Clinical utility of broth cultures of cerebrospinal fluid from patients at risk for shunt infections.

For patients with cerebrospinal fluid (CSF) shunts, culture of the CSF remains the most valuable tool in the evaluation of suspected shunt infections. To detect anaerobic Propionibacterium sp., a well-described cause of these infections, many clinical microbiology laboratories routinely employ a broth medium as an adjunct to solid media. The use of broth, however, creates a diagnostic dilemma since many contaminants also are isolated from broth cultures. Therefore, we retrospectively reviewed the records of 59 patients with CSF shunts in whom an organism was isolated from only broth cultures to assess their utility for the diagnosis of shunt infection. We found that no single clinical or laboratory parameter, including fever, leukocytosis, pleocytosis, or CSF protein and glucose, could reliably predict or exclude a shunt infection. Isolation of coagulase-negative staphylococci only in broth, in the absence of growth on solid media in concurrent or immediately preceding cultures, virtually always represented contamination. The isolation of Propionibacterium sp. from broth only usually represented contamination; however, infection could not be excluded without a repeated CSF culture, even in the absence of pleocytosis. We recommend that specific comments be appended to laboratory reports for isolates from CSF in broth only as an aid to the physician in interpreting the clinical importance of such isolates.