Hostname: page-component-7b7567c8c-kwt9d Total loading time: 0 Render date: 2026-11-13T14:22:12.243Z Has data issue: false hasContentIssue false
  • English
  • Français

Comparison of Antibiotic Susceptibility Patterns of Bacterial Isolates Based on Time From Hospitalization and Culture Source: Implications for Hospital Antibiograms

Published online by Cambridge University Press:  13 November 2015

Yaakov Dickstein,
Yuval Geffen,
Leonard Leibovici and
Mical Paul
Yaakov Dickstein*
Affiliation:
Division of Infectious Diseases, Rambam Health Care Campus, Haifa, Israel
Yuval Geffen
Affiliation:
Medical Microbiology Laboratory, Rambam Health Care Campus, Haifa, Israel
Leonard Leibovici
Affiliation:
Medicine E, Rabin Medical Center, Beilinson Hospital, Sackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel
Mical Paul
Affiliation:
Division of Infectious Diseases, Rambam Health Care Campus, Haifa, Israel
*
Address correspondence to Dr. Yaakov Dickstein, Division of Infectious Diseases, Rambam Health Care Campus, HaAliya HaShniya St. 8, Haifa, Israel 3109601 (y_dickstein@rambam.health.gov.il).
Rights & Permissions [Opens in a new window]

Abstract

We assessed the effects of time from hospitalization and culture source on bacterial susceptibility profiles. Increasing resistance correlated with increasing time from hospitalization for all bacterial groups, with 7 days in hospital representing the best time point for dichotomizing susceptibility rates rather than 48 hours. Antibiograms based on isolates from any source best represented susceptibility profiles.

Infect. Control Hosp. Epidemiol. 2016;37(2):212–214

Information

Type
Concise Communications
Information
Infection Control & Hospital Epidemiology , Volume 37 , Issue 2 , February 2016, pp. 212 - 214
Copyright
© 2015 by The Society for Healthcare Epidemiology of America. All rights reserved 

Antibiograms, which provide data on bacterial resistance patterns, are advocated for antibiotic stewardship. The controversy surrounding the approach to generating antibiograms pertains to avoiding over- or underrepresentation of antibiotic resistance based on isolate inclusion or exclusion.Reference Hindler and Stelling 1 , Reference Bantar, Alcazar and Franco 2 According to current Clinical and Laboratory Standards Institute (CLSI) guidelines, cumulative antibiograms should be prepared by including the first isolate from a given species per patient within a specific time period regardless of culture source. 3 The option of preparing supplemental analyses utilizing variables such as infection site and patient population is presented in the guidelines; however, no guidance is provided on the methods of analysis with regard to time after hospital admission. The purpose of the present study is to determine the effects of time in hospital on antibiograms, considering culture source.

Methods

We included all bacteria from clinical samples at Rambam Health Care Campus, a primary and tertiary, university-affiliated, 900-bed hospital, between January 1, 2009 and December 31, 2014. We included the first isolate per patient based on culture source and antibiotic susceptibilities. We excluded surveillance cultures. We classified the culture sources as follows: blood (group 1); CSF, pleural effusion, peritoneal fluid, abscess and other sterile body fluids (group 2); urine and sputum (group 3); and skin and superficial wounds (group 4).

Susceptibility testing was performed using Vitek 2.0 (Vitek, Valencia CA) and Etest (bioMérieux SA, Marcy l’Etaile, France) based on CLSI criteria. Classifications of resistance were based on the European Centre for Disease Prevention and Control (ECDC) and Centers for Disease Control and Prevention (CDC) definitions for multidrug-resistant (MDR), extensively drug-resistant (XDR) and pandrug-resistant (PDR) bacteria,Reference Magiorakos, Srinivasan and Carey 4 which define resistance categories for Staphylococcus aureus, Enterococcus spp., Enterobacteriaceae, Pseudomonas aeruginosa, and Acinetobacter spp.

We assessed the frequency of resistant bacteria by time in hospital and compared this frequency between the different source groups. Source group 1 (blood) served as the reference for the later comparisons. Comparisons between groups and odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using χ2 tests. A logistic regression analysis was performed to assess association between frequency of MDR phenotype and time from hospital admission. Analyses were conducted using SPSS 21 (IBM, Armonk, NY) and VassarStats (Richard Lowry, Vassar College, Poughkeepsie, NY).

Results

A total of 51,467 patient-unique isolates were analyzed: 4,416 S. aureus; 7,608 Enterococcus spp.; 30,667 Enterobacteriaceae; 5,918 P. aeruginosa; and 2,858 Acinetobacter spp. The percentages of MDR isolates in the respective groups were 33.1%, 2.3%, 36.2%, 16.7%, and 66.3% overall. Resistance increased linearly with time in hospital. The association was weakest for enterococci (OR for MDR increased by 1.004 per day, P=.008) and strongest for Acinetobacter spp. (OR increased by 1.017 per day, P<.001). The optimal dichotomizing time point for resistance was hospital day 7 (Figure 1a). For all groups apart from P. aeruginosa, the increase in resistance thereafter was minimal. Escherichia coli, K. pneumoniae, Proteus mirabilis, and E. cloacae together constituted the majority of Enterobacteriaceae isolates (48.3%, 18.7%, 7.2%, and 6.9%, respectively) and showed the same pattern of greatest increase in resistance at hospital day 7 although levels of resistance at baseline and each time point varied (Figure 1b).

Figure 1

Rates of MDR isolates by time from hospitalization. Arrow points indicate the optimal time point for susceptibility dichotomization. MDR, multidrug resistant.

We compared the culture source group susceptibilities for community-associated isolates (taken within 48 hours of hospital admission), early hospital-acquired isolates (taken between 3 and 7 days of hospital admission), and late hospital-acquired isolates (taken >7 days from hospital admission). The mean times from hospital admission to culture-taken-date were similar for the different sample sources in the late hospital-acquired subgroup for all bacteria apart from Enterococcus spp., allowing a comparison between sample sources in late hospital-acquired isolates. For Enterococcus spp., cultures from urine or sputum and sterile body fluids apart from blood were taken on average after a longer interval from admission than blood (blood, 37.8 days; sterile body fluids, 50.7 days [P=.005]; and urine/sputum, 48.2 days [P=.009]).

Enterobacteriaceae blood isolates were consistently more resistant than isolates from other culture sources (MDR organisms accounted for 48.1%, 41.6%, 34.6%, and 31.1% of groups 1, 2, 3 and 4; OR 0.77, 0.57 and 0.49 for comparison with group 1, respectively, P<.001 for all). This finding remained in analyses of time from hospitalization (Figure 2a–c). Acinetobacter species were the most resistant organisms overall and the only bacterial group in which blood isolates were significantly less resistant than isolates cultured from other sources. MDR organisms accounted for 59.3%, 71.1%, and 74.0% of isolates in groups 1, 2, and 3, respectively (P=.007 and P<.001 for groups 2 and 3 vs group 1). However, isolates from wound cultures displayed less resistance (44.6%; OR, 0.55; 95% CI, 0.43–0.71; P<.001) than blood isolates. These findings were consistent in analyses of time in hospital. S. aureus exhibited a dichotomy between more resistant isolates from blood and sputum or urine (MDR organisms accounted for 42.9% and 42.8%, respectively) and less-resistant isolates from other sterile fluids and wound cultures (27.4% and 21.7% MDR, P<.001 for both). Although the difference in antibiotic susceptibility between blood and other sterile fluids disappeared with time in hospital, wound-site isolates remained significantly more sensitive than blood isolates in the late hospital-acquired subgroup. Pseudomonas and Enterococcus displayed minimal variation in susceptibility based on culture site, a finding consistent in subgroup analyses by time from hospitalization.

Figure 2

Resistance rates by time from hospital admission and culture source. MDR, multidrug resistant.

Discussion

Bacterial resistance increases continuously with time in hospital, eventually plateauing. However, when generating an antibiogram, discrete time points need to be defined because reporting day-by-day susceptibility results relies on too few isolates and provides too much information. We found the time point of >7 days in hospital to be the optimal discrete time point rather than the 48-hour time point defining hospital acquisition.Reference Garner, Jarvis, Emori, Horan and Hughes 5 , Reference Cardoso, Almeida and Friedman 6 Overall in our analysis, relying exclusively on blood isolates for antibiogram generation would have resulted in overestimation of resistance. This held true for all time points from hospitalization.

Consistent with our results, previous studies have found that rates of antibiotic susceptibility were higher among community-associated infections than among nosocomial infections and that rates of susceptibility decreased as time from hospitalization increased for some antibiotics and bacteria but not for others.Reference Kuster, Ruef and Zbinden 7 – Reference Draper, Farland, Heidel, May and Suda 9 A possible explanation for the dichotomy we found at hospital day 7 rather than at 48 hours (as previously described) is that the population at admission represents a mixture of truly community-acquired and healthcare-associated infections and that the ratio between them might have changed over the years. When generating automatic antibiograms, the precise source of acquisition for community-onset isolates is usually unknown. This raises the question of whether the standard use of 48 hours from admission is an accurate cutoff.

As a single-center study, the number of isolates in our study was limited, although our main findings achieved statistical significance with a narrow margin of error. Lack of data on resistance mechanisms makes it unclear whether our findings would remain relevant in locations where the epidemiology or genotypes of resistance are different. Our laboratory reports antibiotic breakpoints using CLSI criteria. The alternative use of European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines has been shown to increase reporting of resistant isolates, and the effect this might have on our findings is uncertain.Reference Hombach, Bloemberg and Böttger 10

We conclude that antibiograms should include a patient’s first isolates of bacterial species, regardless of culture source. If a finer classification cannot be performed, dichotomizing isolates to those obtained in the first week and thereafter will best represent resistance epidemiology.

Acknowledgments

Financial support. Part of this work was funded by the EU-project AIDA (grant no. Health-F3-2011-278348) to YD.

Potential conflicts of interest. All authors report no conflicts of interest relevant to this article.

Footnotes

PREVIOUS PRESENTATION. The data from this article was presented as a poster at the 2015 Conference of the Israeli Society of Infectious Diseases.

References

1. Hindler, JF, Stelling, J. Analysis and presentation of cumulative antibiograms: a new consensus guideline from the Clinical and Laboratory Standards Institute. Clin Infect Dis 2007;44:867–873.Google Scholar
2. Bantar, C, Alcazar, G, Franco, D, et al. Are laboratory-based antibiograms reliable to guide the selection of empirical antimicrobial treatments in patients with hospital-acquired infections? J Antimicrob Chemother 2007;59:140–143.CrossRefGoogle ScholarPubMed
3. Analysis and Presentation of Cumulative Antimicrobial Susceptibility Test Data. 3rd ed. (CLSI publication M39-A3). Wayne, PA: Clinical and Laboratory Standards Institute, 2009.Google Scholar
4. Magiorakos, AP, Srinivasan, A, Carey, RB, et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect 2012;18:268–281.Google Scholar
5. Garner, JS, Jarvis, WR, Emori, TG, Horan, TC, Hughes, JM. CDC definitions for nosocomial infections, 1988. Am J Infect Control 1988;16:128–140.Google Scholar
6. Cardoso, T, Almeida, M, Friedman, ND, et al. Classification of healthcare-associated infection: a systematic review 10 years after the first proposal. BMC Med 2014;12:40.CrossRefGoogle ScholarPubMed
7. Kuster, SP, Ruef, C, Zbinden, R, et al. Stratification of cumulative antibiograms in hospitals for hospital unit, specimen type, isolate sequence and duration of hospital stay. J Antimicrob Chemother 2008;62:1451–1461.Google Scholar
8. Leibovici, L, Schønheyder, H, Pitlik, SD, Samra, Z, Møller, JK. Bacteraemia caused by hospital-type micro-organisms during hospital stay. J Hosp Infect 2000;44:31–36.CrossRefGoogle ScholarPubMed
9. Draper, HM, Farland, JB, Heidel, RE, May, LS, Suda, KJ. Comparison of bacteria isolated from emergency department patients versus hospitalized patients. Am J Health Syst Pharm 2013;70:2124–2128.Google Scholar
10. Hombach, M, Bloemberg, GV, Böttger, EC. Effects of clinical breakpoint changes in CLSI guidelines 2010/11 and EUCAST guidelines 2011 on antibiotic susceptibility test reporting of Gram-negative bacilli. J Antimicrob Chemother 2012;67:622–632.Google Scholar
Figure 0

Figure 1Rates of MDR isolates by time from hospitalization. Arrow points indicate the optimal time point for susceptibility dichotomization. MDR, multidrug resistant.

Figure 1

Figure 2Resistance rates by time from hospital admission and culture source. MDR, multidrug resistant.