Clinical Outcomes of Nalidixic Acid,Ceftriaxone,and Multidrug-Resistant Nontyphoidal Salmonella Infections Compared with Pansusceptible Infections in FoodNet Sites,2006

Abstract

Background:

Nontyphoidal Salmonella causes an estimated 1.2 million infections, 23,000 hospitalizations, and 450 deaths annually in the United States. Most illnesses are self-limited; however, treatment with antimicrobial agents can be life-saving for invasive infections.

Methods:

The Foodborne Diseases Active Surveillance Network and the National Antimicrobial Resistance Monitoring System collaborated on a prospective cohort study of patients with nontyphoidal Salmonella bloodstream and gastrointestinal infections to determine differences in the clinical outcomes of resistant compared with pansusceptible infections. Interviews were conducted within 85 days of specimen collection date.

Results:

Of 875 nontyphoidal Salmonella isolates, 705 (81%) were pansusceptible, 165 (19%) were resistant to at least 1 agent, and 5 (0.6%) had only intermediate resistance. The most common pattern, found in 51 (31%) of resistant isolates, was resistance to at least ampicillin, chloramphenicol, streptomycin, sulfisoxazole, and tetracycline (ACSSuT); 88% of isolates with this pattern were serotype Typhimurium or Newport. Fourteen (52%) of the 27 ceftriaxone-resistant isolates were also ACSSuT resistant. Adjusted for age and serotype, bloodstream infection was significantly more common among patients infected with strains resistant to only two, only three, or only five antimicrobial classes, to ACSSuT with or without other agents, to ACSSuT only, or to nalidixic acid with or without other agents than among patients with pansusceptible isolates. Adjusted for age, serotype, and bloodstream infection, hospitalization was significantly more common among patients infected with strains resistant to only three agents or to ceftriaxone (all ceftriaxone-resistant isolates were resistant to other agents) than among patients with pansusceptible isolates.

Conclusion:

This study extends evidence that patients with antimicrobial-resistant nontyphoidal Salmonella infections have more severe outcomes. Prevention efforts are needed to reduce unnecessary antimicrobial use in patient care settings and in food animals to help prevent the emergence of resistance and infections with resistant nontyphoidal Salmonella.

Introduction

Nontyphoidal Salmonella cause an estimated 1.2 million infections, 23,000 hospitalizations, and 450 deaths annually in the United States (Scallan et al., 2011). Most illnesses are self-limited; however, treatment with antimicrobial agents can be life-saving for invasive infections (Varma et al., 2005b; Scallan et al., 2011). Fluoroquinolones or third-generation cephalosporins are recommended for treatment of invasive Salmonella infections (Guerrant et al., 2001; Crump et al., 2011).

Several studies have shown that patients with Salmonella infections with certain patterns of antimicrobial resistance have more severe illness than those with susceptible infections (Helms et al., 2002; Helms et al., 2004; Molbak, 2005; Varma et al., 2005b; Solghan et al., 2010). Antimicrobial-resistant nontyphoidal Salmonella infections have been associated with increased hospitalization and deaths (Holmberg et al., 1984; Varma et al., 2005a). In a U.S. study, persons with nontyphoidal Salmonella infections resistant to one or more antimicrobial agents (ampicillin, ceftriaxone, ciprofloxacin, gentamicin, or trimethoprim–sulfamethoxazole) were more likely to be hospitalized and have a bloodstream infection than persons with pansusceptible infections (Varma et al., 2005b). Persons with Salmonella serotype Typhimurium infections resistant to ampicillin, chloramphenicol, streptomycin, sulfisoxazole, and tetracycline (ACSSuT) were more likely to be hospitalized and die than persons with pansusceptible infections (Helms et al., 2002; Martin et al., 2004; Solghan et al., 2010). However, other studies have not demonstrated worse clinical outcomes associated with resistance (Threlfall et al., 1998; Devasia et al., 2005). We report findings from a multicenter cohort study designed to examine clinical outcomes of antimicrobial-resistant nontyphoidal Salmonella infections.

Materials and Methods

Surveillance systems

The Foodborne Diseases Active Surveillance Network (FoodNet) of the Centers for Disease Control and Prevention's (CDC) Emerging Infections Program conducts active, population-based surveillance in 10 U.S. sites for all laboratory-confirmed infections with selected enteric pathogens, including Salmonella, that are transmitted commonly through food. The population in the catchment area was approximately 46.4 million persons in 2008, and the distribution of age and sex was similar to that of the U.S. population (CDC, 2010).

The National Antimicrobial Resistance Monitoring System (NARMS) at CDC monitors antimicrobial susceptibility of Salmonella isolated from humans. Clinical laboratories send all Salmonella isolates from ill persons to state public health laboratories for serotyping (Scallan, 2007). From 2003 to 2007, NARMS tested every 20th non-Typhi Salmonella isolate and starting in 2008 tested every 20th nontyphoidal Salmonella isolate (defined as serotypes other than Typhi, Paratyphi A, tartrate-negative Paratyphi B, and Paratyphi C) submitted to participating public health laboratories. For this study, some FoodNet sites enhanced their sampling scheme to submit every fourth (Colorado counties), every fifth (states of Connecticut, Minnesota, New Mexico, and counties in New York), or every tenth (states of Maryland and Tennessee, and counties in California) non-Typhi Salmonella isolate, to achieve the calculated sample size needed to achieve the power necessary to detect key differences in clinical outcomes between infections caused by resistant and pansusceptible isolates. Georgia and Oregon continued to submit every 20th isolate. With rare exception, serotypes were determined by the submitting laboratory.

Antimicrobial susceptibility testing

The CDC NARMS laboratory tested all isolates using broth microdilution (Sensititre^®, Westlake, OH) to determine the minimum inhibitory concentration (MIC) for each of 15 agents representing 8 antimicrobial classes defined by the Clinical and Laboratory Standards Institute (CLSI): aminoglycosides (amikacin, gentamicin, kanamycin, streptomycin), β-lactam/β-lactamase inhibitor combinations (amoxicillin–clavulanic acid), cephems (cefoxitin, ceftiofur, ceftriaxone), folate pathway inhibitors (sulfisoxazole, trimethoprim–sulfamethoxazole), penicillins (ampicillin), phenicols (chloramphenicol), quinolones (ciprofloxacin, nalidixic acid), and tetracyclines (tetracycline). CLSI interpretative criteria were used when available to categorize MIC results as susceptible, intermediate, or resistant. We used current CLSI interpretive criteria for ciprofloxacin: resistance defined as MIC ≥1 μg/mL, intermediate 0.12–0.5 μg/mL, and susceptible ≤0.06 μg/mL (CLSI, 2013). Streptomycin resistance was defined as MIC ≥64 μg/mL, and ceftiofur resistance as MIC ≥8 μg/mL. R1 indicates resistance to one or more agents in one antimicrobial class, R2 resistance to two antimicrobial classes, and so on. ACSSuT resistance was defined as resistance to at least ampicillin, chloramphenicol, streptomycin, sulfisoxazole, and tetracycline. ACSSuT-only indicates resistance to only those agents. Pansusceptible was defined as no resistance to any antimicrobial agent on the panel.

Cohort study design

Eligible persons resided in the FoodNet catchment during the study period, had a nontyphoidal Salmonella strain isolated from blood or stool during a 24-month study period, and the strain was tested for antimicrobial susceptibility by NARMS. The first site started enrolling in January 2006 and the last started in May 2006. Patients were interviewed using a standard questionnaire that collected information on demographics, symptoms, and antimicrobial use. FoodNet staff obtained information about hospitalization and death using a standardized chart extraction form. Interviews were conducted only with persons who could be contacted within 85 days after the initial Salmonella-positive specimen was collected. Patients who refused participation, did not speak English or Spanish, were unable to answer questions, had an incomplete interview, had a co-existing infection with another enteric bacterium, could not be contacted, or were unable to estimate when illness began were excluded.

The study was approved by the Human Subjects Research Committees at CDC and participating sites.

Statistical analysis

Fisher's exact test was used to evaluate differences in proportions of patients by age, sex, previous antimicrobial use, and clinical outcome (diarrhea, specimen collection site, hospitalization, length of hospitalization >3 days, and death) by susceptibility phenotype; differences were deemed significant if the p-value was≤0.05. Analyses were conducted to estimate the relative risk and 95% confidence intervals (CI) for bloodstream infection and hospitalization by comparing resistant with pansusceptible Salmonella isolates. Analyses of bloodstream infection, hospitalization, and hospital stay >3 days were adjusted for age (<5 years, 5–17 years, 18–64 years, and ≥65 years) and serotype (ser. Typhimurium [includes Typhimurium var. O:5-], Enteritidis, Newport, and all others and unknown), comparing patients with pansusceptible isolates to patients with the following resistance categories: R1 through R8, ACSSuT plus other agent(s), ACSSuT-only, ceftriaxone plus other agents, nalidixic acid plus other agent(s), and nalidixic acid-only. Hospitalization and hospital stay >3 days were also adjusted for bloodstream versus stool source in another analysis. All statistical analyses were conducted in SAS (version 9.3, SAS Institute, Cary, NC).

Results

In all, 1057 patients with nontyphoidal Salmonella infection were eligible and 875 (83%) were enrolled; 182 were not enrolled for the following reasons: refused participation (n=123), did not speak English or Spanish (n=18), unable to answer questions (n=17), incomplete interview (n=11), co-existing infection with other enteric bacteria (n=10), could not be contacted (n=2), or unable to estimate when illness began (n=1).

Among the 875 patients enrolled, 831 (95%) had Salmonella isolated from a stool specimen and 44 (5%) from a blood specimen. Fifty percent were female; 26% were <5 years old, 17% were 5–17 years old, 45% were 18–64 years old, and 12% were >64 years old. Ninety-seven percent had 3 or more loose stools in a 24-h period. One patient died (isolate from blood, resistant to only nalidixic acid).

Among the 44 bloodstream infections, known serotypes were Typhimurium (11 isolates; 25%), Heidelberg (9; 20%), Enteritidis (4; 9%), Infantis (2; 5%), Stanley (2; 5%), 1 each of 15 other serotypes, and 1 unknown (Table 1). Twenty-six (59%) were pansusceptible. The 18 resistant isolates fell into 6 nonoverlapping resistance categories: 3 (17% of resistant isolates) R1, 3 (17%) R2, 4 (22%) R3, 1 (5.5%) R4, 6 (33%, including 5 ACSSuT-resistant) R5, and 1 (5.5%) R6 (Table 1). One isolate was resistant to ceftriaxone (also resistant to other agents), and 3 to nalidixic acid (2 also resistant to other agents). Among these 44 patients, 50% were female, 11% were <5 years old, 20% were 5–17 years old, 41% were 18–64 years old, and 27% were >64 years old.

Table 1.

Resistance Patterns of Salmonella Isolates, by Serotype (Three Most Common, Others, and Unknown) and Specimen Source, 2006–2008

			Nonoverlapping categories ^a									Categories that may overlap other categories ^b
		Total	Pansusceptibe	R1	R2	R3	R4	R5	R6	R7	R8	ACSSuT	ACSSuT only	Ceftriaxone	Nalidixic acid	Nalidixic acid only
Serotype	Specimen source	N	(n=705)	(n=41)	(n=28)	(n=29)	(n=12)	(n=39)	(n=3)	(n=11)	(n=2)	(n=51)	(n=34)	(n=27)	(n=21)	(n=13)
Typhimurium	Stool	157	101	5	7	5	7	27	0	2	2	31	24	5	3	1
	Blood	11	6	0	0	0	0	5	0	0	0	5	5	0	0	0
Enteritidis	Stool	157	140	12	2	2	0	0	0	0	0	0	0	1	12	10
	Blood	4	2	1	1	0	0	0	0	0	0	0	0	0	2	1
Newport	Stool	84	75	0	0	0	0	2	0	7	0	9	2	7	0	0
	Blood	1	1	0	0	0	0	0	0	0	0	0	0	0	0	0
Others and unknown^c	Stool	433	363	21	16	18	4	4	2	2	0	6^d	3	13	3	1
	Blood	28	17	2	2	4	1	1	1	0	0	0	0	1	1	0
Total	Stool	831	679	38	25	25	11	33	2	11	2	46	29	26	18	12
	Blood	44	26	3	3	4	1	6	1	0	0	5	5	1	3	1

R1 indicates resistance to only one class, R2 to only 2 classes, etc. All isolates with the R6 through R8 pattern were resistant to ceftriaxone.

The ampicillin, chloramphenicol, streptomycin, sulfisoxazole, and tetracycline (ACSSuT) column shows isolates with resistance to at least ampicillin, chloramphenicol, streptomycin, sulfisoxazole, and tetracycline, with or without other agents, and so includes all 34 isolates in the ACSSuT Only column, and 14 of the isolates in the ceftriaxone column. All ceftriaxone-resistant isolates were resistant to other agents; 16 fell into the R6 through R8 categories, 10 into the R3 category, and 1 into the R2 category (resistant to ampicillin [penicillins], ceftriaxone, and ceftiofur [cephems]). The nalidixic acid column shows isolates with resistance to this agent with or without other agents, and so includes all 13 isolates in the next column.

Stool isolates comprised more than 75 serotypes, including, most commonly, I 4,[5],12,i- (54 isolates), Javiana (28 isolates), Heidelberg and Montevideo (26 isolates each), as well as partially serotyped and not serotyped isolates. Blood isolates were of 17 known and 1 undetermined serotypes.

Known serotypes were Paratyphi B var. L(+) tartrate+ (2 isolates), Agona, Concord, and Oranienburg (1 isolate each); one isolate was partially serotyped as Subspecies I, Group B.

Of the 875 isolates, 705 (81%) were pansusceptible. The most common serotypes of pansusceptible isolates were Enteritidis (20%), Typhimurium (14%), and Newport (11%) (Table 1). No other serotype contributed more than 6% of pansusceptible isolates. Among these 705 patients, 26% were <5 years old, 17% were 5–17 years old, 46% were 18–64 years old, and 11% were >64 years old.

The 165 resistant isolates fell into eight nonoverlapping resistance categories: 41 (25% of the 165 isolates) R1, 28 (17%) R2, 29 (18%) R3, 12 (7.3%) R4, 39 (24%) R5, 3 (1.8%) R6, 11 (6.7%) R7, and 2 (1.2%) R8. Some isolates also had the following resistance categories: 51 (31% of 165) ACSSuT (including 34 ACSSuT-only), 27 (16%) ceftriaxone (all with resistance to other agents), and 21 (13%) nalidixic acid (including 13 nalidixic acid–only) (Table 1). The proportion of isolates with other (some overlapping) resistance categories that included ACSSuT resistance was the following: 37 (95%) R5, 1 of 3 R6, all 13 R7 and R8, 2 (10%) nalidixic acid, and 14 (52%) ceftriaxone. Most of the 27 ceftriaxone-resistant isolates fell into 2 categories: R3 (10, 34% of that category) and R7 (11, all of that category). Two isolates were resistant to both nalidixic acid and ceftriaxone; these were the only isolates resistant to eight classes. Both were resistant to ACSSuT and to amoxicillin-clavulanic acid, cefoxitin, and ceftiofur. One was also resistant to gentamicin and kanamycin. Five (0.6%) isolates had only intermediate resistance; these were not included in further analyses.

The 29 R3-resistant Salmonella had 11 resistance profiles. Two predominated: 35% of isolates (5 ser. I 4,[5],12:i:-, 3 Heidelberg, 2 other) were resistant to ampicillin, amoxicillin-clavulanic acid, cefoxitin, ceftiofur, and ceftriaxone, and 28% (3 ser. Typhimurium, 2 Derby, 2 Stanley, 1 Enteritidis) were resistant to streptomycin, sulfisoxazole, and tetracycline. No other profile contributed more than 7% of isolates with R3 resistance. Three serotypes predominated: Heidelberg (24%), I 4,[5],12:i:- (21%), and Typhimurium (17%). Thirty-one percent of patients were <5 years old, 21% were 5–17 years old, 41% were 18–64 years old, and 7% were >64 years old.

The 39 R5-resistant Salmonella had 6 resistance profiles. One predominated: 87% of isolates (29 ser. Typhimurium, 5 other) were ACSSuT-resistant without resistance to other agents. No other profile contributed more than 3% of isolates with R5 resistance. Serotype Typhimurium predominated (82%). Twenty-eight percent of patients were <5 years old, 18% were 5–17 years old, 36% were 18–64 years old, and 18% were >64 years old.

Of the 16 R6- through R8-resistant isolates, all were resistant to ceftriaxone, and 14 (88%) were ACSSuT-resistant (1 of 3 R6, all 11 R7, and both R8). Serotypes were Newport (7; 44%), Typhimurium (4; 25%), Concord (2; 13%), and 1 each of Agona, Dublin, and Oranienburg. Thirteen percent of patients were <5 years old, 25% were 5–17 years old, 56% were 18–64 years old, and 6% were >64 years old.

The 51 ACSSuT-resistant Salmonella had 11 resistance profiles. Two predominated: 68% of isolates (29 ser. Typhimurium, 5 other) were ACSSuT-resistant without resistance to other agents, and 14% (5 ser. Newport, 2 other) were also resistant to amoxicillin-clavulanic acid, cefoxitin, ceftiofur, and ceftriaxone. No other profile contributed >4% of isolates with ACSSuT resistance. Two serotypes predominated: Typhimurium (71%) and Newport (18%). Twenty-four percent of patients were <5 years old, 22% were 5–17 years old, 39% were 18–64 years old, and 16% were >64 years old.

The 27 ceftriaxone-resistant Salmonella had 11 resistance profiles. Two predominated: 37% of isolates (5 ser. I 4,[5],12:i:-, 3 Heidelberg, 2 other) were resistant to ampicillin, amoxicillin-clavulanic acid, cefoxitin, ceftiofur, and ceftriaxone, and 26% (5 ser. Newport, 2 other) were resistant to all these agents plus chloramphenicol, streptomycin, sulfisoxazole, and tetracycline. No other profile contributed more than 8% of isolates with ceftriaxone resistance. Three serotypes predominated: Newport (26%), I 4,[5],12:i:- (22%), and Typhimurium (19%). Fifteen percent of patients were <5 years old, 30% were 5–17 years old, 48% were 18–64 years old, and 7% were >64 years old.

The 21 nalidixic acid-resistant Salmonella had 8 resistance profiles. One predominated: 62% of isolates (11 ser. Enteritidis, 2 other) were solely resistant to nalidixic acid. No other profile contributed more than 10% of isolates with nalidixic acid resistance. Serotype Enteritidis predominated (67%). Nineteen percent of patients were <5 years old, 24% were 5–17 years old, 48% were 18–64 years old, and 10% were >64 years old.

None of the groups of patients with R1- through R6-, ACSSuT-, ACSSuT-only-, nalidixic acid-, or ceftriaxone-resistant infections were significantly different from those with pansusceptible infections with regard to the proportion of patients in any age group, that had diarrhea (94–98% in all groups), that had taken an antimicrobial agent in the month before specimen was collected, or that lived in a particular FoodNet site. A significantly higher proportion of patients with R7-resistant infections (but no other group) had taken a cephalosporin in the 4 weeks before becoming ill (67% versus 4.6%; 25% missing). Both patients with R8-resistant infections were ages 5–17 years versus 17% of patients with susceptible infections. A significantly lower proportion of patients with nalidixic acid–resistant infections were female (24% versus 51%; p<0.2).

Adjusted for age and serotype, the risk of having a bloodstream infection was higher for patients with isolates in most resistance categories than for those with pansusceptible isolates (Table 2): 2.6 times higher (95% CI 1.3–5.2) for R2, 3.5 times higher (95% CI 2.0–6.0) for R3, 3.9 times higher (95% CI 1.4–11) for R5, 10 times higher (95% CI 4.7–21) for R6, 2.2 times higher (95% CI 1.3–3.8) for ACSSuT, 2.6 times higher (95% CI 1.5–4.6) for ACSSuT-only, and 3.1 times higher (95% CI 1.1–8.8) for nalidixic acid resistance. Adjusted for age and serotype, the risk of hospitalization was higher for patients in several resistance categories than for those with pansusceptible isolates: 1.8 times higher (95% CI 1.2–2.5) for R3, 3.9 times higher (95% CI 1.9–8.1) for R8, and 1.7 times higher (95% CI 1.1–2.5) for ceftriaxone resistance. With the same adjustment, the risk of a hospital stay >3 days was 2.0 times higher (95% CI 1.3–3.0) for patients with R5, 1.7 times higher (95% CI 1.1–2.7) for ACSSuT, and 1.9 times higher (95% CI 1.2–3.2) for ACSSuT-only resistance. Overall, 36 (82%) of patients with bloodstream infections and 207 (25%) of patients with noninvasive infections were hospitalized (relative risk 11, 95% CI 5.4–24). When adjusted for bloodstream infection as well as age and serotype, the risk of hospitalization remained elevated in the same groups as without the adjustment: 1.4 times higher (95% CI 1.0–2.0) for R3, 4.1 times higher (95% CI 2.0–8.2) for R8, and 1.7 times higher (95% CI 1.1–2.4) for ceftriaxone resistance. With the same adjustment, risk of hospital stay >3 days remained elevated only for patients with R5 resistance (1.5 times higher [95% CI 1.0–2.3]).

Table 2.

Relative Risk (RR) for Selected Factors Including Clinical Outcomes Among Patients with Pansusceptible Versus Resistant Nontyphoidal Salmonella Infections

A.
	Pansusceptible		Resistant to only 1 antimicrobial class				Resistant to only 2 antimicrobial classes				Resistant to only 3 antimicrobial classes				Resistant to only 4 antimicrobial classes				Resistant to only 5 antimicrobial classes
			(R1 ^a				(R2)				(R3)				(R4)				(R5)
	(n=705)		(n=41)				(n=28)				(n=29)				(n=12)				(n=39)
Factor	No.	%	No.	%	RR	95% CI	No.	%	RR	95% CI	No.	%	RR	95% CI	No.	%	RR	95% CI	No.	%	RR	95% CI
Adjusted for age and serotype
Bloodstream infection^b	26	3.7	3	7.3	0.9	0.2–3.6	3^c	11	2.6	1.3–5.2	4	14	3.5	2.0–6.0	1	8.3	0.1	0.0–173535	6	15	3.9	1.4–11
Hospitalized	187	27	10	24	0.9	0.5–1.5	9	32	1.2	0.7–1.9	13	45	1.8	1.2–2.5	2	17	0.4	0.1–2.3	17	44	1.2	0.8–1.8
Hospitalized >3 days	77	11	7	17	1.3	0.8–2.1	5	18	1.1	0.7–2.0	4	14	0.8	0.5–1.7	1	8.3	1.0	0.2–4.0	13	33	2.0	1.3–3.0
Died	0	0.0	0	0.0	—^d	—	0	0.0	—	—	0	0.0	—	—	0	0.0	—	—	0	0.0	—	—
Adjusted for age, serotype, and bloodstream infection
Hospitalized	187	27	10	24	0.9	0.6–1.5	9	32	1.1	0.7–1.7	13	45	1.4	1.0–2.0	2	17	0.4	0.1–2.3	17	44	1.0	0.7–1.4
Hospitalized >3 days	77	11	7	17	1.4	0.9–2.1	5	18	1.1	0.7–1.9	4	14	0.8	0.4–1.5	1	8.3	1.0	0.2–4.1	13	33	1.5	1.0–2.3
Died	0	0.0	0	0.0	—	—	0	0.0	—	—	0	0.0	—	—	0	0.0	—	—	0	0.0	—	—

B.
	Resistant to ACSSuT ^e with or without other agents				Resistant to only ACSSuT				Resistant to ceftriaxone ^f				Resistant to nalidixic acid with or without other agents				Resistant to only nalidixic acid
	(n=51)				(n=34)				(n=27)				(n=21)				(n=13)
Factor	No.	%	RR	95% CI	No.	%	RR	95% CI	No.	%	RR	95% CI	No.	%	RR	95% CI	No.	%	RR	95% CI
Adjusted for age and serotype
Bloodstream infection	5	9.8	2.2	1.3–3.8	5	15	2.6	1.5–4.6	1	3.7	0.9	0.1–5.3	3	14	3.1	1.1–8.8	1	7.7	1.2	0.1–11
Hospitalized	21	41	1.2	0.9–1.7	15	44	1.2	0.8–1.8	11	41	1.7	1.1–2.5	7	33	1.5	0.8–2.5	3	28	1.0	0.4–2.6
Hospitalized >3 days	14	27	1.7	1.1–2.7	11	32	1.9	1.2–3.2	5	19	1.0	0.5–1.9	5	24	1.5	0.8–2.6	2	15	1.0	0.4–2.7
Died	0	0.0	—	—	0	0.0	—	—	0	0.0	—	—	1	4.8	—	—	1	7.7	—	—
Adjusted for age, serotype, and bloodstream infection
Hospitalized	21	41	1.0	0.7–1.4	15	44	0.9	0.6–1.4	11	41	1.7	1.1–2.4	7	33	1.4	0.9–2.2	3	28	1.1	0.5–2.5
Hospitalized >3 days	14	27	1.5	1.0–2.3	11	32	1.6	1.0–2.6	5	19	1.1	0.6–1.9	5	24	1.3	0.7–2.1	2	15	1.1	0.5–2.4
Died	0	0.0	—	—	0	0.0	—	—	0	0.0	—	—	1	4.6	—	—	1	7.7	—	—

R1 indicates resistance to only one class, R2 to only 2 classes, etc.

Isolation from blood.

Gray shading indicates statistically significant result.

Model could not converge.

Ampicillin, chloramphenicol, streptomycin, sulfisoxazole, and tetracycline.

All ceftriaxone-resistant isolates were resistant to other agents.

RR, relative risk; CI, confidence interval.

Discussion

Our finding that bloodstream infection was more common in most resistant categories than in pansusceptible nontyphoidal Salmonella infections extends studies that reported a higher rate of bloodstream infections among patients whose isolates were resistant to individual agents or to ACSSuT (Varma et al., 2005b; Crump et al., 2011). Like Varma et al., we found that hospitalization was more likely among persons with ceftriaxone-resistant than pansusceptible Salmonella infections (in our study, all ceftriaxone-resistant isolates were also resistant to other agents). By analyzing isolates both by number of resistance classes and by major resistance phenotypes, we were able to clearly demonstrate that adverse outcomes were associated with several resistance categories.

A major driver of bloodstream infection in our analysis was ACSSuT resistance in serotype Typhimurium. This pattern was described in the 1990s in England and Wales in isolates from persons infected with serotype Typhimurium definitive phage type (DT) 104 (Wall et al., 1994). Human infections were linked to consumption of several food items and to contact with ill farm animals, particularly cattle (Wall et al., 1994). Typhimurium strains with this pattern spread clonally and became common in cattle and in human infections on many continents (Glynn et al., 1998; Poppe et al., 1998; Davis et al., 1999; Besser et al., 2000; Ribot et al., 2002). In the United States, human illnesses were linked to ground beef and raw milk products (Cody et al., 1999; Villar et al., 1999; Olsen et al., 2004; Dechet et al., 2006). The genes that confer this resistance spread to other serotypes, mostly via Inc A/C plasmids (Glenn et al., 2011).

Our study had several limitations. Most U.S. Salmonella infections are acquired by consumption of contaminated food (Scallan et al., 2011). The mode of transmission for the patients in this study, as for most sporadic cases, was not determined. All resistance categories had small numbers, which prevented an examination of clinical outcome by serotype. Although some serotypes have been associated with invasive infection (Crump et al., 2011), the role of resistance versus serotype can be difficult to sort out. However, we controlled for serotype in the analysis. It is possible that some characteristic of patients on which we did not have information (e.g., susceptibility to infection or frequent contact with settings where antimicrobial agents are used, such as hospitals and nursing homes) increased their chances both of acquiring a resistant infection and developing a bloodstream infection, and that the link between resistance and bloodstream infection is not causal. Regardless of causality, patients with resistant infections had more severe outcomes.

Prevention efforts are needed to reduce unnecessary antimicrobial use in patient care settings and in food animals to help prevent the emergence of resistance and infections with resistant nontyphoidal Salmonella.

Footnotes

Acknowledgments

We thank FoodNet and NARMS-participating public health department epidemiology and laboratory groups for sending isolates, interviewing patients, and reviewing charts; Felicita Medalla and Jason Folster for suggestions; the U.S. Food and Drug Administration for contributing funding support for NARMS; and the U.S. Department of Agriculture's Food Safety and Inspection Service and the U.S. Food and Drug Administration for contributing funding support for FoodNet.

Disclosure Statement

No competing financial interests exist.

References

Besser

, Goldoft

, Pritchett

, et al. Multiresistant Salmonella Typhimurium DT104 infections of humans and domestic animals in the Pacific Northwest of the United States. Epidemiol Infect, 2000; 124:193–200.

[CDC] Centers for Disease Control and Prevention. Foodbome Diseases Active Surveillance Network (FoodNet): FoodNet Surveillance Report for 2008 (Final Report) Volume. Atlanta, GA: U.S. Department of Health and Human Services, CDC, 2010. Available at: http://www.cdc.gov/foodnet/AR2008_508Compliant_v3_FINAL.pdf, accessed July 22, 2011 .

[CLSI] Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing; Twenty-Third Informational Supplement. CLSI Document M100-S23. Volume 33. Wayne, PA: CLSI, 2013.

Cody

, Abbott

, Marfin

, et al. Two outbreaks of multidrug-resistant Salmonella serotype Typhimurium DT104 infections linked to raw-milk cheese in Northern California. JAMA, 1999; 281:1805–1810.

Crump

, Medalla

, Joyce

, et al. Antimicrobial resistance among invasive nontyphoidal Salmonella enterica isolates in the United States: National Antimicrobial Resistance Monitoring System, 1996 to 2007. Antimicrob Agents Chemother, 2011; 55:1148–1154.

Davis

, Hancock

, Besser

, et al. Changes in antimicrobial resistance among Salmonella enterica serovar Typhimurium isolates from humans and cattle in the Northwestern United States, 1982–1997. Emerg Infect Dis, 1999; 5:802–806.

Dechet

, Scallan

, Gensheimer

, et al. Outbreak of multidrug-resistant Salmonella enterica serotype Typhimurium definitive type 104 infection linked to commercial ground beef, northeastern United States, 2003–2004. Clin Infect Dis, 2006; 42:747–752.

Devasia

, Varma

, Whichard

, et al. Antimicrobial use and outcomes in patients with multidrug-resistant and pansusceptible Salmonella Newport infections, 2002–2003. Microbial Drug Resistance, 2005; 11:371–377.

Glenn

, Lindsey

, Frank

, et al. Analysis of antimicrobial resistance genes detected in multidrug-resistant Salmonella enterica serovar Typhimurium isolated from food animals. Microb Drug Resist, 2011; 17:407–418.

10.

Glynn

, Bopp

, Dewitt

, Dabney

, Mokhtar

, Angulo

. Emergence of multidrug-resistant Salmonella enterica serotype Typhimurium DT104 infections in the United States. N Engl J Med, 1998; 338:1333–1338.

11.

Guerrant

, Van Gilder

, Steiner

, et al. Practice guidelines for the management of infectious diarrhea. Clin Infect Dis, 2001; 32:331–351.

12.

Helms

, Simonsen

, Molbak

. Quinolone resistance is associated with increased risk of invasive illness or death during infection with Salmonella serotype Typhimurium. J Infect Dis, 2004; 190:1652–1654.

13.

Helms

, Vastrup

, Gerner-Smidt

, Molbak

. Excess mortality associated with antimicrobial drug-resistant Salmonella Typhimurium. Emerg Infect Dis, 2002; 8:490–495.

14.

Holmberg

, Wells

, Cohen

. Animal-to-man transmission of antimicrobial-resistant Salmonella: Investigations of U.S. outbreaks, 1971–1983. Science, 1984; 225:833–835.

15.

Martin

, Fyfe

, Doré

, et al. Increased burden of illness associated with antimicrobial-resistant Salmonella enterica serotype Typhimurium infections. J Infect Dis, 2004; 189:377–384.

16.

Molbak

. Human health consequences of antimicrobial drug-resistant Salmonella and other foodborne pathogens. Clin Infect Dis, 2005; 41:1613–1620.

17.

Olsen

, Ying

, Davis

, et al. Multidrug-resistant Salmonella Typhimurium infection from milk contaminated after pasteurization. Emerg Infect Dis, 2004; 10:932–935.

18.

Poppe

, Smart

, Khakhria

, Johnson

, Spika

, Prescott

. Salmonella typhimurium DT104: A virulent and drug-resistant pathogen. Can Vet J, 1998; 39:559–565.

19.

Ribot

, Wierzba

, Angulo

, Barrett

. Salmonella enterica serotype Typhimurium DT104 isolated from humans, United States, 1985, 1990, and 1995. Emerg Infect Dis, 2002; 8:387–391.

20.

Scallan

. Activities, achievements, and lessons learned during the first 10 years of the Foodborne Diseases Active Surveillance Network: 1996–2005. Clin Infect Dis, 2007; 44:718–725.

21.

Scallan

, Hoekstra

, Angulo

, et al. Foodborne illness acquired in the United States—Major pathogens. Emerg Infect Dis, 2011; 17:7–15.

22.

Solghan

, Dumas

, Root

, et al. Multidrug-resistant nontyphoidal Salmonella in New York state's Foodborne Diseases Active Surveillance Network counties. Foodborne Pathog Dis, 2010; 7:167–173.

23.

Threlfall

, Ward

, Rowe

. Multiresistant Salmonella Typhimurium DT 104 and Salmonella bacteraemia. Lancet, 1998; 352:287–288.

24.

Varma

, Greene

, Ovitt

, Barrett

, Medalla

, Angulo

. Hospitalization and antimicrobial resistance in Salmonella outbreaks, 1984–2002. Emerg Infect Dis, 2005a;11:943–946.

25.

Varma

, Molbak

, Barrett

, et al. Antimicrobial-resistant nontyphoidal Salmonella is associated with excess bloodstream infections and hospitalizations. J Infect Dis, 2005b;191:554–561.

26.

Villar

, Macek

, Simons

, et al. Investigation of multidrug-resistant Salmonella serotype Typhimurium DT104 infections linked to raw-milk cheese in Washington State. JAMA, 1999; 281:1811–1816.

27.

Wall

, Morgan

, Lamden

, et al. A case control study of infection with an epidemic strain of multiresistant Salmonella Typhimurium DT104 in England and Wales. CDR Rev, 1994; 4:R130–R135.

Clinical Outcomes of Nalidixic Acid,Ceftriaxone,and Multidrug-Resistant Nontyphoidal Salmonella Infections Compared with Pansusceptible Infections in FoodNet Sites,2006–2008

Abstract

Background:

Methods:

Results:

Conclusion:

Introduction

Materials and Methods

Surveillance systems

Antimicrobial susceptibility testing

Cohort study design

Statistical analysis

Results

Discussion

Footnotes

Acknowledgments

Disclosure Statement

References