Laboratory-based data underestimate the frequency of several major resistant organisms in patients with hospital-acquired infection. Previous clinical validation of the individual susceptibility reports seems to be a suitable strategy to get more reliable data.

We assessed the impact of antibiotic administration prior to sample collection on the bacterial resistance rates from patients with nosocomial infection. Every individual susceptibility report was assessed in real time at the bedside of the patient by a team composed of infectious diseases and internal medicine specialists as well as clinical microbiologists for clinical significance and appropriateness of the specimen. The report also stated the kind, source and origin of the infection, history of administration of any antibiotic during the last month prior to sample collection. To evaluate the impact of previous antibiotic administration, resistance rates were calculated separately among the group of patients with and without history of antibiotic treatment. A crude univariate analysis was performed to assess the significance of the differences between groups for every species-antibiotic pair. Patients who had received ciprofloxacin showed significantly higher rates of Escherichia coli resistant to ciprofloxacin, broad-spectrum cephalosporins and gentamicin. A higher rate of methicillin-resistant Staphylococcus aureus was observed in patients who were given gentamicin. A stratified analysis showed that the previous antibiotic administration continued to be a risk factor for increased resistance rates regardless of the hospital ward or the source of the infection. This study demonstrates the influence of previous antibiotic administration on bacterial resistance rates although this fact is barely taken into account by the laboratory when constructing the cumulative susceptibility data. Real time clinical validation of the individual susceptibility reports, performed by a multidisciplinary team prior to the data entering, might be a suitable approach to get more reliable susceptibility rates to guide the rational selection of antimicrobial empirical therapy in patients with hospital-acquired infections who have been given antimicrobial treatment prior to specimen collection.

We recently published on the impact of a four-phase hospital-wide intervention program designed to optimize the quality of antibiotic use, where a multidisciplinary team (MDT) could modify prescription at the last phase. Because health care quality was changing during the last 5 years (late 1999 to early 2004), we developed certain indicators to monitor the quality of our intervention over time. Different periods were defined as baseline (pre-intervention), initial intervention-active control, pre-crisis control, crisis control, post-crisis control and end of crisis control. Major indicators were rates of prescription modification by the MDT; prescription for an uncertain infection and a novel index formula (RIcarb) to estimate the rationale for carbapenem use. We assessed 2115 antimicrobial prescriptions. Modification of prescription rate was 30% at the beginning and decreased thereafter up to stable levels. Rate of prescriptions ordered for cases of both uncertain infection and unknown source of infection decreased significantly after intervention (i.e. from baseline to active control). In contrast, a doubling of culture-directed prescriptions was observed between these periods. RIcarb values lower and higher than 60% (modal, cut-off) were assumed as carbapenem overuse and underuse, respectively. Overuse was observed at the pre-intervention, while pronounced underuse was shown during the crisis (RIcarb, 45% and 87%, respectively). The present study demonstrates that certain indicators, other than the widely adopted impact outcomes, are a suitable tool for monitoring the quality of a continuous, long-term, active intervention on antimicrobial prescribing practice, especially when applied in a changing healthcare setting.

In order to estimate the likelihood of success (SL) with the initial empiric antimicrobial therapy, the following formula was constructed with data subjected to prior clinical validation in real time: SL (%) = (Nº isolates susceptible to IEAT/Nº patients with MDI) × 100. Where the numerator of the formula represents the total number of isolates recovered from the assessed type of infection, that was susceptible to any component of empiric antimicrobial therapy (IEAT) used, and the denominator represents the total number of patients with the same assessed, but microbiologically documented infection (MDI). For male hospital-acquired urinary tract infection, only imipenem reached a suitable SL value (i.e. ≥80%). In patients with hospital-acquired peritonitis, imipenem and tigecycline-ceftazidime showed the highest coverage rates. For ventilator-associated pneumonia only imipenem yielded acceptable coverage as a single drug. Implementing the present formula instead of the regular global antibiograms used to guide the selection of the initial treatment may benefit the patient outcome and improve antimicrobial usage.