Using Prior Culture Results to Improve Initial Empiric Antibiotic Prescribing: An Evaluation of a Simple Clinical Heuristic

Machine-learning-based risk assessment tool to rule out empirical use of ESBL-targeted therapy in endemic areas

BACKGROUND

Antimicrobial stewardship focuses on identifying patients who require extended-spectrum beta-lactamase (ESBL)-targeted therapy. 'Rule-in' tools have been researched extensively in areas of low endemicity; however, such tools are inadequate for areas with high prevalence of ESBL-producing pathogens, as almost all patients will be selected.

AIM

To develop a machine-learning-based 'rule-out' tool suitable for areas with high levels of resistance.

METHODS

Gradient-boosted decision trees were used to train and validate a risk prediction model on data from 17,913 (45% ESBL) patients with Escherichia coli and Klebsiella pneumoniae in urine cultures. The predictive power of different sets of variables was evaluated using Shapley values to evaluate the contributions of variables.

FINDINGS

The model successfully identified patients with low risk of ESBL resistance in ESBL-endemic areas (area under receiver operating characteristic curve 0.72). When used to select the 30% of patients with the lowest predicted risk, the model yielded a negative predictive value ≥0.74. A simplified model with seven input features was found to perform nearly as well as the full model. This simplified model is freely accessible as a web application.

CONCLUSIONS

This study found that a risk calculator for antibiotic resistance can be a viable 'rule-out' strategy to reduce the use of ESBL-targeted therapy in ESBL-endemic areas. The robust performance of a version of the model with limited features makes the clinical use of such a tool feasible. This tool provides an important alternative in an era with growing rates of ESBL-producing pathogens, where some experts have called for empirical use of carbapenems as first-line therapy for all patients in areas with high prevalence of ESBL-producing pathogens.

Clinical recommendations for the inpatient management of lower respiratory tract infections in children and adolescents with severe neurological impairment in Germany

Children and adolescents with severe neurological impairment (SNI) require specialized care due to their complex medical needs. In particular, these patients are often affected by severe and recurrent lower respiratory tract infections (LRTIs). These infections, including viral and bacterial etiology, pose a significant risk to these patients, often resulting in respiratory insufficiency and long-term impairments. Using expert consensus, we developed clinical recommendations on the management of LRTIs in children and adolescents with SNI. These recommendations emphasize comprehensive multidisciplinary care and antibiotic stewardship. Initial treatment should involve symptomatic care, including hydration, antipyretics, oxygen therapy, and respiratory support. In bacterial LRTIs, antibiotic therapy is initiated based on the severity of the infection, with aminopenicillin plus a beta-lactamase inhibitor recommended for community-acquired LRTIs and piperacillin-tazobactam for patients with chronic lung disease or tracheostomy. Ongoing management includes regular evaluations, adjustments to antibiotic therapy based on pathogen identification, and optimization of supportive care. Implementation of these recommendations aims to improve the diagnosis and treatment of LRTIs in children and adolescents with SNI. What is Known: • Children and adolescents with severe neurological impairment are particularly affected by severe and recurrent lower respiratory tract infections (LRTIs). • The indication and choice of antibiotic therapy for bacterial LRTI is often difficult because there are no evidence-based treatment recommendations for this heterogeneous but vulnerable patient population; the frequent overuse of broad-spectrum or reserve antibiotics in this patient population increases selection pressure for multidrug-resistant pathogens. What is New: • The proposed recommendations provide a crucial framework for focused diagnostics and treatment of LRTIs in children and adolescents with severe neurological impairment. • Along with recommendations for comprehensive and multidisciplinary therapy and antibiotic stewardship, ethical and palliative care aspects are taken into account.

Can Decision Support Tools Improve Empiric Antibiotic Prescribing?

Decision Support Tools for Antibiotic PrescribingChoosing the right antibiotic is challenging. Unnecessarily broad-spectrum antibiotic treatment promotes antimicrobial resistance; inappropriately narrow-spectrum antibiotic use can lead to treatment failure. A cluster-randomized trial of a model-informed clinical decision support tool is proposed for guiding empiric antibiotic therapy for hospitalized patients with suspected infection.

A sepsis treatment algorithm to improve early antibiotic de-escalation while maintaining adequacy of coverage (Early-IDEAS): A prospective observational study

BACKGROUND

Empiric antibiotic treatment selection should provide adequate coverage for potential pathogens while minimizing unnecessary broad-spectrum antibiotic use. We sought to pilot a sepsis treatment algorithm to individualize antibiotic recommendations, and thereby improve early antibiotic de-escalation while maintaining adequacy of coverage (Early-IDEAS).

METHODS

In this observational study, the Early-IDEAS decision support algorithm was derived from previous Gram- negative and Gram-positive prediction rules and models along with local guidelines, and then applied to prospectively identified consecutive adults within 24 hours of suspected sepsis. The primary outcome was the proportion of patients for whom de-escalation of the primary antibiotic regimen was recommended by the algorithm. Secondary outcomes included: (1) proportion of patients for whom escalation was recommended; (2) number of recommended de-escalation steps along a pre-specified antibiotic cascade; and (3) adequacy of therapy in patients with culture-confirmed infection.

RESULTS

We screened 578 patients, of whom 107 eligible patients were included. The Early-IDEAS treatment recommendation was informed by Gram-negative models in 76 (71%) patients, Gram-positive rules in 64 (59.8%), and local guidelines in 27 (25.2%). Antibiotic de-escalation was recommended in almost half of all patients (n = 52, 48.6%), with a median of 2 steps down the a priori antibiotic treatment cascade. No treatment change was recommended in 45 patients (42.1%), and escalation was recommended in 10 (9.3%). Among the 17 patients with positive blood cultures, both the clinician prescribed regimen and the algorithm recommendation provided adequate coverage for the isolated pathogen in 12 patients (70.6%), (p = 1). Among the 25 patients with positive relevant, non-blood cultures, both the clinician prescribed regimen and the algorithm recommendation provided adequate coverage in 20 (80%), (p = 1).

CONCLUSION

An individualized decision support algorithm in early sepsis could lead to substantial antibiotic de-escalation without compromising adequate antibiotic coverage.

Antibiotic Prescriptions in Critically Ill Patients with Bloodstream Infection Due to ESBL-Producing Enterobacteriaceae: Compliance with the French Guidelines for the Treatment of Infections with Third-Generation Cephalosporin-Resistant Enterobacteriaceae-A Multicentric Retrospective Cohort Study

National and international guidelines were recently published regarding the treatment of Enterobacteriaceae resistant to third-generation cephalosporins infections. We aimed to assess the implementation of the French guidelines in critically ill patients suffering from extended-spectrum β-lactamase-producing Enterobacteriaceae bloodstream infection (ESBL-E BSI). We conducted a retrospective observational cohort study in the ICU of three French hospitals. Patients treated between 2018 and 2022 for ESBL-E BSI were included. The primary assessment criterion was the proportion of adequate empirical carbapenem prescriptions, defined as prescriptions consistent with the French guidelines. Among the 185 included patients, 175 received an empirical anti-biotherapy within 24 h of ESBL-E BSI onset, with a carbapenem for 100 of them. The proportion of carbapenem prescriptions consistent with the guidelines was 81%. Inconsistent prescriptions were due to a lack of prescriptions of a carbapenem, while it was recommended in 25% of cases. The only factor independently associated with adequate empirical carbapenem prescription was ESBL-E colonization (OR: 107.921 [9.303-1251.910], p = 0.0002). The initial empirical anti-biotherapy was found to be appropriate in 83/98 patients (85%) receiving anti-biotherapy in line with the guidelines and in 56/77 (73%) patients receiving inadequate anti-biotherapy (p = 0.06). Our results illustrate the willingness of intensivists to spare carbapenems. Promoting implementation of the guidelines could improve the proportion of initial appropriate anti-biotherapy in critically ill patients with ESBL-E BSI.

Antimicrobial Resistance in Ventilator-Associated Pneumonia: Predictive Microbiology and Evidence-Based Therapy

Ventilator-associated pneumonia (VAP) is a serious intensive care unit (ICU)-related infection in mechanically ventilated patients that is frequent, as more than half of antibiotics prescriptions in ICU are due to VAP. Various risk factors and diagnostic criteria for VAP have been referred to in different settings. The estimated attributable mortality of VAP can go up to 50%, which is higher in cases of antimicrobial-resistant VAP. When the diagnosis of pneumonia in a mechanically ventilated patient is made, initiation of effective antimicrobial therapy must be prompt. Microbiological diagnosis of VAP is required to optimize timely therapy since effective early treatment is fundamental for better outcomes, with controversy continuing regarding optimal sampling and testing. Understanding the role of antimicrobial resistance in the context of VAP is crucial in the era of continuously evolving antimicrobial-resistant clones that represent an urgent threat to global health. This review is focused on the risk factors for antimicrobial resistance in adult VAP and its novel microbiological tools. It aims to summarize the current evidence-based knowledge about the mechanisms of resistance in VAP caused by multidrug-resistant bacteria in clinical settings with focus on Gram-negative pathogens. It highlights the evidence-based antimicrobial management and prevention of drug-resistant VAP. It also addresses emerging concepts related to predictive microbiology in VAP and sheds lights on VAP in the context of coronavirus disease 2019 (COVID-19).

Using Machine Learning to Predict Antimicrobial Resistance-A Literature Review

Machine learning (ML) algorithms are increasingly applied in medical research and in healthcare, gradually improving clinical practice. Among various applications of these novel methods, their usage in the combat against antimicrobial resistance (AMR) is one of the most crucial areas of interest, as increasing resistance to antibiotics and management of difficult-to-treat multidrug-resistant infections are significant challenges for most countries worldwide, with life-threatening consequences. As antibiotic efficacy and treatment options decrease, the need for implementation of multimodal antibiotic stewardship programs is of utmost importance in order to restrict antibiotic misuse and prevent further aggravation of the AMR problem. Both supervised and unsupervised machine learning tools have been successfully used to predict early antibiotic resistance, and thus support clinicians in selecting appropriate therapy. In this paper, we reviewed the existing literature on machine learning and artificial intelligence (AI) in general in conjunction with antimicrobial resistance prediction. This is a narrative review, where we discuss the applications of ML methods in the field of AMR and their value as a complementary tool in the antibiotic stewardship practice, mainly from the clinician's point of view.

Analysis of infections among patients with historical culture positive for extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli or Klebsiella pneumoniae: Is ESBL-targeted therapy always needed?

Objective

Among patients with a history of ESBL infection, uncertainty remains regarding whether all of these patients require ESBL-targeted therapy when presenting with a subsequent infection. We sought to determine the risks associated with a subsequent ESBL infection to help inform empiric antibiotic decisions.

Methods

A retrospective cohort study of adult patients with positive index culture for Escherichia coli or Klebsiella pneumoniae (EC/KP) receiving medical care during 2017 was conducted. Risk assessments were performed to identify factors associated with subsequent infection caused by ESBL-producing EC/KP.

Results

In total, 200 patients were included in the cohort, 100 with ESBL-producing EC/KP and 100 with ESBL-negative EC/KP. Of 100 patients (50%) who developed a subsequent infection, 22 infections were ESBL-producing EC/KP, 43 were other bacteria, and 35 had no or negative cultures. Subsequent infection caused by ESBL-producing EC/KP only occurred when the index culture was also ESBL-producing (22 vs 0). Among those with ESBL-producing index culture, the incidences of subsequent infection caused by ESBL-producing EC/KP versus other bacterial subsequent infection were similar (22 vs 18; P = .428). Factors associated with subsequent infection caused by ESBL-producing EC/KP include history of ESBL-producing index culture, time ≤180 days between index culture and subsequent infection, male sex, and Charlson comorbidity index score >3.

Conclusions

History of ESBL-producing EC/KP culture is associated with subsequent infection caused by ESBL-producing EC/KP, particularly within 180 days after the historical culture. Among patients presenting with infection and a history of ESBL-producing EC/KP, other factors should be considered in making empiric antibiotic decisions, and ESBL-targeted therapy may not always be warranted.

Empirical antibiotic therapy for difficult-to-treat Gram-negative infections: when, how, and how long?

PURPOSE OF REVIEW

To discuss empirical therapy for severe infections due to Gram-negative bacteria with difficult-to-treat resistance (GNB-DTR) in current clinical practice, focusing in particular on the positioning of novel therapeutic agents and rapid diagnostic tests.

RECENT FINDINGS

The current era of novel agents active against GNB-DTR and showing differential activity against specific determinants of resistance is an unprecedented scenario, in which the clinical reasoning leading to the choice of the empirical therapy for treating severe GNB-DTR infections is becoming more complex, but it also allows for enhanced treatment precision.

SUMMARY

Novel agents should be used in line with antimicrobial stewardship principles, aimed at reducing selective pressure for antimicrobial resistance. However, this does not mean that they should not be used. Indeed, excesses in restrictive uses may be unethical by precluding access to the most effective and less toxic treatments for patients with severe GNB-DTR infections. Given these premises (the 'how'), empirical treatment with novel agents should be considered in all patients with risk factors for GNB-DTR and severe clinical presentation of acute infection (the 'when'). Furthermore, empirical novel agents should preferably be continued only for a few hours, until de-escalation, modification, or confirmation (as targeted therapy) is made possible by the results of rapid diagnostic tests (the 'how long').

Initial antimicrobial management of sepsis

Sepsis is a common consequence of infection, associated with a mortality rate > 25%. Although community-acquired sepsis is more common, hospital-acquired infection is more lethal. The most common site of infection is the lung, followed by abdominal infection, catheter-associated blood steam infection and urinary tract infection. Gram-negative sepsis is more common than gram-positive infection, but sepsis can also be due to fungal and viral pathogens. To reduce mortality, it is necessary to give immediate, empiric, broad-spectrum therapy to those with severe sepsis and/or shock, but this approach can drive antimicrobial overuse and resistance and should be accompanied by a commitment to de-escalation and antimicrobial stewardship. Biomarkers such a procalcitonin can provide decision support for antibiotic use, and may identify patients with a low likelihood of infection, and in some settings, can guide duration of antibiotic therapy. Sepsis can involve drug-resistant pathogens, and this often necessitates consideration of newer antimicrobial agents.

Validation of a Community-Acquired Pneumonia Score To Improve Empiric Antibiotic Selection at an Academic Medical Center

The 2019 American Thoracic Society and the Infectious Diseases Society of America community-acquired pneumonia (CAP) guidelines recommend that drug-resistant pathogens (DRP) be empirically covered if locally validated risk factors are present. This retrospective case-control validation study evaluated the performance of the drug resistance in pneumonia (DRIP) clinical prediction score. Two hundred seventeen adult patients with ICD-10 (https://www.who.int/classifications/classification-of-diseases) pneumonia diagnosis, positive confirmed microbiologic data, and clinical signs and symptoms were included. A DRIP score of ≥4 was used to assess model performance. Logistic regression was used to select for significant predictors and create a modified DRIP score, which was evaluated to define clinical application. The DRIP score predicted pneumonia due to a DRP with a sensitivity of 67% and specificity of 73%. The area under the receiver operating characteristic (AUROC) curve was 0.76 (95% confidence interval [CI], 0.69 to 0.82). From regression analysis, prior infection with a DRP and antibiotics in the last 60 days, yielding scores of 2 points and 1 point, respectively, remained local risk factors in predicting drug-resistant pneumonia. Sensitivity (47%) and specificity (94%) were maximized at a threshold of ≥2 in the modified DRIP model. Therefore, prior infection with a DRP remained the only clinically relevant predictor for drug-resistant pneumonia. The original DRIP score demonstrated a decreased performance in our patient population and behaved similarly to other clinical prediction models. Empiric CAP therapy without anti-methicillin-resistant Staphylococcus aureus and antipseudomonal coverage should be considered for noncritically ill patients without a drug resistant pathogen infection in the past year. Our data support the necessity of local validation to authenticate clinical risk predictors for drug-resistant pneumonia.