Previous Article | Next Article ![]()
Journal of Clinical Microbiology, January 2002, p. 26-30, Vol. 40, No. 1
0095-1137/02/$04.00+0 DOI: 10.1128/JCM.40.1.26-30.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Patricia Stewart,1,
and Jane L. Burns2*
Chiron Corporation, Department of Pediatrics, University of Washington, Childrens Hospital and Regional Medical Center, Seattle, Washington,1 Division of Infectious Disease, Department of Pediatrics, University of Washington, Childrens Hospital and Regional Medical Center, Seattle, Washington2
Received 16 July 2001/ Returned for modification 29 August 2001/ Accepted 15 October 2001
|
|
|---|
16 µg/ml, and growth on MAC-100 was considered to be equivalent to a tobramycin MIC of
128 µg/ml. Analysis of method-specific detection rates showed that tobramycin-containing medium was more sensitive than the standard method for the detection of tobramycin-resistant P. aeruginosa, Stenotrophomonas maltophilia, and Achromobacter xylosoxidans but was less sensitive for the detection of Burkholderia cepacia than the standard method. When MICs for P. aeruginosa that grew on tobramycin-containing medium were tested by broth microdilution, the MICs for 28 of 121 strains (23%) growing on MAC-25 and 22 of 56 strains (39%) growing on MAC-100 were MICs <16 and <128 µg/ml, respectively. Addition of a tobramycin-containing MacConkey plate to the routine media for sputum culture may provide additional, clinically relevant microbiologic data. |
|
|---|
Following the reported efficacy of inhaled tobramycin in CF patients (3, 14) the question of increasing tobramycin resistance among sputum isolates of P. aeruginosa and other gram-negative organisms gained importance. If antibiotic selection occurs related to the chronic use of inhaled drug, the ability to detect even small numbers of tobramycin-resistant isolates in sputum may be significant. As noted above, the methods currently in use may underrepresent tobramycin resistance. Thus, development of a technique that would sample the resistance phenotype of all of the organisms present in a given sputum sample was proposed. A previous study by Maduri-Traczewski et al. utilized a culture method in which each P. aeruginosa clone was tested individually for susceptibility, by directly plating sputum to antibiotic-containing medium (12). They concluded that utilization of antibiotic-containing primary media accurately and promptly detected antibiotic-resistant organisms compared to standard quantitative cultures from which colonies were selected for subsequent MIC testing. The current study specifically examines the utility of tobramycin-containing primary media for the detection of tobramycin-resistant organisms from the sputum of CF patients.
(This work was presented in part at the 14th Annual North American Cystic Fibrosis Conference, 9 to 12 November 2000, Baltimore, Md.)
|
|
|---|
![]() View larger version (26K): [in a new window] |
FIG. 1. Diagram of the processing of sputum from 240 patients at seven CF centers in the United States. All samples were sent to a single laboratory (Childrens Hospital and Regional Medical Center, Seattle, Wash.), where they were processed by both the standard method and the antibiotic-containing-medium method (MAC-25 and MAC-100). Tobramycin MICs were determined for all morphologically distinct P. aeruginosa isolates and for all other non-lactose-fermenting gram-negative bacilli.
|
16 µg/ml. Similarly, MAC-100 would suppress the growth of most P. aeruginosa isolates for which the tobramycin MICs are <128 µg/ml while supporting the growth of most P. aeruginosa isolates for which the tobramycin MICs are
128 µg/ml. Thus, growth of P. aeruginosa on MAC-25 was considered to be equivalent to a tobramycin MIC of
16 µg/ml, and growth of P. aeruginosa on MAC-100 was considered to be equivalent to a tobramycin MIC of
128 µg/ml. The method-specific detection rate for each method (standard method [MAC and OFPBL], MAC-25, and MAC-100) was calculated by dividing the number of patients with a given organism detected by that method by the number of patients with that organism detected by all methods.
The percentage of patients whose sputum grew tobramycin-resistant P. aeruginosa detected by the standard method (broth microdilution MIC,
16 µg/ml) was compared with the percentage of patients whose sputum grew P. aeruginosa on MAC-25 by using a two-sided, one-sample McNemar test.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Summary of demographic characteristics and aminoglycoside antibiotic use in this study
|
16 µg/ml. For the 58 patients with resistant P. aeruginosa detected on MAC, the resistant colonies represented, on average, 70% of the total CFU count (median, 97%; range, 0.2 to 100%). Tobramycin-containing medium was much more sensitive for the detection of resistant P. aeruginosa at both the 16-µg/ml level (MAC-25 versus MAC, 98 versus 47%) and the 128-µg/ml levels (MAC-100 versus MAC, 100 versus 21%) (Table 2). Stenotrophomonas maltophilia and Achromobacter xylosoxidans were also detected more frequently on tobramycin-containing media. Conversely, B. cepacia was detected more frequently by the standard method (MAC or OFPBL) than on either MAC-25 or MAC-100. |
View this table: [in a new window] |
TABLE 2. Method-specific detection ratesa(n = 240)
|
16 µg/ml) detected by the standard method (Fig. 2A). Colony counts of resistant P. aeruginosa were higher when it was detected on both MAC and MAC-25 (38 of 55 patients [69%] had counts of >106 CFU/g) than when it was detected on MAC-25 only (13 of 69 patients [19%]). Similar findings were noted for the MAC-100 plates (Fig. 2B).
![]() View larger version (18K): [in a new window] |
FIG. 2. (A) Density of resistant P. aeruginosa detected by the standard method and on MAC-25 (defined as tobramycin MIC of 16 µg/ml by broth microdilution for standard method and as growth on plate for MAC-25). (B) Density of high-level-resistant P. aeruginosa detected by the standard method and by MAC-100 (defined as tobramycin MIC of 128 µg/ml by broth microdilution for standard method and as growth on plate for MAC-100).
|
16 µg/ml, and on MAC-100, 34 of 56 isolates had confirmatory tobramycin MICs of
128 µg/ml, for false-positive rates of 23 and 39%, respectively. Laboratory notes. There were several notable findings regarding the techniques used for this study. First, drug-containing media (MAC-25 and MAC-100) required longer incubation (at least 72 h) for organisms to grow adequately for identification and quantitation. Second, colonies were much smaller and more difficult to differentiate on the drug-containing media. Third, on plates where undiluted sputum was plated, instances occurred where the mucus in the sputum appeared to protect the organisms from the antibiotic in the plate, resulting in false-positive growth.
|
|
|---|
16 µg/ml (as detected by either method) in the current study, compared with the 13% of patients observed from a similar CF population at baseline in the inhaled-tobramycin clinical trials (3). This high number may reflect the fact that 59.6% of patients had received either intravenous or inhaled aminoglycosides within 8 weeks of providing a sputum sample for the study. However, when only the data from the standard method was considered, 24% of patients had P. aeruginosa isolates for which the tobramycin MICs were
16 µg/ml. This number is comparable to the 23% of patients reported to have tobramycin-resistant P. aeruginosa following 6 months of inhaled treatment with tobramycin (3). CF patients typically harbor multiple phenotypically distinct P. aeruginosa organisms, and these organisms frequently have differing antibiotic susceptibility patterns (3, 7, 15, 18). In this study, the relative density of the resistant P. aeruginosa compared to the total colony count of P. aeruginosa was very small for many of the patients from whom tobramycin-resistant P. aeruginosa was isolated. The clinical significance of small numbers of slowly growing tobramycin-resistant P. aeruginosa is unknown. Several published studies of antibiotic treatment for acute exacerbation in CF patients have shown that patients with organisms classified as resistant to the antibiotics they were given responded to treatment as well as those patients with organisms that were classified as susceptible (13, 17; R. Davey, D. Peckham, C. Etherington, and S. Conway. Abstr. 23rd Eur. Cystic Fibrosis Conv., abstr. 409, 2000). This phenomenon is likely the result of multiple factors. First, both susceptible and resistant populations of organisms are present within the lungs of CF patients (18). Second, most aminoglycoside-resistant P. aeruginosa strains from CF patients lack aminoglycoside-modifying enzymes, instead falling into a resistance type termed "impermeability" (11). There is some in vitro evidence that mutants within the impermeability class often show impaired growth in vitro and reduced virulence in animal models (2, 5, 9), suggesting that such strains may be at a competitive disadvantage in the human lung. Finally, antibiotics have been shown to inhibit virulence factors at concentrations below the MIC (4, 6, 8, 10).
There were some patients who would have falsely been reported to have tobramycin-resistant P. aeruginosa had only the results from the tobramycin-containing media been available. These false positives may have been the result of adaptive or transient resistance reverting to susceptibility after a single pass on drug-free media, since a 24 h subculture to blood agar was always performed prior to preparation of the inoculum suspension for the confirmatory broth microdilution MIC test. The phenomenon of adaptive resistance to tobramycin in P. aeruginosa from CF lung infections was reported previously by Barclay et al. (1). Alternatively, when sputum was plated directly on the antibiotic-containing medium, the mucus may have protected the organisms from the antibiotic in the plate, allowing them to grow.
This study focused solely on tobramycin resistance. The previous study by Maduri-Traczewski et al. found that primary culture plates containing ticarcillin and azlocillin increased the sensitivity for detection of resistance to these antibiotics as well (12). Direct susceptibility testing on media containing multiple antibiotics has also been used as a rapid screening test to test for synergy in CF patients; lower counts on drug-containing media compared with non-drug-containing media were considered an indication of synergy (J. Caracciolo, S. Riddell, S. Miller, A. Kerr, and P. Gilligan, Abstr. 96th Gen. Meet. Am. Soc. Microbiol. 1996, abstr. C-309, 1996). It is likely that many CF patients harbor small subpopulations of P. aeruginosa that are resistant to many of the antibiotics that are routinely and repeatedly used for treatment.
Before a decision to add antibiotic-containing plates to the primary plating is made, the costs in both material and technical time should be carefully weighed against the possible benefit of increased detection of resistant organisms. Currently tobramycin-containing MacConkey agar is not available commercially, so plates must either be prepared in-house or be specially ordered. Importantly, because B. cepacia was not detected as well on the tobramycin-containing media and given the significance of this pathogen in the CF patient population, the need to plate to a B. cepacia selective agar would not be obviated by the addition of tobramycin-containing MacConkey agar. In addition, the longer incubation time required for growth on the tobramycin-containing plates and the small, difficult-to-differentiate colonies would increase laboratory technologist time for reading each culture. Finally, because of the false-positive results observed with directly plated sputum, dilution of the sputum in saline may be required prior to plating.
We thank the following principal investigators (PI) and study coordinators for their support of this study: H. Eigen (PI), M. Blagburn, and D. Terrill, Indiana University and Medical Center; R. Anbarm (PI), D. Lindner, and L. Grabowski, SUNY Health Science Center; S. Nasr (PI) and E. Sakmar, University of Michigan; C. Ren (PI) and L. Maffia, University Hospital, Childrens Medical Center at Stonybrook; R. Gibson (PI), S. McNamara, and P. Joy, Childrens Hospital and Regional Medical Center, Seattle, Wash.; P. Hiatt (PI), L. Morris, and D. Treece, Baylor College of Medicine/Texas Childrens Hospital; M. McCarty (PI) and D. Harrington, Deaconess Medical Center, CF Clinic, Spokane, Wash. Most importantly, we thank the patients who volunteered to participate in the study.
Present address: ICOS Corporation, Seattle, Wash. ![]()
Present address: Corixa Corporation, Seattle, Wash. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»