Decreased Ceftriaxone Susceptibility in Emerging (35B and 6C) and Persisting (19A) Streptococcus pneumoniae Serotypes in the United States, 2011-2012: Ceftaroline Remains Active In Vitro among β-Lactam Agents

TEXT

Streptococcus pneumoniae is an important pathogen responsible for community-acquired bacterial pneumonia (CABP), bacteremia, meningitis, and otitis media and continues to be a major cause of morbidity and mortality worldwide (1). In 2000, the seven-valent pneumococcal conjugate vaccine (PCV7) was introduced in the U.S. childhood vaccine schedule, followed by the PCV13 in 2010 (2). Conjugate vaccines were shown to be immunogenic and prevented pneumococcal disease in children and even in immunocompromised patients (3). However, it has been known that vaccine use modifies the epidemiology of pneumococcal disease and colonization, and investigations have documented increases in the rates of carriage and infections caused by non-PCV7 and later non-PCV13 serotypes (1, 4–6).

Recently, the serotype distribution of S. pneumoniae isolates recovered in the United States (2011-2012) was investigated (7). Serotypes 19A, 3, and 35B were found to be the most prevalent serotypes among S. pneumoniae recovered from sampled patients. In addition, serotypes 19A and 35B comprised the majority (81.1%) of S. pneumoniae isolates with elevated ceftriaxone MIC values (≥1 μg/ml). In this study, we investigated the serotypes and geographic distributions of S. pneumoniae isolates that demonstrated elevated MIC results (≥1 μg/ml) for ceftriaxone, a commonly used agent for invasive pneumococcal disease (IPD). Furthermore, the activities of ceftaroline and other comparator agents were quantified when tested against these less-susceptible pneumococcal isolates.

A total of 1,187 S. pneumoniae clinical isolates received from July 2011 through June 2012 as part of the AWARE (Assessing Worldwide Antimicrobial Resistance Evaluation) program, a component of the SENTRY Antimicrobial Surveillance Program, were included in this investigation. These isolates were recovered from hospitalized patients in 63 medical centers located in the nine U.S. Census regions. Isolates were recovered from blood or lower respiratory tract cultures (7). Bacterial identification was performed by the participating microbiology laboratory and confirmed by the central monitoring laboratory (JMI Laboratories, North Liberty, IA, USA). Bacterial identification was confirmed by colony morphology, biochemical algorithms, and the Vitek2 system, as needed. When the bacterial identification was questionable using phenotypic methods or an untypeable serotyping result was obtained by the applied methodology, isolates were subjected to PCR assay for further identification (8).

Isolates were tested for susceptibility by broth microdilution methods, according to the recommendations of the Clinical and Laboratory Standards Institute (CLSI) (9). MIC results for several anti-Gram-positive agents were obtained using panels manufactured by Thermo Fisher Scientific (Cleveland, OH, USA). Validation of the MIC values was performed by concurrent testing of the quality control (QC) strain S. pneumoniae ATCC 49619 (10). In addition, the inoculum density was monitored by colony counts to ensure an adequate number of cells for each testing event. MIC interpretations were based on the CLSI M100-S24 (10) and European Committee on Antimicrobial Susceptibility Testing (EUCAST) criteria (11).

Isolates were subjected to PCR assays for amplification of the cpsB gene as previously described by Leung et al. (12). Amplicons were sequenced on both strands, and the nucleotide sequences were analyzed using the Lasergene software package (DNASTAR, Madison, WI). Sequences were compared to others available via PubMed (see http://www.ncbi.nlm.nih.gov/blast/). Due to sequence homology among certain serotypes, those showing a nucleotide sequence similarity of >99% were grouped (e.g., 9V/9A, 7F/7A, 11A/11D, 15A/15F, 22F/22A, and 15B/15C). All isolates determined to be serogroup 6 by sequencing analysis were subjected to multiplex PCR assays for confirmation and discrimination between 6A/6B and 6C/6D (13).

Overall, ceftriaxone had MIC₅₀ and MIC₉₀ results of ≤0.06 and 1 μg/ml, respectively (Table 1). Totals of 8.7 and 21.0% of all S. pneumoniae isolates were categorized as nonsusceptible to ceftriaxone according to CLSI (≤1 μg/ml for susceptible, nonmeningitis cases) and EUCAST (≤0.5 μg/ml for susceptible) breakpoint criteria, respectively. S. pneumoniae isolates exhibiting ceftriaxone MIC values of ≥1 μg/ml were predominantly (84.3%; 210/249) serotypes 19A, 19F, and 35B, while remaining isolates were associated with 15 other serotypes and 4 nontypeable strains. Serogroup 19 represented 83.5% (86/103) and 54.6% (136/249) of isolates that were nonsusceptible to ceftriaxone according to CLSI and EUCAST criteria, respectively, and these isolates exhibited markedly decreased susceptibility to several agents (0.0 to 36.8% susceptible), except for ceftaroline (≥94.1% susceptible), levofloxacin (99.3% susceptible), linezolid (100.0% susceptible), and vancomycin (100.0% susceptible) (Table 2).

The vast majority of the 35B isolates (95.7%; 88/92) were susceptible to ceftriaxone when applying the CLSI breakpoint (≤1 μg/ml for susceptible) (Table 1). In stark contrast, only 19.6% of the 35B isolates were considered susceptible when the EUCAST criteria were used (≤0.5 μg/ml for susceptible), since this serotype (76.1%; 70/92) showed a ceftriaxone modal MIC result at 1 μg/ml, as previously described (14). The 35B isolates with elevated ceftriaxone MIC values of ≥1 μg/ml had low susceptibility rates to other agents as well, and similar to the results for serogroup 19, ceftaroline, levofloxacin, linezolid, and vancomycin remained active (≥98.6% susceptible), in addition to clindamycin (Table 2). Moreover, penicillin was active against 35B isolates only when a higher CLSI breakpoint (i.e., ≤2 μg/ml for susceptible) was applied. Of note, isolates from serogroup 6 exhibited a bimodal MIC distribution for ceftriaxone, with modal MIC values of ≤0.06 and 0.5 μg/ml (Table 1). In addition, ceftriaxone inhibited 89.9% (62/69) of the emerging 6C/6D serotype isolates at ≤0.5 μg/ml (EUCAST breakpoint).

Among the parenteral cephalosporins, ceftaroline (MIC_50/90, 0.12/0.25 μg/ml) remained active when tested against these isolates with decreased susceptibility to ceftriaxone (i.e., MIC, ≥1 μg/ml), with overall MIC results 8-fold lower than those for ceftriaxone (MIC_50/90, 0.5 μg/ml) (Table 2). Ceftaroline also demonstrated a MIC₅₀ result of 0.12 μg/ml when tested against isolates of serogroup 19 or those of the 35B serotype. S. pneumoniae isolates with elevated ceftriaxone MIC results (≥1 μg/ml) were observed in all U.S. Census regions, but most commonly in the South Atlantic (37.9% of S. pneumoniae isolates within this region) and East South Central (27.8%) regions. Susceptibility rates among other regions varied from 11.9% (West North Central) to 24.9% (East North Central) (Fig. 1). Occurrences of 19A isolates with elevated ceftriaxone MIC results (≥1 μg/ml) varied from 55.6% in the West North Central to 92.3% in the East South Central and South Atlantic regions (Fig. 1), and 35B strains with decreased ceftriaxone susceptibility were also more prevalent in the East South Central and South Atlantic regions (91.7% to 100.0%).

• Open in new tab
• Download powerpoint

FIG 1

Distribution of S. pneumoniae and specific serotypes displaying elevated (≥1 μg/ml) ceftriaxone MIC results within the U.S. Census regions in 2011-2012.

The prevalence of serotype 19A increased in the United States during the post-PCV7 era (2004-2009) (4, 5, 15, 16), but recent investigations have reported that this serotype has decreased since the introduction of PCV13 (6, 17) or possibly prior to its introduction (18). Although serotype 19A has become less prevalent, studies indicated that the nonsusceptibility rates for 19A have increased even further (7, 18), likely due to antimicrobial pressure and consequent expansion of multidrug-resistant (MDR) populations within 19A isolates (i.e., clonal complex 320) (18). PCV7 successfully decreased the prevalence of targeted serotypes and also decreased the nonsusceptibility rates of these serotypes (19). Based on earlier study results (6, 17) and past experience with PCV7, the use of PCV13 should impact pneumococcal disease in a similar fashion and will likely reduce the prevalence of targeted serotypes and nonsusceptibility rates, especially for serotype 19A. However, given the fact that non-PCV13 serotypes seem to be emerging (6, 7, 17), and certain serotypes (i.e., 35B and 6C) exhibit decreased susceptibility to several agents (including ceftriaxone), it appears likely that decreased antimicrobial susceptibility will continue to be a clinical problem (20).

A previous study demonstrated geographic variations in the S. pneumoniae nonsusceptibility rates within the United States, and similar to our findings, higher nonsusceptibility rates were observed in the Southeast (Tennessee and Georgia) (20). The Infectious Diseases Society of America (IDSA) guidelines for the treatment of community-acquired pneumonia in infants and children (2011) recommend empirical treatment with third-generation parenteral cephalosporins (ceftriaxone or cefotaxime) for hospitalized infants and children who are not fully immunized, in regions where the local epidemiology of IPD strains documents high-level penicillin resistance, or for infants and children with life-threatening infection, including those with empyema (21). Additional surveillance studies over time will be required to better understand the pneumococcal epidemiology after introduction of PCV13, but these and other early data suggest the emergence of non-PCV13 serotypes and decreased susceptibility to the recommended parenteral cephalosporins (21). Herein, ceftriaxone covered 79.0 and 91.3% of isolates using current EUCAST and CLSI criteria, respectively (Table 1); therefore, it would seem prudent that revision of the IPD empirical treatment guidelines should be considered, especially for the South Atlantic and East South Central regions.