Article DOI: [Link]
204486
Increased Incidence of Antimicrobial-
Resistant Nontyphoidal Salmonella
Infections, United States, 2004–2016
Appendix
Appendix Table 1. Population estimates, number of isolates reported and tested, and percentage of isolates with clinically
important and multidrug resistance, by year, United States, 2004–2016*
Isolates with clinically Isolates with multidrug
No. reported No. tested in important resistance§ resistance¶
Year Population† to LEDS‡ NARMS‡ No. (%) No. (%)
2004 290,304,689 34,374 1,762 245 (13.9) 200 (11.4)
2005 292,989,788 35,116 2,016 258 (12.8) 238 (11.8)
2006 295,824,198 38,737 2,151 281 (13.1) 254 (11.8)
2007 298,660,828 37,840 2,115 253 (12.0) 232 (11.0)
2008 301,494,062 43,505 2,306 274 (11.9) 223 (9.7)
2009 304,133,689 38,669 2,136 241 (11.3) 208 (9.7)
2010 306,665,034 43,397 2,418 272 (11.2) 225 (9.3)
2011 308,942,304 42,916 2,305 255 (11.1) 212 (9.2)
2012 311,240,143 44,824 2,201 241 (10.9) 192 (8.7)
2013 313,412,383 41,885 2,157 277 (12.8) 214 (9.9)
2014 315,751,397 44,234 2,093 259 (12.4) 194 (9.3)
2015 318,063,375 47,811 2,322 375 (16.1) 286 (12.3)
2016 320,275,892 46,554 2283 315 (13.8) 234 (10.2)
Total 539,862 28,265 3,546 (12.5) 2,912 (10.3)
*LEDS, Laboratory-based Enteric Diseases Surveillance; NARMS, National Antimicrobial Resistance Monitoring System.
†Total estimates from the U.S. Census Bureau for 48 states (Alaska, Hawaii, and District of Columbia excluded).
‡Number of infections reported by the 48 contiguous states to LEDS and number of isolates submitted by the 48 states and tested in NARMS.
§Overall category includes any of 3 clinically important resistance patterns (i.e., resistant to ceftriaxone, resistant to ampicillin, or nonsusceptible to
ciprofloxacin).
¶Resistant to >3 classes of antimicrobial agents.
Page 1 of 8
Appendix Table 2. Estimated changes in the incidence of resistant culture-confirmed nontyphoidal Salmonella infections, by
serotype and resistance category: 2015–2016 versus 2004–2008*
Mean change†‡ in resistance incidence (per 100,000 persons/year) Total All NTS†
Resistance Type of 2015–2016 vs. 2004–2008 (% contribution of serotype) increase/ (net
category change† Enteritidis Typhimurium Newport I 4,[5],12:i:- Heidelberg Other† decrease change)
Any ↑ 0.29‡ (26) – – 0.41‡ (37) – 0.41‡ (37) 1.11 0.68‡
clinically ↓ – -0.33‡ (75) −0.07 (17) – -0.04 (8) – −0.44 –
important
resistance§
Multidrug ↑ 0.13‡ (16) – – 0.40‡ (49) – 0.28 (35) 0.81 0.32
resistance¶ ↓ – -0.37‡ (76) −0.09 (19) – −0.03 (5) – −0.49 –
Amp-only*§ ↑ 0.08 (15) – – 0.35‡ (61) – 0.14 (24) 0.57 0.19
↓ – -0.35‡ (89) −0.01 (3) – −0.03 (8) – −0.39 –
Cef/Amp*§ ↑ NC† 0.003 (2) – 0.02 (13) 0.01 (4) 0.11 (81) 0.14 0.06
↓ NC† – −0.08 (100) – – – −0.08 –
Cipro*§ ↑ 0.19‡ (47) 0.02 (4) NC† 0.04 (11) NC† 0.16‡ (38) 0.41 0.41‡
*Amp-only, resistant to ampicillin but susceptible to ceftriaxone and ciprofloxacin; Cef/Amp, resistant to ceftriaxone and ampicillin; Cipro,
nonsusceptible to ciprofloxacin but susceptible to ceftriaxone; CrI, credible interval; NC, not calculated; NTS, nontyphoidal Salmonella; ↑, increase; ↓,
decrease.
†Mean estimates and 95% CrIs for each resistance and serotype category were derived using Bayesian hierarchical models. Resistance incidence in
2015–2016 was compared with that for 2004–2008 (↑ if 2015–2016 >2004–2008, ↓ if 2015–2016 <2004–2008). Serotypes other than Enteritidis,
Typhimurium, Newport, I 4,[5],12:i:-, and Heidelberg were combined in the “other” category. For all NTS, estimated changes were derived by
summing those for the 6 serotype categories (net increase or decrease). State-year data were too sparse to use in the Bayesian hierarchical models
to estimate resistance incidence for Cef/Amp among Enteritidis and for Cipro among Newport and Heidelberg; thus, estimated changes in resistance
incidence were not calculated (NC).
‡Mean changes are reported as significant (bold font) if the 95% CrIs (rounded to 2 decimals) do not include 0: any clinically important resistance,
Enteritidis (0.29 [95% CrI 0.12, 0.47]), I 4,[5],12:i:- (0.41 [0.27, 0.56]), Typhimurium (−0.33 [-0.58, −0.07]), Other (0.41 [0.12, 0.72]); MDR, Enteritidis
(0.13 [0.04, 0.23]), I,4,[5],12:i:- (0.40 [0.24, 0.56]), Typhimurium (−0.37 [-0.59, −0.14]); Amp-only, I 4,[5],12:i:- (0.35 [0.21, 0.50]), Typhimurium (−0.35
[-0.61, −0.10]); Cipro, Enteritidis (0.19 [0.05, 0.34]), Other (0.16 [0.04, 0.29]);
§An overall category of clinically important resistance includes any of 3 resistance patterns (i.e., resistant to ceftriaxone, resistant to ampicillin, or
nonsusceptible to ciprofloxacin). Amp-only, Cef/Amp, and Cipro are mutually exclusive categories of clinically important resistance
¶Resistant to >3 classes of antimicrobial agents.
Appendix Table 3. Estimated changes in the incidence of resistant culture-confirmed nontyphoidal Salmonella infections, by
serotype and resistance category: 2015–2016 versus 2010–2014*
Mean change†‡ in resistance incidence (per 100,000 persons/year) Total All NTS†
Resistance Type of 2015–2016 vs. 2010–2014 (% contribution of serotype) increase/ (net
category change† Enteritidis Typhimurium Newport I 4,[5],12:i:- Heidelberg Other† decrease change)
Any clinically ↑ 0.20 (30) – – 0.24 (36) – 0.23 (34) 0.67 0.60
important ↓ – −0.04 (56) −0.01 (9) – −0.03 (35)` – −0.07 –
resistance§
Multidrug ↑ 0.09 (16) – – 0.24 (43) – 0.22 (41) 0.55 0.41
resistance¶ ↓ – −0.08 (64) −0.02 (13) – −0.03 (23) – −0.13 –
Amp-only*§ ↑ 0.03 (9) – – 0.20 (61) – 0.10 (30) 0.32 0.23
↓ – −0.08 (85) −0.002 (2) – −0.01 (13) – −0.09 –
Cef/Amp*§ ↑ NC† 0.01 (13) – 0.01 (13) – 0.07 (74)` 0.10 0.08
↓ NC† – −0.02 (83) – −0.004 (17)` – −0.03
Cipro*§ ↑ 0.16‡ (57) 0.02 (7) NC† 0.03 (11) NC† 0.07 (25)` 0.29 0.29‡
*Amp-only, resistant to ampicillin but susceptible to ceftriaxone and ciprofloxacin; Cef/Amp, resistant to ceftriaxone and ampicillin; Cipro, nonsusceptible
to ciprofloxacin but susceptible to ceftriaxone; CrI, credible interval; NC, not calculated; NTS, nontyphoidal Salmonella; ↑, increase; ↓, decrease.
†Mean estimates and 95% credible intervals (CIs) for each resistance and serotype category were derived using Bayesian hierarchical models.
Resistance incidence in 2015–2016 was compared with that for 2010–2014 (↑ if 2015–2016 >2010–2014, ↓ if 2015–2016 <2010–2014). Serotypes other
than Typhimurium, Enteritidis, I 4,[5],12:i:-, Newport, and Heidelberg were combined in the “other” category. For all NTS, estimated changes were derived
by summing those for the 6 serotype categories (net increase or decrease). State-year data were too sparse to use in the Bayesian hierarchical models to
estimate resistance incidence for Cef/Amp among Enteritidis and for Cipro among Newport and Heidelberg; thus, estimated changes in resistance
incidence were not calculated (NC).
‡Mean change is reported as significant (bold font) if the 95% CIs (rounded to 2 decimals) do not include 0: Cipro, Enteritidis (0.16 [95% CrI 0.02. 0.32]).
§An overall category of clinically important resistance includes any of 3 resistance patterns (i.e., resistant to ceftriaxone, resistant to ampicillin,
nonsusceptible to ciprofloxacin). Amp-only, Cef/Amp, and Cipro are mutually exclusive categories of clinically important resistance.
¶Resistant to >3 classes of antimicrobial agents.
Page 2 of 8
Appendix Figure 1. Number of nontyphoidal Salmonella isolates with clinically important resistance, by
mutually exclusive resistance category, 2004–2016. Three mutually exclusive categories of clinically
important resistance were defined: Amp-only as resistant to ampicillin but susceptible to ceftriaxone and
ciprofloxacin; Cef/Amp as resistant to ceftriaxone and ampicillin; and Cipro as nonsusceptible to
ciprofloxacin but susceptible to ceftriaxone. Isolates in each category may have resistance to other
agents.
Page 3 of 8
Appendix Figure 2. Estimated annual incidence of culture-confirmed nontyphoidal Salmonella infections
with multidrug resistance, by serotype and region, 2004–2016. Estimated changes in resistance incidence
(mean and 95% credible intervals of the posterior differences per 100,000 persons/year) were derived
using Bayesian hierarchical models (BHM). Crude resistance incidence rates were derived by multiplying
infection incidence and resistance proportion for state-year. Multidrug resistance (MDR) was defined as
resistance to three or more classes of antimicrobial agents. The “other” category comprised serotypes
other than Enteritidis, Typhimurium, Newport, I 4,[5],12:i:-, and Heidelberg. U.S. Census regions were
used to define 4 geographic regions. NTS, all nontyphoidal Salmonella serotypes.
Page 4 of 8
Appendix Figure 3. Estimated annual incidence of culture-confirmed nontyphoidal Salmonella infections
with ampicillin-only resistance (Amp-only), by serotype and region, 2004–2016. Estimated changes in
resistance incidence (mean and 95% credible intervals of the posterior differences per 100,000
persons/year) were derived using Bayesian hierarchical models (BHM). Crude resistance incidence rates
were derived by multiplying infection incidence and resistance proportion for state-year. Amp-only was
defined as resistant to ampicillin but susceptible to ceftriaxone and ciprofloxacin. The “other” category
comprised serotypes other than Enteritidis, Typhimurium, Newport, I 4,[5],12:i:-, and Heidelberg. U.S.
Census regions were used to define 4 geographic regions. NTS, all nontyphoidal Salmonella serotypes.
Page 5 of 8
Appendix Figure 4. Estimated annual incidence of culture-confirmed nontyphoidal Salmonella infections
with ceftriaxone/ampicillin resistance (Cef/Amp), by serotype and region, 2004–2016. Estimated changes
in resistance incidence (mean and 95% credible intervals of the posterior differences per 100,000
persons/year) were derived using Bayesian hierarchical models (BHM). Crude resistance incidence rates
were derived by multiplying infection incidence and resistance proportion for state-year. Cef/Amp was
defined as resistant to ceftriaxone and ampicillin. The “other” category comprised serotypes other than
Enteritidis, Typhimurium, Newport, I 4,[5],12:i:-, and Heidelberg; estimates for Enteritidis (not included in
the figure) were not derived because state-year data were too sparse to use in the BHM. US Census
regions were used to define 4 geographic regions. NTS, all nontyphoidal Salmonella serotypes.
Page 6 of 8
Appendix Figure 5. Estimated annual incidence of culture-confirmed nontyphoidal Salmonella infections
with ciprofloxacin nonsusceptibility (Cipro), by serotype and region, 2004–2016. Estimated changes in
resistance incidence (mean and 95% credible intervals of the posterior differences per 100,000
persons/year) were derived using Bayesian hierarchical models (BHM). Crude resistance incidence rates
were derived by multiplying infection incidence and resistance proportion for state-year. Cipro was
defined as nonsusceptible to ciprofloxacin but susceptible to ceftriaxone. The “other” category comprised
serotypes other than Enteritidis, Typhimurium, Newport, I 4,[5],12:i:-, and Heidelberg; estimates for
Newport and Heidelberg (not included in the figure) were not derived because state-year data were too
sparse to use in the BHM. US Census regions were used to define 4 geographic regions. NTS, all
nontyphoidal Salmonella serotypes.
Page 7 of 8
Appendix Figure 6. Distribution of ciprofloxacin MICs among Cipro category (i.e., ciprofloxacin-
nonsusceptible and ceftriaxone-susceptible) and ciprofloxacin-nonsusceptible and ceftriaxone-resistant
Salmonella isolates, 2004–2016. Of 854 isolates in the Cipro category, 785 (92%) had MICs within the
intermediate range, 0.12–0.5 µg/mL. Of 78 isolates nonsusceptible to ciprofloxacin and ceftriaxone-
resistant, 71 (91%) had MICs within the intermediate range; these 78 isolates were not included in the
Cipro category.
Page 8 of 8
