Abstract
Empirical antibiotic selection often fails to be optimal targeted in the era of increasingly common resistant organisms. We prospectively evaluated the usefulness or rapid AST for optimal antibiotic selection by infectious disease (ID) physicians in patients with bacteremia of Gram-positive organisms. QMAC-dRAST results led to optimal antibiotic treatment in 33 (89.2%) of the 37 cases receiving non-optimal targeted antibiotics. Optimal targeted treatments based on QMAC-dRAST results were possible in 133 (97.1%) of the 137 cases. In conclusion, the introduction of rapid phenotypic AST can help increase the selection of optimal targeted antibiotics during the early period of bacteremia.
Introduction
Sepsis can cause morbidity, mortality, prolonged hospitalization and high costs for healthcare systems and the most important life-saving action is the prompt administration of an appropriate antibiotic agent.1-5 Clinicians need the definitive identification of the pathogen and antimicrobial susceptibility testing (AST) to adjust antibiotics more precisely and effectively. Since the conventional method for identification and AST takes at least 2 – 3 days, they start an empirical treatment at the beginning of blood stream infection (BSI) diagnosis and adjust antibiotics based on Gram stain results which is still empirical.6-7 However, those empirical treatments often fail to guide for selecting optimal targeted antibiotics in the era of increasingly common resistant organisms.
To provide ID and AST results at the early stage of BSI, many rapid diagnostic tests have been developed directly using positive blood cultures. MALDI-TOF MS has been established for rapid pathogen identification and automated molecular diagnostic tests can identify both pathogens and resistance genes.8-11 Moreover, several rapid phenotypic AST systems have been developed.12-14 Especially, dRAST is a phenotypic AST system for Gram-positive cocci as well as Gram-negative rods utilizing microfluidic technology for bacteria immobilization and microscopic image analysis of bacteria growth.14
In a recent study, we evaluated how well MALDI-TOF MS and dRAST guide ID physicians in the selection of optimal targeted antibiotics for patients with positive blood cultures.15 Here, more specifically, we showed the results for blood stream infection with Gram-positive organisms.
Methods
The current analysis was performed using the data presented in the previous clinical study.15 The data were separated into Gram-positive and Gram-negative cases and then the antibiotic treatments decided by ID physicians based on each BSI diagnosis result were categorized as follows: optimal targeted, unnecessary broad spectrum, suboptimal and ineffective as previously defined.15 To compare with more early stage empirical treatment, we added empirical antibiotics treated at the time of blood collection for blood cultures in this analysis.
Impact of dRAST on ID physician antibiotic selection
In initial empirical antibiotic selection, the ID physicians chose optimal targeted antibiotics in 26 (19.0%) cases, unnecessary broad-spectrum antibiotics in 27 (19.7%) cases and suboptimal or ineffective antibiotics 84 (61.3%) cases (Table 1). After reviewing the Gram stain results, the ID physicians chose optimal targeted antibiotics in 83 (60.6%) cases, unnecessary broad-spectrum antibiotics in 20 (14.6%) cases and suboptimal or ineffective antibiotics in 34 (24.8%) case. After reviewing MALDI-TOF MS results, the ID physicians chose optimal targeted antibiotics in 100 (73.0%) cases, unnecessary broad-spectrum antibiotics in 11 (8.0%) cases and ineffective antibiotics in 26 (19.0%) cases. The MALDI-TOF MS results enabled appropriate antibiotic selection in 119 (86.9%) cases, but the proportion of ineffective antibiotic selection was high for resistant strains (23.1%) and the proportion of unnecessary broad-spectrum antibiotic selection was high for susceptible strains (18.6%).
After reviewing the dRAST results, the ID physicians chose optimal antibiotic treatment for 33 (89.2%) of the 37 patients who were recommended non-optimal antibiotics based on the MALDI-TOF MS results. Specifically, there was no ineffective antibiotic selection for resistant strains and unnecessary broad-spectrum antibiotic selection for susceptible strains was reduced to 2 (3.4%) cases. Overall, optimal targeted treatments based on the dRAST results were possible in 133 (97.1%) of the 137 cases and the proportion of it was not significantly different between susceptible strains and resistant strains.
With the steady increase in resistant strains such as methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus species (VRE), an empirical antibiotic selection is highly possible to be ineffective. And as shown in our data, the rapid pathogen identification alone didn’t play well a role of guiding the empirical antibiotic treatment for resistant strains. In this regards, rapid phenotypic AST should be introduced and used for antibiotic decision-making.
Rapid molecular diagnostic methods enable the rapid detection of antibiotic resistance genes of Gram-positive organisms and may greatly assist with antibiotic selection.10,11,16,17 However, because they mostly detect mecA/C and vanA/B only and can’t tell the susceptibility of the other antibiotics, it is hard to decide de-escalation or escalation of antibiotics properly. In contrast, rapid phenotypic AST gives susceptibility results of various antibiotics and this provides better clinical usefulness.
Conclusion
Our present data show that not only empirical antibiotic selection but also antibiotic selection based on rapid pathogen identification may frequently be inadequate or unnecessary broad-spectrum for Gram-positive organisms. With the increasing number of resistant organisms in clinical practice, the introduction of rapid phenotypic AST can help increase the selection of optimal targeted antibiotics during the early period of bacteremia.
ID physician antibiotic selection based on the results of Gram staining, MALDI-TOF MS and dRAST methods. Resistant strains include MRSA, VRE and MDR strains. Optimal targeted treatment: the most effective and narrowest spectrum antibiotic treatment for the pathogen(s). Unnecessary broad-spectrum treatment: antibiotic treatment that is effective against the pathogen(s), but has broad-spectrum activity requiring de-escalation. Suboptimal treatment: antibiotic treatment to which the pathogen(s) are susceptible, but that has inferior antimicrobial activity compared with that of the optimal targeted treatment. Ineffective treatment: antibiotic treatment to which the pathogen(s) have intermediate susceptibility or are resistant. Appropriate antibiotic treatment: antibiotic treatment to which the pathogen(s) are susceptible.
Data Availability
All data produced in the present study are available upon reasonable request to the authors