MANITOBA AMR ALLIANCE

Correct interpretation of the information presented in hospital antibiograms requires an understanding of their limitations, as follows:

• Hospital antibiograms are institution-specific.  It cannot be assumed that antimicrobial susceptibility rates at one healthcare facility will be identical to those at another. 

• Susceptibility rates presented in hospital antibiograms for common bacterial pathogens should not be extrapolated to isolates obtained from specimens collected in the community (e.g., family physician’s office, outpatient microbiology laboratory). Community isolates are generally excluded from hospital antibiograms.

• It is important to note that when antibiogram data are compiled, only the first isolate of a particular species from a patient is included over the defined time period. Hospital antibiogram susceptibility rates are less helpful in guiding the selection of empiric therapy for patients who have received prior treatment with multiple antibiotic courses during a prolonged hospital stay. 

• Hospital antibiograms do not contain data for all clinically relevant bacterial pathogens.   For a bacterial pathogen to be included, there needs to be a minimum of 30 isolates tested on an annual basis.   

• Antimicrobial resistance patterns vary substantially by age, recent antibiotic treatment history, and underlying medical conditions for a given patient. Always consider individual patient risk factors for antimicrobial resistance when interpreting antibiogram data. A review of previous microbiology results (when available) can be very helpful.

• Data from antibiograms may not be generalizable to specific populations.  As an example, infections occurring in otherwise healthy patients in the community are likely to have less antimicrobial resistance, on average, than infections in hospitalized patients whose data contributed to a hospital antibiogram.

• Susceptibility to a combination of antimicrobials (e.g., predicting that if a combination is used, at least one antimicrobial will be active) cannot be easily extracted from antibiogram data. 

• Beyond antimicrobial resistance, when selecting an empiric antimicrobial, clinicians should also always consider the source of the infection, the most likely pathogen(s) involved, whether the infection is likely to be polymicrobial or monomicrobial, the severity of the infection, additional drug factors (e.g., clinical efficacy, side effect profile, pharmacokinetics, cost), additional patient factors (e.g., age, allergies, medication history, comorbid conditions, social factors), and relevant consensus treatment guidelines.