Authors: Hardik H. Patel, Betcy Benny, Prerna Nahar, Alex Landis
Categories: Original Research, Antimicrobial Stewardship Program, Bloodstream Infection, COVID-19, Escherichia coli, Procalcitonin
Source: Innovations in Pharmacy
Purpose: There are many challenges that pharmacist led antimicrobial stewardship programs can encounter including lack of resources, costs, and inaccurate antimicrobial susceptibility testing (AST) results. The COVID-19 pandemic has led to increased resistance especially with gram negative infections. At a small single center community hospital, gram negative infections, particularly Escherichia coli infections, predominately occur. Therefore, this study aims to address gram negative bacteremia burden and its impact on antimicrobial stewardship efforts for combatting Escherichia coli and ESBL organisms with such barriers during the pandemic.
Methods: In a retrospective cohort study, patients ≥ 18 years of age from a small community hospital were reviewed. Patients were excluded if their blood cultures were not positive for Escherichia coli and if antibiotics were not initiated during hospital stay. The primary endpoint was to determine the length of hospital stay. Critical secondary endpoints include antibiotic de-escalations, duration of antibiotics, time to definitive antibiotic therapy, serum procalcitonin levels, blood culture availabilities, MIC breakpoints, co-infection of COVID-19, and Clostridioides difficile occurrences.
Results: Out of 74 patients with gram negative bacteremia, 41 patients specifically had Escherichia coli bacteremia. The primary endpoint results showed that patients with Escherichia coli bacteremia that stayed in the ICU had a length of stay of 13.6 days. Patients with Escherichia coli bacteremia in the Non-ICU setting has a length of stay of 7.3 days, and patients with ESBL bacteremia in the Non-ICU setting had a length of stay of 6.8 days.
Conclusions: Despite the various challenges that antimicrobial stewardship programs (ASP) face in a single center small community hospital, the ASP at this small community hospital utilizes various policies and tools to increase appropriate antibiotic use and decrease hospital length of stay in patients with Escherichia coli bacteremia.
Key Words: Antimicrobial Stewardship Program, COVID-19, Escherichia coli, Bloodstream Infection, Procalcitonin
Antimicrobial Stewardship Programs (ASP) are designed to prevent any unnecessary use of antibiotics and decrease antibiotic resistance. About 30% of all antibiotics prescribed in U.S acute care hospitals are either unnecessary or suboptimal. In 2014, Centers for Disease Control and Prevention (CDC) implemented Antibiotic Stewardship Program (Core Elements) in hospitals across the United States. The newest update to the Core Elements includes hospital leadership commitment, accountability, pharmacy expertise, action, tracking, reporting and education.^4^ The percentage of hospitals implementing ASP increased from 48% in 2015 to 91% in 2020.^5^ Antimicrobial Stewardship Programs became a necessity for healthcare system across the nation during year 2020-22 in addition to combating COVID-19. Such correlation between COVID-19 pandemic and its impact on rise in antimicrobial resistance required a closer look into local trends in community hospitals. Ultimately, the COVID-19 pandemic has led to the increase in antimicrobial use, difficulty following infection prevention actions, and an overall increase in healthcare-associated, antimicrobial-resistant infections. The pandemic has resulted in hospitals seeing sicker patients requiring a longer length of stay, consequently leading to more resistant infections. In particular, the rate of extended spectrum beta-lactamase (ESBL) producing enterobacterales in the hospital setting has increased by 32% from 2019-2020 alone.^1^ Gram negative bloodstream infections are a major cause of morbidity and mortality and Escherichia coli accounts for majority of gram negative bacteria hospital infections.^2^ The Infectious Disease Society of America (IDSA) addresses treatment of resistant gram negative infections, however we wanted to focus on the management of the most prevalent gram negative organism of Escherichia coli overarching guidelines that direct the management of gram negative bacteremia.^3^ Currently, in a small single center community hospital, gram negative infections are increasing along with antibiotic resistance. Therefore, it is pivotal to assess gram negative bacteremia burden and its impact on antimicrobial stewardship efforts for combatting Escherichia coli and ESBL organisms during the COVID pandemic.
The increased use of ASP programs has resulted in reduced antimicrobial use and cost, and lower incidence of healthcare associated infections.^6^ The small community hospital is governed by DNV which uses the Core Elements as a foundation for antimicrobial stewardship accreditation standards.^7^ The Antimicrobial Stewardship Committee (ASC) is a subcommittee of the Pharmacy and Therapeutics (P&T) committee; ASP efforts are analyzed and reported to Q&P on a quarterly basis as required by DNV. Pharmacist led Antimicrobial Stewardship Programs have shown significant improvements in healthcare. Previous studies have shown that pharmacist led antimicrobial stewardship programs can effectively improve health outcomes such as reducing hospital length of stay, mortality, inappropriate antimicrobial use, and costs.^8,9^ Pharmacists are able to effectively focus on proper antimicrobial utilization while actively working in multidisciplinary groups within the healthcare system.^10^ Overall, contributing to the appropriate use of antimicrobials leading to successful therapeutic outcomes. There are many challenges that an ASP at a small community hospital can face such as lack of resources and costs. This provides an opportunity to evaluate the antimicrobial stewardship program’s current policies and tools that are utilized in a small community hospital specifically in managing gram negative bacteremia. This retrospective review aims to assess current antimicrobial stewardship practices expected at a local hospital level through current practices such as Infectious Diseases Pharmacist led interventions, policy driven measures, and the various tools utilized by the small community hospital’s ASP.
A retrospective cohort analysis study was conducted to evaluate the impact of antimicrobial stewardship practices in treating Escherichia coli bacteremia in both non-ICU and ICU settings. This study was conducted at non-academic small single center community hospital with a 239-bed located in Pennsylvania, USA. All patients that were ≥18 years of age admitted in the hospital with positive blood cultures for Escherichia coli from July 2021 to July 2022 were analyzed. Patients were excluded if their blood cultures were not positive for Escherichia coli and if antibiotics were not initiated during hospital stay. Figure 1 represents flowchart of patient selection process for patients with gram negative bacteremia. During the study 74 patients with gram negative bacteremia were identified and 43 patients had Escherichia coli bacteremia. These patients were further categorized into E. Coli ESBL bacteremia (n=6), E. Coli bacteremia in Non-ICU setting (n=29), and E. Coli bacteremia in ICU setting (n=8). 41 patients were included in the analysis as 2 patients were excluded since antibiotics were not initiated.
Figure 1. Flowchart of patients included in the study
The primary endpoint was to determine the length of hospital stay (Table 2). Critical secondary endpoints include antibiotic de-escalations (Table 3), duration of antibiotics, time to definitive antibiotic therapy, serum procalcitonin levels (Table 5), and blood culture availabilities (Table 6).
At this small community hospital setting, Infectious Diseases Pharmacist-led Antimicrobial Stewardship Program (ASP) is driven by various interventions, policies, and tools specifically targeting gram negative infections/therapeutics listed in Table 1a. A detailed summary of antimicrobial stewardship interventions is also listed in Table 3 which assisted analyzing critical secondary endpoints.
As a means to assess prevalence and resistance patterns during time period from January 2021 to December 2022, ASP collaborated with a multidisciplinary team to develop yearly source specific cumulative antibiograms (urine and non-urine isolates) for 2021 and 2022 (Please refer to Figure 2). In addition, a segregated data analysis for the ESBL organisms was conducted to analyze patterns of resistance for urine and non-urine antibiogram during year 2021 and 2022 (Please refer to Figure 3).
Figure 2. Cumulative Antibiograms
Figure 3. Segregated Data Analysis – ESBL organisms
Data was obtained from a review of the electronic medical record (EMR) and recorded into a de-identified data collection form. Collected data consisted of demographic information such as age and gender, dates of admission, dates of discharge, dates of blood culture collection, dates of blood culture results posted, antibiotic treatments initiated, dates of antibiotics initiated, procalcitonin levels on admission, and IV to PO conversions.
The descriptive statistics and key outcome measures identified are demographics (Table 1b), length of hospital stays (Table 2), broad therapy to de-escalating agents for E. coli bacteremia in non-ICU setting (Table 4), broad therapy to de-escalating agents for E. coli bacteremia in ICU setting (Table 4), agents utilized for ESBL-producing E. coli bacteremia in non-ICU setting, procalcitonin levels on admission (Table 5), presence of positive blood cultures (Table 6). Supporting tools involved comparison of antibiograms for year 2021 and 2022 as listed below.
Figure 2. Cumulative Antibiograms
Figure 3. ESBL organisms URINE AND NON-URINE Antibiograms
The primary endpoint was to determine the impact of antimicrobial stewardship pharmacist-led interventions on the length of hospital stay in patients with Escherichia coli bacteremia during COVID-19 pandemic in year 2021. The primary endpoint results showed that in patients with Escherichia coli bacteremia in the ICU setting the average length of stay was longer at 16.6 days, in patient with Escherichia coli bacteremia in the Non-ICU setting the average length of stay was 7.3 days, and in patients with ESBL E. coli bacteremia the length of stay was 6.8 days according to Table 2.
Critical secondary endpoints include antibiotic de-escalations, duration of antibiotics, time to definitive antibiotic therapy, serum procalcitonin levels, and blood culture availabilities. The most common de-escalating agents are cephalosporins according to Table 4 in both the Non-ICU and ICU settings. De-escalation of agents occurred in 61% of patients through the ASP’s interventions. A detailed summary of antimicrobial stewardship interventions is listed in Table 3. Out of 1061 antimicrobial stewardship interventions focused on targeted broad-spectrum antibiotics (Cefepime, Zosyn, Meropenem, Levofloxacin), highest number of interventions targeted carbapenem de-escalation (13.3%) highlighting increased efforts to one of the broadest spectrums of IV antibiotics to combat resistance.
Out of 27 patients in the E. coli bacteremia in Non-ICU setting, 18 de-escalations occurred and are shown in Table 4. De-escalations did not occur in 9 patients because therapy was broadened or only broad antibiotics was initiated. Average time to de-escalate was 3.3 days. In comparison, out of 8 patients in Escherichia coli bacteremia in ICU setting, 7 de-escalations occurred and are shown in table 4. De-escalations did not occur in 1 patient since therapy was broadened. Average time to de-escalate was 3.6 days.
Six patients had ESBL E. coli bacteremia and the antibiotics that were initiated and switched to are shown in Table 4. Three patients ended up on the appropriate antibiotics (carbapenems) and the other three patients ended up with inappropriate antibiotics.
Admission serum procalcitonin levels were available for 29 patients. One patient had a PCT level greater than a 100 and was not included in the PCT analysis. Blood cultures were collected in all patients included in analysis. Repeat blood cultures were only collected for 13 patients.
According to the segregated data analysis, the antibiogram for non-urine isolates accounting for gram negative organisms reported between January 2021 to December 2021 indicated that Escherichia coli remained the most prevalent organism (number of isolates identified 122) at our institution. The antibiogram analysis demonstrated that a total of 30 isolates which were ESBL producer in Non-Urine group. Of these, 20 isolates were identified as E. Coli – ESBL producer. In view of the high prevalence of Escherichia coli-ESBL producer in non-urine group, we focused on de-escalation efforts and therapeutics utilized for patients experiencing bacteremia with Escherichia coli, gram negative bacilli, since these organisms are a major threat in hospitalized patients. They have a mortality rate of 12-38% depending on appropriate antibiotic use.^2^ Hence this study focused primarily on ASP led interventions, procalcitonin monitoring, and blood culture assays.
Reduction in hospital length of stay improves bed turnover, improving patient outcomes, decreasing mortality, and decreasing costs.^11^ Antimicrobial stewardship programs have an impact on patient outcomes such as length of stay, re-admissions, and mortality. According to a Cochrane database systematic review on interventions to improve antibiotic prescribing practices for hospital in patients in 2017, lower use of antibiotics decreases mortality and reduces length of stay.^12^ The CDC also emphasizes that ASPs can help improve clinical outcomes and minimize patient harms by improving antibiotic prescribing.^4^ In our patient population the length of hospital stay for ICU patients were longer which is expected due to disease burden and complications. Data from CDC’s National Hospital Discharge survey showed that patients hospitalized with sepsis have an average length of stay that was 43% higher than that of other patients.^13^ However, the average length of stay for Escherichia coli bacteremia and ESBL Escherichia coli in the non-ICU setting is much shorter possibly due to the fact that these patients are not as sick and therefore require less antibiotic use overall. The average duration of any antibiotic that was ever given to a patient was 2.6 days. Typical duration of treatment for uncomplicated gram-negative bloodstream infections ranges from 7-14 days. Many retrospective studies and reviews have shown that there are no differences in clinical outcomes in patients treated for bloodstream infections with shorter courses of antibiotic therapy compared to prolonged courses especially in patients with urinary sources of infection.^14,15^ Majority of the patients in our study had other sources of infection observed in the urine. Treatment duration was much shorter compared to a duration of 7 days and this is because the data does not account the duration of days in which a patient may have been sent home with on antibiotics as it only captures antibiotics given on an inpatient level. However, the ASP program has automatic 7 day stop dates for any antibiotics initiated in patients to avoid unnecessary prolonged antibiotic use. The automatic 7 day stop dates along with infectious disease stewardship interventions has led to an overall decrease in the antibiotic utilization and duration, thus overall decreasing selective pressure of antibiotics and resistance.
The most common de-escalating agents were cephalosporins according to Table 3 in both the Non-ICU and ICU settings. De-escalation of agents occurred in 61% of patients through the ASP’s interventions. De-escalation of antibiotics did not occur in all patients since it depends on the severity of the infection and other comorbidities. The ASP assesses appropriateness of antibiotic use and de-escalates therapy as soon as blood culture panels and culture and susceptibility reports come back. The average time for culture and susceptibility reports to come back is 2.8 days, allowing ample time for the infectious disease stewardship team to de-escalate therapy for a narrower coverage in order to decrease resistance. The average time to de-escalate therapy was 3 days. Studies have shown that de-escalating antibiotic therapy leads to lower mortality and decreased resistance.^16,17^ There is very limited data available on gram-negative bacteremia oral step-down therapy with cephalosporins. However, there are studies conducted stating that oral antibiotics with high bioavailability were effective treatments in hospitalized patients with gram negative bacteremia.^18^ Out of all oral cephalosporins, cephalexin (Keflex) has the highest bioavailability of nearly 100%. De-escalating therapy from IV to PO is important to prevent cannula related infections and improve patient safety. Patients that transitioned from IV to oral antibiotics experienced similarly low rates of treatment failure than those who received only IV therapy.^19^ Therefore, it’s important to consider de-escalating patients from IV to oral antibiotics. Among 43 patients IV to PO oral step-down therapy occurred in 16 (37.2%) patients with cephalexin and oral fluoroquinolones being the most common oral-step down agents. Cephalexin and oral fluoroquinolones have high bioavailability’s which are effective in treating hospitalized patients with gram-negative bacteremia due to their ability to reach adequate blood concentrations. Less than half of the patients ended up on oral antibiotic therapy, however that may be due the patients’ clinical stability and bacteremia burden. The ASP has an IV to PO policy in place that allows the infectious disease stewardship team to safely switch hospitalized patients initially on intravenous antibiotics to an oral equivalent once the patient is clinically stable. The ASP’s efforts are heavily focused on decreasing complications that may arise from IV antibiotic use and decreasing overall costs as well.
Procalcitonin (PCT) is an essential biomarker utilized in the early detection of bacterial infections. Other biomarkers such as C-reactive protein (CRP) lack specificity in differentiating between bacterial or non-bacterial infections, whereas serum PCT levels are not elevated in viral infections which makes it an ideal biomarker in identifying bacterial infections. Serum levels greater than 0.25 ng/mL can indicate a bacterial infection. Early detection of bacterial infections with PCT can lead decrease in morbidity, mortality, and antibiotic overuse. High serum levels of PCT correlate to positive blood cultures and sepsis.^20^ Therefore, PCT levels at baseline were evaluated in this study as well to address antimicrobial stewardship efforts in patients with Escherichia coli bacteremia in Non-ICU and ICU settings at a small community hospital. Admission serum procalcitonin levels were collected for 29 patients. 1 patient had a PCT level greater than a 100 and was not included in the PCT analysis. The average length of stay of patients in which PCT levels exceeded 2ng/mL was 9.6 days. This increase could be due other etiologies such as compromised renal function, cancer, or spinal cord injuries and this study did not excluded patients with other co-morbidities. Average PCT levels in the ICU were higher (23 ng/mL) compared to PCT levels in the Non-ICU setting (10 ng/mL). This is expected since patients in the ICU were more likely to be treated with sepsis or septic shock. The ASP policy regarding procalcitonin levels for systemic bacterial infections indicates that patients with PCT levels greater than 2.0 ng/mL have a high risk for sepsis and/or septic shock. The average time it took to de-escalate antibiotics when PCT levels in patients were above 2 ng/mL were 3 and 4 days in ICU and Non-ICU patients respectively. Studies have shown that using procalcitonin levels to guide de-escalation can decrease duration of antibiotics without complications.^21-23^ The ASP implements the procalcitonin policy to further evaluate antibiotic use and its appropriateness.
Blood culture identification panels allow clinicians to easily interpret results of a positive blood culture within a short time frame. The BCID panel tests for bacterial pathogens along with their antimicrobial resistance genes such as the CTX-M gene for ESBL producing Enterobacterales. BCID panels are used to guide empiric selection of therapy.^24^ According to the HNL Lab Medicine BCID panel’s guidance for interpreting blood culture results, if Escherichia coli with no resistance markers is detected then ceftriaxone should be used as empiric therapy. If Escherichia coli with CTX-M is detected, then meropenem should be used as empiric therapy.^25^ Our institution utilizes blood culture identification panels (BCID) to further assess optimal antibiotic use by ruling out resistant organisms. According to Table 6 the average time for BCID panels to get posted is 1.2 days, allowing the ASP sufficient time to de-escalate antibiotics. This is crucial as the rapid initiation of appropriate antibiotic therapy has led to reduction morbidity and mortality.^26^ With a turnaround time of approximately 1 day, the antimicrobial stewardship team can promptly start appropriate antibiotics or even de-escalate antibiotics. This can help decrease any unwanted adverse drug events such as Clostridioides difficile infections as well. These rapid diagnostic tests improve patient care since they provide important and reliable pathogen-related information in a short time frame.^26^
Minimum inhibitory concentrations (MIC) provide clinicians information to select the most optimal antibiotic for their patients. CLSI defines MIC categories as Susceptible (S) meaning isolates of the organism are inhibited by the normal achievable concentrations of the antibiotic when the recommended dose is used. Susceptible dose dependent (SDD) is when isolates of the organism are inhibited depending on the dose of the antibiotic. Intermediate (I) is defined as isolates of the organism with response rates that may be lower than the susceptible isolates. This category shows clinical efficacy in sites of the body where the antibiotic is concentrated in or when a higher than normal dose of a drug is used. Resistant (R) is defined as when isolates of the organism are not inhibited by the normal achievable concentrations of the antibiotic when the recommended dose is used.^27^ Among all the de-escalating agents, cephalosporins were the most common agent utilized as seen in Table 3. Ceftriaxone and cefazolin were the most common cephalosporin de-escalating agents. Clinical MIC breakpoints are currently set and published by two organizations, the European EUCAST (European Committee on Antimicrobial Susceptibility Testing) and the American CLSI (Clinical and Laboratory Standards Institute). Cefazolin is recognized as a surrogate test agent for oral cephalosporins. The CLSI standards on AST reports an MIC breakpoint of cefazolin for treating uncomplicated urinary tract infections as ≤16 μg/ml when a dose of 1g q8h is given. Co-infections with Escherichia coli in the urine were observed in 19 patients. Even though some patients also had urine infections, we focused on bloodstream infections considering that CLSI guidelines for MIC breakpoints were updated to include infections other than urinary tract infections. According CLSI standards regarding AST, the MIC breakpoint based on the dose of 2g q8h for cefazolin when treating infections other than uncomplicated urinary tract infections (UTIs) due to Escherichia coli was ≤ 2 μg/ml.^27^ The HNL lab does susceptibility testing for the small community hospital setting and the MIC of cefazolin with systemic infections due to Escherichia coli was reported as ≤ 4 μg/ml in 29 patients. Our local lab reports a higher MIC value than the recommendation by CLSI. This is because there are many barriers when implementing new MIC breakpoints in labs that use FDA or commercial MIC susceptibility testing systems. It could take several years before new MIC breakpoints are implemented as the manufacturers of these systems cannot ship the equipment needed to assess these breakpoints, since it will take time for the FDA to clear the susceptibility testing systems that include the revised CLSI breakpoints.^28^ Despite this barrier, ID specialists and pharmacists play an important role in assessing the impact of CLSI revised breakpoints on patient management. Through working with our local HNL lab, the infectious disease stewardship team implements ways to increase infection control measures and optimize antibiotic use to overall help decrease hospital length of stay.
Despite the various challenges that antimicrobial stewardship programs (ASP) face in a single center small community hospital, the ASP at this small community hospital utilizes various policies and tools to increase appropriate antibiotic use and decrease hospital length of stay in patients with Escherichia coli bacteremia. Pharmacists have a great impact in the ASP, as they play pivotal roles in utilizing various strategies to improve health outcomes. Limitations of the study include the small population size, retrospective study design, and different primary diagnoses can be confounding factors. These limitations may make it difficult to predict whether the results can be extrapolated to other facilities. Future research can focus on a more rigorous study design with a multicenter approach on a larger population sample. Despite its limitations, this study has shown the various pharmacist led interventions a small community hospital utilizes to improve patients’ healthcare.
Conflict of Interest Statement: No potential conflicts of interests were noted by the authors of this paper.
Ethics Approval Statement: The research was approved by the Holy Redeemer Hospital’s Institutional Review Board.
Disclaimer: The statements, opinions, and data contained in all publications are those of the authors.