The installation of a Peritoneal Ventricular Shunt device (VPS) is one of the most frequent procedures in neurosurgery [1]. Dysfunction and infection of these devices are common complications, which add morbidity and mortality and increase treatment costs. The incidence is estimated to be between 10–20%, which varies according to the case definition, studied period, etc [2, 3]. In our center the incidence is 7.8% in the first installation and 23% in the following ones [4].
Ventricular infection should be suspected in a patient with VPS who presents with fever, usually associated with symptoms of valvular dysfunction, such as headache, nausea or compromised consciousness [5]. In addition, they may present with focal neurological deficits, nuchal rigidity, seizures and photophobia. In a patient with a peritoneal shunt, abdominal signs should raise suspicion of device infection. Demonstration of bacteremia without other cause, in a patient with Ventricular Atrial Shunt, raises suspicion of infection [6].
Routine CSF cytochemical tests are moderately useful, since some alterations may be secondary to infection or because of the shunt (elevated protein, hemorrhage). In turn, a CSF without alterations does not completely rule out a ventricular infection [7, 8].
Lactate has emerged as a marker of inflammation in CSF, which allows distinguishing an infection of the central nervous system by bacteria versus a viral infection or inflammation by another cause (aseptic meningitis) [9, 10]. In infections, lactate increases due to inflammation and cerebral ischemia, which increases anaerobic metabolism [11]. Lactate can also be used for monitoring therapy [12].
For neurosurgical patients, this relationship is less clear. In adult patients, lactate has better positive predictive value than hypoglycorrhachia or pleocytosis for the diagnosis of post neurosurgical meningitis [13, 14]. A cut-off point close to 3.45 mmol/L lactate has a sensitivity close to 90% and a specificity over 85%[15], and is not altered by the presence of blood in the CSF[16]. Recent studies in ventriculostomy users have shown low sensitivity and specificity [17], but with a good negative predictive value [18] (if it is low, bacterial infection can be ruled out).
In pediatrics the current evidence is less. In external ventricular drainage, one study found no significant difference (although the mean lactate in infection was higher)[19].
Meanwhile, another study found a higher mean lactate level in patients with suspected infection of the peritoneal ventricular drainage, positive culture and a negative predictive value of 96% with a cut-off point of 2.95 mmol/L[20]. There is a good correlation between lactate taken from a lumbar puncture to the sample obtained directly from the ventricle [21].
Currently the evidence of the usefulness of lactate as a marker of infection in children with ventriculoperitoneal shunts is scarce. Our objective is to provide evidence on the usefulness of lactate as a marker of ventricular infection secondary to VPS.