Cerebrospinal fluid (CSF) shunting has been used for the treatment of hydrocephalus for over 40 years. Drainage of CSF into the peritoneal cavity via a ventriculoperitoneal (VP) shunt is the system used most frequently.
Infection of CSF shunts is an important cause of morbidity. Although reported shunt infection rates range from 2% to 30%, the infection rate at most high volume institutions does not exceed 5%. Risk factors associated with the development of CSF shunt infection include shunts placed for hydrocephalus secondary to intra-ventricular hemorrhage, replacement of an extraventricular drain (EVD) placed for prior infection, patient age younger than 4 months and the experience of the surgeon. EVD catheters are used for short-term monitoring of intracranial pressure or drainage of CSF, and the rate of infection approaches 11%. The exposed exit site of the catheter serves as the portal of entry, and the risk of infection increases with increased duration of catheter use.
Coagulase-negative staphylococci and Staphylococcus aureus cause about 75% of shunt infections. Coagulase-negative staphylococci produce a mucoid substance that facilitates adherence of the bacteria to the shunt apparatus and protects the organisms from host-defense factors and antimicrobial therapy. Gram-negative bacilli cause 20% of shunt infections. The remainder of infections is caused by a variety of organisms, including Micrococcus species, Streptococcus species, fungi (usually Candida species) and anaerobes such as Propionibacterium acnes.
Contamination of the apparatus at the time of surgical placement is the most common mechanism by which shunts become infected. Only 20% of infections due to skin flora are caused by an organism present before surgery, suggesting that the hospital environment may also provide a potential reservoir of microorganisms.
Shunt infections are also caused by retrograde infection originating at the distal end of the shunt. The abdomen is the usual source, and infection can occur by several mechanisms. Peritonitis resulting from penetration of the bowel by the distal end of the catheter or arising spontaneously may cause a secondary ascending infection of the shunt. Less commonly, infection follows transient bacteremia with secondary seeding of the shunt apparatus, or organisms gain direct access from overlying skin or wound tissue that has become infected.
Pediatric Neurosurgery, edited by David Frim and Nalin Gupta. ©2006 Landes Bioscience.
Approximately 80% of shunt infections occur within 3 months of shunt placement or revision. Patients may present with shunt malfunction and associated symptoms such as headache, vomiting, irritability and lethargy. Fever and meningeal signs may not be present. Patients with abdominal signs often appear acutely ill upon presentation, and are more likely to have gastrointestinal involvement and infection caused by a Gram-negative organism. Coagulase-negative staphylococci infection may also present with signs of acute abdominal infection, although a subacute presentation with nonspecific symptoms is more common.
Diagnosis is made by culture and Gram-stain of ventricular CSF. Evaluation of CSF for protein and glucose is also helpful. Ventricular CSF typically has a pleocy-tosis of several hundred white blood cells per ^l, increased CSF protein and a normal CSF glucose concentration. Abdominal ultrasound or computed tomography (CT) is indicated if abdominal signs are present.
Treatment of a shunt infection consists of a combination of antimicrobial therapy and surgery. Because the success rate of antibiotic therapy alone is low, a combined medical and surgical approach is required. This includes antimicrobial therapy and externalization of the distal end of the shunt (for an isolated abdominal infection) or removal of the shunt apparatus with placement of a temporary EVD until sterilization of the CSF is documented. This latter approach results in cure of 96% of treated patients. Eradication of infection is less likely when any component of the infected apparatus is left in place or if the shunt is replaced immediately before the CSF has sterilized. Complications from the infection such as relapse once antibiotics are discontinued are less likely if the entire shunt apparatus is removed early in the course of treatment. If an abdominal pseudocyst is present, drainage is usually not required. Removal of the shunt and antibiotic therapy is sufficient in most cases to result in disappearance of the fluid collection. Exploratory laparotomy is usually only necessary when an acute abdominal infection is present.
Initial empiric antimicrobial therapy consists of an antistaphylococcal antibiotic, such as nafcillin or vancomycin. The initial Gram-stain smear of the CSF should be used to determine if additional antibiotics are necessary. A third-generation cephalosporin and an aminoglycoside are added if abdominal symptoms are present or the Gram-stain smear of the CSF shows Gram-negative bacilli. Once the antimicrobial susceptibilities of the organism are known, empiric therapy is adjusted to target the causative organism(s). Vancomycin is often indicated for the treatment of coagulase-negative staphylococci infections because of high rates of resistance to multiple antibiotics. Rifampin has been used for treatment of S. aureus or coagulase-negative staphylococcal infection that persists following shunt removal and optimal antibiotic therapy.
The use of intrathecal antibiotics in the treatment of CSF shunt infections is controversial. Many antibiotics, including vancomycin, have poor CSF penetration when given parenterally, even in the presence of inflamed meninges. Therefore, direct installation of the antibiotic into the meninges has been used to improve
delivery. However, the pharmacokinetics of intraventricular antibiotics are not well studied, and the potential exists for neurotoxicity and chemical ventriculitis. Use of intraventricular antibiotics should be reserved for infections that have failed standard treatment.
Frequent CSF cultures are used to monitor response to therapy. If the shunt was left in place and cultures remain positive after several days, removal of the shunt with EVD placement should be considered. The duration of antibiotic therapy is usually 10 to 14 days following sterilization. Several factors, such as the infecting organism, time to sterilization and presence of distal complications, should be considered when determining length of therapy. A new shunt can be placed once the CSF is sterile, but there are no data as to the optimal timing of shunt replacement.
Was this article helpful?