University of Groningen
Antimicrobial Resistance Differs Significantly Between Hospitals
Berends, Matthijs S.; Meijer, Bart; Ott, Alewijn; Roelofs, Yvonne; Arends, Jan; Hendrix, Ron; Glasner, Corinna; Friedrich, Alexander
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Early version, also known as pre-print
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Berends, M. S., Meijer, B., Ott, A., Roelofs, Y., Arends, J., Hendrix, R., Glasner, C., & Friedrich, A. (2018). Antimicrobial Resistance Differs Significantly Between Hospitals: The Advantage of a Regional Analysis of Blood Culture Isolates. Poster session presented at European Congress of Clinical Microbiology and Infectious Diseases (ECCMID) 2018, Madrid, Spain.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Antimicrobial Resistance Differs Significantly Between Hospitals:
The Advantage of a Regional Analysis of Blood Culture Isolates
I N T R O D U C T I O N
R E S U LT S
Berends MS
1,3, Meijer BC
1, Ott A
1, Roelofs YC
2, Arends JP
3, Hendrix MGR
3, Glasner C
3, Friedrich AW
3Contact: m.berends@certe.nl
1 Department of Medical Microbiology, Certe Medical Diagnostics & Advice, Groningen, the Netherlands
2 Izore Centre for Infectious Diseases Friesland, Leeuwarden, the Netherlands
3 Department of Medical Microbiology and Infection Prevention, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands
Sepsis is a life-threatening syndrome caused by a dysregulated host response to infection. In case of a suspected case of sepsis empirical antimicrobial treatment is needed. The choice of empiric sepsis treatment is primarily an “informed guess” based on national guidelines, but also on knowledge of local and regional resistance of typically isolated microorganisms. Therapy based on guidelines works best when there are no major differences in antimicrobial resistance. For the determination of the most successful therapy, intrinsic resistance of isolates must also be analysed.
M AT E R I A L S / M E T H O D S
In this study, we analysed bug-drug combinations of all bacterial blood culture isolates from patients from all 14 secondary and tertiary care hospitals in the Northern Netherlands detected in the last 15 years (from 2002 to 2016). The Netherlands consists of 12 provinces, of which the Northern three (Friesland, Groningen and Drenthe) contain a total of 14 hospitals all within a 30 km range of the other nearest hospital. Only first isolates were included. The isolates were selected using a novel selection algorithm, based on the M39-A4 guideline of CLSI, taking into account isolate-specific resistance to key antibiotics which were chosen based on the genus and Gram stain. Subsequently, we compared the resistance of blood culture isolates between hospitals using a full-region approach, by grouping the hospital locations by province (Friesland, Groningen, Drenthe). This allowed for reliable and anonymous analysis of antimicrobial resistance for any bug/drug combination. To calculate differences in antimicrobial resistance, the number of susceptible isolates (“S”) for a specific antibiotic was divided by the total number of test results for that specific antibiotic (“S”, “I” or “R”). Then results were compared with a G-test of goodness-of-fit. The
D I S C U S S I O N
Using the novel selection algorithm, more than 10% more bacterial strains were included than with the application of the CLSI guideline, resulting in a total of about 113,000 instead of 98,000 isolates. For the same microbial species, we found significant differences in antimicrobial resistance between hospitals. However, as the years went by, the differences decreased. For instance, figure 2 shows all isolates tested for amoxicillin, illustrating that the resistance to amoxicillin varied greatly between 2002 and 2010 in the three provinces (p < 0.05), but since 2011 this difference diminished (p > 0.05, yellow bars).
Geel: c3 m5 y95 k0 Groen: c51 m24 y93 k8 Magenta: c6 m83 y17 k0 Blauw: c90 m48 y33 k18 Roze: c14 m33 y0 k0 Geel: c3 m5 y95 k0 Groen: c51 m24 y93 k8 Magenta: c6 m83 y17 k0 Blauw: c90 m48 y33 k18 Roze: c14 m33 y0 k0 73 3190 937 31 690 231 38 990 426 53 1307 470 86 2606 1259 181 4047 2230 294 7973 5187 248 7221 5594 152 5853 5437 6 793 808 4 7 0% 25% 50% 75% 100% 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99 100-109 Age group Percen ta ge Gram negative Gram positive Other (e.g. Fungi)
Without coagulase negative Staphylococcus
Age vs. proportion of Gram stain
135 4757 937 52 967 231 66 1426 426 77 1860 470 124 3832 1259 242 5643 2230 385 10973 5187 316 10161 5594 214 8501 5437 15 1048 808 7 7 0% 25% 50% 75% 100% 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99 100-109 Age group Percen ta ge Gram negative Gram positive Other (e.g. Fungi)
Age vs. proportion of Gram stain
Empiric therapy for patients with septicaemia might be improved in single hospitals, by analysing the regional inter-institutional epidemiology of bacteriaemic isolates comprising the correspondent resistance. It has been shown that there are major epidemiological differences in microbial resistance between provinces within our region. This allows for a more specific empiric therapy of first choice using local hospital protocols, rather than using general national guidelines. We recommend analysing bacterial surveillance and prevalence of resistance in a full-region approach instead of single hospitals or the national level, using the novel selection algorithm to include more relevant isolates and improve reliability of the analysis.
Figure 1: Map of the Northern part of the Netherlands, showing all 14 hospitals included in the study. All hospitals
are localised within a 30 km radius from another hospital in the study (mean nearest: 17.4 km, sd: 6.5 km).
Figure 4a and 4b: Age group (per 10 years) compared to the prevalence of Gram stains of first weighted isolates. Figure 4a includes
all microorganisms, in figure 4b all coagulase negative Staphylococci were excluded.
Table 1: Resistance percentages (both intrinsic [according to EUCAST] and measured) per microorganisms group and antibiotic for
the whole region. A gray field means that less than 30 isolates were available for that bug/drug combination. Only first weighted isolates were analysed (n = 113,332).
0% / 0 5% / 0.05 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Year % R esist ant / p v alue of G -te st of g oodne ss-of-fit
(yellow: p value) Drenthe Friesland Groningen
First weighted isolates, n = 28553
Resistance of Amoxicillin of blood culture isolates per province
Gram Group amox amcl cfur cfta cftr mero peni pita gent tobr trim trsu cipr vanc teic
Negative E. coli 52.5% 15.0% 8.0% 3.8% 3.2% 0.0% 100.0% 6.2% 5.1% 5.0% 40.5% 32.2% 14.1% 100.0% 100.0%
Klebsiella spp. 100.0% 8.2% 8.0% 3.9% 2.6% 0.1% 100.0% 6.5% 3.1% 3.2% 15.9% 10.4% 3.8% 100.0% 100.0%
Proteus spp. 31.2% 6.8% 7.1% 1.0% 2.0% 0.0% 100.0% 1.3% 6.7% 2.9% 45.8% 31.4% 4.1% 100.0% 100.0%
Pseudomonas spp. 100.0% 99.8% 100.0% 5.0% 100.0% 2.3% 100.0% 6.9% 2.3% 1.0% 100.0% 97.4% 6.9% 100.0% 100.0%
Other Gram negatives 87.9% 51.3% 74.1% 40.7% 31.0% 4.4% 90.0% 42.1% 14.8% 15.8% 24.3% 7.0% 6.2% 99.6% 100.0%
Positive Enterococcus spp. 29.9% 29.5% 99.8% 100.0% 100.0% 87.9% 57.1% 98.5% 98.5% 100.0% 99.8% 99.6% 69.7% 2.2% 0.0%
S. aureus 87.8% 0.0% 0.0% 100.0% 79.5% 0.3% 2.2% 12.7% 3.4% 7.6% 0.0% 0.1%
MRSA 100.0% 100.0% 30.8% 25.6% 36.8% 0.0%
S. non-aureus (esp. CNS) 98.3% 46.8% 39.3% 100.0% 81.5% 73.5% 19.7% 46.1% 54.7% 21.8% 30.8% 0.1% 25.3%
Streptococcus spp. 0.8% 0.9% 1.0% 77.8% 0.5% 0.0% 1.6% 100.0% 100.0% 20.5% 5.1% 4.4% 0.1% 0.8%
Other Gram positives 32.1% 9.1% 27.6% 60.7% 40.1% 8.2% 22.0% 16.4% 7.3% 18.0% 63.0% 13.2% 7.6% 15.3% 73.9%
The distribution of Gram-positive and Gram-negative bacteria was related to the age of the patients. Older patients showed a higher prevalence of septicaemia caused by Gram-negative bacteria than younger patients (figure 4a). Figure 4b shows this difference without all coagulase negative Staphylococci.
R E S U LT S ( C O N T. )
Figure 2: Resistance of amoxicillin in blood culture isolates between 2002 and 2016 of all 14 hospitals, divided by province. Yellow
bars reflect the p value as a result of G-tests of goodness-of-fit. A total of 28,553 first weighted isolates were analysed for amoxicillin.
Notably, the prevalence of Quinolone and Aminoglycoside Resistant Enterobacteriaceae (QARE) rose from 0.3% in 2002 to 5% in tertiary care in 2016 (figure 3). In particular, prevalence in tertiary care increased since 2008. Nevertheless, prevalence of QARE remains lower than in neighbouring countries.
Figure 3: Prevalence of QARE between 2012 and 2016, divided by type of hospital care.
To determine the best empiric therapy for sepsis, an overview of antibiotics that can be used are displayed in tabel 1, with their according resistance percentages.
0.0% 0.3% 0.6% 0.6% 0.0% 1.5% 0.0% 1.2% 0.0% 0.8% 0.0% 1.4% 1.4%1.5% 5.2% 1.2% 4.1% 2.1% 3.0% 2.9% 5.6% 2.1% 4.8% 2.5% 4.6% 2.1% 2.9% 2.2% 4.9% 2.3% 0% 2% 4% 6% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Year % QARE / f irs t wei gh ted Ent er obact er iace ae Type of care: Secondary Care Tertiary Care
First weighted isolates from blood cultures in the last 15 years (total of these from the family of Enterobacteriaceae: n = 18646)
