis, and measures for its prevention and control. Issue date March 2006.

19. Wakai A, Roberts IG. Mannitol for acute traumatic brain injury. EBM Reviews - Cochrane Database of Systematic Reviews. 00075320-100000000-00509 Cochrane Database of Systematic Reviews 2009:4.

20. Core laboratory investigations for the metabolic causes of reduced conscious level. Section 13. of summary of recommendation:

http://www.nottingham.ac.uk/paediatric-guideline/recdoc.pdf

21. Guidance note on Perimortem / Postmortem Specimen Collection for Paediatric Patients suspected of unknown infectious diseases issued on 3 July 2009, Hospital Authority (Ref: HA 752/10/38/5/2).

22. Reece A, Cohn A, Heckmatt J. A suggested update for coma guideline. Arch Dis Child 2010;95:570-1.