Continuing Education Activity

Ventriculitis is the inflammation of the ependymal lining of the cerebral ventricles, usually secondary to infection (for example meningitis, device-related or a complication of trauma). There is no clear definition for ventriculitis and no accepted diagnostic criteria, which makes the assessment of incidence difficult. Early diagnosis is essential for appropriate treatment; this requires a clinical history, CSF sample, and imaging. Antibiotics are the mainstay of treatment. This activity will review the causes of ventriculitis and the different microbes responsible. It will evaluate the different investigations required for diagnosis and review the treatment options available. It will briefly highlight some of the controversies and gaps in the literature and identifies the role of the interprofessional team in managing this condition. This activity describes the pathophysiology of ventriculitis and highlights the role of the interprofessional team in its management.

Objectives:

• Identify the etiology of ventriculitis.
• Describe the clinical presentation of ventriculitis.
• Explain the treatment and management options available for ventriculitis.
• Review interprofessional team strategies for improving the diagnosis and management of ventriculitis to improve outcomes.

Introduction

Ventriculitis is the inflammation of the ependymal lining of the cerebral ventricles, usually secondary to infection. It has other names, such as ependymitis, ventricular empyema, pyocephalus, and pyogenic ventriculitis.[1] It is an indolent but lethal infection and a source of persistent infection following meningitis treatment. Early diagnosis is essential for appropriate treatment. It is of particular concern in patients with external ventricular drains (EVDs) or intraventricular shunts.

Etiology

Ventriculitis can be secondary to[2][3]:

1. Meningitis – bacterial and viral
2. Cerebral abscess with intraventricular rupture
3. Catheter-related – shunt or EVD related
4. Trauma
5. Cerebrospinal fluid (CSF) leak
6. Complication of neurosurgery
7. Complications of intrathecal chemotherapy

Epidemiology

There is no clear definition for ventricular infection and no accepted diagnostic criteria. It is unclear whether ventriculitis, catheter-related infections, and positive CSF cultures describe the same condition. True infection, contamination, colonization, or suspected ventriculostomy-related infection are hard to distinguish (Table 1).[2][4]  This ambiguity means it is difficult to quantify the incidence.

Ventriculitis secondary to meningitis (commonly known as pyogenic ventriculitis) is more common in infants. Risk factors are related to low host immunity (cancer, HIV, diabetes, alcoholism, etc.) and higher virulence of the causative organism. Often, when meningitis fails to respond to antibiotics, or when it recurs, ventriculitis should be considered.[5] Suggested mechanisms include direct hematogenous spread to the choroid plexus. Chronically, septations within the ventricles can develop, resulting in multiloculated hydrocephalus, which worsens prognosis and is more common with bacterial infections.

Typical organisms include gram-negative species followed by Staphylococcus species.[1] The incidence of gram-negative bacillary meningitis has increased, likely reflecting an increase in nosocomial meningitis, which presents a challenge due to its indolent course and its tendency to recur.[1][6]

The incidence of ventricular catheter-related ventriculitis (or healthcare-associated ventriculitis) ranges from 0 to 45% depending on the insertion technique and management (commonly less than 10%).[4][7][8] CSF shunt infection has ranged from 4 to 41% (usually in the range 4 to 17%), EVD ventriculitis has ranged from 0 to 22%, and lumbar drain meningitis rates are up to 5%.[2] This has been difficult to assess due to the lack of clear definitions, the severity of the underlying illness, skin flora contamination, and the possibility that the indwelling catheter could induce a CSF pleocytosis. Most of the reported studies are single-center, retrospective studies with a small number of patients. A recent multi-center, UK-based study on EVDs found rates vary between 3 to 18% and from 4.8 to 12.7 per 1000 EVD days.[9]

Catheter-related ventriculitis is associated with significant morbidity and mortality, especially with gram-negative organisms (approaching 58% in some studies).[4][7][8] Gram-positive cocci consistent with skin flora present as isolates in 50 to 60% of infections, including coagulase-negative Staphylococcus (most common), Corynebacterium, Bacillus, Micrococcus, or Propionibacterium species.[9][10] The rise in gram-negative organisms (Escherichia coli, Klebsiella, Enterobacter, Pseudomonas aeruginosa and Acinetobacter baumannii) and drug-resistant organisms has been attributed to the use of antibiotic prophylaxis targeting gram-positive bacteria and prolonged hospitalization.[10]

In ventriculitis following head trauma, Streptococcus pneumoniae and gram-negative rods are the most common pathogens.[10] Oral flora bacteria (Streptococcus pneumoniae, Haemophilus influenzae, and Streptococcus pyogenes) cause infections in patients with skull base fractures and persistent CSF leaks.[10]

Pathophysiology

Ventriculitis, either alone or in the setting of meningitis, had choroid plexitis, with an inflammatory response to the ependymal lining of the ventricles.

Risk factors for catheter-related ventriculitis can categorize into three groups; (1) patient characteristics and the underlying condition; (2) events that break the integrity of the closed system; (3) environmental influences.[8] Risk factors include subarachnoid hemorrhage, neurosurgical operations, concurrent infection, external catheters (i.e., an EVD over a shunt), frequent manipulation of the EVD system, non-adherence to insertion and maintenance protocols and extended duration of EVD.[11][12][13][14] There appears to be no association with age, sex, or race.[8] CSF leaks are a significant risk factor for infection, allowing a long-standing conduit for retrograde microorganism migration.[8] 

Four mechanisms exist by which CSF shunts can become infected. Most commonly, colonization occurs at the time of surgery. Other mechanisms include; retrograde infection from the distal end of the shunt (e.g., bowel perforation), through the skin (e.g., after inserting a needle into the reservoir), and hematogenous seeding.[2] For EVDs, the introduction of infection is most likely at the time of placement, but retrograde infection also plays a role.[2] Biofilm formation on devices protects the microorganisms from the host immune response and antimicrobial therapy.[10]

History and Physical

Symptoms include fever and signs of meningism (nuchal rigidity, headache, photophobia, decreased mental status, seizures, or moribund).

With ventriculitis secondary to catheters, trauma, or neurosurgery, the onset is more subtle, and often there is a relative lack of fever and severe presenting symptoms, reflecting the predilection for immunocompromised patients.[4][13] The patients may have fever unrelated to the infection (e.g., central fever, drug fever, chemical meningitis).[2]  Erythema or tenderness over the subcutaneous shunt tubing is suggestive of infection.[2]

Patients can present with features of obstructive hydrocephalus, particularly in infants with inflammatory aqueduct obstruction.  

Evaluation

Investigations for ventriculitis include CSF sampling and imaging.

The CSF sample can demonstrate an elevated protein count (greater than 50mg/dL). This can be related to a decrease in CSF production, as seen in rabbit models of E.Coli ventriculitis.[15] CSF can have low glucose (less than 25mg/dL), pleocytosis (over 10cells/microL with 50% or more polymorphonuclear neutrophils) and a positive culture or Gram stain.[2] Cultures may be negative following antibiotic therapy, despite active ventriculitis. They may require several days or weeks of incubation, for example for low virulence organisms like Propionibacterium acnes which require at least ten days, however, treatment should not be delayed.[2]

The presence of oligoclonal immunoglobulin G or M bands, CSF lactate, procalcitonin, and lysozymes help make an early diagnosis.[16] An increase in CSF lactate, procalcitonin and lysozymes suggest bacterial over a viral infection.[3][17] A meta-analysis reported a pooled sensitivity and specificity of 93% and 96% for CSF lactate, although this is reduced when antibiotics are administered prior to CSF collection.[18]

CSF cultures are the most important test for establishing the diagnosis of healthcare-associated ventriculitis.[2] However, repeated sampling from EVDs, in the absence of clinical signs of ventriculitis, has been associated with higher infection rates and a low predictive value for diagnosis.[19][20][21] Catheter-related ventriculitis is hard to diagnose on CSF as results are often subtle, and it is difficult to determine if the abnormalities are related to infection, device placement, or following neurosurgery.[2]

Ultrasound, for neonates, can be performed using a high-frequency transducer through the anterior fontanelle in coronal and sagittal planes. Ventriculitis has increased thickness, irregularity and echogenicity of ependyma, with echogenic debris in the ventricle.[5] Ependymal irregularity results from denudation of segments of ependyma, resulting in glial proliferation.[5] At a later stage, when the inflammatory exudate organizes, there can be septa formation (composed of denuded and detached segments of ependyma), compartmentalization, intraventricular cysts, and obstructive hydrocephalus.[5] Increased echogenicity may be seen in the periventricular region as a result of subependymal infiltration with lymphocytes and plasma cells and swollen subependymal astrocytes.[1][5] Additionally, inflammation of the choroid plexus can show increased echogenicity and irregularity.[5] Ultrasound can also be used to detect CSF loculations at the shunt terminus in infected ventriculoperitoneal shunts.

Non-contrast CT demonstrated non-specific findings, including dependent hyperdense ventricular debris, hydrocephalus, periventricular low density as well as features of the underlying abnormality (e.g. signs of meningitis – pial or dura/arachnoid signal abnormality or enhancement).[1] With contrast, the ependymal lining of the ventricles enhances homogeneously.[1] Denuding of the ependyma could be a component of the breakdown of the blood-brain barrier, resulting in enhancement.  

MRI shows the same features as CT, with ventricular debris hyperintense to CSF on T1-weighted images and hypointense to CSF on T2- weighted images.[1] This is the most commonly observed imaging finding for ventriculitis and can be seen in up to 94% of cases.[1] There can be intensely restricted diffusion of the intraventricular debris on DWI/ ADC as seen with cerebral abscesses, but not always.[1][22] FLAIR images are sensitive to subtle periventricular hyperintense signal in 78% of cases.[1]

Ventricular uptake in radionuclide brain scintigraphy using technetium-99 can be used.

Treatment / Management

Antimicrobial therapy that can reach effective concentrations in the CSF is required. Immunocompromised patients need aggressive treatment. Initially, empirical therapy is used based on the patient’s age and etiology. For catheter-related ventriculitis, this is generally vancomycin and an anti-pseudomonal beta-lactam (cefepime, ceftazidime or meropenem).[2]

Specific antibiotics are chosen based on in vitro susceptibility and penetration into CSF when meningeal inflammation is present.[3] A summary appears in Table 2 below. The duration of antibiotics depends on the microbe isolated, the CSF findings, and the clinical symptoms but varies between 10 and 21 days.[2] 

Intraventricular antibiotics are an option if the ventriculitis is refractory to systemic therapy. Commonly used antibiotics include vancomycin (5 to 20mg/d) or gentamicin (1 to 8mg/d) but depend on local microbial policies.[2] This regimen results in higher levels achieved in the ventricular CSF than by intravenous administration.[23][24] Antibiotic dosages have been used empirically, with dose adjustments and intervals based on the ability to achieve adequate CSF concentrations.[3] After the initial dose, doses can be determined by the calculation of the inhibitory quotient (the trough CSF concentration divided by the minimal inhibitory concentration (MIC) of the agent for the isolated bacterial pathogen), which should exceed 10 to 20 for consistent CSF sterilization.[3] 

Removing all components of the infected shunt or EVD, in combination with antimicrobial therapy, is recommended for catheter-related infection.[2][3] This action allows the infection to clear more rapidly, as the microorganisms can adhere to prostheses and survive despite antimicrobe therapy. The decision to re-implant the device depends on the patient, microorganism, severity of infection, and CSF findings.[3] A summary appears in table 3.

Differential Diagnosis

Ependymal lining enhancement can present in primary CNS lymphoma, the ependymal spread of glioblastoma, metastases, or germinoma.

Pertinent Studies and Ongoing Trials

Antibiotics-impregnated (0.054% rifampicin and 0.15% clindamycin) and silver-impregnated ventriculostomy catheters have been developed to reduce infection rates with some conflicting results.[11][25][26] Silver-coated catheters provide an antimicrobial surface, which can inhibit the growth of bacteria, and biofilm formation, over an extended period without systemic side effects.[27]  One study found complete growth inhibition of all microorganisms but P. aeruginosa after 72 hours, with almost complete inhibition of bioform formation for E.Coli, S. aureus, and C.albicans, and reached more than 50% for Enterococcus, coagulase-negative Staphylococci, and P. aeruginosa after 72 hours.[27] There was a significantly lower infection rate with the silver-coated catheter in a randomized controlled trial (RCT), but not in the pooled non-RCTs on meta-analysis.[28] An RCT comparing antibiotic-impregnated catheters with an untreated catheter found that CSF cultures were seven times less frequent in patients with the antibiotic catheters (1.3% versus 9.4%).[29]

The British antibiotic and silver-impregnated catheters for ventriculoperitoneal shunts multi-center randomized controlled trial (BASICS) Trial is a 1200 patient, 17 center trial comparing silicone, antibiotic-coated or silver catheters.[30] It is designed to determine whether impregnated catheters reduce early shunt infections, and results are forthcoming. 

Controversy exists about whether the regular exchange of EVDs can reduce infection. There is a body of literature reporting that infection rates increase after 4 to 5 days of catheter insertion.[12] This data is the basis for the recommendation that EVD catheters should be removed and inserted at a different site if required for longer than five days. Other studies showed no benefit of routine catheter exchanges.[8][31][32][33] Holloway’s study shows a rising risk of infection over the first ten days, but then infection becomes very unlikely, despite a population still at risk.[32] This research has led to the recommendation of removing the catheters at the earliest opportunity clinically and routinely exchanging the catheter if the device becomes obstructed or if an infection develops.[8][31][32]

Prognosis

If untreated, ventriculitis could lead to poor neurology, hydrocephalus, and death.[34] Early recognition and treatment are essential. High-quality studies evaluating prognosis are lacking.

Complications

Due to the risk of recurrence and hydrocephalus, long-term follow up is recommended. The ventricles and choroid plexus can serve as a reservoir of infection, even when the lumbar puncture yields sterile cultures.[35]

Deterrence and Patient Education

The literature is inconclusive about antibiotic prophylaxis for patients with ventriculostomies.[4][36][37] Periprocedural prophylactic antibiotics are currently recommended for patients undergoing CSF shunt or drain insertion.[2] Studies have compared perioperative antibiotics only, or prolonged antibiotics for the duration of the ventricular catheter. Findings show a reduction in the incidence of serious CSF infections, but selecting resistant or opportunistic organisms, including candida and methicillin-resistant Staphylococcus aureus (MRSA).[37] Improved catheter maintenance techniques have been shown to reduce infections.[38]

Experience and the use of aseptic techniques reduce the risk of catheter-related ventriculitis, necessitating the need for education.[7] The surgical technique is important, with lower rates when the device tunnels under the skin, with a distal skin puncture.[14] Healthcare bundles are being increasingly used to minimize EVD- associated infections, including measures such as education, meticulous handling, sampling only when clinically necessary, pre-operative prophylactic antibiotics, with positive results.[9]

Antibiotics prophylaxis is ineffective in trauma patients with CSF fistula.[39] Prophylactic antibiotics for craniotomies help prevent surgical site infections, but has no effect on meningitis prevention and predisposes the patient to more resistant organisms.[40]

Pearls and Other Issues

A need exists for an agreed set of definitions, and increased awareness of ventriculitis, to adequately assess incidence and risk factors in an attempt to reduce infection rates. Recommendations are currently based on expert opinion as rigorous clinical data are not yet available.[2] 

New tests and needed to rapidly and reliably detect the common causes of device-related ventriculitis, including those involving indolent bacteria such a P. acnes. Nucleic acid amplification tests may help accomplish this.[10]

Care is necessary when sampling CSF as there is a risk of introducing infection, and it should be done based on clinical need only.[19][20]

Enhancing Healthcare Team Outcomes

A systematic review highlights the need for a collaborative, interprofessional, and multi-pronged approach to reduce infections, including ventriculitis.[41] Through persistence and teamwork, a reduction in infections is possible.[38][41] Predisposing factors for catheter-related infections are non-adherence to maintenance protocols, frequent manipulation, and catheter irrigation.[13]

The nurse looking after patients with ventriculostomy must be vigilant about signs of infection and report any abnormality to the interprofessional team. The infectious disease nurse and consultant should be involved early in the care to help select the appropriate antibiotics and make recommendations regarding the catheter.  The pharmacist should ensure that the antibiotics selected are sensitive to the organism isolated and that the drugs are safe, reporting any concerns to the team. The wound care nurse should recommend appropriate cleaning solutions for the ventriculostomy site. Only through education and collaboration of the whole interprofessional team, can management and outcomes be improved.[7] [Level V]