JOURNAL OF CLINICAL MICROBIOLOGY, Aug. 1992, p. 2191-2194 0095-1137/92/082191-04$02.00/0 Copyright © 1992, American Society for Microbiology
Vol. 30, No. 8
Proposed Interpretive Criteria and Quality Control Parameters for Testing Susceptibility of Neisseria gonorrhoeae to 13-Lactam-Clavulanate Combinations PETER C.
FUCHS,'*
ARTHUR L. BARRY,2 CAROLYN N. BAKER,3 PATRICK R. MURRAY,4 AND JOHN A. WASHINGTON5
St. Vincent Hospital and Medical Center, Portland, Oregon 97225'; The Clinical Microbiology Institute, Tualatin, Oregon 970622; Centers for Disease Control, Atlanta, Georgia 303333; Washington University, St. Louis, Missouri 6311(4; and The Cleveland Clinic Foundation, Cleveland, Ohio 441955 Received 23 January 1992/Accepted 29 April 1992
To support future clinical studies, in vitro susceptibility tests were examined to determine whether Neisseria gonorrhoeae could be tested reliably against two 1-lactam-clavulanate combinations. All isolates that were tested appeared to be susceptible to amoxicillin and ticarcillin in combination with clavulanic acid. In the absence of resistant isolates, only a breakpoint for a susceptible category could be defined for agar dilution tests with amoxicillin-clavulanic acid (MIC of s 2.0/1.0 ,ug/ml is tentatively proposed). For disk diffusion tests, a corresponding breakpoint zone diameter of 228 mm is suggested. The validity of the breakpoints for penicillinase-negative penicillin-resistant strains awaits clinical data. Proposed quality control limits for testing amoxicillin-clavulanic acid by agar dilution and disk diffusion methods are a MIC of 0.25/0.125 to 1.0/0.5 ,ug/ml and zones of 30 to 40 mm in diameter for N. gonorrhoeae ATCC 49226, a MIC of 0.125/0.06 to 0.5/0.25 jg/ml for Staphylococcus aureus ATCC 29213, and zones of 30 to 38 mm for S. aureus ATCC 25923. Ticarcillinclavulanate is currently tested against other species by preparing doubling dilutions of ticarcillin with a constant 2 ,ug of clavulanate per ml. By that method, all gonococci were susceptible to low concentrations. However, the amount of clavulanic acid that is included (2 jug/ml) will, by itself, inhibit many strains of N. gonorrhoeae. Consequently, the role of ticarcillin in the combination cannot be determined, and such tests are not recommended.
Penicillinase-producing Neisseria gonorrhoeae (PPNG) first recognized and reported in the mid-1970s (1, 2, 20). Since then, the prevalence of PPNG has increased to the point that 3-lactams such as penicillin and ampicillin are unreliable and no longer the drugs of choice in the empirical therapy of gonorrhea (3). The 1-lactamases of PPNG are readily inhibited by sodium clavulanate (12). The natural extension of this was the treatment of gonorrhea with ,-lactam-clavulanate combinations. However, most of the reported studies used amoxicillin (3 g) with either 125 or 250 mg of clavulanate, which is substantially less clavulanate than that used in the current American formulations of this combination. In earlier studies, the cure rate for uncomplicated PPNG infections ranged from 63 to 96% (5-9, 13, 17). In one study involving anogenital and rectal gonorrhea, the cure-rate was 36% (21). In the only report we found in which PPNG infections were treated with a 2:1 formulation of amoxicillin-clavulanate, the cure rate for 134 patients was 100% (18). Thus, it appears that amoxicillin-clavulanate as currently formulated may be an effective drug for the treatment of uncomplicated gonorrhea, including cases caused by PPNG. The other P-lactam currently available in combination with clavulanate is ticarcillin. Although this combination would not normally be considered in the treatment of uncomplicated gonorrhea, it has been reported to be efficacious in the treatment of various deep pelvic infections, including pelvic inflammatory disease (10, 11, 19). Since pelvic inflam-
matory disease is often initiated by N. gonorrhoeae, the continued efficacy of ticarcillin-clavulanate against this agent
was
*
should be monitored periodically. The reliability of in vitro tests that measure susceptibility to that combination needs to be documented.
Although neither of these combination drugs is currently those recommended for the treatment of gonorrhea, clinical studies evaluating amoxicillin-clavulanate for that purpose are extant. We undertook in vitro studies to determine whether the resistance or susceptibility of gonococci could be measured reliably. The present study describes interpretive criteria and quality control parameters for use in clinical studies when such combination drugs are tested against N. gonorrhoeae. To determine interpretive criteria for amoxicillin-clavulanic acid, 102 N. gonorrhoeae isolates were tested. These included 21 PPNGs (penicillin MICs: 2.0 to >32 jug/ml) and 81 penicillinase-negative N. gonorrhoeae (PNNG) isolates, which included 20 penicillin-resistant (penicillin MICs: 1.0 to 8.0 ,ug/ml) and 61 penicillin-susceptible or moderately susceptible strains (penicillin MICs: c0.03 to 0.5 jug/ml). Each isolate was tested in triplicate by both agar dilution and disk diffusion methods, following the procedures recommended by the National Committee for Clinical Laboratory Standards (14-16). Disks contained 20/10 ,ug of amoxicillinclavulanate, and for agar dilution tests, the drugs were combined in 2:1 ratios with twofold dilution concentrations ranging from 4/2 to 0.004/0.002 ,ug/ml. Ticarcillin-clavulanate was tested by using disks containing 75 jug of ticarcillin and 10 jug of clavulanate and agar dilution concentrations of ticarcillin ranging from 4.0 to 0.003 ,ug/ml with a constant 2.0 among
Corresponding author. 2191
J. CLIN. MICROBIOL.
NOTES
2192
40
0 OOOA A A
2D
A
A 0 0-
A A A
A A 0
0
z
A A
0 0 AO0 A A ** 0 A
00
(112
0
A
C2 0Q25
S0 00 0
A
A
A
0
0.0 0
0 0 0 0 0 000 0 0 0 0
A 0
0
0 0
0 0000 0 000 00 0 00 0 00 0 0 0 0 0 0 0 00 0
0
a1 0
r>
0
OL02
0 0 0
0 0 0
36
30
40
50
45
ZONE DIAMETER IN MM FIG. 1. Scattergram depicting the mode of three amoxicillin-clavulanate MICs (as micrograms of amoxicillin per milliliter in 2:1 mixtures) plotted against the mean of three zone diameters around disks containing 20/10 ,ug of amoxicillin-clavulanate for each of 102 clinical isolates of N. gonorrhoeae. A, PPNG; *, penicillin-resistant PNNG; 0, penicillin-susceptible or moderately susceptible PNNG.
,ug of clavulanate per ml (14, 15). Benzylpenicillin was tested simultaneously for control purposes. All isolates were susceptible to amoxicillin-clavulanate at .2.0/1.0 ,ug/ml (Fig. 1). Since all of the PPNG isolates reported by Osato et al. (18) were inhibited by .4.0/2.0 ,ug/ml and all 134 responded to amoxicillin-clavulanate therapy, we conclude that all the isolates in our study were truly susceptible. Although 29% of these isolates were inhibited by 1.0 ,ug of clavulanic acid alone per ml, that amount is four to eight times greater than the concentration of clavulanate
in the MIC of combination drug for most isolates. By using 2.0/1.0 ,ug/ml as a MIC breakpoint for the susceptible category, the corresponding zone diameter breakpoint for the disk diffusion test is 228 mm. It is noteworthy that among the penicillin-resistant PNNG isolates, many had penicillin MICs of 4.0 to 8.0 p,g/ml, but all had amoxicillin-clavulanate MICs of .2.0/1.0 ,ug/ml. The MICs of amoxicillin alone were usually the same as (or within 1 dilution of) those of amoxicillin-clavulanate in this subgroup. We are unaware of any studies in which infections due to chromosomally medi-
TABLE 1. Amoxicillin-clavulanate MICs reported for two quality control organisms Organism
No. of occurrences of the following MICs
Laboratory '0.03/0.015
N. gonorrhoeae ATCC 49226
A B C D E
0.125/0.06
(5) 0 (0)
gob
10 (5) 20
(5) 0 (5)
20
7 37 (10) 47
20 (0)
23 (10) 33
3 (3) 7 (5)
17 (2) 13 19 (5) 13 (5)
1 7
62 (12) 74
8 (0) 8
C
Total
10 0 (5)
1.0/0.5
20
A B
D E
0.5/0.25
0(5)
%b S. aureus ATCC 29213
0.25/0.05
13 (5) 0 (5)
Total
0.06/0.03
(p.g/ml)a:
20 (0) 20
20 20 (0)
10 (8) 18
a The number of times that each MIC was reported by each laboratory with the common lot is shown in parentheses; all other reported occurrences were obtained with the lot unique to each laboratory. b Note that data from laboratory E were not included for the percentage calculations. See the text for details.
VOL. 30, 1992
NOTES
2193
TABLE 2. Median and range of zone diameters obtained with 20 and 10 pLg, respectively, of amoxicillin-clavulanate disks and two quality control organismsa Unique lot
Organism
Laboratory
No. of tests
N. gonorrhoeae ATCC 49226
S. aureus ATCC 25923
Common lot
Zone diam (mm) Median Range
No. of tests
Zone diam (mm) Median Range
quality control
% within
range (mm)
A B C D E
150 150 150 150 150
34 35 33 37 38
31-40 32-37 29-41 34-41 34-43
15 15 15 15 15
33 34 33 35 34
32-35 32-35 32-34 34-35 32-36
Combined
750
35
29-43
75
34
32-36
A B C D E
150 150 150 150 150
34 33 35.5 34 34
30-38 31-36 31-38 32-36 31-38
15 15 15 15 15
35 32 34 33 34
33-36
31-33 32-35 32-35 32-35
100 100 100 100
Combined
750
34
30-38
75
34
31-36
100
30-40
100 100 97.3 96.7 90.6 96.9
30-38
100
a Calculated by the method of Gavan et al. (4).
ated intrinsically penicillin-resistant N. gonorrhoeae were treated with amoxicillin-clavulanate, although there are suggestions that such strains may not respond (22). If this proves to be true, then the in vitro susceptibility tests with amoxicillin-clavulanate, as currently performed, will not be able to differentiate between the two types of penicillinresistant gonococci, i.e., the endpoints by both testing methods completely overlap for those two subgroups. Although all isolates were susceptible to low concentrations of ticarcillin-clavulanate, a high percentage (80% of 30 isolates, including PPNG) were inhibited by the constant 2 ,ug of clavulanate per ml in the test system. The effect of ticarcillin in the combination could not be determined, although most strains were inhibited by the lowest concentrations tested (0.003/2.0 p,g/ml). For that reason, susceptibility of N. gonorrhoeae to ticarcillin-clavulanate cannot be measured in a meaningful way by current testing formulations. If testing the susceptibility of N. gonorrhoeae to ticarcillin-clavulanate is ever deemed necessary, a lower concentration of clavulanate in the test system should be considered. To develop quality control parameters for amoxicillinclavulanate, tests were conducted in the five laboratories directed or supervised by each of the authors, following procedures recommended by the National Committee for Clinical Laboratory Standards (16). Each laboratory was supplied with antibiotic powders of certified potency, an XV supplement devoid of cysteine, and two lots of GC agar base (one lot common to all participants and one lot unique to each). The six lots were obtaiined from three different manufacturers: three from Becton-Dickinson Microbiology Systems, two from Difco Laboratories, and one from Oxoid Laboratories. For agar dilution tests, agar plates containing twofold dilutions of amoxicillin-clavulanate at a 2:1 ratio were prepared on the day of te§ting. Replicate tests (each representing a different inoculating preparation) were performed with N. gonorrhoeae ATCC 49226 and Staphylococcus aureus ATCC 29213. Each laboratory generated 20 MIC results by using its unique lot and 5 MIC results by using the common lot with each organism. The results of these tests are summarized in Table 1. Laboratory E reported unusually
low MICs that fell outside the proposed control limits. Since these values involved the laboratory's unique lot of media as well as the common lot, these aberrant results appear to be a laboratory-specific problem rather than medium-related disparities. The exact problem(s) could not be identified conclusively. Analysis of all reported data led us to recommend the following MIC limits (mode + 1 doubling dilution) for agar dilution tests with amoxicillin-clavulanate: 0.25/ 0.125 to 1.0/0.5 ,ug/ml for N. gonorrhoeae ATCC 49226 and 0.125/0.06 to 0.5/0.25 p,g/ml for S. aureus ATCC 29213. All of the results from four laboratories were within these control limits, and the fifth laboratory consistently recorded MICs outside of these limits. For the disk diffusion quality control tests, each laboratory performed 55 separate tests with each of the aforementioned control strains by using the same allotted GC agar bases described above (50 tests with the unique lot and 5 tests with the common lot). Each test involved a disk containing 30-,g of tetracycline for control purposes and three disks containing 20/10 p,g of amoxicillin-clavulanate, each from a different lot (two from Becton-Dickinson Microbiology Systems and one from Difco Laboratories). Thus, each laboratory reported 150 zone diameter measurements for its unique lot of agar and 15 measurements for the common agar lot. No appreciable differences in zone diameters were reported for the three different lots of disks. The results of these tests are summarized in Table 2. By calculating the median zone diameter in millimeters + one-half of the median ranges from individual laboratories as described by Gavan et al. (4), we propose the following control limits: 30 to 40 mm for N. gonorrhoeae ATCC 49226 and 30 to 38 mm for S. aureus ATCC 25923. From all five participants, 97% of the zone diameters with N. gonorrhoeae and 100% of the zone diameters with S. aureus fell within the proposed limits. REFERENCES 1. Ashford, W. A., R. G. Golash, and V. G. Hemmings. 1976. Penicillinase-producing Neisseria gonorrhoeae. Lancet ii:657658. 2. Center for Disease Control. 1976. Penicillinase-producing Neis-
2194
NOTES
seria gonorrhoeae. Morbid. Mortal. Weekly Rep. 23:261. 3. Centers for Disease Control. 1989. Sexually transmitted diseases treatment guidelines. Morbid. Mortal. Weekly Rep. 38(Suppl. 8):1-43. 4. Gavan, T. L., R. N. Jones, A. L. Barry, P. C. Fuchs, E. H. Gerlach, J. M. Matsen, L. B. Reller, C. Thormsberry, and L. D. Thrupp. 1981. Quality control limits for ampicillin, carbenicillin, mezlocillin, and piperacillin disk diffusion susceptibility tests: a collaborative study. J. Clin. Microbiol. 14:67-72. 5. Key, P. R., B.' S. Azadian, and B. A. Evans. 1985. Augmentin compared with amoxycillin in treating uncomplicated gonorrhea. Genitourin. Med. 61:165-167. 6. Latif, A. S., J. Sithole, S. Brumbe, B. Gumbo, M. Kawemba, and R. S. Summers. 1984. Treating gonococcal urethritis in men: oral amoxycillin potentiated by clavulanate compared with intramuscular procaine penicillin. Br. J. Vener. Dis. 60:29-30. 7. Lawrence, A. G., and D. C. Shanson. 1985. Single dose oral amoxycillin 3 g with either 125 mg or 250 mg clavulanic acid to treat uncomplicated congenital gonorrhea. Genitourin. Med. 61:168-171. 8. Lim, K. B., T. Thirumoorthy, C. T. Lee, E. H. Sng, and T. Tan. 1986. Three regimens of procaine penicillin G, Augmentin and probenecid compared for treating acute gonorrhea in men. Genitourin. Med. 62:82-85. 9. Lim, K. B., T. Thirumoorthy, C. T. Lee, E. H. Sng, and T. Tan. 1986. Clinical experience in the use of clavulanic acid/penicillin regimens in the treatment of uncomplicated gonorrhea. Ann. Acad. Med. Singapore 15:258-261. 10. Matsuda, S., M. Suzuki, R. Miyazaki, N. Cho, K. Kunii, K. Fukunaga, S. Hayashi, H. Nakamura, M. Tateno, H. Okada, et al. 1987. Clinical studies of BAL 28500 (clavulanic acid/ticarcillin) in the treatment of pelvioferitoritin and Douglas' abscess. Jpn. J. Antibiot. 40:951-968. 11. McGregor, J. A. 1990. Ticarcillin/clavulanate for the treatment of female genital tract infections. Efficacy, safety and comparative microbiology. J. Reprod. Med. 35(Suppl. 3):333-338. 12. Miller, J. M., C. N. Baker, and C. Thornsberry. 1978. Inhibition of P-lactamase in Neisseria gonorrhoeae by sodium clavulanate. Antimicrob. Agents Chemother. 14:794-796.
J. CLIN. MICROBIOL. 13. Munday, P. E., J. S. Bingham, C. A. Ison, Y. S. Erdman, J. R. Harris, and C. S. Easmon. 1985. Treatment of gonorrhea with clavulanate-potentiated amoxicillin. Sex. Transm. Dis. 12:163-165. 14. National Committee for Clinical Laboratory Standards. 1990. Performance standards for antimicrobial disk susceptibility tests. M2-A4. National Committee for Clinical Laboratory Standards, Villanova, Pa. 15. National Committee for Clinical Laboratory Standards. 1990. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. M7-A2. National Committee for Clinical Laboratory Standards, Villanova, Pa. 16. National Committee for Clinical Laboratory Standards. 1990. Development of in vitro susceptibility test criteria and quality control parameters. M23-T. National Committee for Clinical Laboratory Standards, Villanova, Pa. 17. Osaba, A. O., A. E. Oyelese, J. 0. Ashiru, C. C. Ekweozor, and J. Ochei. 1985. Single dose treatment of gonococcal urethritis with Augmentin in Ibadan. Afr. J. Med. Sci. 14:169-173. 18. Osato, K., H. Tsugami, K. Harada, and J. Maruyama. 1986. Incidence of gonorrhea due to penicillinase producing Neisseria gonorrhoeae in Japan 1981-3 and treatment using a new antibiotic combination, BAL 25000 (amoxycillin and clavulanic acid). Genitourin. Med. 62:158-162. 19. Pastorer, J. G., Jr., K. E. Aldridge, G. L. Cunningham, S. Faro, S. Graffeo, G. S. McNeeley, and J. S. Tan. 1985. Comparison of ticarcillin plus clavulanic acid with cefoxitin in the treatment of female pelvic infections. Am. J. Med. 79(Suppl. 5B):161-163. 20. Phillips, I. 1976. ,-lactamase producing penicillin-resistant gonococcus. Lancet ii:656-657. 21. Schift, R., J. VanUlsen, M. C. Ansink.Schipper, T. vanJoost, M. F. Michel, R. K. Woudstra, and E. Stolz. 1986. Comparison of oral treatment of uncomplicated anogenital and rectal gonorrhea with cefuroxime axetil ester or clavulanic acid potentiated amoxicillin (Augmentin). Genitourin. Med. 62:313-317. 22. Tapsall, J. W., E. A. Phillips, and L. M. Morris. 1987. Chromosomally mediated intrinsic resistance to penicillin of penicillinase producing strains of Neisseria gonorrhoeae isolated in Sydney: guide to treatment with Augmentin. Genitourin. Med. 63:305-308.
Vol. 30, No. 8
Proposed Interpretive Criteria and Quality Control Parameters for Testing Susceptibility of Neisseria gonorrhoeae to 13-Lactam-Clavulanate Combinations PETER C.
FUCHS,'*
ARTHUR L. BARRY,2 CAROLYN N. BAKER,3 PATRICK R. MURRAY,4 AND JOHN A. WASHINGTON5
St. Vincent Hospital and Medical Center, Portland, Oregon 97225'; The Clinical Microbiology Institute, Tualatin, Oregon 970622; Centers for Disease Control, Atlanta, Georgia 303333; Washington University, St. Louis, Missouri 6311(4; and The Cleveland Clinic Foundation, Cleveland, Ohio 441955 Received 23 January 1992/Accepted 29 April 1992
To support future clinical studies, in vitro susceptibility tests were examined to determine whether Neisseria gonorrhoeae could be tested reliably against two 1-lactam-clavulanate combinations. All isolates that were tested appeared to be susceptible to amoxicillin and ticarcillin in combination with clavulanic acid. In the absence of resistant isolates, only a breakpoint for a susceptible category could be defined for agar dilution tests with amoxicillin-clavulanic acid (MIC of s 2.0/1.0 ,ug/ml is tentatively proposed). For disk diffusion tests, a corresponding breakpoint zone diameter of 228 mm is suggested. The validity of the breakpoints for penicillinase-negative penicillin-resistant strains awaits clinical data. Proposed quality control limits for testing amoxicillin-clavulanic acid by agar dilution and disk diffusion methods are a MIC of 0.25/0.125 to 1.0/0.5 ,ug/ml and zones of 30 to 40 mm in diameter for N. gonorrhoeae ATCC 49226, a MIC of 0.125/0.06 to 0.5/0.25 jg/ml for Staphylococcus aureus ATCC 29213, and zones of 30 to 38 mm for S. aureus ATCC 25923. Ticarcillinclavulanate is currently tested against other species by preparing doubling dilutions of ticarcillin with a constant 2 ,ug of clavulanate per ml. By that method, all gonococci were susceptible to low concentrations. However, the amount of clavulanic acid that is included (2 jug/ml) will, by itself, inhibit many strains of N. gonorrhoeae. Consequently, the role of ticarcillin in the combination cannot be determined, and such tests are not recommended.
Penicillinase-producing Neisseria gonorrhoeae (PPNG) first recognized and reported in the mid-1970s (1, 2, 20). Since then, the prevalence of PPNG has increased to the point that 3-lactams such as penicillin and ampicillin are unreliable and no longer the drugs of choice in the empirical therapy of gonorrhea (3). The 1-lactamases of PPNG are readily inhibited by sodium clavulanate (12). The natural extension of this was the treatment of gonorrhea with ,-lactam-clavulanate combinations. However, most of the reported studies used amoxicillin (3 g) with either 125 or 250 mg of clavulanate, which is substantially less clavulanate than that used in the current American formulations of this combination. In earlier studies, the cure rate for uncomplicated PPNG infections ranged from 63 to 96% (5-9, 13, 17). In one study involving anogenital and rectal gonorrhea, the cure-rate was 36% (21). In the only report we found in which PPNG infections were treated with a 2:1 formulation of amoxicillin-clavulanate, the cure rate for 134 patients was 100% (18). Thus, it appears that amoxicillin-clavulanate as currently formulated may be an effective drug for the treatment of uncomplicated gonorrhea, including cases caused by PPNG. The other P-lactam currently available in combination with clavulanate is ticarcillin. Although this combination would not normally be considered in the treatment of uncomplicated gonorrhea, it has been reported to be efficacious in the treatment of various deep pelvic infections, including pelvic inflammatory disease (10, 11, 19). Since pelvic inflam-
matory disease is often initiated by N. gonorrhoeae, the continued efficacy of ticarcillin-clavulanate against this agent
was
*
should be monitored periodically. The reliability of in vitro tests that measure susceptibility to that combination needs to be documented.
Although neither of these combination drugs is currently those recommended for the treatment of gonorrhea, clinical studies evaluating amoxicillin-clavulanate for that purpose are extant. We undertook in vitro studies to determine whether the resistance or susceptibility of gonococci could be measured reliably. The present study describes interpretive criteria and quality control parameters for use in clinical studies when such combination drugs are tested against N. gonorrhoeae. To determine interpretive criteria for amoxicillin-clavulanic acid, 102 N. gonorrhoeae isolates were tested. These included 21 PPNGs (penicillin MICs: 2.0 to >32 jug/ml) and 81 penicillinase-negative N. gonorrhoeae (PNNG) isolates, which included 20 penicillin-resistant (penicillin MICs: 1.0 to 8.0 ,ug/ml) and 61 penicillin-susceptible or moderately susceptible strains (penicillin MICs: c0.03 to 0.5 jug/ml). Each isolate was tested in triplicate by both agar dilution and disk diffusion methods, following the procedures recommended by the National Committee for Clinical Laboratory Standards (14-16). Disks contained 20/10 ,ug of amoxicillinclavulanate, and for agar dilution tests, the drugs were combined in 2:1 ratios with twofold dilution concentrations ranging from 4/2 to 0.004/0.002 ,ug/ml. Ticarcillin-clavulanate was tested by using disks containing 75 jug of ticarcillin and 10 jug of clavulanate and agar dilution concentrations of ticarcillin ranging from 4.0 to 0.003 ,ug/ml with a constant 2.0 among
Corresponding author. 2191
J. CLIN. MICROBIOL.
NOTES
2192
40
0 OOOA A A
2D
A
A 0 0-
A A A
A A 0
0
z
A A
0 0 AO0 A A ** 0 A
00
(112
0
A
C2 0Q25
S0 00 0
A
A
A
0
0.0 0
0 0 0 0 0 000 0 0 0 0
A 0
0
0 0
0 0000 0 000 00 0 00 0 00 0 0 0 0 0 0 0 00 0
0
a1 0
r>
0
OL02
0 0 0
0 0 0
36
30
40
50
45
ZONE DIAMETER IN MM FIG. 1. Scattergram depicting the mode of three amoxicillin-clavulanate MICs (as micrograms of amoxicillin per milliliter in 2:1 mixtures) plotted against the mean of three zone diameters around disks containing 20/10 ,ug of amoxicillin-clavulanate for each of 102 clinical isolates of N. gonorrhoeae. A, PPNG; *, penicillin-resistant PNNG; 0, penicillin-susceptible or moderately susceptible PNNG.
,ug of clavulanate per ml (14, 15). Benzylpenicillin was tested simultaneously for control purposes. All isolates were susceptible to amoxicillin-clavulanate at .2.0/1.0 ,ug/ml (Fig. 1). Since all of the PPNG isolates reported by Osato et al. (18) were inhibited by .4.0/2.0 ,ug/ml and all 134 responded to amoxicillin-clavulanate therapy, we conclude that all the isolates in our study were truly susceptible. Although 29% of these isolates were inhibited by 1.0 ,ug of clavulanic acid alone per ml, that amount is four to eight times greater than the concentration of clavulanate
in the MIC of combination drug for most isolates. By using 2.0/1.0 ,ug/ml as a MIC breakpoint for the susceptible category, the corresponding zone diameter breakpoint for the disk diffusion test is 228 mm. It is noteworthy that among the penicillin-resistant PNNG isolates, many had penicillin MICs of 4.0 to 8.0 p,g/ml, but all had amoxicillin-clavulanate MICs of .2.0/1.0 ,ug/ml. The MICs of amoxicillin alone were usually the same as (or within 1 dilution of) those of amoxicillin-clavulanate in this subgroup. We are unaware of any studies in which infections due to chromosomally medi-
TABLE 1. Amoxicillin-clavulanate MICs reported for two quality control organisms Organism
No. of occurrences of the following MICs
Laboratory '0.03/0.015
N. gonorrhoeae ATCC 49226
A B C D E
0.125/0.06
(5) 0 (0)
gob
10 (5) 20
(5) 0 (5)
20
7 37 (10) 47
20 (0)
23 (10) 33
3 (3) 7 (5)
17 (2) 13 19 (5) 13 (5)
1 7
62 (12) 74
8 (0) 8
C
Total
10 0 (5)
1.0/0.5
20
A B
D E
0.5/0.25
0(5)
%b S. aureus ATCC 29213
0.25/0.05
13 (5) 0 (5)
Total
0.06/0.03
(p.g/ml)a:
20 (0) 20
20 20 (0)
10 (8) 18
a The number of times that each MIC was reported by each laboratory with the common lot is shown in parentheses; all other reported occurrences were obtained with the lot unique to each laboratory. b Note that data from laboratory E were not included for the percentage calculations. See the text for details.
VOL. 30, 1992
NOTES
2193
TABLE 2. Median and range of zone diameters obtained with 20 and 10 pLg, respectively, of amoxicillin-clavulanate disks and two quality control organismsa Unique lot
Organism
Laboratory
No. of tests
N. gonorrhoeae ATCC 49226
S. aureus ATCC 25923
Common lot
Zone diam (mm) Median Range
No. of tests
Zone diam (mm) Median Range
quality control
% within
range (mm)
A B C D E
150 150 150 150 150
34 35 33 37 38
31-40 32-37 29-41 34-41 34-43
15 15 15 15 15
33 34 33 35 34
32-35 32-35 32-34 34-35 32-36
Combined
750
35
29-43
75
34
32-36
A B C D E
150 150 150 150 150
34 33 35.5 34 34
30-38 31-36 31-38 32-36 31-38
15 15 15 15 15
35 32 34 33 34
33-36
31-33 32-35 32-35 32-35
100 100 100 100
Combined
750
34
30-38
75
34
31-36
100
30-40
100 100 97.3 96.7 90.6 96.9
30-38
100
a Calculated by the method of Gavan et al. (4).
ated intrinsically penicillin-resistant N. gonorrhoeae were treated with amoxicillin-clavulanate, although there are suggestions that such strains may not respond (22). If this proves to be true, then the in vitro susceptibility tests with amoxicillin-clavulanate, as currently performed, will not be able to differentiate between the two types of penicillinresistant gonococci, i.e., the endpoints by both testing methods completely overlap for those two subgroups. Although all isolates were susceptible to low concentrations of ticarcillin-clavulanate, a high percentage (80% of 30 isolates, including PPNG) were inhibited by the constant 2 ,ug of clavulanate per ml in the test system. The effect of ticarcillin in the combination could not be determined, although most strains were inhibited by the lowest concentrations tested (0.003/2.0 p,g/ml). For that reason, susceptibility of N. gonorrhoeae to ticarcillin-clavulanate cannot be measured in a meaningful way by current testing formulations. If testing the susceptibility of N. gonorrhoeae to ticarcillin-clavulanate is ever deemed necessary, a lower concentration of clavulanate in the test system should be considered. To develop quality control parameters for amoxicillinclavulanate, tests were conducted in the five laboratories directed or supervised by each of the authors, following procedures recommended by the National Committee for Clinical Laboratory Standards (16). Each laboratory was supplied with antibiotic powders of certified potency, an XV supplement devoid of cysteine, and two lots of GC agar base (one lot common to all participants and one lot unique to each). The six lots were obtaiined from three different manufacturers: three from Becton-Dickinson Microbiology Systems, two from Difco Laboratories, and one from Oxoid Laboratories. For agar dilution tests, agar plates containing twofold dilutions of amoxicillin-clavulanate at a 2:1 ratio were prepared on the day of te§ting. Replicate tests (each representing a different inoculating preparation) were performed with N. gonorrhoeae ATCC 49226 and Staphylococcus aureus ATCC 29213. Each laboratory generated 20 MIC results by using its unique lot and 5 MIC results by using the common lot with each organism. The results of these tests are summarized in Table 1. Laboratory E reported unusually
low MICs that fell outside the proposed control limits. Since these values involved the laboratory's unique lot of media as well as the common lot, these aberrant results appear to be a laboratory-specific problem rather than medium-related disparities. The exact problem(s) could not be identified conclusively. Analysis of all reported data led us to recommend the following MIC limits (mode + 1 doubling dilution) for agar dilution tests with amoxicillin-clavulanate: 0.25/ 0.125 to 1.0/0.5 ,ug/ml for N. gonorrhoeae ATCC 49226 and 0.125/0.06 to 0.5/0.25 p,g/ml for S. aureus ATCC 29213. All of the results from four laboratories were within these control limits, and the fifth laboratory consistently recorded MICs outside of these limits. For the disk diffusion quality control tests, each laboratory performed 55 separate tests with each of the aforementioned control strains by using the same allotted GC agar bases described above (50 tests with the unique lot and 5 tests with the common lot). Each test involved a disk containing 30-,g of tetracycline for control purposes and three disks containing 20/10 p,g of amoxicillin-clavulanate, each from a different lot (two from Becton-Dickinson Microbiology Systems and one from Difco Laboratories). Thus, each laboratory reported 150 zone diameter measurements for its unique lot of agar and 15 measurements for the common agar lot. No appreciable differences in zone diameters were reported for the three different lots of disks. The results of these tests are summarized in Table 2. By calculating the median zone diameter in millimeters + one-half of the median ranges from individual laboratories as described by Gavan et al. (4), we propose the following control limits: 30 to 40 mm for N. gonorrhoeae ATCC 49226 and 30 to 38 mm for S. aureus ATCC 25923. From all five participants, 97% of the zone diameters with N. gonorrhoeae and 100% of the zone diameters with S. aureus fell within the proposed limits. REFERENCES 1. Ashford, W. A., R. G. Golash, and V. G. Hemmings. 1976. Penicillinase-producing Neisseria gonorrhoeae. Lancet ii:657658. 2. Center for Disease Control. 1976. Penicillinase-producing Neis-
2194
NOTES
seria gonorrhoeae. Morbid. Mortal. Weekly Rep. 23:261. 3. Centers for Disease Control. 1989. Sexually transmitted diseases treatment guidelines. Morbid. Mortal. Weekly Rep. 38(Suppl. 8):1-43. 4. Gavan, T. L., R. N. Jones, A. L. Barry, P. C. Fuchs, E. H. Gerlach, J. M. Matsen, L. B. Reller, C. Thormsberry, and L. D. Thrupp. 1981. Quality control limits for ampicillin, carbenicillin, mezlocillin, and piperacillin disk diffusion susceptibility tests: a collaborative study. J. Clin. Microbiol. 14:67-72. 5. Key, P. R., B.' S. Azadian, and B. A. Evans. 1985. Augmentin compared with amoxycillin in treating uncomplicated gonorrhea. Genitourin. Med. 61:165-167. 6. Latif, A. S., J. Sithole, S. Brumbe, B. Gumbo, M. Kawemba, and R. S. Summers. 1984. Treating gonococcal urethritis in men: oral amoxycillin potentiated by clavulanate compared with intramuscular procaine penicillin. Br. J. Vener. Dis. 60:29-30. 7. Lawrence, A. G., and D. C. Shanson. 1985. Single dose oral amoxycillin 3 g with either 125 mg or 250 mg clavulanic acid to treat uncomplicated congenital gonorrhea. Genitourin. Med. 61:168-171. 8. Lim, K. B., T. Thirumoorthy, C. T. Lee, E. H. Sng, and T. Tan. 1986. Three regimens of procaine penicillin G, Augmentin and probenecid compared for treating acute gonorrhea in men. Genitourin. Med. 62:82-85. 9. Lim, K. B., T. Thirumoorthy, C. T. Lee, E. H. Sng, and T. Tan. 1986. Clinical experience in the use of clavulanic acid/penicillin regimens in the treatment of uncomplicated gonorrhea. Ann. Acad. Med. Singapore 15:258-261. 10. Matsuda, S., M. Suzuki, R. Miyazaki, N. Cho, K. Kunii, K. Fukunaga, S. Hayashi, H. Nakamura, M. Tateno, H. Okada, et al. 1987. Clinical studies of BAL 28500 (clavulanic acid/ticarcillin) in the treatment of pelvioferitoritin and Douglas' abscess. Jpn. J. Antibiot. 40:951-968. 11. McGregor, J. A. 1990. Ticarcillin/clavulanate for the treatment of female genital tract infections. Efficacy, safety and comparative microbiology. J. Reprod. Med. 35(Suppl. 3):333-338. 12. Miller, J. M., C. N. Baker, and C. Thornsberry. 1978. Inhibition of P-lactamase in Neisseria gonorrhoeae by sodium clavulanate. Antimicrob. Agents Chemother. 14:794-796.
J. CLIN. MICROBIOL. 13. Munday, P. E., J. S. Bingham, C. A. Ison, Y. S. Erdman, J. R. Harris, and C. S. Easmon. 1985. Treatment of gonorrhea with clavulanate-potentiated amoxicillin. Sex. Transm. Dis. 12:163-165. 14. National Committee for Clinical Laboratory Standards. 1990. Performance standards for antimicrobial disk susceptibility tests. M2-A4. National Committee for Clinical Laboratory Standards, Villanova, Pa. 15. National Committee for Clinical Laboratory Standards. 1990. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. M7-A2. National Committee for Clinical Laboratory Standards, Villanova, Pa. 16. National Committee for Clinical Laboratory Standards. 1990. Development of in vitro susceptibility test criteria and quality control parameters. M23-T. National Committee for Clinical Laboratory Standards, Villanova, Pa. 17. Osaba, A. O., A. E. Oyelese, J. 0. Ashiru, C. C. Ekweozor, and J. Ochei. 1985. Single dose treatment of gonococcal urethritis with Augmentin in Ibadan. Afr. J. Med. Sci. 14:169-173. 18. Osato, K., H. Tsugami, K. Harada, and J. Maruyama. 1986. Incidence of gonorrhea due to penicillinase producing Neisseria gonorrhoeae in Japan 1981-3 and treatment using a new antibiotic combination, BAL 25000 (amoxycillin and clavulanic acid). Genitourin. Med. 62:158-162. 19. Pastorer, J. G., Jr., K. E. Aldridge, G. L. Cunningham, S. Faro, S. Graffeo, G. S. McNeeley, and J. S. Tan. 1985. Comparison of ticarcillin plus clavulanic acid with cefoxitin in the treatment of female pelvic infections. Am. J. Med. 79(Suppl. 5B):161-163. 20. Phillips, I. 1976. ,-lactamase producing penicillin-resistant gonococcus. Lancet ii:656-657. 21. Schift, R., J. VanUlsen, M. C. Ansink.Schipper, T. vanJoost, M. F. Michel, R. K. Woudstra, and E. Stolz. 1986. Comparison of oral treatment of uncomplicated anogenital and rectal gonorrhea with cefuroxime axetil ester or clavulanic acid potentiated amoxicillin (Augmentin). Genitourin. Med. 62:313-317. 22. Tapsall, J. W., E. A. Phillips, and L. M. Morris. 1987. Chromosomally mediated intrinsic resistance to penicillin of penicillinase producing strains of Neisseria gonorrhoeae isolated in Sydney: guide to treatment with Augmentin. Genitourin. Med. 63:305-308.
