Determination of the antimicrobial activity of 29 clinically important compounds tested against fastidious HACEK group organisms

NOTES

Determination of the Antimicrobial Activity of 29 Clinically Important Compounds Tested Against Fastidious HACEK Group Organisms Kari C. Kugler, Douglas J. Biedenbach, and Ronald N. Jones

HACEK group organisms are very fastidious organisms (Haemophilus spp., Actinobacillus actinomycetemcomitans, Cardiobacterium hominus, Eikenella corrodens, Kingella spp.) that can produce serious invasive infections such as endocarditis. Problems with susceptibility testing methods and their rarity of isolation limit available information of therapeutic choices, particularly among newer antimicrobial agents. Forty-two HACEK strains were tested by the Etest (AB BIODISK, Solna, Sweden) method against 29 anti-

microbial agents. Nearly all compounds exhibited activity with best potency observed among the tested b-lactamase inhibitor combinations, “third- or fourth-generation” cephems, meropenem, fluoroquinolones, and rifampin. Numerous therapeutic options appear possible for initial parenteral treatment followed by oral “step-down” or switch therapy. Each case of HACEK infection therapy should be guided by accurate susceptibility tests, for which the Etest seems preferred for these fastidious species. © 1999 Elsevier Science Inc.

INTRODUCTION

broad range of serious infections such as bacteremias, abscesses, and peridontal infections, but are most recognized as a cause of endocarditis. Geraci and Wilson (1982) found these organisms to be responsible for 57% of the reported cases of infective Gram-negative endocarditis. Even though the overall incidence of endocarditis is low, studies have cited that there has been an increase in rates in the past two decades (Jenny et al., 1987; Wormser and Bottone, 1983). However, it has not been determined whether the increase is due to a higher rate of infection or that the methodologies and awareness in the laboratory has improved detection. The traditional and effective therapy of HACEK organisms had been ampicillin or penicillin alone or in combination with an aminoglycoside (Geraci et al., 1977; Sofianou and Kolokotronis, 1990; Graham et al., 1990; Zinner et al., 1973). Empiric treatment of HACEK cases is most widely used because organism identification and susceptibility testing is problematic. Furthermore, standardization of susceptibility

The acronym HACEK describes a group of Gramnegative organisms that are considered normal flora of the human oral cavity, but upon disturbance of the mucosal integrity can become invasive and pathogenic. The HACEK organisms include Haemophilus spp. (except H. influenzae), Actinobacillus actinomycetemcomitans, Cardiobacterium hominus, Eikenella corrodens, and Kingella spp. (Geraci and Wilson, 1982; Graham et al., 1990; Jenny et al., 1987; Kaplan et al., 1989; Morrison and Wagner, 1989; Stoloff and Gillies, 1986; Wormser and Bottone, 1983; Yagupsky and Dagon, 1997). These organisms are known to cause a From the Medical Microbiology Division, Department of Pathology, University of Iowa College of Medicine, Iowa City, Iowa. Address reprint requests to Dr. Ronald N. Jones, Medical Microbiology Division, C606 GH, Department of Pathology, University of Iowa College of Medicine, Iowa City, IA 52242. Received 1 December 1997; revised and accepted 4 January 1998.

DIAGN MICROBIOL INFECT DIS 1999;34:73–76 © 1999 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010

0732-8893/99/$–see front matter PII S0732-8893(98)00165-5

74 test methods for these organisms has not been defined. Applied testing methodologies have ranged from the Stokes comparative disk method (Stoloff and Gillies, 1986) to agar dilution (Goldstein et al., 1980). Media choices, incubation requirements, and inoculation concentrations have also varied, but it is agreed that enriched media incubated in a carbondioxide rich environment with an extended incubation period aides the growth of these organisms (Goldstein et al., 1980; Wormser and Bottone, 1983). It is the goal of this study to report the antimicrobial patterns of 29 clinically important antimicrobial agents against HACEK organisms using a methodology acknowledged to be accurate for fastidious organisms. The Etest (AB BIODISK, Solna, Sweden) methodology is not compromised by media, incubation, or inoculum size requirements in most types of antimicrobial susceptibility testing. To our knowledge, this study is the first published work that combines susceptibilities from all HACEK organisms with an antimicrobial breadth and test accuracy significant for analysis. A collection of clinically derived HACEK organisms were requested from the American Type Culture Collection (ATCC; Rockville, MD, USA) and supplemented with nine Haemophilus parainfluenzae strains from various antimicrobial studies (University of Iowa, 1994 –96) and five clinical isolates of Eikenella corrodens collected from patients at the University of Iowa Hospitals and Clinics (Iowa City, IA, USA). The final collection tested consisted of Haemophilus aphrophilus (two strains), Haemophilus hemolyticus (two strains), Haemophilus parahaemolyticus (one strain), H. parainfluenzae (15 strains), Haemophilus seguis (one strain), Actinobacillus actinomycetemcomitans (five strains), Cardiobacterium hominus (five strains), E. corrodens (six strains), Kingella dentrificans (one strain), Kingella kingae (three strains), and Kingella indologenes (one strain). For the insurance of test quality control (QC), the following ATCC strains were tested using the guidelines recommended by National Committee for Clinical Laboratory Standards (NCCLS; 1998): H. influenzae ATCC 49247, 49766, and 10211. All 42 HACEK isolates and three ATCC QC strains were tested against 29 different antimicrobial agents: penicillins (six), cephalosporins (12), carbapenems (two), fluoroquinolones (seven), and other agents (two) using the Etest methodology (see Table 1). Using previous experience derived from work with HACEK organisms (University of Iowa, 1995), the isolate inocula were prepared using a 48-h incubated subculture suspended in 0.95% saline to achieve a 1.0 McFarland standard then swabbed onto either chocolate agar (Haemophilus spp.) or Brucella Blood Agar (all other organisms) plates. Agar plates were prepared by AB BIODISK using chocolatized

Notes Mueller–Hinton or Brucella agar base media supplied by BBL (Cockeysville, MD, USA). Placing no more than five Etest strips onto a plate, organisms were incubated in 35°C with 5% CO2 for 48 h. Reading and recording of the minimum inhibitory concentration (MIC) was performed at both 24 and 48 h. A potency comparison of Etest MIC results for 29 antimicrobials can be found in Table 1. MIC frequency distributions were determined for each antimicrobial and presented in this table format using ranges and MIC50 to summarize the central tendency of activity for each antimicrobial agent (with the exception of Haemophilus spp., where MIC50 and MIC90 were presented for species with $ 10 strains). By averaging the MIC50s of all the organisms generated, a potency rank order was determined for each class of antimicrobial compounds. In cases where the averages were the same, the MIC range was utilized to determine the final rank. Within the class of carbapenems, meropenem was clearly more active than imipenem (see Haemophilus results) and had up to a four log2 dilution potency advantage. The activity rank order among fluoroquinolones was as follows: clinafloxacin (average MIC50, 0.006 mg/mL) . ciprofloxacin (0.014 mg/mL) . sparfloxacin (0.029 mg/ mL) . levofloxacin (0.035 mg/mL) . trovafloxacin (0.040 mg/mL) . grepafloxacin (0.047 mg/mL), and ofloxacin (0.069 mg/mL). The rank order of potency for the penicillins was: piperacillin/tazobactam (0.11 mg/mL) . ampicillin/sulbactam 5 ticarcillin (0.21 mg/mL) . ampicillin (0.25 mg/mL) . penicillin (1.5 mg/mL) . oxacillin (155 mg/mL). Cephalosporin activity rank order was: ceftriaxone (0.019 mg/mL) . cefixime (0.024 mg/mL) . cefotaxime (0.040 mg/mL) . cefpodoxime (0.060 mg/mL) . ceftibuten (0.070 mg/mL) . ceftazidime (0.094 mg/mL) 5 cefpirome (0.094 mg/mL) . cefdinir (0.100 mg/mL) . cefepime (0.130 mg/mL) . cefoxitin (0.710 mg/mL) . cefuroxime (0.900 mg/mL) . cefprozil (1.560 mg/mL). MIC50 results for rifampin ranged from 0.064 mg/mL to 0.75 mg/mL, and trimethoprim/sulfamethoxazole was most active against Kingella spp. (MIC50, 0.064 mg/mL) but inactive against some strains of Haemophilus spp. (MICs, . 32 mg/mL). The activity of most drugs tested, with the exception of oxacillin, appeared to be sufficient for therapy against all HACEK organisms. However, when separating by organism genus, some interesting trends were observed. C. hominis appeared to respond quite favorably to all antimicrobial classes; susceptibilities were predominately off-scale (#0.016 mg/mL). These findings are consistent with those reported by Wormser and Bottone (1983) who found C. hominus to be inhibited by nearly all antimicrobials tested at that time. Kaplan et al. (1989) report that A. actinomycetem-

Antimicrobial Agent (no.) Penicillins (6) Ampicillin Oxacillin Penicillin Ticarcillin Ampicillin/Sulbactam Piperacillin/Tazobactam Cephalosporins (12) Cefprozil Cefixime Ceftibuten Cefpodoxime Cefdinir Cefuroxime Cefoxitin Cefotaxime Ceftriaxone Ceftazidime Cefepime Cefpirome Carbapenems (2) Imipenem Meropenem Fluoroquinolones (7) Ciprofloxacin Clinafloxacin Grepafloxacin Levofloxacin Ofloxacin Sparfloxacin Trovafloxacin Other agents (2) Rifampin Trimethoprim/ Sulfamethoxazole

Haemophilus (n 5 21)

Actinobacillus (n 5 5)

Cardiobacterium (n 5 5)

Eikenella (n 5 6)

Kingella (n 5 5)

Range

MIC50/90

Range

MIC50

Range

MIC50

Range

MIC50

0.125–.256 6–.256 0.094–.32 0.047–.256 0.047–6 #0.016–3

0.5/.256 .256/.256 3/.32 0.25/16 0.5/3 0.25/2

0.19–1 .256 1.5–8 0.25–0.5 0.125–0.75 0.19–0.75

0.5 .256 2 0.38 0.38 0.25

#0.016–0.047 0.5–4 0.006–0.064 #0.016–0.75 #0.016–0.47 #0.016–0.25

#0.016 3 0.012 #0.016 #0.016 #0.016

0.094–0.5 12–.256 0.38–8 0.19–1 #0.016–0.38 #0.016–0.25

0.25 .256 2 0.38 0.125 #0.016

0.5–16 #0.016–0.19 #0.016–0.5 #0.016–0.75 0.032–1.5 0.047–6 0.25–4 0.004–0.19 #0.002–0.19 #0.016–1 0.023–0.25 #0.016–0.5

2/6 0.023/0.064 0.032/0.125 0.047/0.38 0.25/0.75 0.25/1.5 2/4 0.016/0.125 0.016/0.125 0.064/0.125 0.094/0.125 0.064/0.25

0.38–3 #0.016–0.064 0.023–0.19 0.032–0.19 0.047–0.19 0.125–0.38 0.25–1.5 0.023–0.064 0.006–0.023 0.064–0.25 0.064–0.25 0.064–0.25

1.5 0.032 0.032 0.125 0.125 0.19 1 0.032 0.006 0.125 0.19 0.125

0.094–0.25 #0.016–0.5 #0.016–0.032 #0.016–0.032 #0.016–0.032 #0.016–0.023 #0.016–0.023 0.047–0.125 0.012–0.064 0.064–0.5 0.064–0.25 0.032–0.19

0.125 #0.016 #0.016 #0.016 #0.016 #0.016 #0.016 0.064 0.032 0.094 0.19 0.125

0.047–8 0.023–0.19 #0.016–0.38 0.047–0.38 0.047–0.19 0.75–8 0.19–1.5 0.023–0.125 0.016–0.064 0.047–0.75 0.012–0.125 #0.016–0.19

0.25–16 0.016–1

1.5/4 0.125/0.25

0.25–0.75 0.064–0.19

0.38 0.125

#0.002–0.016 #0.002–0.016

0.016 #0.002

0.004–0.19 #0.002–0.125 0.023–6 0.008–0.38 0.023–0.75 0.008–3 0.023–3

0.023/0.047 0.016/0.023 0.125/0.25 0.064/0.125 0.125/0.19 0.064/0.19 0.094/0.25

0.004–0.023 #0.002–0.008 0.008–0.19 0.012–0.047 0.032–0.125 #0.002–0.064 0.008–0.25

0.012 0.003 0.064 0.023 0.064 0.023 0.047

0.004–0.5 #0.002–0.094 0.004–0.19 0.012–1 0.032–8 0.006–0.25 0.008–0.094

0.008 #0.002 0.008 0.016 0.047 0.012 0.012

0.25–2

0.75/1.5

0.094–2

0.75

0.047–0.125

0.064

0.19/0.25

0.023–0.125

0.047

0.5–1.5

1

0.064–.32

Range

MIC50

#0.016–6 0.25–96 0.003–8 #0.016–16 #0.016–4 #0.016

#0.016 2 0.012 #0.016 #0.016 #0.016

4 0.032 0.19 0.094 0.094 4 0.5 0.047 0.032 0.094 0.047 0.094

#0.016–.256 #0.016–3 #0.016–4 #0.016–4 #0.016–8 #0.016–256 #0.016–48 #0.002–1 #0.002–4 #0.016–4 0.032–3 #0.016–.256

0.19 #0.016 0.094 #0.016 0.032 0.023 #0.016 0.016 0.008 0.094 0.125 0.064

0.023–0.19 0.012–0.064

0.125 0.023

0.023–0.094 0.004–0.25

0.047 0.008

0.003–0.012 #0.002–0.004 0.008–0.064 0.008–0.023 0.008–0.47 0.012–0.064 0.016–0.094

0.004 0.003 0.016 0.012 0.016 0.016 0.023

0.006–0.032 0.004–0.012 0.012–0.032 0.023–0.064 0.047–0.125 0.012–0.047 0.012–0.047

0.023 0.006 0.023 0.047 0.094 0.032 0.023

0.38–2

0.75

0.064–0.5

0.125

0.19

0.032–0.5

0.064

0.094–0.38

K.C. Kugler et al.

TABLE 1 Antimicrobial Activity of 29 Antimicrobial Agents Tested Against HACEK Group Fastidious Pathogens by the Etest Method

75

76 comitans was generally more susceptible to “thirdgeneration” cephalosporins than “first or second generation” products; our findings corroborate their results. For instance, ceftriaxone (MIC50, 0.006) was 250-fold more active than the last ranked “secondgeneration” oral agent (cefprozil; MIC50 5 1.5), however differences among “third- or fourth-generation” cephalosporins were minimal. With an increased clinical use of cephalosporins for empiric treatment of these organisms (Goldstein et al., 1978), these findings support their continued efficacy. With reported resistance to beta-lactams in E. corrodens (Sofianou and Kolokotronis, 1990) and patient allergies to penicillins, the need for other chemotherapy options increases. The results generated from

Notes these 42 HACEK organisms indicate that current chemotherapy options are vast, and that the “gold standard” use of penicillins should be reconsidered when HACEK organisms are suggested or proven, especially with the variable results among Haemophilus spp. Newer fluoroquinolones appear to be excellent choices for treatment as well as broad-spectrum b-lactams (meropenem, cefepime, and b-lactamase inhibitor combinations). Continued studies with recently isolated HACEK organisms need to be continued in order to properly address any emergence of resistance among these species. The availability of the Etest allows these studies to progress with confidence in the results from surveillance work and in the clinical case settings.

REFERENCES Berbari EF, Cockeril FR III, Steckelberg JM (1997) Infective endocarditis due to unusual or fastidious microorganisms. Mayo Clin Proc 72:532–542. Geraci JE, Wilkowske CJ, Wilson WR, Washington JA (1977) Haemophilus endocarditis: Report of 14 patients. Mayo Clin Proc 52:209–215. Geraci JE, Wilson WR (1982) Endocarditis due to Gramnegative bacteria. Report of 56 cases. Mayo Clin Proc 57:145–148. Goldstein EJC, Gombert ME, Agyare EO (1980) Susceptibility of Eikenella corrodens to newer beta-lactam antibiotics. Antimicrob Agents Chemother 18:832–833. Goldstein EJC, Sutter VL, Finegold SM (1978) Susceptibility of Eikenella corrodens to ten cephalosporins. Antimicrob Agents Chemother 14:639–641. Graham DR, Band JD, Thornsberry C, Hollis DG, Weaver RE (1990) Infections caused by Moraxella, Moraxella urethralis, Moraxella-like groups M-5 a, Moraxella-like groups M-5 and M-6, and Kingella kingae in the United States, 1953-1980. Rev Infect Dis 12:423–431. Jenny DB, Letendre PW, Iverson G (1987) Endocarditis caused by Kingella indologenes. Rev Infect Dis 9:787–789.

Kaplan AH, Weber DJ, Oddone EZ, Perfect JR (1989) Infection due to Actinobacillus actinomycetemcomitans: 15 cases and review. Rev Infect Dis 11:46–63. Morrison VA, Wagner KF (1989) Clinical manifestations of Kingella kingae infections: Case report and review. Rev Infect Dis 11:776–782. Sofianou D, Kolokotronis A (1990) Susceptibility of Eikenella corrodens to antimicrobial agents. J Chemother 2:156–158. Stoloff AL, Gillies ML (1986) Infections with Eikenella corrodens in a general hospital: A report of 33 cases. Rev Infect Dis 8:50–53. Wormser GP, Bottone EJ (1983) Cardiobacterium hominis: Review of microbiologic and clinical features Rev Infect Dis 5:680–690. Yagupsky P, Dagan R (1997) Kingella kingae: An emerging cause of invasive infections in young children. Clin Infect Dis 24:860–866. Zinner SH, Daly AK, McCormack WM (1973) Isolation of Eikenella corrodens in a general hospital. Appl Microbiol 25:705–708.