Strategies for antibiotic selection in empirical therapy

The PEST (Pathology, Epidemiology, Severity, Treatment) approach to optimizing antimicrobial therapy

BACKGROUND

Selecting an empiric antimicrobial regimen can be difficult for early learners and misuse of antibiotics can lead to adverse events and antimicrobial resistance. There have been few interventions that have focused on improving antibiotic decision making, as a form of therapeutic reasoning, for post-graduate trainees. We describe here an approach to aid internal medicine interns in their therapeutic reasoning, particularly when it comes to diagnosing and empirically treating infections.

METHODS

The PEST (pathology, epidemiology, severity, treatment) model was created as a four-step approach to therapeutic reasoning and choosing an appropriate antimicrobial regimen for a given infectious disease syndrome. In February 2020, we conducted two independent teaching sessions for interns on the PEST approach. We assessed pre-and post-teaching responses to five clinical vignette-based questions. Results were presented as a percentage of interns who chose an appropriate antibiotic and provided sufficient therapeutic reasoning as defined by using at least three out of the four PEST criteria. Statistical analysis occurred via Fischer's exact test to determine the level of statistical significance between responses.

RESULTS

Twenty-seven interns participated in the activity. At baseline, several interns had incorporated aspects of the PEST approach in their pre-teaching responses. Ten interns commented on the usefulness of such a systematic approach. While there was no statistically significant difference in antibiotic selection, the teaching session demonstrated a trend towards significance in improving therapeutic reasoning as defined by the PEST strategy.

CONCLUSION

Our results suggested an improvement in using a structured cognitive tool such as the PEST approach to reinforce therapeutic reasoning, but the method did little to improve antibiotic selection. Some interns used select "PEST" concepts prior to the intervention suggesting that the PEST approach may enhance prior knowledge or clinical reasoning skills. Continued incorporation of the PEST approach using a case-based framework may solidify conceptual and practical knowledge of antimicrobial selection. Further studies are needed to assess the impact of such teaching interventions.

What to Do with the New Antibiotics?

Multidrug-resistant Gram-negative bacteria-related infections have become a real public health problem and have exposed the risk of a therapeutic impasse. In recent years, many new antibiotics have been introduced to enrich the therapeutic armamentarium. Among these new molecules, some are mainly of interest for the treatment of the multidrug-resistant infections associated with Pseudomonas aeruginosa (ceftolozane/tazobactam and imipenem/relebactam); others are for carbapenem-resistant infections associated with Enterobacterales (ceftazidime/avibactam, meropenem/vaborbactam); and finally, there are others that are effective on the majority of multidrug-resistant Gram-negative bacilli (cefiderocol). Most international guidelines recommend these new antibiotics in the treatment of microbiologically documented infections. However, given the significant morbidity and mortality of these infections, particularly in the case of inadequate therapy, it is important to consider the place of these antibiotics in probabilistic treatment. Knowledge of the risk factors for multidrug-resistant Gram-negative bacilli (local ecology, prior colonization, failure of prior antibiotic therapy, and source of infection) seems necessary in order to optimize antibiotic prescriptions. In this review, we will assess these different antibiotics according to the epidemiological data.

Antibiotic de-escalation therapy in patients with community-acquired nonbacteremic pneumococcal pneumonia

Background De-escalation therapy is recommended as an effective antibiotic treatment strategy for several infectious diseases. While there is limited evidence supporting its clinical and cost-effective outcomes in patients with community-acquired bacteremic pneumonia, there is no evidence in patients with nonbacteremic pneumonia. Objective This study aimed to evaluate the antibiotic costs in patients who did and did not receive de-escalation therapy, based on the 2017 Japanese guidelines for the management of community-acquired nonbacteremic pneumococcal pneumonia of the Japanese Respiratory Society (JRS). Setting Kobe university hospital, Japan. Methods A retrospective case series review including antibiotic use and length of hospital stay was conducted using the medical records from April 2008 to May 2019 at a university hospital in Japan. Main outcome measure Impact of antibiotic de-escalation therapy on the antibiotic costs. Results Among 55 patients who were eligible, the treating physicians de-escalated antibiotics in 28 (51%). The differences in the median length of hospital stay and the incidence of adverse drug reactions between the two groups were not statistically significant (p = 0.67 and 1.0, respectively). However, the median total antibiotic cost per infected patient in the de-escalated group was significantly lower than that in the non-de-escalated group [$269.8 ($195-$389) vs. $420.5 ($221-$799), p = 0.048]. Conclusion Antibiotic de-escalation based on the 2017 JRS guidelines leads to a reduction in total antibiotic costs for the management of community-acquired nonbacteremic pneumococcal pneumonia.

Spectrum and patterns of antimicrobial resistance of uropathogens isolated from a sample of hospitalised Lebanese patients with urinary tract infections

A retrospective study was carried out to assess the current distribution and antimicrobial susceptibility patterns of uropathogens isolated from patients with community-acquired (CA) or hospital-acquired (HA) urinary tract infections (UTIs) who were admitted to the largest governmental hospital in Beirut City (Lebanon) in 2009-2012. Overall, 292 uropathogens were isolated. Enterobacteriaceae constituted 81% of the isolated uropathogens, with Escherichia coli, Klebsiella pneumoniae and Proteus mirabilis being the most reported species. Moreover, 37% of the isolated uropathogens produced extended-spectrum β-lactamases (ESBLs). ESBL-producing strains were more prevalent in CA-UTIs than in HA-UTIs. In the CA-UTI group, ESBL-producing strains were more prevalent in older patients (age >60 years), male patients and patients with diabetes mellitus, urinary catheterisation and benign prostatic hyperplasia. Uropathogens isolated from CA-UTI and HA-UTI had similar resistance patterns. The highest resistance rates were found for ampicillin, amoxicillin/clavulanic acid, trimethoprim/sulfamethoxazole and ciprofloxacin. Imipenem, nitrofurantoin and piperacillin/tazobactam showed good overall activity against the isolated uropathogens. Of the isolated non-ESBL-producing E. coli, 29% were found to be multidrug-resistant. These data suggest that there is an urgent need to implement a national policy regulating the use of antibiotics, notably in the community, in order to limit the widespread increase in antimicrobial resistance.

The study of prevalence and antimicrobial susceptibility of tracheal bacterial strains isolated from pediatric patients

The aim of this study was to investigate the prevalence and antimicrobial susceptibility of bacterial strains isolated from tracheal specimens obtained from pediatric patients admitted to a major children hospital in Tehran, in 2007. Tracheal specimens were cultured on the appropriate bacteriological media. Bacterial isolates were identified by standard biochemical and serological tests. Antimicrobial susceptibility testing was performed according to Clinical and Laboratory Standards Institute (CLSI) guidelines. Pseudomonas spp. was identified as the most prevalent bacterial isolate (32%) followed by Staphylococcus aureus (27.6%). Thirty strains (16%) were identified as Klebsiella spp., 18 (9.6%) as Enterobacter spp. and the rest belonged to coagolase negative Staphylococci, Streptococcus viridans, Acinetobacter spp., Escherichia coli and Neisseria spp. All Pseudomonas spp. were resistant to ampicllin, kanamycin and ceftizoxime. Staphylococcus and Klebsiella spp. showed high degree of resistance to 40% of examined antibiotics.