AAC Accepts, published online ahead of print on 16 December 2013 Antimicrob. Agents Chemother. doi:10.1128/AAC.01908-13 Copyright © 2013, American Society for Microbiology. All Rights Reserved.
1
Trends in Antibiotic Resistance in Coagulase Negative Staphylococci, United States, 1999–
2
2012
3 4
Larissa May,a* Eili Y. Klein,b,c#* Richard E. Rothman,b Ramanan Laxminarayanc,d
5 6
Department of Emergency Medicine, George Washington University, Washington, DC, USAa;
7
Department of Emergency Medicine, Johns Hopkins University, Baltimore, MD, USAb; Center
8
for Disease Dynamics, Economics & Policy, Washington, DCc; Princeton Environmental
9
Institute, Princeton University, Princeton, NJ, USAd
10 11
*
These authors contributed equally to this work
12 13
Running Head: Trends in Antibiotic Resistance in CoNS
14 15 16 17 18
# Address correspondence to Eili Y. Klein, [email protected]
19
Abstract
20
Coagulase-negative staphylococci (CoNS) are important bloodstream pathogens typically
21
resistant to multiple antibiotics. Despite concern about increasing resistance, there have been no
22
recent studies describing the national prevalence in CoNS pathogens. We used national
23
resistance data over a period of 13 years (1999–2012) from The Surveillance Network (TSN) to
24
determine prevalence and assess trends in resistance for S. epidermidis, the most common CoNS
25
pathogen, as well as all CoNS pathogens. Over the course of the study period, S. epidermidis
26
resistance to ciprofloxacin and clindamycin increased steadily from 58.3% to 68.4% and from
27
43.4% to 48.5%, respectively. Resistance to levofloxacin increased rapidly from 57.1% in 1999
28
to a high of 78.6% in 2005, followed by a decrease to 68.1% in 2012. Multidrug resistance for
29
CoNS followed a similar pattern, and this rise and small decline in resistance was found to be
30
strongly correlated with levofloxacin prescribing patterns. Resistance patterns were similar for
31
the aggregate of CoNS pathogens. The results from our study demonstrate that antibiotic
32
resistance in CoNS pathogens has increased significantly over the past 13 years. These results
33
are important as CoNS can serve as sentinels for monitoring resistance, and they play a role as a
34
reservoir of resistance genes that can be transmitted to other pathogens. The link between the
35
levofloxacin prescription rate and resistance levels suggest a critical role for reducing
36
inappropriate use of fluoroquinolones and other broad-spectrum antibiotics in both healthcare
37
settings and the community to help curb the reservoir of resistance in these colonizing pathogens.
38
39
Introduction
40
Coagulase-negative staphylococci (CoNS) are normal flora of the human skin and mucosa.
41
Because they have long been considered contaminants rather than true clinical pathogens, there
42
are few systematic studies describing their epidemiology in human infections. Nonetheless,
43
colonizing CoNS pathogens have been reported to be responsible for infections in humans,
44
particularly in immune-compromised hosts (1-3) and neonates (4, 5). Seventy-three percent of all
45
neonatal bacteremia in the United States is caused by CoNS pathogens (4, 5). In addition, CoNS
46
are typically resistant to multiple drug classes (3). Clinical presentations for patients infected
47
with CoNS are often distinct from those caused by Staphylococcus aureus infections, and these
48
occur more commonly in patients on corticosteroid therapy, those undergoing hemodialysis, or
49
those with implanted catheters or prosthetic valves (6). Of the CoNS pathogens, S. epidermidis is
50
the most abundant colonizer of human skin and mucous membranes, as well as the most common
51
cause of catheter-associated bloodstream infections (7).
52 53
CoNS infections are most commonly treated with glycopeptides, including vancomycin, but
54
there has been increasing concern regarding emerging resistance to these agents (3). Up to 90%
55
of S. epidermidis strains are now resistant to methicillin, with resistance to aminoglycosides and
56
macrolides noted among hospital-associated strains. Resistance to linezolid also occurs, but is
57
much less frequent (8, 9). Despite hospital-level reports of increasing resistance, there are few
58
recent national studies describing the prevalence of antibiotic resistance in clinically important
59
CoNS pathogens. For example, previously reported surveillance data (10-12) is more than a
60
decade old.
61
62
In this study, we examined national trends in antibiotic resistance of clinical CoNS isolates. We
63
paid particular attention to S. epidermidis isolates obtained from blood specimens in hospitalized
64
patients, given the important role of this pathogen in hospital-acquired infections associated with
65
medical devices. Finally, we compared resistance patterns in S. epidermidis with overall
66
resistance trends in CoNS pathogens, and examined correlations with antibiotic prescribing
67
patterns.
68 69
Methods
70
Data
71
We analyzed national trends in the frequency of resistance for S. epidermidis and all CoNS
72
pathogens (S. saprophyticus, S. lugdunensis, S. schleiferi, and S. caprae), using antibiotic
73
susceptibility data from The Surveillance Network (TSN) Database-USA (managed by Eurofins
74
Medinet, Chantilly, Virginia). TSN is an electronic repository of susceptibility test results
75
collected from more than 300 microbiology laboratories in the United States, representing a
76
national sample of isolates tested for resistance. Laboratories were selected to be geographically
77
and demographically representative of hospital and patient characteristics (bed size, age, race,
78
sex) in each of the nine US Census Bureau regional divisions (13, 14). Testing of isolates occurs
79
on-site as part of routine diagnostic testing for susceptibility to different antibiotic agents using
80
standards established by the Clinical and Laboratory Standards Institute (CLSI). Data from TSN
81
have been used extensively to evaluate patterns and trends of antibiotic drug resistance (13-21).
82 83
84
Our data included CoNS isolates collected from inpatient areas of healthcare facilities from
85
January 1999 to June 2012. To assess longitudinal trends in resistance, we examined two sets of
86
isolates: (1) all S. epidermidis bacterial isolates where the isolate source was blood; and (2) all
87
CoNS isolates regardless of source. The former reduces the likelihood of contamination, though
88
without clinical indicators it cannot be ruled out. The latter forms the full reservoir of CoNS
89
isolates, which can serve as an important indicator of possible changes in resistance levels.
90
Isolates were tested for resistance to oxacillin (a proxy for all β-lactam antibiotic drugs,
91
including methicillin), ciprofloxacin, clindamycin, levofloxacin, linezolid, and vancomycin
92
individually. While both ciprofloxacin and levofloxacin are fluoroquinolones, and thus have
93
similar mechanisms of resistance (22), we examined them separately because levofloxacin has a
94
broader spectrum of activity against gram positive organisms (see e.g. 23, 24), and because
95
differences in their resistance rates have been noted (25). In addition to individual resistance
96
profiles, we operationally defined isolates as multidrug resistant if they were resistant to
97
oxacillin, ciprofloxacin, clindamycin, and levofloxacin. Isolates were categorized for resistance
98
according to the CLSI breakpoint criteria. Between 2004 and 2005 the resistance breakpoints for
99
levofloxacin were modified, however no other changes to data collection or processing methods
100
occurred during the study period. Isolates which had intermediate resistance were classified as
101
resistant because of the change in resistance breakpoints for levofloxacin (the post MIC values
102
for resistance include the intermediate values for the prior years). For comparison, we also
103
examined levofloxacin resistance rates in S. aureus (which were subject to the same breakpoint
104
changes) and Escherichia coli (no breakpoint change) blood isolates using the same methods.
105 106
Trend Analysis
107
Resistance rates for each year were calculated as the proportion of phenotypes resistant of the
108
number of isolates tested. Confidence intervals were calculated by using the Wilson score
109
method incorporating continuity correction as detailed by Newcombe (26). The Mann-Kendall
110
test for trend and the Wilcox-Mann-Whitney rank sum test, which tests whether the distribution
111
from the first half of the study was the same as the second, were used to evaluate the statistical
112
significance of observed changes in the frequency of resistance. All data analysis was done using
113
Stata version 10 software (StataCorp LP, College Station, Texas).
114 115
Correlation Analysis
116
We also investigated the correlation between the yearly levofloxacin prescription rate and the
117
percentage of Staphylococcus epidermidis isolates that were resistant. However, as has often
118
been noted in the statistical literature, two time series can appear highly correlated without any
119
statistical significance (or meaningful connection) (21). Accordingly, we used a time series
120
analysis method similar to Sun and colleagues (21), which addresses this issue, to ascertain the
121
statistical correlation. Briefly, we applied the Box–Jenkins approach to fit time-series data to
122
autoregressive moving average (ARIMA) statistical models. This method removes the trend from
123
the data and transforms it into a series of independent, identically distributed random variables.
124
Time series were differenced and the best ARIMA model was selected based on Akaike
125
information criterion (AIC). The residuals of the model were then diagnosed for acceptability
126
using the Box–Ljung white noise test. The residuals from prescription and resistance models
127
were then cross-correlated to examine their association.
128
129
Prescription data were obtained from IMS Health’s Xponent database, which contains the
130
number of dispensed drug prescriptions collected from retail pharmacies (chain, mass
131
merchandisers, independents, and food stores) in the United States by month. The database
132
covers more than 70% of all prescriptions filled in the United States, and records are then
133
weighted to project 100% of total prescriptions dispensed. Because the fluoroquinolones may
134
have similar mechanisms of action and thus contribute to cross-resistance, we examined the
135
correlation between levofloxacin resistance in S. epidermidis and both ciprofloxacin and all
136
fluroquinolone (ciprofloxacin, levofloxacin, moxifloxacin, ofloxacin) prescriptions as well.
137 138
Results
139
Trend Analysis
140
We evaluated the resistance of S. epidermidis isolates from blood specimens obtained from
141
hospitalized patients during the study period and compared their resistance patterns with those
142
for all CoNS pathogens obtained from all clinical sites. Between January 1, 1999, and June 30,
143
2012, more than 540,000 CoNS isolates and 80,000 S. epidermidis blood isolates were submitted
144
to TSN and included in our analysis.
145 146
Over the course of the study, resistance to S. epidermidis increased to ciprofloxacin,
147
levofloxacin, and clindamycin individually, as did resistance to a multi-drug resistant phenotype
148
(Figure 1). However, patterns of resistance differed among drugs. Resistance to ciprofloxacin
149
and clindamycin increased from 58.3% (95% confidence Interval [CI], 56.8–59.8) in 1999 to
150
68.4% (95% CI, 64.9–71.6) in 2012, and from 43.4% (95% CI, 42.1–44.8) in 1999 to 48.5%
151
(95% CI, 45.9–51.0) in 2012, respectively. Resistance to levofloxacin exhibited a different
152
pattern. The percentage of isolates resistant to levofloxacin increased rapidly from 57.1% (95%
153
CI 54.8–59.3) in 1999 to a high of 78.6% (95% CI, 77.4–79.7) in 2005, followed by a significant
154
decrease to 68.1% (95% CI, 65.1–70.9) in 2012. Multidrug resistance followed a similar pattern,
155
ranging from 33.6% (95% CI, 31.3–36.0) in 1999 to a high of 56.3% (95% CI, 54.4–58.2) in
156
2005, and a subsequent decrease to 41.1% (95% CI, 37.3–45.0) in 2012. Oxacillin resistance
157
remained generally constant at ~80% during the study period. Resistance to linezolid was first
158
detected in 2004, and by 2012 had reached 1.1% (95% CI, 0.6–1.9) (Supplementary Figure 1).
159
Resistance to vancomycin was not observed over the study period.
160 161
Positive trends observed for ciprofloxacin and clindamycin were statistically significant (p
1
Trends in Antibiotic Resistance in Coagulase Negative Staphylococci, United States, 1999–
2
2012
3 4
Larissa May,a* Eili Y. Klein,b,c#* Richard E. Rothman,b Ramanan Laxminarayanc,d
5 6
Department of Emergency Medicine, George Washington University, Washington, DC, USAa;
7
Department of Emergency Medicine, Johns Hopkins University, Baltimore, MD, USAb; Center
8
for Disease Dynamics, Economics & Policy, Washington, DCc; Princeton Environmental
9
Institute, Princeton University, Princeton, NJ, USAd
10 11
*
These authors contributed equally to this work
12 13
Running Head: Trends in Antibiotic Resistance in CoNS
14 15 16 17 18
# Address correspondence to Eili Y. Klein, [email protected]
19
Abstract
20
Coagulase-negative staphylococci (CoNS) are important bloodstream pathogens typically
21
resistant to multiple antibiotics. Despite concern about increasing resistance, there have been no
22
recent studies describing the national prevalence in CoNS pathogens. We used national
23
resistance data over a period of 13 years (1999–2012) from The Surveillance Network (TSN) to
24
determine prevalence and assess trends in resistance for S. epidermidis, the most common CoNS
25
pathogen, as well as all CoNS pathogens. Over the course of the study period, S. epidermidis
26
resistance to ciprofloxacin and clindamycin increased steadily from 58.3% to 68.4% and from
27
43.4% to 48.5%, respectively. Resistance to levofloxacin increased rapidly from 57.1% in 1999
28
to a high of 78.6% in 2005, followed by a decrease to 68.1% in 2012. Multidrug resistance for
29
CoNS followed a similar pattern, and this rise and small decline in resistance was found to be
30
strongly correlated with levofloxacin prescribing patterns. Resistance patterns were similar for
31
the aggregate of CoNS pathogens. The results from our study demonstrate that antibiotic
32
resistance in CoNS pathogens has increased significantly over the past 13 years. These results
33
are important as CoNS can serve as sentinels for monitoring resistance, and they play a role as a
34
reservoir of resistance genes that can be transmitted to other pathogens. The link between the
35
levofloxacin prescription rate and resistance levels suggest a critical role for reducing
36
inappropriate use of fluoroquinolones and other broad-spectrum antibiotics in both healthcare
37
settings and the community to help curb the reservoir of resistance in these colonizing pathogens.
38
39
Introduction
40
Coagulase-negative staphylococci (CoNS) are normal flora of the human skin and mucosa.
41
Because they have long been considered contaminants rather than true clinical pathogens, there
42
are few systematic studies describing their epidemiology in human infections. Nonetheless,
43
colonizing CoNS pathogens have been reported to be responsible for infections in humans,
44
particularly in immune-compromised hosts (1-3) and neonates (4, 5). Seventy-three percent of all
45
neonatal bacteremia in the United States is caused by CoNS pathogens (4, 5). In addition, CoNS
46
are typically resistant to multiple drug classes (3). Clinical presentations for patients infected
47
with CoNS are often distinct from those caused by Staphylococcus aureus infections, and these
48
occur more commonly in patients on corticosteroid therapy, those undergoing hemodialysis, or
49
those with implanted catheters or prosthetic valves (6). Of the CoNS pathogens, S. epidermidis is
50
the most abundant colonizer of human skin and mucous membranes, as well as the most common
51
cause of catheter-associated bloodstream infections (7).
52 53
CoNS infections are most commonly treated with glycopeptides, including vancomycin, but
54
there has been increasing concern regarding emerging resistance to these agents (3). Up to 90%
55
of S. epidermidis strains are now resistant to methicillin, with resistance to aminoglycosides and
56
macrolides noted among hospital-associated strains. Resistance to linezolid also occurs, but is
57
much less frequent (8, 9). Despite hospital-level reports of increasing resistance, there are few
58
recent national studies describing the prevalence of antibiotic resistance in clinically important
59
CoNS pathogens. For example, previously reported surveillance data (10-12) is more than a
60
decade old.
61
62
In this study, we examined national trends in antibiotic resistance of clinical CoNS isolates. We
63
paid particular attention to S. epidermidis isolates obtained from blood specimens in hospitalized
64
patients, given the important role of this pathogen in hospital-acquired infections associated with
65
medical devices. Finally, we compared resistance patterns in S. epidermidis with overall
66
resistance trends in CoNS pathogens, and examined correlations with antibiotic prescribing
67
patterns.
68 69
Methods
70
Data
71
We analyzed national trends in the frequency of resistance for S. epidermidis and all CoNS
72
pathogens (S. saprophyticus, S. lugdunensis, S. schleiferi, and S. caprae), using antibiotic
73
susceptibility data from The Surveillance Network (TSN) Database-USA (managed by Eurofins
74
Medinet, Chantilly, Virginia). TSN is an electronic repository of susceptibility test results
75
collected from more than 300 microbiology laboratories in the United States, representing a
76
national sample of isolates tested for resistance. Laboratories were selected to be geographically
77
and demographically representative of hospital and patient characteristics (bed size, age, race,
78
sex) in each of the nine US Census Bureau regional divisions (13, 14). Testing of isolates occurs
79
on-site as part of routine diagnostic testing for susceptibility to different antibiotic agents using
80
standards established by the Clinical and Laboratory Standards Institute (CLSI). Data from TSN
81
have been used extensively to evaluate patterns and trends of antibiotic drug resistance (13-21).
82 83
84
Our data included CoNS isolates collected from inpatient areas of healthcare facilities from
85
January 1999 to June 2012. To assess longitudinal trends in resistance, we examined two sets of
86
isolates: (1) all S. epidermidis bacterial isolates where the isolate source was blood; and (2) all
87
CoNS isolates regardless of source. The former reduces the likelihood of contamination, though
88
without clinical indicators it cannot be ruled out. The latter forms the full reservoir of CoNS
89
isolates, which can serve as an important indicator of possible changes in resistance levels.
90
Isolates were tested for resistance to oxacillin (a proxy for all β-lactam antibiotic drugs,
91
including methicillin), ciprofloxacin, clindamycin, levofloxacin, linezolid, and vancomycin
92
individually. While both ciprofloxacin and levofloxacin are fluoroquinolones, and thus have
93
similar mechanisms of resistance (22), we examined them separately because levofloxacin has a
94
broader spectrum of activity against gram positive organisms (see e.g. 23, 24), and because
95
differences in their resistance rates have been noted (25). In addition to individual resistance
96
profiles, we operationally defined isolates as multidrug resistant if they were resistant to
97
oxacillin, ciprofloxacin, clindamycin, and levofloxacin. Isolates were categorized for resistance
98
according to the CLSI breakpoint criteria. Between 2004 and 2005 the resistance breakpoints for
99
levofloxacin were modified, however no other changes to data collection or processing methods
100
occurred during the study period. Isolates which had intermediate resistance were classified as
101
resistant because of the change in resistance breakpoints for levofloxacin (the post MIC values
102
for resistance include the intermediate values for the prior years). For comparison, we also
103
examined levofloxacin resistance rates in S. aureus (which were subject to the same breakpoint
104
changes) and Escherichia coli (no breakpoint change) blood isolates using the same methods.
105 106
Trend Analysis
107
Resistance rates for each year were calculated as the proportion of phenotypes resistant of the
108
number of isolates tested. Confidence intervals were calculated by using the Wilson score
109
method incorporating continuity correction as detailed by Newcombe (26). The Mann-Kendall
110
test for trend and the Wilcox-Mann-Whitney rank sum test, which tests whether the distribution
111
from the first half of the study was the same as the second, were used to evaluate the statistical
112
significance of observed changes in the frequency of resistance. All data analysis was done using
113
Stata version 10 software (StataCorp LP, College Station, Texas).
114 115
Correlation Analysis
116
We also investigated the correlation between the yearly levofloxacin prescription rate and the
117
percentage of Staphylococcus epidermidis isolates that were resistant. However, as has often
118
been noted in the statistical literature, two time series can appear highly correlated without any
119
statistical significance (or meaningful connection) (21). Accordingly, we used a time series
120
analysis method similar to Sun and colleagues (21), which addresses this issue, to ascertain the
121
statistical correlation. Briefly, we applied the Box–Jenkins approach to fit time-series data to
122
autoregressive moving average (ARIMA) statistical models. This method removes the trend from
123
the data and transforms it into a series of independent, identically distributed random variables.
124
Time series were differenced and the best ARIMA model was selected based on Akaike
125
information criterion (AIC). The residuals of the model were then diagnosed for acceptability
126
using the Box–Ljung white noise test. The residuals from prescription and resistance models
127
were then cross-correlated to examine their association.
128
129
Prescription data were obtained from IMS Health’s Xponent database, which contains the
130
number of dispensed drug prescriptions collected from retail pharmacies (chain, mass
131
merchandisers, independents, and food stores) in the United States by month. The database
132
covers more than 70% of all prescriptions filled in the United States, and records are then
133
weighted to project 100% of total prescriptions dispensed. Because the fluoroquinolones may
134
have similar mechanisms of action and thus contribute to cross-resistance, we examined the
135
correlation between levofloxacin resistance in S. epidermidis and both ciprofloxacin and all
136
fluroquinolone (ciprofloxacin, levofloxacin, moxifloxacin, ofloxacin) prescriptions as well.
137 138
Results
139
Trend Analysis
140
We evaluated the resistance of S. epidermidis isolates from blood specimens obtained from
141
hospitalized patients during the study period and compared their resistance patterns with those
142
for all CoNS pathogens obtained from all clinical sites. Between January 1, 1999, and June 30,
143
2012, more than 540,000 CoNS isolates and 80,000 S. epidermidis blood isolates were submitted
144
to TSN and included in our analysis.
145 146
Over the course of the study, resistance to S. epidermidis increased to ciprofloxacin,
147
levofloxacin, and clindamycin individually, as did resistance to a multi-drug resistant phenotype
148
(Figure 1). However, patterns of resistance differed among drugs. Resistance to ciprofloxacin
149
and clindamycin increased from 58.3% (95% confidence Interval [CI], 56.8–59.8) in 1999 to
150
68.4% (95% CI, 64.9–71.6) in 2012, and from 43.4% (95% CI, 42.1–44.8) in 1999 to 48.5%
151
(95% CI, 45.9–51.0) in 2012, respectively. Resistance to levofloxacin exhibited a different
152
pattern. The percentage of isolates resistant to levofloxacin increased rapidly from 57.1% (95%
153
CI 54.8–59.3) in 1999 to a high of 78.6% (95% CI, 77.4–79.7) in 2005, followed by a significant
154
decrease to 68.1% (95% CI, 65.1–70.9) in 2012. Multidrug resistance followed a similar pattern,
155
ranging from 33.6% (95% CI, 31.3–36.0) in 1999 to a high of 56.3% (95% CI, 54.4–58.2) in
156
2005, and a subsequent decrease to 41.1% (95% CI, 37.3–45.0) in 2012. Oxacillin resistance
157
remained generally constant at ~80% during the study period. Resistance to linezolid was first
158
detected in 2004, and by 2012 had reached 1.1% (95% CI, 0.6–1.9) (Supplementary Figure 1).
159
Resistance to vancomycin was not observed over the study period.
160 161
Positive trends observed for ciprofloxacin and clindamycin were statistically significant (p
