PAEDIATRIC ALTERED CONSCIOUS LEVEL GUIDELINE
HOME	AIMS	SCOPE	CLINICAL QUESTIONS	EVIDENCE-BASED SEARCH	DELPHI PROCESS	RECOMMENDATIONS	COMMENTS	GUIDELINE DEVELOPMENT GROUP	DRAFT GUIDELINE

DELPHI PROCESS

Round one

12. Intracranial infections

Statement 12a (Intracranial infections)

(i) Bacterial meningitis can be diagnosed in the presence of at least one of the following:

Diagnosis of bacterial meningitis

Cerebral spinal fluid (CSF) positive bacterial culture

CSF positive Gram's stain with negative cultures in the presence of clinical manifestations of meningitis.

Positive bacterial culture or Gram's stain from blood, petechial lesion or sputum in the presence of clinical manifestations of meningitis

Polymerase chain reaction result positive for bacteria in the CSF or blood in the presence of clinical manifestations of meningitis

CSF profile demonstrating significant cytochemical changes such as white blood cells ³ 1,000/mm³, neutrophils ³ 60%, protein ³ 100 mg/dL and glucose £ 50 mg/dL in the presence of clinical manifestations of meningitis

% Agree	% Disagree	Result
90.6%	3.1%	Included

Position	Comments
ED P	Yes except PCR would not be available acutely
Radiol	Purpuric rash = treat first
Endo	Need to change protein and glucose to “English” mmol/l
PICU	are you after diagnosis of, possible diagnosis – considered likely?
Paed	mmol/l for gluc/prot please
Micro	I presume by including sputum you are suggesting pneumococcal meningitis. We rarely receive this specimen type – usually ET secretions. Some blood PCRs e.g. pneumococcal need to be interpreted with caution.

(ii) Bacterial meningitis should be suspected in children who score 8 or more in the following clinical decision rule:

Instructions Sum the scores of the symptoms/signs present. If the total is 8 or more then the chances of having bacterial meningitis is high.

Symptom/sign Score If GCS < 9 = 8 Neck stiffness present = 7.5 Duration of symptoms =1 /each 24 hrs Vomiting = 2 Cyanosis = 6.5 Petechiae = 4 Serum CRP = CRP value (g/dl) divided by 100

% Agree	% Disagree	Result
46.7%	26.7%	Discussed in round 2

Position	Comments
Metab	I have no idea how or if this has been validated
PICU	ALL TOO NON SPECIFIC
Neuro	Not specific enough
Neuro	I don’t think the diagnosis of likely bacterial meningitis requires such a sophisticated scoring system. It looks to me as if someone had to much time on their hands!
Endo	Sorry, too complex for my liking
ED P	I am not familiar with this scoring system, presumably if only the CRP score was greater than 8, this is not suggestive of meningitis only. I cannot comment on this system
	Never seen this before but it doesn’t look very discriminating between bacterial and viral meningitis
Neuro	CRP can be 80…and the score would be 8…the leve of CRP is not related to bact meningitis…it is non-specific…
ED P	Not familiar with this
PICU N	never seen before but sounds fair
Paed	I don’t know this score or it’s sensitivity
PICU	would be interested to read paper
Neuro	What about fever? The mistakes I have seen made include LPs to exclude meningitis in patients with GCS<8 but without fever
Micro	Any child with cyanotic heart disease who vomits has meningitis

(iii) If bacterial meningitis is suspected, dexamethasone 0.4mg / kg should be administered before or with the first dose of antibiotics

% Agree	% Disagree	Result
65.2%	21.7%	Discussed in round 2

Position	Comments
Paed	I don’t know this score or it’s sensitivity
Paed	Excluding those with meningococcal rash
Neuro	evidence
ED P	For pneumococcal
Metab	In our centre the recommended dose is 0.15 mg/kg 4 times a day for 4 days
Neuro	Depends on type
ED P	If possible
ED	Local protocol may be important – I’m not sure if there is a national policy
ED N	Depends if they have had Hib
Paed	Good for HIB/Strep. No proof in meningococcus
Paed	only haemophilus
Micro	Only of benefit in Haemophilus / pneumococcal meningitis. Many doses would need to be given to be worthwhile.

(iv) If bacterial meningitis is suspected, dexamethasone 0.4 mg / kg should be given twice a day for 48 hours

% Agree	% Disagree	Result
52.4%	28.6%	Discussed in round 2

Position	Comments
Paed	Good for HIB/Strep. No proof in meningococcus
Paed	Excluding those with meningococcal rash
Metab	In our centre, dose is 0.15 mg/kg
Neuro	Depends on type
Neuro	evidence

(v) If bacterial meningitis is suspected, broad spectrum antibiotics should be started without waiting for a lumbar puncture to be performed if it is contraindicated

% Agree	% Disagree	Result
96.9%	0%	Included

Position	Comments
Paed	If it’s contraindicated then you aren’t going to perform it. Thus waiting for nothing

(vi) If bacterial meningitis is suspected, broad spectrum antibiotics should be continued until further advice is available from microbiology

% Agree	% Disagree	Result
96.9%	3.0%	Included

Position	Comments
Paed	paediatric decision

Statement 12b

(i) Herpes simplex encephalitis (HSE) can be diagnosed by the presence of at least one of the following

Diagnosis of HSE

A positive PCR for herpes simplex virus DNA from a cerebrospinal fluid sample

Specific herpes simplex virus antibody production in a cerebrospinal fluid sample

Herpes simplex virus from a brain biopsy sample

% Agree	% Disagree	Result
100%	0%	Included

Position	Comments
Metab	Our centre has had false positives for the PCR test alone
Micro	PCR is the technique of choice but a positive result may not be present very early (3 days from symptom onset) or after 14 days. Only if PCR is unavailable or timing is inappropriate is antibody in CSF of value but needs to be compared with serum antibody response to ensure there is no leakage across the BB barrier. Brain biopsy is no longer justified.
Endo	Is PCR perfect?
PICU	3. v rarely done now. MRI + LP usual indicators… EEG helps
Paed	mostly presumptive diagnosis because of results taking many weeks
Micro	Usually require both serum and CSF to demonstrate production of antibodies within CSF.

(ii) HSE can be excluded if a magnetic resonance image is normal 3 days after presentation

% Agree	% Disagree	Result
27.8%	55.6%	Excluded

Position	Comments
Micro	MRI must definitely not be used to exclude HSE at any stage, but particularly not on day 3.
Neuro	But it does make it very unlikley
	Both CT and MRI can be normal for both encephalitis and meningitis
PICU	unlikely but if strong clinical suspicion because of story/contact continue aciclovir

(iii) HSE should be suspected clinically in a child with reduced conscious level if two or more of the following are present: Fever in the history or a temperature > 38C on admission

Seizures (either focal or generalised)

Headache in the history or on admission

Vomiting in the history or on admission

Focal neurological signs

% Agree	% Disagree	Result
65.5%	6.9%	Discussed in round 2

Position	Comments
Paed	These are simply signs of meningitis and not of herpes
Neuro	But, I think the early signs of HSE are so non-specific that it should be suspected in any child with reduced conscious level without other obvious cause. I can think of children with HSE who have not been pyrexial or had any of the above at presentation
Neuro	temrperature not necessary high
Paed	Several other things in differential diagnosis
Metab	Seems reasonable ,but I do not know how or if this has been validated
Micro	I agree, but fever is not always reliable in my experience, perhaps because nursing staff are less well trained or as diligent as they should be. Focal seizures in the neonate, which may become generalised, are a common presenting feature
Metab	I find this question confusing. What about the wider differential diagnosis
Neuro	Should be suspected but not sensitive enough
Endo	“suspected” = “considered” ? all febrile convulsions getting treated for HSE!
ED P	But I would weight the finding as the focal signs are more important. What about other signs such as fluctuating conscious level, presence of/or contact with herpetic lesions?
Paed	- prolonged seizure, - varying conscious level
Neuro	Seizures is probably the most specific, followed by focal signs

(iv) If HSE is suspected clinically then intravenous aciclovir 10mg / kg (or 500mg/m² if aged 3 months to 12 years) twice a day should be administered, without waiting to perform a lumbar puncture if a lumbar puncture is contraindicated

% Agree	% Disagree	Result
90%	6.7%	Included

Position	Comments
Paed	Suspicion must be more than (iii) above
Pharm	FREQUENCY should be THREE times a day, DOSE may require increasing to 20mg/kg in babies <3months
Paed	Should be thrice a day
ED	BNF recommends 8 hourly dosing
Micro	I agree, although doses divided over an 8-hour period are more usual. There is sometimes a tendency (erroneous) to observe an infant, particularly in neonates, to see if they improve following one or more seizures. This results in unnecessary delay in administering aciclovir with medico-legal consequences. In addition to now recommending more prolonged administration (see below) higher doses are now recommended, e.g. 60/mg/kg/daily divided every 8 hours (Gutierrez K and ArvinA.M. Lon term antiviral suppression after treatment for neonatal herpes infection. Paediatric Infect Dis J. Vol. 22(4) April 2003/371-372)
Neuro	Yes, except there is debate as to whether a higher dose is appropriate (15mg/kg)
ED	I’d look it up but agree it should be treated stat
Micro	SHOULD BE GIVEN 8 HOURLY

(v) If HSE is confirmed or highly suspected then intravenous aciclovir should continue for 14 days

% Agree	% Disagree	Result
80.8%	11.5%	Included

Position	Comments
Neuro	I would do three weeks
Neuro	Relapses known at 14 days treatment: 21 days?
PICU	LONGER COURSE
Micro	Since recurrences may occur, there are now recommendations for treatment for up to 21 days. Studies have shown that morbidity may be due to intermittent reactivation of HSE in the months, or even years, after neonatal treatment
Metab	Isn’t it usually 10 days?
Neuro	AT A MINIMUM, much longer treatment may be appropriate
ED P	Probably, although we have been advised 3 weeks if confirmed and then to stop, and if deterioraton then to restart
Paed	or even longer
PICU	some would say 3/52. I think 2/52 adequate if response adequate + immuno OK

(vi) Intravenous aciclovir can be stopped if the cerebrospinal fluid sample results are negative for either PCR or HSV specific antibody

% Agree	% Disagree	Result
40.1%	27.3%	Discussed in round 2

Position	Comments
Paed	Depends also on clinical picture and whether other diagnosis has been confirmed
Neuro	False negatives are well recognised. It depends on the clinical picture as a whole and the results of scans, etc
Micro	See above comments relating to timing of CSF specimens for PCR. It may take up to 14-21 days for antibodies to appear in the CSF, a little shorter for serum antibodies. If in doubt, continue ACV for at least 14 days. The drug is extremely well tolerated. Bear in mind that PCR is extremely sensitive and false positive results may occur if laboratory "hygiene" is not of the highest order. False negatives may also occur, particularly if samples are collected early or late. If in doubt, always ask the laboratory to repeat investigations using, if necessary, fresh samples
ED P	In conjunction with scans and EEG
ED P	… and no other definite indicators clinically of HSV
Neuro	False negatives known
Neuro	Depends on clinical condition
Endo	Depends on clinical suspicion/recovery
Neuro	continue if clinically suspect. Consoider repeating LP, doing a biopsy etc
PICU	+ suspicion low/other diagnoses made
Paed	results never back in time
Micro	False negatives well recognised

(vii) Intravenous aciclovir can be stopped if a magnetic resonance image performed 3 days after presentation is normal

% Agree	% Disagree	Result
35.3%	47.1%	Excluded

Position	Comments
Neuro	Certainly helps, but need clinical context
Micro	MRI must definitely not be used to exclude HSE at any stage, but particularly not on day 3.
Neuro	This is one factor which should be taken into account, but not the only one
Endo	Presumably!
PICU	+ suspicion low/other diagnoses made
Paed	interesting if this is true

Statement 12c

(i) An intracranial abscess can be diagnosed from the results of cranial imaging

% Agree	% Disagree	Result
90.3%	0%	Included

Position	Comments
Neuro	small abcess can be missed….but repeated imaging should detect it
ED	can be difficult
Metab	Recall a case where imaging and history strongly suspected tumour , but which turned out to be an abscess
Paed	In the presence of good history too
Neuro S	Appearances can be subtle in early/evolving cerebritis – first 3 or 4 days. Sometimes necessitating repeat scan
Endo	Can “usually” be…
PICU	contrast MRI + CT
Micro	MRI more sensitive than CT scan, latter needs to be with contrast. Not always diagnostic

(ii) If an intracranial abscess is diagnosed, broad spectrum antibiotics should be administered after blood cultures have been taken.

% Agree	% Disagree	Result
93.1%	0%	Included

Position	Comments
Renal	Can’t see how you would diagnose this without imaging and you would be likely to give antibiotics before then
PICU	seek micro cons advice
Micro	May need to be more specific on choice of antibiotics, we would not use cefotaxime alone

(iii) If an intracranial abscess is diagnosed, advice from a paediatric neurosurgeon should be obtained urgently

% Agree	% Disagree	Result
100%	0%	Included

Position	Comments
PICU	+transfer to PICU

Statement 12d

(i) Tuberculous (TB) meningitis can be diagnosed from a cerebrospinal fluid sample by a positive TB culture or a positive PCR for TB DNA.

% Agree	% Disagree	Result
92.6%	0%	Included

Position	Comments
Neuro	May also be missed
Endo	?PCR perfect

(ii) If the microscopy of a cerebrospinal fluid sample is abnormal, request a Zeihl-Neelsen stain.

% Agree	% Disagree	Result
87.5%	4.2%	Included

Position	Comments
Neuro	if indicated
ED	?If TB infection a possibility
Neuro	Depends on clinical circumstances
Paed	Only if high risk patient and/or cultures neg
Paed	Not in Norwich
Neuro	Depends on abnormality
Micro	Consider if cultures negative, microscopy suggestive etc etc

-

(iii) If the microscopy of a cerebrospinal fluid sample is abnormal, request a PCR for TB DNA.

% Agree	% Disagree	Result
80.9%	4.8%	Included

Position	Comments
Neuro	and c/ w tb
Neuro	Depends on clinical circumstances
Paed	Only if high risk and/or cultures neg
Paed	Not in Norwich
Neuro	Depends on abnormality
Endo	need to be more specific eg raised lymphocytes with decreased glucose – otherwise could be all with raised WCC ie clear bacterial
Micro	Consider if cultures negative, microscopy suggestive etc etc

(iv) If the microscopy of a cerebrospinal fluid sample is abnormal seek urgent advice from the microbiology department

% Agree	% Disagree	Result
81.8%	0%	Included

Position	Comments
ED P	Unless clear cut
Neuro	Comments within reason
Endo	Unless standard local guidelines for 1^st line management of meningitis in place
Paed	but use paediatric experience as well
Micro	Helpful in discussing priorities in terms of further investigation.