| |||||||||||||||||||||||||||||||||||||||||
We report on three cases of meningococcal disease caused by ciprofloxacin-resistant Neisseria meningitidis, one in North Dakota and two in Minnesota. The cases were caused by the same serogroup B strain. To assess local carriage of resistant N. meningitidis, we conducted a pharyngeal-carriage survey and isolated the resistant strain from one asymptomatic carrier. Sequencing of the gene encoding subunit A of DNA gyrase (gyrA) revealed a mutation associated with fluoroquinolone resistance and suggests that the resistance was acquired by means of horizontal gene transfer with the commensal N. lactamica. In susceptibility testing of invasive N. meningitidis isolates from the Active Bacterial Core surveillance system between January 2007 and January 2008, an additional ciprofloxacin-resistant isolate was found, in this case from California. Ciprofloxacin-resistant N. meningitidis has emerged in North America.
Nasopharyngeal carriage of N. meningitidis precedes meningococcal disease. Approximately 8 to 20% of the general population are asymptomatic carriers,3 and the vast majority remain well. Close contacts of case patients are at increased risk for disease, and chemoprophylaxis is an urgent intervention for prevention of disease.4 The currently recommended chemoprophylactic antibiotics are rifampin, ciprofloxacin, and ceftriaxone.4 The fluoroquinolone ciprofloxacin is frequently prescribed, and ciprofloxacin-resistant N. meningitidis is rare.5
Between January 2007 and January 2008, three cases of infection with ciprofloxacin-resistant N. meningitidis were detected in the United States, one in North Dakota and two in Minnesota. We report on the epidemiologic investigation, the pharyngeal-carriage survey used to assess local circulation of the bacterium, and the molecular characterization of the strain. Recent N. meningitidis isolates from a population-based surveillance system were screened for resistance.
Methods
Case Investigation
Cases of meningococcal disease are reportable to state public health authorities. Confirmed cases were defined as those that were clinically compatible with isolation of N. meningitidis in a specimen from a normally sterile site, such as blood or cerebrospinal fluid.6 Probable cases were defined as those that were clinically compatible with polymerase-chain-reaction (PCR), immunohistochemical, or latex-agglutination evidence of N. meningitidis at a normally sterile site. Sterile cerebrospinal fluid or blood cultures from clinically compatible cases in North Dakota and Minnesota were tested by the Minnesota Department of Health with the use of a PCR assay for N. meningitidis DNA performed according to published methods.7 Cases were investigated by public health staff to identify close contacts for chemoprophylaxis.4
Carriage Survey
A survey of N. meningitidis carriage was performed 2 to 3 weeks after the third case patient (Patient 3) presented with the disease. The survey included this patient's close contacts and a convenience sample of volunteers, 18 years of age or older, from three local universities and a pub. These sites were chosen after in-depth interviews of Patient 3 and associated close contacts, who identified social networks at these sites. The survey was conducted as part of a public health response and was not considered research; thus, institutional review for protection of human subjects was not required.
University students were invited to participate at designated locations and specific times, and passersby were also recruited. Nonstudents who were associated with the universities were also eligible to participate. The pub's owner organized a meeting during which employees and regular patrons were invited to participate. All participants provided written or oral informed consent and answered questions regarding demographic characteristics and risk factors for carriage (for details see the Supplementary Appendix, available with the full text of this article at NEJM.org). Specimens were obtained by swabbing the posterior pharynx and tonsils with a sterile Dacron or rayon swab. Specimens were inoculated directly onto modified Thayer Martin agar (Remel) and incubated in carbon dioxide–enriched, sealed containers at 37°C. The plates were inspected after 24, 48, and 72 hours, and suspected colonies underwent further characterization.
Identification of Isolates and Antimicrobial Susceptibility Testing
N. meningitidis isolates were identified and characterized with the use of conventional microbiologic methods. Serogroup results were determined by slide agglutination and real-time PCR assays (for details on the latter, see the Supplementary Appendix). All isolates were tested for susceptibility to ciprofloxacin, azithromycin, penicillin G, rifampin, tetracycline, and ceftriaxone with the use of Etest (AB Biodisk) or broth-microdilution panels (PML Microbiologicals). The minimum inhibitory concentration (MIC) breakpoints used were those established by the Clinical and Laboratory Standards Institute (CLSI) for broth microdilution.8 Nonsusceptible isolates were confirmed with the use of broth microdilution.
Susceptibility testing was performed with Etest on 155 N. meningitidis isolates received at the Centers for Disease Control and Prevention (CDC) from state and local health departments through the Active Bacterial Core surveillance system from January 2007 through January 2008. Active Bacterial Core surveillance is an active, laboratory, and population-based surveillance system in 10 states, including Minnesota, covering 13% of the U.S. population.9
Molecular Characterization
Ciprofloxacin-resistant N. meningitidis isolates were compared by means of pulsed-field gel electrophoresis with the use of the restriction endonuclease NheI (Promega), as previously described.10 Multilocus sequence typing and typing for the alleles for porin A (porA), porin B (porB), and the ferric enterochelin receptor (fetA) were performed as previously described11,12,13,14 (for details see the Supplementary Appendix). Sequence types were assigned by querying the Neisseria Multi Locus Sequence Typing and neisseria.org Web sites.14,15,16,17 The quinolone-resistance–determining region (QRDR) of the genes encoding subunit A of both DNA gyrase (gyrA) and DNA topoisomerase IV (parC) from neisseria species were amplified by means of PCR and were sequenced with the use of novel primers derived from identical regions in N. meningitidis, N. gonorrhoeae, and N. lactamica. The entire gyrA sequences from selected isolates were also determined by means of PCR with the use of primers derived from identical regions in N. meningitidis, N. gonorrhoeae, and N. lactamica. (For details on sequencing, see the Supplementary Appendix.) Sequence data were analyzed with the use of the Genetics Computer Group package, version 10.3,18 and Lasergene 7 (DNASTAR). Phylogenetic analysis was performed with the use of MEGA3.19
Results
Case Ascertainment
Three cases of meningococcal disease caused by ciprofloxacin-resistant N. meningitidis were detected in North Dakota and Minnesota from January 2007 through January 2008, constituting 9% of a total of 33 confirmed cases reported in these states during that period (Figure 1). The first case patient (Patient 1) was a toddler attending a child-care center in eastern North Dakota, where in August 2006 an adult child-care worker died from probable meningococcal disease. A culture of cerebrospinal fluid from the child-care worker was sterile but was PCR-positive for N. meningitidis. Ciprofloxacin was administered to adults who were close contacts of the child-care worker and rifampin to children at the care center, including Patient 1. There were no secondary cases. In January 2007, Patient 1 was hospitalized with culture-confirmed meningococcal disease and recovered with ceftriaxone treatment.
|
Carriage-Survey and Laboratory Results
Pharyngeal swabs were obtained from 530 carriage-survey participants, including 5 close contacts of Patient 3 and a convenience sample of 525 volunteers from the universities and the pub. The median age was 21 years (range, 18 to 65). Other demographic and risk-factor data from the questionnaire are summarized in the Supplementary Appendix. N. meningitidis was isolated from 40 survey participants (7.5%), including 1 close contact of Patient 3. N. lactamica was isolated from nine participants.
Serogroup Identification and Antimicrobial Susceptibility Testing
The N. meningitidis isolates from all three patients belonged to serogroup B, and bacterial DNA amplified from the cerebrospinal fluid of the child-care worker was positive for the same serogroup (Table 1). Of the 40 N. meningitidis carriage isolates, 9 were serogroup B (including 1 close contact of Patient 3), 2 were serogroup Y, and 29 could not be grouped with slide agglutination methods. The isolates from Patients 1, 2, and 3 and the close contact of Patient 3 were resistant to ciprofloxacin (MIC, 0.25 µg per milliliter, with susceptibility defined as
0.03 µg per milliliter, intermediate susceptibility defined as 0.06 µg per milliliter, and resistance defined as
0.12 µg per milliliter) and susceptible to penicillin, ceftriaxone, rifampin, tetracycline, and azithromycin. All carriage isolates were susceptible to azithromycin; three (8%) had an MIC at the limit of susceptibility (2.0 µg per milliliter).
|
Molecular Characterization
The ciprofloxacin-resistant N. meningitidis isolates from all three case patients and the close contact of Patient 3 had indistinguishable patterns of pulsed-field gel electrophoresis, belonged to the same clonal complex (sequence type [ST]–162), and had the same multilocus sequence type (ST-162) and porA, porB, and fetA types (Table 1). N. meningitidis DNA from the child-care worker's cerebrospinal fluid matched these isolates for porA, fetA, and clonal complex; there was insufficient DNA to characterize porB and the shikimate dehydrogenase gene (aroE), the latter being necessary for multilocus sequence typing. The 40 carriage-study isolates had varied multilocus-sequence-typing profiles, and the only ST-162 isolate was from the close contact of Patient 3.
Sequences of the gyrA gene from the North Dakota–Minnesota ciprofloxacin-resistant N. meningitidis isolates were identical. The QRDR of the gyrA gene had a nucleotide change leading to a threonine-to-isoleucine substitution at amino acid 91 (T91I), a mutation associated with fluoroquinolone resistance.5 The gyrA QRDR amplified from the child-care worker's cerebrospinal fluid revealed no mutations associated with fluoroquinolone resistance. Sequencing of the parC QRDR from ciprofloxacin-resistant isolates did not reveal any resistance-associated mutations.
The first 1265 nucleotides of the gyrA gene from the North Dakota–Minnesota ciprofloxacin-resistant isolates had only 94% similarity to the same region amplified from the child-care worker's cerebrospinal fluid, whereas the final 1486 nucleotides were identical, suggesting horizontal gene transfer (Figure 2A). The gyrA sequence of the isolate from the California patient also showed the T91I mutation; however, the gene was 99.7% similar to the gyrA amplified from the child-care worker's cerebrospinal fluid, which is consistent with intraspecies variation (Figure 2B). The gyrA genes from three N. lactamica isolates in the North Dakota–Minnesota carriage survey were sequenced. As compared with the gyrA gene from the North Dakota–Minnesota ciprofloxacin-resistant N. meningitidis isolates, the gyrA gene from one of the N. lactamica isolates had a single nucleotide difference in the first 1265 nucleotides at the position encoding the T91I mutation in the N. meningitidis isolates (99.9% similarity). The next 1441 nucleotides had 90% similarity, suggesting that an N. lactamica strain was the donor for the horizontal gene transfer (Figure 2C). Phylogenetic analysis of the gyrA QRDR from 132 N. meningitidis, N. gonorrhoeae, and N. lactamica strains indicated clustering by species, except for the North Dakota–Minnesota ciprofloxacin-resistant N. meningitidis isolates, which were grouped within the N. lactamica cluster (for details, see the Supplementary Appendix).
|
We report the emergence of ciprofloxacin-resistant N. meningitidis in North America. Since the case patients had no epidemiologic links and the duration of asymptomatic meningococcal carriage is often limited to a few months or less,3,21 we suspect that the ciprofloxacin-resistant strain was probably maintained in the North Dakota–Minnesota area by multiple carriers. Although our carriage survey identified only one carrier of the ciprofloxacin-resistant strain, this low prevalence must be interpreted cautiously, since virulent N. meningitidis strains often represent a minority of carried strains.21
Fluoroquinolone antibiotics target bacterial DNA gyrase subunits A and B (encoded by gyrA and gyrB, respectively) and topoisomerase IV subunits A and B (encoded by parC and parE, respectively).22 The gyrA T91I mutation in the ciprofloxacin-resistant isolates is analogous to gyrA mutations observed in fluoroquinolone-resistant N. gonorrhoeae23 and has been described in ciprofloxacin-resistant N. meningitidis outside the United States.5 The gyrA sequencing suggests that the ciprofloxacin-resistant strains from California and North Dakota–Minnesota acquired the mutation differently. The California strain most likely acquired resistance by means of point mutation. In contrast, the North Dakota–Minnesota strain probably acquired resistance through horizontal gene transfer from N. lactamica, since a large gyrA segment containing the T91I mutation closely matched the sequence from a locally carried N. lactamica strain. Although this N. lactamica sequence lacked the T91I mutation, we believe that horizontal gene transfer remains the most likely mechanism of introduction (as opposed to mutation after transfer), since recombination events are at least 80 times as likely as mutation to introduce a change of any one nucleotide in a N. meningitidis housekeeping gene.24 N. lactamica is a commensal of the human upper respiratory tract that is rarely pathogenic, and the horizontal gene transfer probably occurred in a person who simultaneously carried both neisseria species. N. lactamica carriage is common among infants,25 and the child-care center might have been an ideal setting for interspecies genetic exchange.
Our findings add evidence to the theory that N. meningitidis can acquire resistance from the wider gene pool of related species. In all likelihood, horizontal gene transfers from other neisseria species introduced sulfonamide and β-lactam antibiotic resistance into some N. meningitidis strains.26 N. lactamica with decreased susceptibility to ciprofloxacin has been suggested as a potential source for ciprofloxacin resistance in N. meningitidis.27 Since most meningococci do not carry fluoroquinolone-resistance mutations, it seems unlikely that the mutation leads to any survival advantage in the absence of fluoroquinolone exposure. Therefore, use of fluoroquinolones, one of the most commonly prescribed classes of broad-spectrum antibiotics,28 probably facilitated the emergence of fluoroquinolone-resistant strains.
Emerging fluoroquinolone resistance raises concern about current treatment and chemoprophylaxis recommendations for meningococcal disease. Practice guidelines from the Infectious Diseases Society of America consider fluoroquinolones an alternative therapy for presumptive or confirmed meningococcal meningitis.29 Chemoprophylaxis recommendations evolved in the 1960s and 1970s in response to emerging resistance to sulfonamide and frequent adverse reactions to minocycline.30 Cases of rifampin-resistant N. meningitidis, although rare, have been associated with chemoprophylaxis failures.31 Among currently recommended agents, ciprofloxacin is often prescribed as chemoprophylaxis for adults (men and nonpregnant women) because the regimen is simple (a single oral dose) and because it is associated with a low rate of adverse events and relatively few drug–drug interactions. The isolation of the resistant strain from a close contact 2 weeks after receipt of ciprofloxacin might represent failure to clear carriage, although acquisition after chemoprophylaxis is possible.
Out of concern about the potential effect of resistance on the efficacy of chemoprophylaxis, a regional health advisory was issued for eastern North Dakota and western Minnesota, recommending that ciprofloxacin chemoprophylaxis not be used.32 Rifampin or ceftriaxone is recommended in its place. A single dose of azithromycin is effective in eradicating carriage20 and can be considered in areas where the ciprofloxacin-resistant strain is established. However, our finding that 8% of N. meningitidis isolates from the carriage survey had MICs for azithromycin that were at the upper limit of susceptibility is a matter of concern. Therefore, areas with a single sporadic case of ciprofloxacin-resistant N. meningitidis should continue to follow current recommendations to use rifampin, ciprofloxacin, or ceftriaxone. Research is needed to evaluate the prophylactic efficacy of other antibiotics, including oral third-generation cephalosporins.33
Since antimicrobial susceptibility testing for N. meningitidis has not been routinely recommended in the United States, resistant cases may go undetected. At this time, widespread resistance to ciprofloxacin seems unlikely because secondary cases remain rare in the presence of routine ciprofloxacin use. It is unclear whether the ciprofloxacin resistance reported here represents a sporadic phenomenon or the early stage of wider dissemination. Previous experience with the widespread emergence of highly fluoroquinolone-resistant N. gonorrhoeae34 raises concern. Health care providers should report suspected chemoprophylaxis failures to their local health departments. Wider surveillance for antimicrobial-resistant N. meningitidis would help inform future recommendations for chemoprophylaxis and treatment.
No potential conflict of interest relevant to this article was reported.
We thank Karen Anderson and David Lonsway of the Division of Healthcare Quality Promotion, National Center for Preparedness, Detection, and Control of Infectious Diseases, and Luis Lowe of the Division of Viral Diseases, National Center for Immunization and Respiratory Diseases, both at the Centers for Disease Control and Prevention, Atlanta; Joanne Bartkus, John Besser, Richard Danila, and Joseph Mariotti from the Minnesota Department of Health, St. Paul; the clinical-section staff of the Minnesota Department of Health Public Health Laboratory, St. Paul; Stacy Lovelace, Jill Slettland, Lisa Elijah, Lisa Well, John Baird, Terry Dwelle, Jan Trythall, Mike Trythall, Myra Kosse, and Sarah Perius from the North Dakota Department of Health, Bismarck; Kathy McKay, Kathy Anderson, Isaac Triebold, and Jessica Broten from Clay County Public Health, Moorhead, MN; Robyn Litke, Chelsea Matter, and Brady Scribner from Fargo Cass Public Health, Fargo, ND; Pamala D. Kirley from the California Emerging Infections Program, Oakland; Rosilyn Ryals from the Alameda County Public Health Department, Oakland, CA; Susan Farley from Contra Costa Health Services, Martinez, CA; Kathryn E. Arnold from the Georgia Department of Human Resources, Division of Public Health, Atlanta; the Active Bacterial Core Surveillance Team, Emerging Infections Program, Atlanta; and all institutions that participated in the carriage survey.
Source Information
From the Epidemic Intelligence Service Program, Office of Workforce and Career Development (H.M.W., S.C.W.), and the Division of Bacterial Diseases, National Center for Immunization and Respiratory Diseases, Coordinating Center for Infectious Diseases (H.M.W., B.H.H., C.P.H., S.C.W., R.T.N., X.W., S.S., R.D.M., M.J.T., A.C.C., T.A.C., N.E.M., L.W.M.), Centers for Disease Control and Prevention, Atlanta; the Emerging Infections Program, Minnesota Department of Health, St. Paul (B.A.J., A.G., D.J.B., J.R., R.L.); and the North Dakota Department of Health, Bismarck (M.A.S., T.K.M., K.K.).
Address reprint requests to Dr. Wu at the Division of Bacterial Diseases, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, 1600 Clifton Rd., MS C-09, Atlanta, GA 30333, or at hwu{at}cdc.gov.
References
| |||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2010 Massachusetts Medical Society. All rights reserved. |