Selective medium for Branhamella catarrhalis with acetazolamide as a specific inhibitor of Neisseria spp

JOURNAL OF CLINICAL MICROBIOLOGY, Dec. 1988, p. 2544-2548 0095-1137/88/122544-05$02.00/0 Copyright © 1988, American Society for Microbiology

Vol. 26, No. 12

Selective Medium for Branhamella catarrhalis with Acetazolamide as a Specific Inhibitor of Neisseria spp. MARIO VANEECHOUTTE,* GERDA VERSCHRAEGEN, GEERT CLAEYS, AND ANNE-MARIE VAN DEN ABEELE Department of Medical Microbiology, University Hospital, De Pintelaan 185, B-9000 Ghent, Belgium Received 21 June 1988/Accepted 1 September 1988

Several semiselective media for Branhamella catarrhalis have been proposed. These media allow growth of all members of the family Neisseriaceae, and further differentiation is necessary. By addition of 10 ,g of acetazolamide, a carbonic anhydrase inhibitor, per ml and incubation in air, a medium was created which reduced growth of Neisseria spp. When saliva samples from 178 healthy schoolchildren were screened for the presence of B. catarrhalis, the carrier rate for this organism was estimated to be 48.9% with the selective medium compared with 12.4% when a semiselective medium, which contains only 10 ,ug of vancomycin, 5 ,ug of trimethoprim, and 2 ,ug of amphotericin B per ml, was used and 6.2% when a nonselective blood agar plate was used. The number of Neisseria spp. isolated dropped from 297 on the semiselective agar to 55 on the selective agar.

Branhamella catarrhalis is now generally considered an important pathogen in different diseases. It is responsible for an overall rate of 2 to 26% of the cases of respiratory diseases (8, 11, 23, 28, 29, 35) and was found to be present in 22 of 40 cases of acute laryngitis in otherwise healthy adults in Sweden (27) and in acute maxillary sinusitis (6, 21). It is also reported as the pathogenic agent in 6 to 17% of cultures of acute and chronic otitis media (19, 21, 22, 23). Several reports of ophthalmia neonatorum (15), conjunctivitis (25), and keratitis (16) have been published. Finally, sporadic cases of urethritis (12), peritonitis (10), and systemic diseases such as meningitis, endocarditis, and septicemia (13) are known. Several authors have proposed media which allow growth of members of the family Neisseriaceae and which can differentiate between Neisseria spp. and B. catarrhalis. Jennison et al. (18) and Berger (3) used, respectively, glucose and maltose in the presence of a pH indicator for differentiation. Corkill and Makin (9) proposed the combination of the antibiotics from Thayer-Martin medium (31) and DNase test agar, and Soto-Hernandez et al. (30) developed a medium based on this idea. Van Hare et al. (33) used a semiselective broth to study nasopharyngeal samples in acute otitis media patients. This paper describes the development of a fully selective medium for B. catarrhalis based on the addition of acetazolamide (2-acetylamino-1,3,4-thiodiazol-5-sulfonamide), a synthetic sulfonamide (4, 14, 26), which inhibits the growth of Neisseria spp. The use of this medium in a screening study of healthy schoolchildren is evaluated. (This work was presented in part at the 28th Interscience Conference on Antimicrobial Agents and Chemotherapy, Los Angeles, Calif., 1988.) MATERIALS AND METHODS Organisms. B. catarrhalis strains were obtained from E. Falsen, Culture Collection of the University of Goteborg, Goteborg, Sweden; E. Van Dyck, Institute for Tropical Medicine, Antwerp, Belgium; B. Davies, De Wever Ziekenhuis, Heerlen, The Netherlands; M. A. Calder, City Hospi*

Corresponding author.

tal, Edinburgh, Scotland; U. Berger, University of Heidelberg, Heidelberg, Federal Republic of Germany; T. Nagatake, Institute for Tropical Medicine, Nagasaki, Japan; I. Eliasson, University of Lund, Lund, Sweden; C. Gilquin, Institute for Hygiene and Epidemiology, Brussels, Belgium; M. Piessens-Denef, O. L. V. Ziekenhuis, Mechelen, Belgium; and from our own laboratory. B. catarrhalis (ATCC 25238), B. caviae (ATCC 14659), B. cuniculi (ATCC 14688), and B. ovis (ATCC 19575) type strains and Neisseria spp. were kindly supplied by U. Berger, C. Gilquîn, and E. Van Dyck. Other members of the Neisseriaceae were isolated from clinical samples sent to our laboratory. These strains were used to determine MICs of the antibiotics uspd in the media and to find the optimal combinations of basic media, antibiotics, and conditions of incubation. Antibiotic susceptibility testing. MICs of vancomycin, trimethoprim, amphotericin B, and acetazolamide were determined by the agar dilution technique with Trypticase soy agar (BBL Microbiology Systems, Cockeysville, Md.), supplemented with 5% human or ovine blood. For acetazolamide, plates were incubated in both air and a C02-enriched atmosphere. P-Lactamase production was tested with nitrocefin (Oxoid Ltd., Basingstoke, England). Composition of selective and semiselective media and incubation conditions. (i) Selective and semiselective agars. The basic media tested were Trypticase soy agar and brucella agar (BBL), which were supplemented with 5% human blood or 5% ovine blood. Antibiotics added were 10 ,ug of vancomycin (Vancocin HCI; Eli Lilly Benelux, Brussels, Belgium), 5 ,ug of trimethoprim (Wellcome Reagents, Erembodegem, Belgium), 2 ,ug of amphotericin B (Sigma Chemical Co., St. Louis, Mo.), and 10 ,ug of sodium acetazolamide (Diamox; Lederle, Etten-Leur, The Netherlands) per ml. When these media are incubated in air, Neisseria spp. are inhibited, while incubation in 5% C02 neutralizes the inhibition of carbonic anhydrase by acetazolamide, resulting in normal growth of Neisseria spp. Thus, according to the incubation conditions, a selective or semiselective medium is created. (ii) Selective and semiselective broths. Brain heart infusion broth (Difco Laboratories, Detroit, Mich.) and tryptone soya 2544

SELECTIVE MEDIUM FOR B. CATARRHALIS

VOL. 26, 1988

2545

TABLE 1. Results of susceptibility testing of the antibiotics used in selective medium MIC (,ug/ml)"

Antibiotic

Vancomycin Trimethoprim Amphotericin B Acetazolamide

tn

Organism(s)

Incubation in air Range

50%

905%

Incubation in 5 % C 0 (range)

32 64 >32 100

64 >64 >32 200

25-100

B. catarrhalis B. (atarrlalis B. catairhlalis B. catbarrhalis Other Neisseriac eae

60 60 60 60

32-128 32->64 >32 50->200

Acinetobacter c(alco(ac eticits

4

50->200

Branhamella caviae B. (Ufittic-iili B. ovis Moraxella osloensis Neisseria canis N. cinerea N. flavesc ens N. gonorrhoeae N. laatalnica N. mneningitidis N. ilrnosa N. polysacc/harea N. siecca N. subflav'a subsp. petflava N. subflava subsp. flava

1 1 2 2

>200 100 200 6 0.8 3-6 1.5-3

1 5 4 2 4 13 2 1 3 3 1

50->200 >200

100 100 6->200 >200 100->200 50->200 50-100 0.2-100 0.2-100 50->200

Ob

0-0.8 0-0.8 0.4-1.5 0 0.8-1.5 0.4-0.8 0.4

0.2 100->200 100->200 >200

"50% and 90%, MIC for 50 and 90% of strains tested, respectively. h No growth when incubated in air.

(Oxoid Ltd.) to which 5 jtg of vancomycin, 3 p.g of 2 Fg of amphotericin B, and 10 or 15 p.g of acetazolamide per ml were added were evaluated as selec-

broth

trimethoprim,

tive broths. Omission of acetazolamide creates a semiselective broth, comparable to the one used by Van Hare et al. (33). Broths were incubated on a shaking platform in air. After 18 h, the broths were inoculated onto blood agar with a 10-,tl calibrated loop and incubated in 5% CO,. Study group. Saliva samples from a group of 178 schoolchildren (65 3 year olds, 84 7 year olds, and 29 12 year olds) were obtained in April 1988 to compare the selective broth and agar with the semiselective broth and agar and with nonselective blood agar plates. Paired saliva and throat samples were taken from 75 children. The children were asked to spit saliva into a sterile receptacle. One- to 3-ml portions were obtained. Throat swabs were taken from the tonsils with sterile cotton swabs. A transport medium was not used, since specimens were inoculated within 4 h. Isolation and identification of members of the Neisseriaceae.

All oxidase-positive colonies which were isolated on the various media used to screen the study groups were Gram stained. Gram-negative diplococci positive for oxidase, 4methylumbelliferyl butyrate (32), and DNase and glucose negative were identified as B. catarrhalis. Other gram-

ml when incubated in air. When the incubation atmosphere was enriched with carbon dioxide, most strains could grow at concentrations of 100 ,ug of acetazolamide per ml. Laboratory testing of selective agar. Different combinations of agar base, blood supplement, and antibiotics were inoculated with 59 B. catarrhalis strains and 33 Neisseria sp. strains. The addition of acetazolamide reduced the number of Neisseria spp. able to grow from 30 (medium 2) to 1 (media 3, 4, and 5) (Table 2). A combination of brucella agar with 5% sheep blood supported the growth of all B. cattarrhalis strains. Sheep blood was an important factor in enhancing the growth of B. catarrhalis in air, as can be concluded from the significant rise of viable cultures when Trypticase soy agar with human blood is compared with that containing sheep blood (X2 = 13.55; P < 0.005). TABLE 2. Number of strains of B. catarrhalis and Neisseria spp. able to grow on different combinations of agar, blood supplement, and antibiotics" Medium Medium no.

base

negative, oxidase-positive diplococci were considered Neisseria spp., without further identification to species level. Statistical analysis. Statistical analysis was done by x2 with

Susceptibility testing. Table 1 presents the results of the susceptibility testing of the various antibiotics. All B. catarrhalis strains were highly resistant to the four antibiotics used in the selective medium. All strains belonging to the genus Neisseria were susceptible to