Given the preeminence of the MIC in modern PK/PD concepts, it is obviously important, for the outcome of critical infections, that the laboratory can accurately measure the MIC of causative organisms in real time, in order to guide appropriate therapy and dosing schedules. In the absence of a pathogen or on preliminary identification, an estimate of the MICs of likely pathogens, based on data from the previous year's isolates, can be used to inform decisions on therapy. For instance, if a quinolone is being used, will low dose suffice (e.g., for a fully susceptible Escherichia coli)? If possible pathogens include A. baumannii, which is likely to have borderline susceptible MICs, then high-dose quinolone will be more efficacious, perhaps in combination with an aminoglycoside or p-lactam.
The greatest challenge for the microbiology laboratory in the treatment of infection on the ICU is to make results available on a time scale that can influence treatment and there is reasonable hope that molecular tests will facilitate this. For bacterial infections, however, the only widely available molecular test, commonly relevant to the ICU, is PCR for the mecA gene of MRSA.44 No molecular test is generally available for rapid diagnosis of invasive fungal infection, and while PCR for viral infections is making great inroads, it is outside the scope of this review.
Nevertheless, conventional culture, microscopy, and antigen-based tests generate results that are generally deliverable to the clinician in a useful time frame.45 Sometimes the seemingly most simple problems are the most difficult to solve. Rapid delivery of specimens to the laboratory is crucial, with immediate processing either through an on-call or shift system operating in the laboratory. Then results of cell counts, Gram stains and antigen tests on CSF, urine, and other body fluids can be reported within the hour. Urine cultures (with direct susceptibility testing) can usually be read at 6 h for a clinically useful, if preliminary result. Unless received in the laboratory late the previous evening, most bronchoalveolar lavage samples (BAL), sputums, and wound cultures can be read the following morning (although anaerobic incubation takes longer) and susceptibilities predicted based on recent trends reported from the laboratory. Ninety percent of clinically significant blood cultures for bacteria will be positive within 24 h of receipt in the laboratory and a continuous monitoring system for positive results with telephoned results of Gram stain to the ICU should be standard practice.45,46 Provisional susceptibilities and identification, including coagulase and mecA status where appropriate, can usually be made available in a further 6 h.
The laboratory should provide susceptibility summaries of the previous year's data for the ICU, split into community acquired, hospital acquired, ICU acquired and possibly also by body site, as susceptibilities and organisms may differ markedly depending on source.3-5
These results can be used to inform antibiotic guidelines for the ICU if a 24-h on-call medical microbiologist or other infection specialist is not available to advise on empiric antibiotic therapy. Rationalization of therapy at each stage of receipt of further information from the laboratory is crucial to control escalating antibiotic use. In particular, the use of broad-spectrum agents for initial empiric therapy should be changed to narrow spectrum agents at the earliest opportunity. Daily ward rounds with an infection specialist in possession of the very latest laboratory results are crucial to this process.47,48
The microbiologist can also help with the early detection of outbreaks, by close observation of the routine data and sometimes by the use of surveillance cultures. It is important, however, not to place too much emphasis on surveillance cultures for the choice of empiric therapy as these can have poor reliability in predicting the cause of infection and lead to overuse of antibiotics.14, 49
It is common now to screen all admissions to the ICU for MRSA. Isolation precautions can then be taken to prevent spread and early decolonization of MRSA-positive patients attempted.50 Early assessment of MRSA carriage status can also inform empiric therapy should it be necessary.
Was this article helpful?