Match the culture failure to the test
- Culture will be too slow and the patient is unstable: use a multiplex syndromic panel to get a result in hours
- Specimen was collected after antibiotics were started: use direct-specimen PCR, which does not require viable organisms
- Organism does not grow on available media or grows too slowly: use broad-range sequencing such as 16S ribosomal RNA gene sequencing
- Syndrome is well defined and the likely pathogens are known: use a targeted single-organism PCR, which is faster and cheaper than a broad panel
Why sequencing is the fallback, not the default
16S ribosomal RNA gene sequencing amplifies a region of the bacterial ribosome that is conserved enough to be present in nearly all bacteria and variable enough within that region to distinguish species. Because it does not require a prior guess about the organism, it can identify something no panel covers. The cost is time, expense, and a result that still does not tell you whether the organism is alive or susceptible.
Two patients, two different choices
A patient with suspected meningitis has already received a dose of ceftriaxone before the lumbar puncture. The cerebrospinal fluid Gram stain is negative and culture is likely to be negative as well, because the antibiotic has already acted. A cerebrospinal fluid multiplex panel is the right test here: it detects nucleic acid from organisms that are no longer viable, and it returns within hours. A second patient has a chronic, slowly progressive infection and no organism has grown after weeks of culture on standard media. A broad-range 16S ribosomal RNA gene sequencing approach is appropriate, because the organism may be one that no routine medium supports and no panel targets.