Antimicrobial use and other risk factors for infections with antimicrobial-resistant bacteria and fungi bacteria and fungi in an intensive care unit

Abigail W Cheng; Jesse Chou; Robert G Sawyer

doi:10.4103/wjsi.wjsi_12_22

ORIGINAL ARTICLE

Year : 2022 | Volume : 1 | Issue : 2 | Page : 50-57

Antimicrobial use and other risk factors for infections with antimicrobial-resistant bacteria and fungi bacteria and fungi in an intensive care unit

Abigail W Cheng, Jesse Chou, Robert G Sawyer
Department of Surgery, Western Michigan University Homer Stryker M.D. School of Medicine, Kalamazoo, Michigan, USA

Date of Submission	07-Dec-2022
Date of Decision	20-Dec-2022
Date of Acceptance	21-Dec-2022
Date of Web Publication	15-Feb-2023

Correspondence Address:
Robert G Sawyer
1000 Oakland Drive Kalamazoo, Michigan 49008
USA

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/wjsi.wjsi_12_22

Abstract

Introduction: Resistant infections are especially problematic in the intensive care unit (ICU), but risk factors remain unclear. We hypothesized that the risk factors for resistant Gram-negative rods (rGNR), resistant Gram-positive cocci (rGPC), and secondary fungal infections differed.
Materials and Methods: A single-center cohort study of patients with ICU-acquired infections from 1997 to 2017 was performed. Inclusion was conditioned on the presence of rGNR, rGPC, or fungi. Risk factors studied included demographics, medical comorbidities, Acute Physiology and Chronic Health Evaluation (APACHE) II score, and previous antimicrobial exposure.
Results: Four thousand three hundred and nineteen ICU-acquired infections were identified. One thousand nine hundred and ninety-eight were considered resistant and 2321 were considered nonresistant. Identification of any resistant organism was significantly associated with female sex, nontrauma diagnosis, APACHE II score, liver disease, chronic steroid use, history of any prior infection, and history of a resistant infection, but not days of prior antimicrobial use. Infections with rGNR were associated with days of therapeutic antimicrobials given for a previous infection, but not total prior antimicrobial days during hospitalization. rGPC infections were associated with both previous infections treated with antimicrobials and total prior antimicrobial days during hospitalization. Fungal infections were not associated with any measure of prior antimicrobial exposure. Controlling for the severity of illness and demographics, resistant infections were not associated with mortality compared to nonresistant infections.
Conclusions: The likelihood of rGNR infection is closely linked to recent antimicrobial exposure, while rGPC infection appears to be associated with prior antimicrobial exposure. Fungal infections may not be associated with prior antimicrobial exposure. These findings suggest disparate mechanisms of dysbiosis for different classes of resistant pathogens.

Keywords: Antimicrobial resistance, infection, intensive care unit

How to cite this article:
Cheng AW, Chou J, Sawyer RG. Antimicrobial use and other risk factors for infections with antimicrobial-resistant bacteria and fungi bacteria and fungi in an intensive care unit. World J Surg Infect 2022;1:50-7

How to cite this URL:
Cheng AW, Chou J, Sawyer RG. Antimicrobial use and other risk factors for infections with antimicrobial-resistant bacteria and fungi bacteria and fungi in an intensive care unit. World J Surg Infect [serial online] 2022 [cited 2023 Mar 30];1:50-7. Available from: https://www.worldsurginfect.com/text.asp?2022/1/2/50/369704

Introduction

Health-care-associated infections are estimated to occur in 1 in every 25 acute hospitalizations and represent a significant cost burden in the billions of dollars.^[1] It is estimated that 30%–70% of these infections are preventable by the implementation of best practices in infection control.^[2] Among these, the best practice is antimicrobial stewardship, which aims to reduce inappropriate antimicrobial use while maintaining treatment efficacy.^[3] However, despite having an array of antimicrobials available to tackle pathogens, the development of antimicrobial resistance has allowed many of these pathogens to render the efficacy of much of our arsenal obsolete.^[4] Resistant infections pose a great challenge, especially in the intensive care unit (ICU), with up to 40% incidence and an association with high mortality.^[5] Numerous independent risk factors for the development of resistant infections among ICU patients include central venous access, various forms of catheterization, mechanical ventilation, ICU length of stay, and previous history of resistant organisms.^[5],[6] Studies have found the rate of antimicrobial resistance to be dependent on various characteristics that ICU patients harbor sicker patient population, intensity of infection control, pathogens involved, and method of transmission.^{[5],[7],[8],[9]}

The relationship between antimicrobial usage and resistance has been demonstrated on multiple occasions to be linear, where antimicrobials can generate a breeding ground for resistance, especially if taken for a duration of > 7 days, and resistance is associated with worse outcomes.^[6],[7] However, newer theories suggest that resistance rates may only be increased in areas where antimicrobial use triggers selection pressures after a minimum threshold is exceeded, resulting in the development of resistance genes.^[7] This threshold has been described as the temporal relationship between the use of certain antimicrobials and subsequent rates of resistant microbes, antimicrobial substitutions, and the “total use thresholds hypothesis” incorporating potentially lasting effects on the microbiome.^[7] Numerous studies have looked at the optimal duration and dosage of antimicrobials to achieve the best balance between cure rate and continued antimicrobial efficacy.^{[10],[11],[12]}

Although general risk factors for antimicrobial resistance development have been identified, predictors of specific infections such as resistant Gram-negative rods (rGNR), resistant Gram-positive cocci (rGPC), and fungal infections have yet to be delineated.^[10] Clostridiodes (née Clostridium difficile), Staphylococcus aureus, and Escherichia ^{More Details} coli are common pathogens causing health-care-associated infections, with 45% of S. aureus demonstrating methicillin resistance S. aureus (MRSA), and 5% of Gram-negative rods demonstrating carbapenem resistance in one study.^[6] In another study, E. coli and S. aureus resistance demonstrated a nonlinear relationship with fluoroquinolone use.^[7] ICU patients frequently have antimicrobials in their treatment plan, be it for an established infection or surgical prophylaxis. Therefore, determining the effects of antimicrobial treatment on the subsequent cultivation of resistant organisms is crucial to further improve antimicrobial stewardship. Our study aims to investigate the different risk factors for the isolation of these resistant organisms and their associated mortality.

Materials and Methods

We performed a retrospective analysis of prospectively collected data from patients requiring intensive care from 1997 to 2017 in a single academic surgical-trauma ICU. Patients with ICU-acquired infections were separated into the type of infection present: nonresistant, rGNR, rGPC, or fungi. Fungi were included to assess the relationship, if any, between receipt of antibacterial agents and development of secondary fungal infections. The local Institutional Review Board waived the need for consent for this study due to its observational nature. All patients with ICU-acquired infections, using the Centers for Disease Control and Prevention definitions, were included.^[13],[14] Demographics, culture, and treatment data were recorded for each infection. The Acute Physiology and Chronic Health Evaluation (APACHE) II scoring was performed at the time of initiation of treatment for infection.^[15] Data specifically collected to assess prior antimicrobial exposure included history of prior infection treated during hospitalization, days from admission to treatment for the index infection, history of any prior infection with a resistant pathogen, days of continuous antimicrobial therapy preceding the qualifying infection if the patient was diagnosed with an ICU-acquired infection during the treatment of a previously diagnosed infection, and total days of antimicrobial therapy during hospitalization prior to the qualifying ICU-acquired infection. Patients were followed until hospital discharge.

Resistant infections included MRSA, vancomycin-resistant enterococci, any Gram-negative pathogen resistant to all members of at least one major class of antimicrobials (aminoglycosides, cephalosporins, fluoroquinolones, carbapenems, penicillin/beta-lactamase inhibitor combinations), any Stenotrophomonas maltophilia, C. difficile, and any fungus (inherently resistant to antimicrobial agents).

Statistical analysis

Continuous variables were compared using the Student's t-test, and categorical variables were compared using the Chi-square test. Independent predictors of the presence of a resistant pathogen and mortality were determined by logistic regression analysis. IBM^® SPSS^® version 27 was used for statistical analysis (IBM, Inc., Armonk, NY, USA).

Results

Patient characteristics, index infection, and infection biography

Of the 4319 ICU-acquired infections identified, 1998 (46.3%) had resistant infections and 2321 (53.7%) had nonresistant infections [Table 1]. Resistant organisms were significantly associated with increased age and female sex, as well as multiple comorbidities and nontrauma-related admission diagnosis. Resistant infections were associated with a higher white blood cell count, lower maximum temperature, and a longer duration of hospital stay prior to diagnosis of infection. The mean APACHE II score for the index infection was 20.7 ± 0.1 for the resistant group and 19.5 ± 0.1 for the nonresistant group (P < 0.0001). However, there was no significant association between site of infection and resistance. Resistant organisms were significantly associated with any history of prior resistant or nonresistant infections, as well as days of antimicrobials for the most recently treated prior infection and total days of antimicrobials received in the hospital prior to the diagnosis of the qualifying ICU-acquired infection.

Table 1: Baseline demographics and clinical characteristics*

Click here to view

Resistance profile

The 1998 resistant infections identified consisted of 24.1% with Gram-positive cocci, 44.0% with Gram-negative rods, 5.6% with C. difficile, and 36.4% with fungi [Table 2]. A comprehensive list of all organisms isolated is provided in [Table 3]. E. coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, S. maltophilia, Serratia spp., and Acinetobacter spp. were the most common resistant Gram-negative bacteria isolated, and the most common resistant fungi isolated were Candida spp.

Table 2: Most common resistant bacteria isolated and fungi*^†

Click here to view

Table 3: Organisms isolated*^†

Click here to view

Risk factors for infection with resistant pathogens

For all infections, female sex (odds ratio [OR] = 1.18, 95% confidence interval [CI] = 1.03–1.35, P = 0.02), APACHE II score per point (OR = 1.02, 95% CI = 1.01–1.03, P = 0.003), liver disease (OR = 1.40, 95% CI = 1.08–1.81, P = 0.01), chronic steroid use (OR = 1.35, 95% CI = 1.08–1.69, P = 0.008), history of any infection (OR = 1.82, 95% CI = 1.47–2.24, P = 0.0001), history of resistant infection (OR = 1.48, 95% CI = 1.22–1.79, P = 0.0001), and days from admission to diagnosis of infection (OR = 1.02, 95% CI = 1.47–2.24, P = 0.0001) were significantly associated with identification of any resistant organism. An admitting diagnosis of trauma was associated with sensitive infections (OR = 0.75, 95% CI = 0.63–0.89, P = 0.001), with a C statistic of 0.72 and Hosmer–Lemeshow test of 0.001. However, prior antimicrobial use did not have a significant association with identifying any resistant organism.

We hypothesized that even though there was no association between days of antimicrobial exposure and subsequent infection with any resistant pathogen, there could still be a relationship between days of antimicrobials and infection with each of the different classes of resistant organisms. [Table 4] shows the results of multivariate analyses assessing the independent predictors of ICU-acquired infections with rGNR, rGPC, and fungi. Risk factors were different between classes of resistant organisms, with only days from admission to diagnosis of infection shared by all the three. In terms of prior infection and exposure to antimicrobials, prior infection was associated with rGNR and fungal infections, history of infection with resistant pathogen was associated with rGNR and rGPC infections, days of antimicrobials used for the most recent previous infection was associated with rGNR infections, and fewer total days of antimicrobials during hospitalization was associated with rGPC infections.

Table 4: Multivariate analysis for resistant Gram-negative rods, resistant Gram-positive cocci, and fungi*^†

Click here to view

Antimicrobials, outcomes, and risk factors for mortality

By univariate analysis, resistant infections were significantly associated with increased mortality, length of stay, and total days on antimicrobials [Table 5]. Results for the multivariate analysis for all-cause, inhospital mortality are given in [Table 6]. Mortality was associated with age, severity of illness, greater white blood cell count, and lower maximum temperature. It was also associated with liver disease, malignancy, and ventilator use. Greater total days of antimicrobials prior to identification of resistant infection, as well as fewer days of antimicrobials for the most recent prior infection, were associated with mortality. However, after controlling for the severity of illness and demographics, there was no association between isolation of a resistant pathogen resistance and mortality.

Table 5: Overall treatments and patient outcomes*

Click here to view

Table 6: Multivariate analysis for mortality*^†

Click here to view

Discussion

Despite traditional theories defining a linear relationship between antimicrobial use and the development of resistance in an organism, our data suggest a more nuanced relationship. Our data were consistent with previous studies in that the risk factors that confer resistance differ substantially between rGNR, rGPC, and fungi.^[6],[7],[16] Among patients with infections, our data suggested that general resistance was associated with nonmodifiable risk factors such as increased age and female sex. However, on subsequent analysis, there were differences in risk factors on the classification of infectious etiology by the organism type: rGNR, rGPC, and fungi.

These differences in risk factors imply that rGNR, rGPC, and fungi utilize different mechanisms to promote resistance. In particular, previous studies have suggested that GNR, such as E. coli, may have an increased rate of mutations and subsequent resistance as a result of the oxidative stress placed by antimicrobial use.^[16],[17] In addition, some authors have described how GNRs are more capable of developing resistance de novo after exposure to antimicrobial therapy.^[18] These observations dovetail with our data and observations that any history of infection and days of antimicrobials for prior infection are significant predictors of rGNR infection. Similarly, the pattern for rGPC may be multifactorial as well, as demonstrated by our data showing significantly different results from the multivariate analysis of risk factors compared to rGNR and fungi. These strongly suggest a unique process for the development of resistance that varies at the organism level.

Furthermore, previous studies have found an association between prior antimicrobial exposure and the development of secondary fungal infections.^[19] However, our data found no associations between secondary fungal infections and any measure of prior antimicrobial exposure. This could be attributed in part to a different degree of selective pressure associated with each antimicrobial and institutional variation in the employment of empiric antimicrobial treatment.^[7],[8],[9] Studies are inconsistent in the documentation of fungal infections, namely, whether the presence of a fungal organism denotes a nonpathogenic colonizer or disseminated disease. Furthermore, there is little consensus over whether empiric antifungal treatment is warranted in any or all instances of empiric antimicrobial treatment. In addition, our data suggested a history of prior infection being a risk factor for the development of fungal infections. This highlights the use of antimicrobials themselves, rather than the temporal relationship, as the initiator of resistance development.

Evidently, the relationship between pathogen resistance and the presence or absence of risk factors is no longer univariable, such as being solely attributable to antimicrobial usage. Rather, the mechanisms for developing resistance remain elusive. Delineating these mechanisms will aid greatly in the development of meaningful interventions to prevent and treat resistant infections, as well as enable the tailoring of treatments, whether antimicrobial, microbiome based, or nonmicrobial based.

Finally, patients in the ICU represent a population particularly vulnerable to infection.^[5],[8],[9] Often, these infections are caused by antimicrobial-resistant organisms, as evidenced by our rate of resistant infections (53.7%), which was higher than predicted by general review of the literature.^[3],[5],[6] However, our data bore no association between the presence of resistance and mortality, even after adjusting for the severity of illness. A partial explanation could be due to other factors such as severe underlying disease, increased age, and longer hospitalization time.

Several limitations may affect the interpretation of the data. For one, our study reflects data from a single tertiary care institution's ICU. Resistance patterns from one institution may not be broadly generalizable to other institutions, especially if there are substantively different demographics. In addition, this is a retrospective analysis. A useful prospective study could include more preillness data such as preoperative microbiome or other preillness assessment of patients. Third, antimicrobial prescription practices are not standardized between institutions, let alone physicians. It is unclear whether these differences contributed to any meaningful difference in our findings.

Conclusions

Demographic and treatment risk factors for resistance to antimicrobial agents vary by resistant pathogen class. The likelihood of rGNR infection appears to be most closely linked to recent antimicrobial exposure, while rGPC infection appears to be associated with the totality of prior antimicrobial exposure. Fungal infections may not be associated with prior exposure. These findings suggest disparate mechanisms of dysbiosis for different classes of resistant pathogens, and the need for more studies to further investigate these mechanisms.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Ben-Ami R, Olshtain-Pops K, Krieger M, Oren I, Bishara J, Dan M, et al. Antibiotic exposure as a risk factor for fluconazole-resistant Candida bloodstream infection. Antimicrob Agents Chemother 2012;56:2518-23.
2.	Cirz RT, O'Neill BM, Hammond JA, Head SR, Romesberg FE. Defining the pseudomonas aeruginosa SOS response and its role in the global response to the antibiotic ciprofloxacin. J Bacteriol 2006;188:7101-10.
3.	Händel N, Hoeksema M, Freijo Mata M, Brul S, ter Kuile BH. Effects of stress, reactive oxygen species, and the SOS response on de novo acquisition of antibiotic resistance in Escherichia coli. Antimicrob Agents Chemother 2015;60:1319-27.
4.	Haque M, Sartelli M, McKimm J, Abu Bakar M. Health care-associated infections – An overview. Infect Drug Resist 2018;11:2321-33.
5.	Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008;36:309-32.
6.	Knaus WA, Zimmerman JE, Wagner DP, Draper EA, Lawrence DE. APACHE-acute physiology and chronic health evaluation: A physiologically based classification system. Crit Care Med 1981;9:591-7.
7.	Lawes T, López-Lozano JM, Nebot C, Macartney G, Subbarao-Sharma R, Dare CR, et al. Turning the tide or riding the waves? Impacts of antibiotic stewardship and infection control on MRSA strain dynamics in a Scottish region over 16 years: Non-linear time series analysis. BMJ Open 2015;5:e006596.
8.	Lawes T, Lopez-Lozano JM, Nebot CA, Macartney G, Subbarao-Sharma R, Dare CR, et al. Effects of national antibiotic stewardship and infection control strategies on hospital-associated and community-associated meticillin-resistant Staphylococcus aureus infections across a region of Scotland: A non-linear time-series study. Lancet Infect Dis 2015;15:1438-49.
9.	Laxminarayan R, Duse A, Wattal C, Zaidi AK, Wertheim HF, Sumpradit N, et al. Antibiotic resistance-the need for global solutions. Lancet Infect Dis 2013;13:1057-98.
10.	López-Lozano JM, Lawes T, Nebot C, Beyaert A, Bertrand X, Hocquet D, et al. A nonlinear time-series analysis approach to identify thresholds in associations between population antibiotic use and rates of resistance. Nat Microbiol 2019;4:1160-72.
11.	Magill SS, O'Leary E, Janelle SJ, Thompson DL, Dumyati G, Nadle J, et al. Changes in prevalence of health care-associated infections in U.S. Hospitals. N Engl J Med 2018;379:1732-44.
12.	Martínez JL, Baquero F, Andersson DI. Predicting antibiotic resistance. Nat Rev Microbiol 2007;5:958-65.
13.	Neely AN, Holder IA. Antimicrobial resistance. Burns 1999;25:17-24.
14.	Schmier JK, Hulme-Lowe CK, Semenova S, Klenk JA, DeLeo PC, Sedlak R, et al. Estimated hospital costs associated with preventable health care-associated infections if health care antiseptic products were unavailable. Clinicoecon Outcomes Res 2016;8:197-205.
15.	Schreiber PW, Sax H, Wolfensberger A, Clack L, Kuster SP, Swissnoso . The preventable proportion of healthcare-associated infections 2005-2016: Systematic review and meta-analysis. Infect Control Hosp Epidemiol 2018;39:1277-95.
16.	Talan DA, Stamm WE, Hooton TM, Moran GJ, Burke T, Iravani A, et al. Comparison of ciprofloxacin (7 days) and trimethoprim-sulfamethoxazole (14 days) for acute uncomplicated pyelonephritis pyelonephritis in women: A randomized trial. JAMA 2000;283:1583-90.
17.	Tosi M, Roat E, De Biasi S, Munari E, Venturelli S, Coloretti I, et al. Multidrug resistant bacteria in critically ill patients: A step further antibiotic therapy. J Emerg Crit Care Med 2018;2:105-13. [doi: 10.21037/jeccm. 2018.11.08].
18.	Uranga A, España PP, Bilbao A, Quintana JM, Arriaga I, Intxausti M, et al. Duration of antibiotic treatment in community-acquired pneumonia: A multicenter randomized clinical trial. JAMA Intern Med 2016;176:1257-65.
19.	Verduri A, Luppi F, D'Amico R, Balduzzi S, Vicini R, Liverani A, et al. Antibiotic treatment of severe exacerbations of chronic obstructive pulmonary disease with procalcitonin: A randomized noninferiority trial. PLoS One 2015;10:e0118241.