Neurosurgical shunt infections

• Coagulase-negative staphylococci (Staphylococcus epidermidis) account for the majority of infections in VP shunts.
• Other organisms include Staphylococcus aureus, Cutibacterium (Propionibacterium) acnes, streptococci, enterococci, corynebacteria.
• Gram-negative organisms including Enterobacteriaceae and Pseudomonas aeruginosa are more common in EVDs.
• Rare causes include fungi, Mycobacteria and Nocardia.

• Contamination (at implantation)
• Externalisation (erosion of shunt through skin)
• Retrograde infection (bowel perforation)
• Haematogenous (rare) 

For perioperative prophylaxis contact the Paediatric ID team for advice.

• A single EVD can be used for as long as clinically indicated, unless a change is necessary because of persistent CSF infection or catheter malfunction.
• Prophylactic exchange of EVD does not significantly decrease the likelihood of CSF infection.
• The use of post-placement prophylactic prolonged systemic antimicrobials for prevention of infection is not recommended.
• Current practice at OUH is for the use of rifampicin-impregnated EVDs

These are variable and depend on the patient’s age, responsible microorganism, the site of infection (proximal vs distal) and loss of shunt function.

• New headache, nausea, lethargy, and/or change in mental status.
• Erythema and tenderness over the subcutaneous shunt tubing.
• Symptoms and signs of peritonitis or abdominal tenderness, in patients with VP shunts.
• Symptoms and signs of pleuritis in patients with V-pleural shunts.
• Demonstration of blood stream infection in a patient with a VA shunt, in the absence of another clear source of blood stream infection.
• Meningeal signs are uncommon and the absence of fever does not exclude infection.

Collect specimens for microbiology before starting antimicrobial therapy where possible.

• Blood tests. Full blood count, ESR, CRP, U&E, creatinine, liver function tests, glucose. A normal white cell count, ESR and CRP do not exclude shunt infection. Unexplained renal disease in a patient with a CSF shunt  (particularly VA shunt) may be due to “shunt nephritis”. 
• Blood cultures. In patients with VA shunt infections, the positivity approaches 90% when multiple cultures are drawn. Blood cultures are rarely positive with VP shunt infection.
• CSF examination-before starting antimicrobial therapy if possible.
  • Lumbar puncture/shunt tap should be discussed with the neurosurgical team (OARS referral). CSF samples are usually taken from the shunt reservoir and occasionally via lumbar puncture.
  • Request urgent CSF analysis for MC&S, protein and glucose. For other tests (e.g. suspected TB or fungal infection) liaise with Microbiology.
  • None of the classical CSF findings in meningitis are specific for shunt infection. CSF white cell counts in samples obtained by shunt aspiration tend to be lower than when CSF is obtained after lumbar puncture.
  • In patients with obstructive hydrocephalus, CSF obtained by LP may not be in communication with the ventricular space, therefore may not be interpretable.
  • A negative CSF Gram stain does not exclude the presence of infection.
  • CSF should be cultured for at least 10 days in an attempt to identify slower growing organisms such as Propionibacterium (Cutibacterium) acnes.
  • If a CSF shunt or drain is removed in patients suspected of having infection, cultures of shunt and drain components are recommended. This will include the proximal catheter, valve or reservoir, and a distal catheter.

• Peritoneal fluid should be sent for culture if there is evidence of peritoneal inflammation in a VP shunt

• Contrast-enhanced CT scan or Magnetic resonance imaging (with gadolinium enhancement and diffusion-weighted (DW) imaging) is recommended for detecting abnormalities such as hydrocephalus, material with restricted water diffusion in the ventricular space and ependimal enhancement. The use of contrast and DW images may help to identify ventriculitis. Contrast enhanced scans may not be needed if there is strong clinical suspicion of infection.
• Ultrasound/CT of the abdomen are recommended to detect CSF loculations at the VP shunt terminus in patients with abdominal symptoms.
• Chest XR. If ventriculo-atrial or ventriculo-pleural shunt.
• Echocardiogram. Consider if ventriculo-atrial shunt to reveal potential thrombi at the distal end of the atrial catheter and/or tricuspid endocarditis.

CNS shunt infection (VA/VP, EVD or LD associated)

• Standard practice for management of infection is shunt externalisation.
• Cerebrospinal fluid should be sent for microscopy and culture.
• Modify treatment according to sensitivity test results.  
• Treatment duration is pathogen dependent. Consult Paediatric Infectious Diseases.

ceftriaxone 50mg/kg iv BD

AND vancomycin iv

OR if at risk of resistant infection (previous infection, prolonged hospital stay, known colonised by organism resistant to ceftriaxone)

ceftazidime 1 month or over: 50mg/kg iv TDS (Max 6g per day)

AND vancomycin iv

For penicillin allergy (severe)

ciprofloxacin* 10mg/kg iv TDS (Max 400mg per dose) 

AND vancomycin iv

 *ciprofloxacin may induce convulsions in patients with or without a history of convulsions – use with caution

ceftazidime 1 month or over: 50mg/kg iv TDS (Max 6g per day)

AND vancomycin iv

AND vancomycin intrathecally see below for doses

vancomycin intrathecal paediatric dosages:

Under 1 month: 5mg -10mg intrathecally OD. Discuss with Paed Infectious Diseases

1 month–18 years:

• 10mg intrathecally OD (with CNS trough levels)
• Lower dosage if slit ventricles (5mg intrathecally daily)
• Higher dosage if large ventricles (15-20mg intrathecally daily)
• More frequent dosing if CSF loss exceeds 100ml/24hours.

This is an unlicensed product and route of administration, but is the most appropriate product and administration method to use.

The intrathecal preparation must be used, NOT the intravenous preparation.

Choices should be guided by susceptibility results and after discussion with Paeds ID team

Typical dosing: always discuss with Paeds ID; please consult BNFC

• Management of shunt infections without removal of shunt hardware is associated with a poor outcome, and a high mortality rate.
• Once a shunt infection is diagnosed including infected EVD, for the majority of cases, the entire device should be removed and replaced with an interim external ventricular drain, ideally at a different site.
• A conservative management plan with device retention (systemic antibiotics plus intraventricular antimicrobial therapy through a separate ventricular access device) may be appropriate for selected patients (usually those with high perioperative surgical risk or poor short-term outcome) who may have CSF shunt infections caused by less virulent organisms, such as coagulase-negative staphylococci. and Cutibacterium (Propionibacterium) acnes.

Always consult Paeds ID team

Intraventricular antimicrobial therapy should not be used as a substitute for prompt and effective source control and shunt removal. Pharmacokinetics of intraventricular antimicrobials are complex and depend on ability to redistribute after instillation, CSF clearance rates, presence of hydrocephalus/haemorrhage and rate of instillation. 

Intraventricular antimicrobial therapy may be considered an option for the following groups of patients:

• patients with healthcare-associated ventriculitis in which the infection responds poorly to systemic antimicrobial therapy alone (ongoing clinical signs after 48h or persistent positive culture on day 5),
• suboptimal systemic antimicrobial options usually due to resistance
• shunt retention/ salvage when removal is delayed/ not possible 

It is recommended that when an antibiotic is given intrathecally, it should also be given IV. 

Rationale for administration of intraventricular therapy is based on a less pronounced ability of antimicrobials to penetrate the BBB in ventriculitis because of a mild inflammatory response. The theoretical benefits of IVT antimicrobial administration should be weighed against any potential for toxicity. The efficacy and safety of this route of administration have not been demonstrated in controlled trials and there is insufficient evidence to recommend their general use. Penicillins, cephalosporins and ciprofloxacin should not be given by the intrathecal route because they have been associated with significant neurotoxicity and/or seizures.

Intraventricular antimicrobial therapy may be considered an option for the following groups of patients:

• patients with healthcare-associated ventriculitis in which the infection responds poorly to systemic antimicrobial therapy alone (ongoing clinical signs after 48h or persistent positive culture on day 5),
• suboptimal systemic antimicrobial options usually due to resistance
• shunt retention/ salvage when removal is delayed/ not possible 

• The optimal timing of shunt reimplantation is unclear and it should be individualized based on the isolated organism, severity of ventriculitis, and improvement of CSF. Typical duration of therapy ranges from 7 to 21 days. Longer durations may be warranted in cases of difficult to treat microorganisms, persistent positive CSF cultures, brain abscesses, fulminant ependymitis and in immunocompromised patients.
• The 2017 IDSA Practice Guidelines for Healthcare-Associated Ventriculitis and Meningitis recommends the following durations, however this needs to be discussed on a case by case basis with the Adult or Paediatric ID team:

** If no associated CSF abnormalities and with negative CSF cultures for 48 hours after externalisation, a new shunt could be reimplanted as soon as the third day after removal

• When reimplanting an EVD, it is typically placed into a different hole through the skull to the previous EVD, but not necessarily into the contralateral ventricle. The two ventricles are connected anatomically and there is no evidence that placement in the other ventricle will reduce infection risks.
• For reimplantation of VP shunt, the new shunt is typically placed back into the original hole/ track once the EVD has been placed for a few days in a different location.
• The risk of infection does increase with EVD duration, hence prompt removal of the external ventricular drain when no longer needed is strongly recommended. Should the EVD become infected in a shunt-dependent patient, an exchange of the ventriculostomy is warranted.
• If device reimplantation is postponed for up to 2 days of the initially recommended antimicrobial plan, antibiotics should be continued until the procedure has been carried out but, if it is delayed for longer periods, antibiotics should be discontinued and CSF  samples monitored in order to verify clearing of the infection 2-3 days before shunt reimplantation.