Epidemiological Analysis of Antimicrobial Resistance in Staphylococcus epidermidis in Scotland,2014

Abstract

Aims:

This study aimed to investigate recent national surveillance trends in Staphylococcus epidermidis antibiotic resistance in Scotland and to draw conclusions on the potential clinical and public health impact of multidrug-resistant isolates.

Results:

Resistance in S. epidermidis isolates to individual agents was broadly stable over the past 5 years. Isolates from sterile sites, and therefore those most likely to be associated with clinical infection, were found to be more resistant to the majority of reported agents, than isolates from nonsterile sites. Increased resistance to a number of important antibiotics was observed in rifampicin-, vancomycin-, and daptomycin-resistant isolates, suggesting limited treatment options for infections caused by these isolates.

Conclusions:

Although S. epidermidis resistance to individual agents has been broadly stable over the past 5 years nationally, of particular concern is the association of multidrug resistance in rifampicin-resistant isolates, which has been reported elsewhere. This has the potential to result in treatment failures in significant device-related infections.

Introduction

Staphylococcus epidermidis is one of the predominant bacterial species associated with skin colonization in healthy individuals.^1–3 A propensity to form biofilms enables the organism to contaminate a wide variety of surfaces, making it an important opportunistic pathogen. S. epidermidis is associated with a number of infections and is one of the leading causes of medical device-associated infections, particularly prosthetic valve endocarditis, prosthetic joint, and central line-associated bloodstream infections.^3–8 Because of the nature of these infections, clinical management is typically complex and often requires removal of the contaminated device, alongside antibiotic therapy.^5,9–12

Coagulase-negative staphylococci, including S. epidermidis, are frequently associated with resistance to multiple antibiotics.^1,3,8,13,14 Furthermore, isolates from cases of clinical infection are typically more resistant than those associated with commensal carriage.^4,6 A European Centre for Disease Prevention and Control (ECDC) Rapid Risk Assessment (RRA) from November 2018¹⁵ highlighted that several endemic health care-associated multidrug-resistant strains have been identified globally. A number of European health care outbreaks associated with multidrug-resistant S. epidermidis have also been described.^8,13,16

Isoxazolyl penicillin, typically flucloxacillin (alone or in combination with other antibiotics), is recommended as first-line therapy for a number of medical device-associated infections caused by staphylococci, including endocarditis^9,10 and prosthetic joint infections.^11,12 Isoxazolyl penicillin resistance is common in coagulase-negative staphylococci.^3,6–8 In addition, the transfer of the mecA gene from S. epidermidis to Staphylococcus aureus has been demonstrated and may potentially lead to increased occurrence of methicillin-resistant S. aureus, an important pathogen, particularly in health care settings.¹⁷

Because of high levels of resistance to penicillins, and contraindications for use in documented penicillin allergy, vancomycin is typically indicated for the treatment of S. epidermidis infections.¹ Vancomycin resistance is considered to be rare in this species.^1,3,8,14 Although the clinical significance of heteroresistance to vancomycin is not well established, this has been described and worryingly, cannot be detected by standard susceptibility testing methods.^7,18,19

Scotland has one of the highest rates of vancomycin-resistant enterococci in Europe.²⁰ Although considered very rare and appearing to be limited to S. aureus, plasmid-mediated transfer of vancomycin resistance from enterococci to staphylococci has been demonstrated.²¹ Global emergence of transferable vancomycin resistance in S. aureus and coagulase-negative staphylococci would have significant consequences for the treatment of infections caused by these species. It is recognized that in Scotland, hospital prescribing of both flucloxacillin and glycopeptides has increased annually over the past 5 years,²⁰ which may further drive the development of resistance.

Other antibiotics including rifampicin, daptomycin, and linezolid are also indicated as alternative agents or as part of combination therapy.^9–12 Rifampicin resistance has been well documented, and the agent should therefore always be used as part of combination therapy.^5,7,19

Recently, both rifampicin resistance and glycopeptide heteroresistance in S. epidermidis isolates was identified among two common multilocus sequence types, ST2 and ST23, found globally. Furthermore, it has been described that rifampicin resistance may act as a marker for identifying these multidrug-resistant clones.¹⁹ Resistance to daptomycin and linezolid occurs rarely,^3–5,7,14 although a number of outbreaks related to linezolid-resistant S. epidermidis infections have recently been described.^8,11,16 Increasing rates of resistance in this species will result in even more limited therapeutic options for infections that are already difficult to manage.

The ECDC RRA¹⁵ outlined the need for further epidemiological studies to determine geographical presence of multidrug-resistant S. epidermidis in European countries. In Scotland, national microbiology results are captured by the Electronic Communication of Surveillance in Scotland (ECOSS) system and interrogated by Health Protection Scotland (HPS).

In light of the findings from the ECDC RRA,¹⁵ the purpose of this study was to describe the prevalence of resistance in S. epidermidis in Scotland, with particular focus on resistance to first-line agents used for the treatment of important infections.

Materials and Methods

Data sources

S. epidermidis identification and susceptibility results were obtained from diagnostic laboratories in Scotland represented by all 14 Health Boards nationally. It is considered that the majority of results were obtained by use of VITEK-2 analyzers and reported to each laboratory's Laboratory Information Management System. Limitations associated with use of the VITEK-2 system for susceptibility testing are outlined in the Discussion section. Susceptibility testing results were considered to have been interpreted by laboratories by use of the European Committee on Antimicrobial Susceptibility Testing (EUCAST) methodology.

Data analysis

Diagnostic laboratory results were captured directly into the ECOSS system. Data were collected for a period of 60 months, during the years 2014–2018. For the purpose of this report, S. epidermidis isolates from unique patients obtained within 14 days of the original sample were treated as the same episode and duplicates were deleted before analysis. Isolates from all specimen sites, nonsterile sites, and sterile sites were assessed separately to investigate potential differences in resistance, associated with sample type. Multidrug resistance was defined as resistance to two or more agents. Isolates found to have intermediate resistance were not assessed. Of note, although submissions from Scottish laboratories to ECOSS for results from sterile samples are considered reliable, results from nonsterile sites may in some instances be less reliable.

Statistical analysis

Statistical analysis was carried out using R version 3.5.1 (2018-07-02). Likelihood ratio chi-square tests were used to assess trends in resistance and Wald tests were used to determine percentage change over the assessed time period.

For comparisons of resistance between sterile versus nonsterile site isolates and vancomycin-, daptomycin-, or rifampicin-susceptible isolates versus isolates resistant to these agents, the chi-square or Fisher's exact tests were used as appropriate and p ≤ 0.05 was considered to be statistically significant.

Ethical approval

As only aggregated demographic data are provided and the study analyzed microbiological data, ethical approval was not required.

Results

Descriptive epidemiology

Resistance data relating to a total of 16,110 isolates over the 5-year period (2014–2018) were extracted for analysis. The majority, 10,475 (65.0%) were from sterile site specimens and of these, 6,836 (65.3%) were from blood cultures.

The majority of isolates, 9,185 (57.0%) were from male patients and 1,880 (11.7%), 7,861 (48.8%), and 6,369 (39.5%) were from patients ≤16, 17–64, and ≥65 years of age, respectively.

Resistance to commonly used antibiotics and trends over time

The proportion of isolates resistant to various agents over the 5-year period and 5-year trends in resistance to each antibiotic were assessed for isolates from all sample types (Table 1) and those from sterile site samples (Table 2).

Table 1.

Number of Staphylococcus epidermidis Susceptibility Results Reported for Each Antibiotic, Percentage Resistance to Each Antibiotic by Year and 5-Year Percentage Change Trend for Isolates from All Sample Types

All isolates
ABX	2014		2015		2016		2017		2018		5 year % change
ABX	n	% R	n	% R	n	% R	n	% R	n	% R	5 year % change
CHL	2,364	3.6	2,870	3.9	3,194	3.8	3,091	4.6	3,003	4.3	5.4
CIP	2,500	47.5	2,971	45.6	3,356	47.8	3,236	48.2	3,074	48.4	1.9
CLR	1,761	67.6	1,960	67.8	2,223	69.5	2,163	69.6	2,098	69.0	2.1
CLI	2,508	51.9	2,946	51.3	3,267	51.1	3,158	50.9	2,980	51.4	−0.5
DAP	2,096	1.5	2,726	1.9	3,157	1.3	3,073	1.5	2,990	0.9	−10.8
DOX	1,324	47.8	1,811	42.8	2,312	38.7	2,276	37.4	2,316	37.2	−9.8^a
ERY	1,849	66.0	1,641	63.1	1,765	63.3	1,642	63.2	1,510	65.2	−1.0
FLX	2,535	67.5	2,582	66.8	2,968	65.7	2,896	65.3	2,810	64.8	−3.0
FA	2,489	61.3	2,951	61.2	3,321	61.1	3,216	60.7	3,069	60.6	−0.8
GEN	2,724	50.9	3,014	52.0	3,369	51.9	3,263	53.5	3,132	51.9	1.4
LZD	2,361	<0.1	2,832	<0.1	3,165	<0.1	3,095	<0.1	3,006	0.1	N/A
MUP	2,429	11.4	2,913	10.8	3,276	9.2	3,087	10.3	2,971	10.8	−1.5
PEN	2,734	90.6	2,875	90.6	3,145	91.1	3,033	91.5	3,011	90.8	1.3
RIF	2,512	11.3	2,928	11.5	3,283	9.5	3,231	9.1	3,063	8.9	−7.8^a
TMP	2,438	59.6	2,850	58.8	3,245	58.1	3,109	60.8	2,938	59.3	0.7
VAN	2,656	0.5	2,833	0.5	3,176	0.3	3,107	0.5	2,980	0.3	−6.4

Statistically significant (p < 0.001) change observed. The Bonferroni adjustment was applied to p-values for all antibiotics to correct for multiple testing.

ABX, antibiotic; CHL, chloramphenicol; CIP, ciprofloxacin; CLR, clarithromycin; CLI, clindamycin; DAP, daptomycin; DOX, doxycycline; ERY, erythromycin; FLX, flucloxacillin; FA, fusidic Acid; GEN, gentamicin; LZD, linezolid; MUP, mupirocin; PEN, penicillin; RIF, rifampicin; TMP, trimethoprim; VAN, vancomycin.

Table 2.

Sterile site isolates
ABX	2014		2015		2016		2017		2018		5 year change (%)
ABX	n	% R	n	% R	n	% R	n	% R	n	% R	5 year change (%)
CHL	1,565	3.7	1,890	4.0	2,012	4.2	1,945	4.8	1,915	5.0	8.7
CIP	1,665	50.2	1,949	48.0	2,100	51.0	2,043	50.2	1,965	51.9	2.3
CLR	1,190	71.7	1,314	69.5	1,441	74.5	1,433	72.5	1,383	72.9	2.8
CLI	1,690	54.9	1,944	54.1	2,058	55.7	2,005	54.6	1,895	54.6	0.0
DAP	1,421	1.2	1,810	1.5	1,987	1.4	1,942	1.5	1,910	0.9	−5.1
DOX	898	48.1	1,244	41.3	1,567	38.5	1,573	37.4	1,526	38.1	−8.6^a
ERY	1,274	68.1	1,047	64.1	1,052	66.2	966	65.6	938	68.3	0.4
FLX	1,740	71.4	1,699	70.8	1,875	70.6	1,862	69.1	1,791	70.0	−2.2
FA	1,667	64.3	1,950	63.3	2,095	62.2	2,045	61.6	1,969	61.4	−3.2
GEN	1,870	53.2	1,991	55.0	2,119	56.4	2,073	57.2	2,006	56.7	3.8
LZD	1,574	0.1	1,871	0.1	2,001	<0.1	1,950	0.1	1,918	<0.1	N/A
MUP	1,639	12.0	1,929	10.9	2,077	10.8	1,953	11.7	1,917	12.0	0.9
PEN	1,841	91.5	1,911	91.9	2,018	92.1	1,961	92.0	1,947	92.4	2.6
RIF	1,703	11.8	1,948	12.0	2,082	10.2	2,064	9.4	1,967	9.9	−6.6
TMP	1,659	62.2	1,935	61.5	2,078	61.4	2,006	62.2	1,927	61.7	−0.1
VAN	1,846	0.3	1,946	0.5	2,073	0.4	2,028	0.6	1,973	0.2	N/A

Statistically significant (p < 0.001) change observed. The Bonferroni adjustment was applied to p-values for all antibiotics to correct for multiple testing.

Resistance to the majority of antibiotics has remained broadly stable over the assessed time period as observed in Tables 2 and 3. In addition, significant decreases in resistance trends over the 5-year period were observed for doxycycline and rifampicin. Only the decrease in doxycycline resistance in isolates from all sample types showed a linear trend. The decrease in rifampicin resistance could only be observed when assessing isolates from all sample types and was no longer significant when considering sterile site isolates only.

Table 3.

Cumulative 5-Year Resistance in Staphylococcus epidermidis Sterile and Nonsterile Site Isolates

ABX	Sterile site isolates, n (% R)	Nonsterile site isolates, n (% R)	p
CHL	9,327 (4.4)	5,195 (3.5)	0.009
CIP	9,722 (50.3)	5,415 (42.6)	<0.001
CLR	6,761 (72.3)	3,444 (61.8)	<0.001
CLI	9,592 (56.8)	5,079 (46.6)	<0.001
DAP	9,070 (1.3)	4,972 (1.5)	NS
DOX	5,583 (48.7)	2,760 (46.9)	NS
ERY	5,277 (66.5)	3,130 (60.3)	<0.001
FLX	8,967 (70.3)	4,824 (57.9)	<0.001
FA	9,726 (62.5)	5,320 (58.2)	<0.001
GEN	10,059 (55.8)	5,443 (45.3)	<0.001
LZD	9,314 (0.04)	5,145 (0.04)	NS
MUP	9,317 (11.7)	5,058 (8.8)	<0.001
PEN	9,678 (92.0)	5,120 (88.9)	<0.001
RIF	9,750 (10.6)	5,249 (8.8)	<0.001
TMP	9,593 (61.9)	4,969 (54.6)	<0.001
VAN	9,866 (0.4)	4,886 (0.5)	NS

NS, not significant.

Comparison of resistance in isolates from sterile versus nonsterile sites

Resistance to individual agents in isolates from sterile sites was compared with resistance in isolates from nonsterile sites cumulatively for the 5-year period. The majority of isolates from sterile sites were significantly more resistant, with the exception of resistance to doxycycline, daptomycin, vancomycin, and linezolid, which was comparable between the two groups (Table 3).

Assessment of multidrug resistance in rifampicin-, vancomycin-, and daptomycin-resistant isolates

Resistance to individual agents in rifampicin-, vancomycin-, and daptomycin–resistant isolates was assessed cumulatively for the 5-year period, in comparison with isolates reported as being susceptible to each of these three antibiotics.

As given in Table 4, isolates that were resistant to rifampicin were typically found to also be significantly more resistant to other agents, in comparison with rifampicin-susceptible isolates, over the assessed time period. Exceptions included chloramphenicol and mupirocin resistance, for which no significant difference was observed between the two groups. Furthermore, conversely, ciprofloxacin resistance was significantly higher in rifampicin-susceptible isolates.

Table 4.

Multidrug Resistance in Rifampicin-Resistant Isolates Compared with Rifampicin-Susceptible Isolates

ABX	RIF resistant, n (% R)	RIF susceptible, n (% R)	p
CHL	1,439 (4.8)	12,834 (4.0)	NS
CIP	1,487 (41.1)	13,336 (48.1)	<0.001
CLR	1,052 (76.0)	8,897 (67.9)	<0.001
CLI	1,475 (87.5)	12,615 (49.0)	<0.001
DAP	1,401 (2.7)	12,452 (1.2)	<0.001
DOX	583 (71.7)	7,613 (46.6)	<0.001
ERY	675 (70.2)	7,236 (63.8)	<0.001
FLX	1,369 (97.6)	11,975 (62.7)	<0.001
FA	1,492 (92.8)	13,353 (57.4)	<0.001
GEN	1,495 (89.2)	13,466 (48.0)	<0.001
LZD	1,449 (0.3)	12,876 (0.02)	0.0013
MUP	1,254 (9.3)	13,014 (10.7)	NS
PEN	1,455 (99.5)	12,943 (90.0)	<0.001
TMP	1,470 (91.7)	12,710 (55.6)	<0.001
VAN	1,478 (0.8)	12,867 (0.4)	0.03

In addition, vancomycin-resistant isolates were found to be significantly more resistant to daptomycin, linezolid, and rifampicin than vancomycin-susceptible isolates. In contrast, vancomycin-susceptible isolates were significantly more resistant to trimethoprim. Clarithromycin, ciprofloxacin, clindamycin, doxycycline, erythromycin, flucloxacillin, fusidic acid, mupirocin, and penicillin resistance was comparable between vancomycin-resistant and -susceptible isolates (Table 5).

Table 5.

Multidrug Resistance in Vancomycin-Resistant Isolates Compared with Vancomycin-Susceptible Isolates

ABX	VAN resistant, n (% R)	VAN susceptible, n (% R)	p
CHL	54 (3.7)	13,715 (4.1)	NS
CIP	59 (42.4)	14,235 (47.6)	NS
CLR	41 (61.0)	9,376 (69.5)	NS
CLI	55 (54.5)	13,541 (53.6)	NS
DAP	42 (100)	13,272 (1.2)	<0.001
DOX	34 (41.2)	8,101 (48.6)	NS
ERY	34 (58.8)	8,111 (64.5)	NS
FLX	53 (54.7)	12,794 (67.3)	NS
FA	59 (64.4)	14,240 (60.9)	NS
GEN	61 (42.6)	14,416 (53.1)	NS
LZD	56 (5.4)	13,769 (0.02)	<0.001
MUP	53 (13.2)	13,703 (10.7)	NS
PEN	58 (84.5)	14,054 (91.2)	NS
RIF	60 (20.0)	14,285 (10.3)	<0.001
TMP	58 (44.8)	14,233 (59.5)	0.03

Daptomycin-resistant isolates were found to be significantly more resistant to vancomycin and rifampicin in comparison with daptomycin-susceptible isolates. In contrast, daptomycin-susceptible isolates were more resistant to ciprofloxacin, flucloxacillin, and trimethoprim. Chloramphenicol, clarithromycin, clindamycin, doxycycline, erythromycin, fusidic acid, gentamicin, linezolid, mupirocin, and penicillin resistance was comparable between daptomycin-resistant and -susceptible isolates (Table 6).

Table 6.

Multidrug Resistance in Daptomycin-Resistant Isolates Compared with Daptomycin-Susceptible Isolates

ABX	DAP resistant, n (% R)	DAP susceptible, n (% R)	p
CHL	195 (6.2)	13,761 (4.0)	NS
CIP	196 (34.7)	13,816 (47.8)	<0.001
CLR	109 (65.1)	9,874 (69.0)	NS
CLI	186 (48.9)	13,101 (53.7)	NS
DOX	86 (45.3)	7,590 (48.9)	NS
ERY	143 (62.9)	6,779 (64.7)	NS
FLX	142 (58.5)	12,343 (66.7)	0.04
FA	197 (61.4)	13,814 (61.4)	NS
GEN	197 (47.2)	13,823 (52.7)	NS
LZD	194 (0.5)	13,710 (0.02)	NS
MUP	180 (10.0)	13,216 (10.4)	NS
PEN	190 (92.1)	13,300 (91.2)	NS
RIF	192 (19.8)	13,661 (10.0)	<0.001
TMP	196 (52.0)	13,179 (60.0)	0.03
VAN	195 (21.5)	13,119 (0.0)	<0.001

Discussion

One of the recommendations from the ECDC RRA indicated that treatment of S. epidermidis infections should be guided by local epidemiological surveillance data and individual antimicrobial susceptibility results.¹⁵

To our knowledge, this is the first retrospective study to describe trends in national S. epidermidis resistance data, associated with a variety of clinical sample types. Over the assessed 5-year time period, resistance to all reported antibiotics was found to be relatively stable. In fact, statistically significant decreasing trends were observed for doxycycline and rifampicin resistance; however, these trends were only observed for doxycycline resistance when assessing sterile site isolates. In addition, resistance to the majority of agents was comparable or lower to that reported in a number of other European studies.^1,3–5,7,14

It is accepted that there is often difficulty in determining the significance of identifying S. epidermidis in clinical samples, as the bacteria is considered both a commensal and an opportunistic pathogen. However, recent studies have indicated that S. epidermidis strains from clinical infections often represent a pathogenic subpopulation that have acquired genetic elements (including resistance genes) that promote infection.^2,22 Therefore, in addition to assessing resistance in isolates from all sample types, resistance in isolates from sterile sites was separately assessed owing to the recognition that these isolates are more likely to be associated with cases of true clinical infection.

Resistance to most antibiotics was significantly higher in sterile site isolates versus nonsterile sites isolates, as has been previously described for S. epidermidis elsewhere.^4,6 This is potentially because of the fact that patients with invasive infections will have been more likely to have experienced longer hospital stays and been exposed to previous antibiotics.

The exceptions to this were doxycycline, daptomycin, vancomycin, and linezolid resistance, which were found to be comparable between sterile and nonsterile site isolates. The reason for no difference in daptomycin, vancomycin, and linezolid resistance in isolates from sterile and nonsterile sites is probably because of the small numbers of resistant isolates reported, which impacts on the analysis of statistical significance.

A recent study described that rifampicin resistance is a potentially useful marker for identifying multidrug-resistant S. epidermidis clones, with a number of these having been recognized globally, including in the United Kingdom. Rifampicin-resistant isolates were more likely to be resistant to multiple antibiotics and in particular; heteroresistance to vancomycin was identified in these isolates owing to specific rpoB mutations. The authors concluded that clinical vigilance is warranted and that current guidelines recommending rifampicin and vancomycin combination therapy for the treatment of staphylococci may not be appropriate.¹⁹

Vancomycin heteroresistance cannot currently be detected in S. epidermidis isolates by diagnostic laboratories in Scotland. This is a concern as glycopeptides, particularly vancomycin, remain the first-line therapy option for the treatment of S. epidermidis infections. In addition, there is a recognized limitation associated with susceptibility testing for teicoplanin using the VITEK 2 system and in practice, if used clinically, diagnostic laboratories confirm these results by use of alternative susceptibility testing methods.

Based on local (unpublished) data from one large testing laboratory in Scotland, confirmed teicoplanin resistance in S. epidermidis is considered to occur in <5% of isolates. It is anticipated that in 2020, new software will be introduced to VITEK 2 machines, which will allow for a national rule set including automatic suppression of teicoplanin susceptibility analysis for S. epidermidis. This will ensure that teicoplanin results will only be reported once confirmed by an alternative testing method and will lead to more accurate reporting of resistance to this agent.

In agreement with Lee et al.,¹⁹ rifampicin resistance in our study was associated with significantly higher resistance to vancomycin, penicillins, macrolides, and aminoglycosides, in addition to the majority of other reported agents. The most notable differences between rifampicin-resistant and -susceptible isolates were associated with clindamycin (87.5% vs. 49.0%), doxycycline (71.7% vs. 46.6%), flucloxacillin (97.6% vs. 62.7%), fusidic acid (92.8% vs. 57.4%), gentamicin (89.2% vs. 48.0%), and trimethoprim resistance (91.7% vs. 55.6%). In contrast, no association was found between rifampicin and ciprofloxacin resistance. Therefore, we agree that rifampicin is a potential important marker for identifying multidrug-resistant isolates.

In this study, we were also interested in assessing whether vancomycin or daptomycin resistance may be associated with increased resistance to typical first-/second-line agents for the treatment of prosthetic valve endocarditis or prosthetic joint infections.^9–12 Vancomycin-resistant isolates were significantly more resistant to daptomycin, linezolid, and rifampicin, than vancomycin-susceptible isolates. In addition, all were daptomycin resistant. Only 5% were linezolid resistant, although this represents 50% of the total linezolid-resistant isolates in the study.

Daptomycin-resistant isolates were significantly more resistant to vancomycin and rifampicin, in comparison with isolates susceptible to daptomycin. Although these results need to be interpreted with caution because of potential bias associated with small numbers of resistant isolates, increased resistance to multiple important agents is worrying. Specifically, all vancomycin-resistant isolates were also found to be resistant to daptomycin and ∼20% of daptomycin-resistant isolates were resistant to vancomycin (compared with 0.1% of all isolates).

Resistance to daptomycin, vancomycin, and linezolid is considered to be rare in coagulase-negative staphylococci, and has been reported in this study as 1.4%, 0.4%, and 0.04%, respectively, for S. epidermidis, over the assessed time period. It is considered that a proportion of resistant results associated with these isolates are likely to have occurred because of unrecognized testing/reporting errors and the true proportion of resistance in Scotland is expected to be even lower than that reported in this study.

Other European studies indicate that resistance to these agents was not reported, although all assessed much smaller numbers of isolates than this study.^5,7,14 It is currently recommended in Scotland that isolates with exceptional resistance phenotypes are sent to a reference laboratory for confirmation of resistance. HPS, the Scottish Reference Laboratory Service, and other experts have recently developed guidance for referral of isolates with rare resistance, with the aim of increasing vigilance and improving testing and reporting of exceptional resistance in S. epidermidis and other organisms.

Because only six isolates were reported as being linezolid resistant over the assessed time period, no analysis of multidrug resistance could be performed. A number of outbreaks associated with linezolid-resistant isolates have been reported.^8,16 In one of these, tedizolid resistance was also described, further limiting treatment options. Isolates were found to be susceptible to ceftobiprole and ceftaroline.⁸ Previous linezolid exposure was considered to play a role in the emergence of resistance.^8,16 Although use of whole genome sequencing was shown to be vital within these outbreak studies, molecular typing of isolates may not always be helpful because of various determinants associated with pathogenicity and resistance.²

A significant strength of our study relates to the large volume of resistance data available for analysis, which encouragingly shows no increase in resistance to any of the reported antibiotics, over the assessed time period. In addition, it is considered that the majority of susceptibility results have arisen owing to use of the same methodology throughout Scotland (use of VITEK-2 analyzers and EUCAST interpretation), which provides standardization and consistency to the data received by the ECOSS system.

Limitations include that only a small amount of directly comparable resistance data is currently available from European studies and a bias in published literature exists for reporting of more resistant isolates, typically owing to assessment of susceptibility in clinical cases of infection or in outbreak situations.

A further limitation around the time of data capture was that the ECOSS system did not routinely capture all positive results, from all sample types. Improvement work is currently ongoing to ensure all positive results are captured and it is expected that only a very small proportion of isolates may have been missed during the period of data capture. Furthermore, the system relies on complex mapping algorithms, and on occasion may have resulted in incorrect classifications in recognizing sterile versus nonsterile sites. Laboratories in Scotland currently do not speciate all coagulase-negative staphylococci from nonsterile sites and do not routinely perform susceptibility testing on these isolates.

Because isolates from nonsterile sites are likely to represent colonization rather than infection, and as described in this study, because sterile site isolates were associated with significantly higher resistance to a number of antibiotics, the result of missing nonsterile site isolate data is not considered to be significant from a national public health perspective. Furthermore, it is not possible to ascertain whether all isolates from sterile site specimens were from cases of true clinical infection, as S. epidermidis is a recognized common culture contaminant.

Nevertheless, the results of this study provide useful information on overall national resistance in isolates obtained from clinical specimens. Multidrug resistance associated with agents other than rifampicin, vancomycin, or daptomycin was not assessed because of limited clinical implications for use of these agents alone or as part of combination therapy. In addition, because ∼65% of isolates were reported as being resistant to flucloxacillin (which equated to ∼10,000 isolates), multidrug resistance associated with this agent was also not investigated. This is broadly comparable with resistance reported from other European studies^3,7 and further exemplifies that alternative agents should be used for the treatment of infections owing to this organism.^1,9–12

In conclusion, this study highlights that resistance in S. epidermidis isolates has been broadly stable over the past 5 years in Scotland. As expected, resistance was higher in isolates from sterile sites, which are more likely to have arisen from cases of clinical infection. Of particular concern is the association of multidrug resistance found in rifampicin-resistant isolates, which we have confirmed and has been reported elsewhere. In addition, methods to identify glycopeptide resistance are currently not sufficiently robust to recognize vancomycin heteroresistance. Increased vigilance and a focus on appropriate antibiotic stewardship are necessary in light of the emergence of multidrug-resistant strains reported globally.

Footnotes

Acknowledgments

The authors acknowledge the Scottish microbiology diagnostic laboratories for submitting results to the Electronic Communication of Surveillance in Scotland System.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received.

References

Trueba

, Garrabe

, Hadef

, Fabre

, Cavallo

J.D

, Tsvetkova

, and Chesneau

. 2006. High prevalence of teicoplanin resistance among Staphylococcus epidermidis strains in a 5-year retrospective study. J. Clin. Microbiol. 44:1922–1923.

Meric

, Mageiros

, Pensar

, Laabei

, Yahara

, Pascoe

, Kittiwan

, Tadee

, Post

, Lamble

, Bowden

, Bray

, Morgenstern

, Jolley

, Maiden

, Feil

, Didelot

, Miragaia

, de Lencastre

, Moriarty

T.F

, Rohde

, Massey

, Mack

, Corander

, and Sheppard

. 2018. Disease-associated genotypes of the commensal skin bacterium Staphylococcus epidermidis. Nat. Commun. 9:5034.

Deplano

, Vandendriessche

, Nonhoff

, Dodemont

, Roisin

, and O

Denis

. 2016. National surveillance of Staphylococcus epidermidis recovered from bloodstream infections in Belgian hospitals. J. Antimicrob. Chemother. 71:1815–1819.

Cherifi

, Byl

, Deplano

, Nonhoff

, Denis

, and Hallin

. 2013. Comparative epidemiology of Staphylococcus epidermidis isolates from patients with catheter-related bacteremia and from healthy volunteers. J. Clin. Microbiol. 51:1541–1547.

Hamad

, Hellmark

, Nilsdotter-Augustinsson

, and Soderquist

. 2015. Antibiotic susceptibility among Staphylococcus epidermidis isolated from prosthetic joint infections, with focus on doxycycline. APMIS. 123:1055–1060.

Cremniter

, Slassi

, Quincampoix

J.C

, Sivadon-Tardy

, Bauer

, Porcher

, Lortat-Jacob

, Piriou

, Judet

, Herrmann

J.L

, Gaillard

J.L

, and Rottman

. 2010. Decreased susceptibility to teicoplanin and vancomycin in coagulase-negative Staphylococci isolated from orthopedic-device-associated infections. J. Clin. Microbiol. 48:1428–1431.

Tevell

, Claesson

, Hellmark

, Soderquist

, and Nilsdotter-Augustinsson

. 2014. Heterogeneous glycopeptide intermediate Staphylococcus epidermidis isolated from prosthetic joint infections. Eur. J. Clin. Microbiol. Infect. Dis. 33:911–917.

Dortet

, Glaser

, Kassis-Chikhani

, Girlich

, Ichai

, Boudon

, Samuel

, Creton

, Imanci

, Bonnin

, Fortineau

, and Naas

. 2018. Long-lasting successful dissemination of resistance to oxazolidinones in MDR Staphylococcus epidermidis clinical isolates in a tertiary care hospital in France. J. Antimicrob. Chemother. 73:41–51.

Gould

F.K.

, Denning

D.W

, Elliott

T.S

, Foweraker

, Perry

J.D

, Prendergast

B.D

, Sandoe

J.A

, Spry

M.J

, and Watkin

R.W.

. 2012. Guidelines for the diagnosis and antibiotic treatment of endocarditis in adults: a report of the Working Party of the British Society for Antimicrobial Chemotherapy. J. Antimicrob. Chemother. 67:269–289.

10.

Habib

, Lancellotti

, Antunes

M.J

, Bongiorni

M.G

, Casalta

J.P

, Del Zotti

, Dulgheru

, El Khoury

, Erba

P.A

, Iung

, and Miro

J.M.

. 2015 ESC guidelines for the management of infective endocarditis: the task force for the management of infective endocarditis of the European Society of Cardiology (ESC) endorsed by: European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Nuclear Medicine (EANM). Eur. Heart J. 21;36:3075–3128.

11.

Osmon

D.R.

, Berbari

E.F

, Berendt

A.R

, Lew

, Zimmerli

, Steckelberg

J.M

, Rao

, Hanssen

, and Wilson

W.R.

. 2013. Diagnosis and management of prosthetic joint infection: clinical practice guidelines by the Infectious Diseases Society of America. Clin. Infect. Dis. 56:e1–e25.

12.

, Renz

, and Trampuz

. Management of periprosthetic joint infection. 2018. Hip. Pelvis. 30:138–146.

13.

O'Connor

, Powell

, Finnegan

, O'Gorman

, Barrett

, Hopkins

K.L

, Pichon

, Hill

, Power

, Woodford

, Coffey

J.C

, Kearns

, O'Connell

, and Dunne

C.P.

. 2015. Incidence, management and outcomes of the first cfr-mediated linezolid-resistant Staphylococcus epidermidis outbreak in a tertiary referral centre in the Republic of Ireland. J. Hosp. Infect. 90:316–321.

14.

Decousser

J.W.

, Desroches

, Bourgeois-Nicolaos

, Potier

, Jehl

, Lina

, Cattoir

, Vandenesh

, and Doucet-Populaire

. 2015. Susceptibility trends including emergence of linezolid resistance among coagulase-negative staphylococci and meticillin-resistant Staphylococcus aureus from invasive infections. Int. J. Antimicrob. Agents. 46:622–630.

15.

European Centre for Disease Prevention and Control. 2018. Multidrug-resistant Staphylococcus epidermidis. Available at www.ecdc.europa.eu/en/publications-data/rapid-risk-assessment-multidrug-resistant-staphylococcus-epidermidis (accessed December 2, 2019).

16.

Lazaris

, Coleman

D.C

, Kearns

A.M

, Pichon

, Kinnevey

P.M

, Earls

M.R

, Boyle

, O'Connell

, Brennan

, and Shore

. 2017. Novel multiresistance cfr plasmids in linezolid-resistant methicillin-resistant Staphylococcus epidermidis and vancomycin-resistant Enterococcus faecium (VRE) from a hospital outbreak: co-location of cfr and optrA in VRE. J. Antimicrob. Chemother. 72:3252–3257.

17.

Hanssen

A.M.

, and Ericson Sollid

J.U.

. 2006. SCC mec in staphylococci: genes on the move. FEMS Immunol. Med. Microbiol. 46:8–20.

18.

Chong

, Quach

, Blanchard

A.C

, Poliquin

P.G

, Golding

G.R

, Laferriere

, and Lévesque

. 2016. Molecular epidemiology of a vancomycin-intermediate heteroresistant Staphylococcus epidermidis outbreak in a neonatal intensive care unit. Antimicrob. Agents Chemother. 60:5673–5681.

19.

Lee

J.Y.

, Monk

I.R

, da Silva

A.G

, Seemann

, Chua

K.Y

, Kearns

, Hill

, Woodford

, Bartels

M.D

, Strommenger

, Laurent

, Dodémont

, Deplano

, Patel

, Larsen

, Korman

T.M

, Stinear

T.P

, and Howden

B.P.

. 2018. Global spread of three multidrug-resistant lineages of Staphylococcus epidermidis. Nat. Microbiol. 3:1175.

20.

Health Protection Scotland. 2019. Scottish One Health Antimicrobial Use and Antimicrobial Resistance Report in 2018. Available at www.hps.scot.nhs.uk/web-resources-container/scottish-one-health-antimicrobial-use-and-antimicrobial-resistance-in-2018 (accessed December 2, 2019).

21.

Sujatha

, and Praharaj

. 2012. Glycopeptide resistance in gram-positive cocci: a review. Interdiscip. Perspect. Infect. Dis. 2012:781679.

22.

Shelburne

S.A.

, Dib

R.W

, Endres

B.T

, Reitzel

, Li

, Kalia

, Sahasrabhojane

, Chaftari

A.M

, Hachem

, Vargas-Cruz

N.S

, Jiang

, Garey

, Fowler

V.G

, Holland

T.L

, Gu

, Miller

, Sakurai

, Arias

C.A

, Aitken

S.L

, Greenberg

D.E

, Kim

, Flores

A.R

, and Raad

. 2020. Whole genome sequencing of Staphylococcus epidermidis bloodstream isolates from a prospective clinical trial reveals that complicated bacteremia is caused by a limited number of closely related sequence types. Clin. Microbiol. Infect. 26:646.e1–646.e8.

Epidemiological Analysis of Antimicrobial Resistance in Staphylococcus epidermidis in Scotland,2014–2018

Abstract

Aims:

Results:

Conclusions:

Introduction

Materials and Methods

Data sources

Data analysis

Statistical analysis

Ethical approval

Results

Descriptive epidemiology

Resistance to commonly used antibiotics and trends over time

Comparison of resistance in isolates from sterile versus nonsterile sites

Assessment of multidrug resistance in rifampicin-, vancomycin-, and daptomycin-resistant isolates

Discussion

Footnotes

Acknowledgments

Disclosure Statement

Funding Information

References