In Australia, meningococcal disease, particularly that caused by serogroup C, has declined since introduction of the meningococcal C conjugate vaccine (MenCCV) into the National Immunisation Program (NIP) schedule. The most common N. meningitidis serogroup causing invasive disease in those under 19 years is serogroup B (91%), followed by serogroup C (2–3%).¹ Until the introduction of 7-valent conjugate pneumococcal vaccine (7vPCV) into the NIP schedule, the most common pneumococcal serotypes causing invasive disease were those contained in 7vPCV (14, 6B, 18C, 19F, 4, 23F, 9V).² However, since then, it appears that serotype replacement is occurring; nasal carriage of 7PCV serotypes has been replaced by others, namely 19A and 16F,³ and the rate of invasive pneumococcal disease caused by 7vPCV serotypes has decreased significantly.⁴ However, the rate of invasive pneumococcal disease caused by serotype 19A increased in non-Indigenous people and in the population overall.⁴ Antibiotic resistance is almost exclusively restricted to these serotypes (and remaining 7PCV serotypes).

Viruses

Enteroviruses, including coxsackie and echoviruses, cause 85–95% of cases of viral meningitis. Herpes simplex viruses 1 and 2 (HSV-1 and HSV-2) and other herpes viruses (human herpesviruses 6, 7 and 8, varicella-zoster virus, cytomegalovirus and Epstein–Barr virus) tend to cause meningoencephalitis. HSV-1 and HSV-2 are perhaps the most important causes to consider, as meningoencephalitis caused by these viruses is associated with high morbidity and mortality, which may be reduced with early treatment with antiviral medications.

Fungi

Cryptococcus neoformans is the most common fungal cause of meningitis, but occurs almost exclusively in immunocompromised hosts.

Clinical findings

The classical features of meningitis comprise fever, headache, vomiting, neck stiffness, photophobia and altered mentation. However, the clinical manifestations are often non-specific, particularly in infants and young children. They may include fever, irritability, lethargy, poor feeding or vomiting. Up to 58% of children with meningitis have received antibiotics before the emergency department (ED) presentation.⁵ This may modify the clinical presentation of meningitis.⁶ It is therefore important to consider the possibility of central nervous system (CNS) infection in any sick infant or child, particularly if they are already taking antibiotics.

If the fontanelle is still open, it may be bulging when examined with the infant in a sitting position. Photophobia is difficult to ascertain in young children, and other signs of meningeal irritation may be absent or difficult to elicit. Resistance to being picked up or distress on walking may be the only clues. Kernig’s sign (inability to extend the knee when the leg is flexed at the hip), Brudzinski’s sign (bending the head forward produces flexion movements of the legs) and nuchal rigidity may be present in older children, but have even been shown to have low positive and negative predictive value in adults with meningitis.⁷

Rashes may occur with any bacterial meningitis, although are less common with pneumococcal infection. Petechiae or purpura are suggestive of meningococcal sepsis, but may also occur in Hib and viral meningitis. Enteroviral meningitis may even be associated with florid purpura fulminans.

It is impossible to reliably differentiate between bacterial and viral meningitis on clinical grounds. However, children with enteroviral meningitis are more likely to present in summer or autumn with gradual onset of non-specific constitutional symptoms including diarrhoea, cough and myalgia, in addition to the more typical features.

Investigations

Definitive diagnosis of meningitis relies on examination of the CSF with biochemical analysis, microscopy and culture. Children with suspected meningitis should have a lumbar puncture (LP) performed, unless there is a clear contraindication. The only absolute contraindication is raised intracranial pressure (ICP). It may be difficult to determine whether ICP is raised, but the following signs may be indicative:

• coma (absent or non-purposeful response to painful stimulus);

• abnormal pupillary responses;

• abnormal posturing;

• focal neurological signs or seizures;

• recent (within 30 minutes), prolonged (over 30 minutes) or tonic seizures;

• papilloedema – although this is an unreliable and late sign of raised ICP.

A bulging fontanelle, in the absence of other signs of raised ICP, is not a contraindication to LP.

The threshold for performing an LP should be lower in young children with less specific signs or those who have been taking antibiotics prior to presentation.

Cerebral computerised tomography (CT) should not be used to decide whether it is safe to proceed with LP or not. In a prospective study of bacterial meningitis, CT findings obtained during the acute stages failed to reveal any clinically significant abnormalities that were not suspected on neurological examination.⁸ Moreover, cerebral herniation can occur with a normal CT^9,¹⁰ and the true cause of coning and relationship to prior lumbar puncture is not clearly established.

Other relative contraindications to LP include:

• cardiovascular compromise or shock

• respiratory compromise

• coagulopathy or thrombocytopenia.

Examination of the CSF

A CSF specimen should always be sent for urgent microscopy to help guide empiric treatment. Normal CSF is clear and contains few cells (and no neutrophils). As few as 200 × 10⁶ cells per litre will cause CSF to appear turbid. The CSF profile may help differentiate between bacterial and viral meningitis, but the findings vary. Table 8.7.1 indicates the typical profiles in normal children and those with meningitis. However, these should always be interpreted in the context of the clinical picture. In early bacterial meningitis, the CSF cell count may be normal. In enteroviral meningitis, there is typically a neutrophil predominance early, and this may remain so for more than 24 hours.¹¹

Organisms are seen on CSF Gram stain in 60–80% of cases of bacterial meningitis, provided that prior antibiotics have not been given. The sensitivity is highest in pneumococcal meningitis. Prior antibiotics may preclude culture of the causative organism, but the biochemistry and white cell count remain abnormal for several days whether or not antibiotics have been given.

A traumatic tap occurs in 15–20% of LP in children.^12,¹³ Several formulae have been devised for interpretation of CSF contaminated with blood, but the safest practice is to disregard the red cell count and treat if meningitis is suspected until culture and other studies are clearly negative.

Seizures do not result in increased CSF cell count in the absence of meningitis.

Bacterial or viral DNA can be detected in blood and/or CSF using polymerase chain reaction analysis (PCR). Sensitivity and specificity are high, particularly for N. meningitidis, HSV and enterovirus. PCR for N. meningitidis is particularly useful in patients with a clinical picture consistent with meningococcal meningitis, but who have received prior antibiotics.

Latex agglutination allows rapid detection of bacterial antigens in CSF and urine. However, it lacks sensitivity and specificity, other than for Hib, and is therefore not clinically useful.

Other investigations

• Culture of blood, throat swab, stool/rectal swab or of skin lesion may yield a causative organism.

• Gram stain on blood smear may be positive.

• Blood glucose should be measured at the same time as CSF glucose.

• A baseline sodium should be measured. Hyponatraemia occurs in about one third of children with meningitis, and may be due to increased antidiuretic hormone secretion, increased urine sodium losses, and excessive electrolyte-free water intake or administration.

• Full blood count and acute phase reactants (e.g. C-reactive protein) may provide supportive information, but are more useful when measured serially to monitor the response to therapy.

Management

Antibiotics

Following initial fluid resuscitation, the emphasis is on prompt commencement of parenteral antibiotics. Delay in antibiotic therapy has been associated with adverse clinical outcome in adults with bacterial meningitis.¹⁴

Age <3 months: Amoxicillin 50 mg kg⁻¹ intravenously (IV) 12-hourly (week 1 of life), 8-hourly (week 2–4 of life), 4–6-hourly (> week 4 of life) plus Cefotaxime 50 mg kg⁻¹ IV 12-hourly (week 1 of life), 6-hourly (>1 week of age).

Because of the morbidity associated with neonatal Gram-negative meningitis, and the high rates of recrudescence, some neonatologists add gentamicin to the above regimen.

Age >3 months: Cefotaxime 50 mg kg⁻¹ (max 2 g) IV 6-hourly

Rates of resistance to penicillin and cephalosporins amongst pneumococci have fallen in Australia since the introduction of the conjugate pneumococcal vaccine.^3,^4,¹⁵ Vancomycin should be added if rates of resistance are high.

Steroids

Routine administration of steroids as adjunctive therapy has been controversial in the past. The evidence that they protect against neurological (particularly audiological) complications of childhood meningitis is strongest for Hib meningitis, when dexamethasone is given before the first dose of antibiotics, and when a third generation cephalosporin is used. A recent large European trial in adults with meningitis showed reduction in mortality and severe morbidity with pneumococcal meningitis.¹⁶ A recent Cochrane meta-analysis including adult and paediatric trials concluded that adjuvant steroids are beneficial for children with bacterial meningitis.¹⁷ Evidence from animal studies shows that dexamethasone reduces penetration of vancomycin into infected cerebrospinal fluid.¹⁸ Thus, there is concern that use of dexamethasone with vancomycin could compromise the efficacy of vancomycin in third generation cephalosporin-resistant strains. Fortunately, the majority of cases of pneumococcal meningitis are still caused by strains that are susceptible to penicillin and third-generation cephalosporins.

Accordingly, children (>4 weeks old) who are being treated for possible meningitis (who have not yet received parenteral antibiotics, or who have received their first dose less than 1 hour previously) should be given dexamethasone 0.25 mg kg⁻¹ IV (max 10 mg) (followed by 0.25 mg kg⁻¹ 6-hourly). Steroids should preferably be given 15–30 minutes before antibiotics, although antibiotic administration should not be delayed for more than 30 minutes.

Fluids

Careful management of fluid and electrolyte balance is important in the treatment of meningitis. Over or under hydration are associated with adverse outcomes.¹⁹^–²¹ Many children have increased antidiuretic hormone secretion, and some will have dehydration due to vomiting, poor fluid intake or septic shock. Assessment of the clinical signs of hydration, including weight, measurement of the serum sodium, documentation of urine output, and clinical assessment of the neurological state should be monitored closely, and the total fluid intake adjusted accordingly.

Initial fluid resuscitation to treat shock should be with 20 mL kg⁻¹ of isotonic (normal) saline. Thereafter, isotonic fluids should be given to maintain systemic blood pressure (and thereby cerebral blood flow). Previous guidelines have emphasised the importance of fluid restriction.¹⁹ It may be necessary to restrict fluids if the serum sodium is <130 mmol L⁻¹, or if there are signs of fluid overload. However, fluid restriction does not generally improve outcome²⁰ and has even been associated with a worse neurological outcome.²¹ Thus, most children should receive normal maintenance fluid volumes.

Prevention

Contacts of patients with meningococcal meningitis may require chemoprophylaxis to prevent secondary spread. Those who should receive chemoprophylaxis include:

• the index case if treated only with penicillin (does not eradicate carriage);

• all intimate, household or day-care contacts who have been exposed to the index case within 10 days of onset;

• any person who gave mouth-to-mouth resuscitation to the index case, or had direct contact with their airway secretions.

Although nosocomial transmission of meningococcal infection is rare, droplet precautions are recommended until the patient has received 24 hours of antibiotic therapy. As the aetiology is usually unknown during this period, all cases of suspected bacterial meningitis should be managed in this way. Preferred placement is in a single room. However, if a spatial separation of >1 metre can be achieved and curtains can be drawn between the infected patient and other patients and visitors, this may be sufficient.

Chemoprophylaxis

Rifampicin 10 mg kg⁻¹ (5 mg kg⁻¹ <1 month old) orally 12-hourly (max 600 mg) for 2 days or

Ceftriaxone 125 mg (≤12 years)/250 mg (>12 years) mg intramuscularly as a single dose or

Ciprofloxacin 500 mg orally as a single dose.

7vPCV and MenCCV vaccines are part of the routine NIP schedule for all children and should have a continuing impact on the number of cases of childhood meningitis.

Complications

Bacterial meningitis is associated with a 4.5% mortality rate, and intellectual, cognitive and auditory impairment in 10–20% of survivors.²² The risk for sequelae is greatest in those who experience acute neurological complications at the time of their illness.²³

Viral meningitis

Most cases are self-limiting, and treatment is symptomatic.

Brain abscess

Brain abscess classically presents with fever, headache and focal neurological deficit. Although rare, early recognition is vital, as early antibiotic treatment and drainage improve outcome. Diagnosis is by cerebral CT or magnetic resonance imaging (MRI). Aspiration for diagnosis and neurosurgical intervention are usually required. The most common causes are oral viridans streptococci, anaerobes, Gram-negatives and S. aureus.

Empiric treatment is:

Flucloxacillin 50 mg kg⁻¹ (max 2 g) IV 4-hourly plus

Cefotaxime 50 mg kg⁻¹ (max 2 g) IV 6-hourly plus

Metronidazole 15 mg kg⁻¹ (max 1 g) IV stat, then 7.5 mg kg⁻¹ (max 500 mg) IV 8-hourly.

Encephalitis

Aetiology

HSV is the most common cause of non-seasonal encephalitis in Australia. Other causes include enteroviruses, influenza virus, other herpes viruses and Mycoplasma pneumoniae. Several other viruses cause seasonal epidemics of encephalitis in specific geographic areas. Examples include Murray Valley encephalitis in the east Kimberley region, and Japanese B encephalitis in the Torres Strait Islands.

Clinical findings

There is considerable overlap between the features of meningitis and encephalitis, and the two frequently co-exist. Encephalitis should be suspected if there is encephalopathy, that is, a change of behaviour, altered conscious state or extreme drowsiness. Convulsions (particularly focal) and focal neurological deficits are also more common in encephalitis.

Investigations

CSF biochemistry and microscopy findings are non-specific (see Table 8.7.1). PCR testing for HSV and enterovirus is sensitive and specific. Cerebral CT or MRI of the brain and EEG may be suggestive of HSV encephalitis, particularly if they show temporal lobe involvement.