RESEARCH NOTE

Isolated anti-HBV core phenotype in anti-HCV-positive patients is associated with hepatitis C virus replication H. Wedemeyer, M. Cornberg, B. Tegtmeyer, H. Frank, H. L. Tillmann and M. P. Manns Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, Hannover, Germany

ABSTRACT Recovery from hepatitis B virus (HBV) infection is associated with the presence of antibodies against HBV surface (HBs) antigen and HBV core (HBc) antigen. However, anti-HBs antibodies are lost in many cases, and only anti-HBc antibodies persist. A higher frequency of the anti-HBc-alone pattern has been demonstrated for anti-hepatitis C virus (HCV)-positive patients. In this report, 1126 antiHCV-positive ⁄ anti-HBc-positive patients were studied, and the role of HCV replication in influencing the presence or absence of anti-HBs antibodies was investigated. The anti-HBc-alone phenotype was significantly more frequent in HCV-viraemic than in HCV-recovered patients. This finding represents new information regarding the immunopathogenesis of chronic HCV infection and supports previous data indicating impaired humoral immune responses in HCV infection. Keywords

Antibodies, hepatitis B, hepatitis C

Original Submission: 26 November 2002; Revised Submission: 17 February 2002; Accepted: 11

March 2003 Clin Microbiol Infect 2004; 10; 70–72 The presence of antibodies against both hepatitis B virus (HBV) surface antigen (HBsAg) and HBV core antigen (HBcAg) are serological markers of recovery from HBV infection. However, in many cases, anti-HBc, but no anti-HBs, antibodies can be detected. The clinical relevance of this pheCorresponding author and reprint requests: H. Wedemeyer, Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, 30623 Hannover, Germany Tel: +49 511 532 2093 Fax: +49 511 532 2093 E-mail: [email protected]

nomenon is controversial, and the mechanisms responsible for the loss of anti-HBs antibodies are not known. Previously, it has been shown that subjects with antibodies only to HBV core (antiHBc alone) were more often co-infected with hepatitis C virus (HCV) than subjects who also had antibodies to HBsAg in addition to HBcAg [1]. Recently, Greub and Frei reported a 30% frequency of the anti-HBc-alone pattern among anti-HBc- ⁄ anti-HCV-positive patients [2]. However, about 20% of the anti-HCV-positive patients tested in routine diagnostic laboratories are negative for HCV RNA in serum, indicating recovery from HCV infection. The earlier studies investigating patients with antibodies against HBV core alone did not distinguish between HCV-viraemic and HCV RNA-negative patients [1,2]. Since it is not known whether active HCV replication is associated with the anti-HBc-alone phenotype in anti-HCV-positive subjects, the present study addressed this issue in 1126 anti-HCV-positive and anti-HBc-positive individuals. In total, 3153 consecutive anti-HCV-positive sera collected from patients between 1992 and 2001 at the Hannover Medical School, Germany were investigated for serological markers of ongoing or previous HBV infection. All patients tested positive for anti-HCV in the Laboratory of Gastroenterology and Hepatology. Patients comprised inpatients admitted to Hannover Medical School and outpatients seen in the clinic of our department. Patients with HIV co-infection were excluded from the analysis. None of the patients was tested while receiving antiviral treatment. Routine serological markers for hepatitis viruses consisted of anti-hepatitis A virus IgG, anti-HBs, anti-HBc, HBsAg, anti-HCV and antihepatitis D virus. Serological testing was done as described previously with commercial immunoenzymatic assays (Abbott Laboratories, Chicago, IL, USA) [3,4]. Anti-HCV was tested from 1992 on by the second-generation assay, and then after 1995 by the third-generation assay. Detection and quantification of HCV RNA, as well as HCV genotyping, was also performed exactly as described previously [5,6]. Sixty-eight patients underwent liver biopsy. The liver biopsy material was fixed in formalin and embedded in paraffin for routine staining with haematoxylin and eosin. The biopsy material was examined by a single experienced pathologist, who

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases

Research Note 71

was unaware of the clinical and virological data. Biopsy specimens were assessed for fibrosis (staging, score 0–6) and activity (grading, score 0–18) according to the hepatitis activity index of Knodell and co-workers, as modified by Ishak et al. [7]. In total, 1126 anti-HCV-positive individuals were identified who also tested positive for antiHBc antibodies in the absence of HBsAg. HCV RNA was detectable by PCR (detection limit 600 copies ⁄ mL) in 887 (79%) of these patients. As expected for German patients, the dominant HCV genotype was genotype 1 (303 of 426 sera tested; 71%). Antibodies against both HBsAg and HBcAg were found in 881 (79%) patients, while the anti-HBc-alone phenotype was present in 245 (21%) individuals. This frequency is about onethird lower than that found by Greub and Frei [2]. It can be speculated that a different proportion of patients with additional risk factors, e.g., intravenous drug abuse or haemodialysis, might have contributed to the higher number of patients lacking anti-HBs in that study. In addition, genetic factors could potentially contribute to the ability of patients to produce anti-HBs antibodies. Importantly, the anti-HBc-alone pattern was found significantly more often in HCV-viraemic patients (214 of 887; 22%) than in HCV RNAnegative individuals (31 of 249; 13%; p < 0.0001). In addition, among anti-HBs-positive ⁄ HCV RNAviraemic patients, HCV RNA levels, as investigated by Cobas Amplicor, tended to be higher in patients with anti-HBs titres < 50 IU ⁄ mL than in Table 1. Relationship between antibody production and other patient characteristics HCV RNA positive HCV genotyped (n) Genotype 1 Genotype 2 ⁄ 3 Other genotypes Age (years) Female ⁄ male ALT (U ⁄ L)a AST (U ⁄ L)a Bilirubin (mmol ⁄ L)a Inflammationb Fibrosisb

patients with anti-HBs titres > 50 IU ⁄ mL (p ¼ 0.12, data not shown). In contrast, no significant association was evident between the anti-HBcalone antibody phenotype and HCV genotypes, age, sex, previous exposure to hepatitis A virus, liver function tests or histological grading and staging (Table 1). Thus, active HCV replication was associated with decreased anti-HBs antibody production in HBcAg-positive ⁄ HBsAg-negative patients. The present data are in line with previous reports of impaired humoral immune responses after vaccination against HBV or hepatitis A virus in patients with chronic HCV infection [8,9]. What could be the explanation for these observations? First, humoral and cellular immune responses are induced by activation of dendritic cells. This cell type has been shown to be functionally impaired in chronic HCV infection, but not after recovery from HCV infection [10,11]. Second, HCV replication is associated with decreased HBV replication [12]. Thus, some of the patients might not have recovered totally from HBV infection, but HBV could instead have been suppressed by HCV co-infection. In this scenario, latent ongoing chronic HBV infection could be the reason for undetectable anti-HBs antibodies that are captured by low levels of HBsAg. Finally, cellular immune responses against HCV might interfere with HBV-specific B-cell and T-cell responses. Specific cross-reactivity or non-specific bystander mechanisms could be involved in

Anti-HBs+/anti-HBc+ (n = 881; 79%)

Anti-HBs–/anti-HBc+ (‘anti-HBc alone’) (n = 245; 21%)

76% 311 70% 27% 3% 50.4 ± 15.9 335 ⁄ 546 60 ± 57 38 ± 24 15.0 ± 12.1 6.7 ± 2.8 2.8 ± 2.2

87% 115 75% 24% 1% 52.2 ± 15.4 86 ⁄ 159 48 ± 55 29 ± 24 12.6 ± 8.0 5.2 ± 2.7 3.0 ± 2.5

a

p value < 0.001 NS NS NS NS NS NS NS NS NS NS

Liver functions tests were available for 192 patients. Histological grading (inflammatory score 0–18) and staging (fibrosis score 0–6) was performed according to the Ishak score (11) in 68 patients (56 anti-HBs positive and 12 anti-HBc alone). ALT, alanine aminotransferase; AST, aspartate aminotransferase. b

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

72 Clinical Microbiology and Infection, Volume 10 Number 1, January 2004

suppressing anti-HBs-producing B-lymphocytes. Alternatively, a stronger anti-HBV humoral immune response, as indicated by the presence of anti-HBs antibodies, might in part suppress HCV replication [13]. Heterologous immunity and viral interactions are much more frequent than previously thought, and cross-reactivity between influenza A virus and HCV determinant-specific cytotoxic T-cells has been described [14]. The potential clinical relevance of viral co-infections has been emphasised by reports on the improved survival of HIV-positive patients who are co-infected with GB virus C [15]. Future studies will be needed to clarify the detailed mechanisms by which HCV replication suppresses anti-HBs production in anti-HBc-positive patients. Understanding the pathogenesis of HBV–HCV co-infection is important not only for hepatologists, but also for virologists in general, since infection with multiple hepatitis viruses represents a unique and clinically well defined model to study viral interactions and heterologous immunity in humans. REFERENCES 1. Jilg W, Sieger E, Zachoval R, Schatzl H. Individuals with antibodies against hepatitis B core antigen as the only serological marker for hepatitis B infection: high percentage of carriers of hepatitis B and C virus. J Hepatol 1995; 23: 14–20. 2. Greub G, Frei PC. Isolated antibody to hepatitis B core is associated with hepatitis C virus co-infection. Clin Microbiol Infect 2000; 6: 629. 3. Wedemeyer H, Pethig K, Wagner D et al. Long-term outcome of chronic hepatitis B in heart transplant recipients. Transplantation 1998; 66: 1347–1353. 4. Wedemeyer H, Boker KH, Pethig K et al. Famciclovir treatment of chronic hepatitis B in heart transplant recipients: a prospective trial. Transplantation 1999; 68: 1503–1511. 5. Ockenga J, Tillmann HL, Trautwein C, Stoll M, Manns MP, Schmidt RE. Hepatitis B and C in HIV-infected patients. Prevalence and prognostic value. J Hepatol 1997; 27: 18–24. 6. Tillmann HL, Heringlake S, Trautwein C et al. Antibodies against the GB virus C envelope 2 protein before liver transplantation protect against GB virus C de novo infection. Hepatology 1998; 28: 379–384. 7. Ishak K, Baptista A, Bianchi L et al. Histological grading and staging of chronic hepatitis. J Hepatol 1995; 22: 696–699. 8. Keeffe EB, Iwarson S, McMahon BJ et al. Safety and immunogenicity of hepatitis A vaccine in patients with chronic liver disease. Hepatology 1998; 27: 881–886. 9. Wiedmann M, Liebert UG, Oesen U et al. Decreased immunogenicity of recombinant hepatitis B vaccine in chronic hepatitis C. Hepatology 2000; 31: 230–234.

10. Bain C, Fatmi A, Zoulim F, Zarski JP, Trepo C, Inchauspe G. Impaired allostimulatory function of dendritic cells in chronic hepatitis C infection. Gastroenterology 2001; 120: 512–524. 11. Kanto T, Hayashi N, Takehara T et al. Impaired allostimulatory capacity of peripheral blood dendritic cells recovered from hepatitis C virus-infected individuals. J Immunol 1999; 162: 5584–5591. 12. Wedemeyer H, Tegtmeier B, Tillmann HL et al. Impact of HBV coinfection on hepatitis C virus infection: analysis of 3170 anti HCV-positive patients. J Hepatol 2000; 34: 4A. 13. Liaw YF. Concurrent hepatitis B and C virus infection: is hepatitis C virus stronger? J Gastroenterol Hepatol 2001; 16: 597–598. 14. Wedemeyer H, Mizukoshi E, Davis AR, Bennink JR, Rehermann B. Cross-reactivity between hepatitis C virus and influenza A virus determinant-specific cytotoxic T cells. J Virol 2001; 75: 11392–11400. 15. Tillmann HL, Heiken H, Knapik-Botor A et al. Infection with GB virus C and reduced mortality among HIVinfected patients. N Engl J Med 2001; 345: 715–724.

RESEARCH NOTE

Relationship between ciprofloxacin resistance and extended-spectrum b-lactamase production in Escherichia coli and Klebsiella pneumoniae strains V. Tolun1, O¨. Ku¨c¸u¨kbasmacı2, D. To¨ru¨mku¨ney-Akbulut1, C¸. C¸atal1, M. Ang˘-Ku¨c¸u¨ker1 and O¨. Ang˘1 1

Department of Microbiology and Clinical Microbiology, Istanbul Faculty of Medicine and 2Institute of Experimental Medicine Research, Istanbul University, C ¸ apa, Istanbul, Turkey

ABSTRACT Resistance to fluoroquinolones has increased markedly since their introduction. Mechanisms of resistance to any antibiotic class might play a role in resistance to an unrelated antibiotic class. Corresponding author and reprint requests: M. Ang˘-Ku¨c¸u¨ker, Department of Microbiology and Clinical Microbiology, Istanbul Faculty of Medicine, University of Istanbul, 34390, C ¸ apa, Istanbul, Turkey Tel: + 90 0212 6351186 Fax: + 90 212 6351186 E-mail: [email protected]

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

Research Note 73

This study evaluated the relationship between extended-spectrum b-lactamase (ESBL) production and ciprofloxacin resistance in Escherichia coli and Klebsiella pneumoniae strains isolated as causative agents of urinary tract infection. ESBL-producing strains were significantly more frequent among ciprofloxacin-resistant E. coli strains than among ciprofloxacin-susceptible E. coli strains (p ¼ 0.015), but the difference was not significant among K. pneumoniae strains (p ¼ 0.276). Keywords b-Lactamase, ciprofloxacin, Escherichia coli, extended-spectrum b-lactamases, Klebsiella pneumoniae, resistance Original Submission: 6 May 2002; Revised Submission: 12 September 2002; Accepted: 15

October 2002 Clin Microbiol Infect 2004; 10; 72–75 Fluoroquinolones are potent antimicrobial agents used for the treatment of a wide variety of community-acquired and nosocomial infections. However, resistance to fluoroquinolones has increased significantly since their introduction in the late 1980s [1,2]. The mechanisms of fluoroquinolone resistance involve chromosomal mutations and decreased cellular accumulation [3]. Extended-spectrum b-lactamases (ESBLs) confer resistance to newer b-lactam agents possessing an oxyimino group, such as ceftazidime, ceftriaxone, cefotaxime and aztreonam. The genes encoding ESBL production may be chromosomal or plasmid-mediated, and it is well-known that plasmids carrying genes encoding ESBLs may also carry genes encoding resistance to many of the aminoglycosides and trimethoprim–sulphamethoxazole [4]. In this study, the relationship between ESBL production and ciprofloxacin resistance was investigated in Escherichia coli and Klebsiella pneumoniae strains isolated as causative agents of urinary tract infection.

In total, 258 E. coli and 50 K. pneumoniae strains isolated from January to June 2001 from urine samples of outpatients with community-acquired urinary tract infections were investigated. All the strains were isolated from different patients and were stored at )70 C in 10% v ⁄ v glycerol. Susceptibility testing was performed by standard disk diffusion tests [5]. ESBL production was detected by a double disk synergy method [6] in which ceftriaxone, ceftazidime and aztreonam disks were placed 2–3 cm away from an amoxycillin–clavulanate disk. A clear extension of the edge of the inhibition zone of any of the antibiotics towards the disk containing clavulanate was interpreted as positive for ESBL production. The statistical significance of differences in ESBL production between ciprofloxacin-susceptible and ciprofloxacin-resistant strains was evaluated using Fischer’s exact chi-square test. Of the 258 E. coli strains tested, 137 (53.1%) were ciprofloxacin-resistant. Seven (5.1%) of the ciprofloxacin-resistant E. coli strains produced ESBL. In contrast, none of the 121 ciprofloxacinsusceptible E. coli strains were ESBL producers. Of the 50 K. pneumoniae strains tested, 17 (34%) were found to be resistant to ciprofloxacin. Five (29.4%) of the ciprofloxacin-resistant and five (15.1%) of the ciprofloxacin-susceptible K. pneumoniae strains produced ESBL. Statistical analyses showed that the incidence of ESBL production was significantly higher among ciprofloxacin-resistant E. coli strains than among ciprofloxacin-susceptible E. coli strains (p ¼ 0.015). In contrast, the difference was not significant in K. pneumoniae strains (p ¼ 0.276) (Table 1). Paterson et al. [4] found that 15 (60%) of 25 ciprofloxacin-resistant K. pneumoniae strains from patients with bacteraemia were ESBL producers, but only 68 (16%) of 427 ciprofloxacinsusceptible K. pneumoniae isolates produced ESBL. In the same study, the proportion of ESBL producers that were also ciprofloxacin-resistant

Table 1. ESBL production among ciprofloxacin-resistant and -sensitive E. coli and K. pneumoniae strains E. coli (n ¼ 258) Ciprofloxacin-resistant Ciprofloxacin-sensitive K. pneumoniae (n ¼ 50) Ciprofloxacin-resistant Ciprofloxacin-sensitive

ESBL+ n (%)

ESBL– n (%)

Two-sided significance

Total

7 (5.1) 0 (0)

130 (94.9) 121 (100)

p ¼ 0.015 p ¼ 0.015

137 121

5 (29.4) 5 (15.2)

12 (70.6) 28 (84.8)

p ¼ 0.276 p ¼ 0.276

17 33

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

74 Clinical Microbiology and Infection, Volume 10 Number 1, January 2004

was as high as 56% in Turkey, and it was suggested that ciprofloxacin resistance in K. pneumoniae is closely associated with ESBL production. In a European collaborative study [7], ciprofloxacin resistance in ESBL producer and non-producer E. coli and K. pneumoniae strains isolated in 1994 and 1997 was compared; it was shown that ciprofloxacin resistance did not change among ESBL-producing K. pneumoniae, but that ciprofloxacin resistance had increased among ESBL-non-producing K. pneumoniae strains. In Brazil, 94% of 72 ESBL-producing K. pneumoniae strains were found to be ciprofloxacinsusceptible [8]. In a Turkish study of intensive care and renal transplantation patients [9], 40% of the E. coli strains were found to produce ESBL, and in these strains the incidence of ciprofloxacin resistance was as high as 56%. In contrast, 52% of the K. pneumoniae strains were ESBL producers, but the ciprofloxacin resistance rate was only 2%. Several mechanisms are known to determine resistance to fluoroquinolones in Gram-negative bacteria, including mutations in the topoisomerase (II and IV) genes, and decreased accumulation because of outer-membrane alterations and ⁄ or expression of efflux pumps [3,10]. Among these mechanisms, alterations in the GyrA subunit of DNA gyrase have a central role in conferring high-level quinolone resistance in Gram-negative bacteria, such as E. coli and K. pneumoniae [10]. Sanders et al. [11] showed that most quinolone-selected mutants were cross-resistant only to other drugs within this class, but certain mutants of K. pneumoniae selected by quinolones were also less susceptible to b-lactam antibiotics. It was thought that this unusual pattern of multiple drug resistance might be associated with changes in outermembrane proteins. However, it is well-known that the multiple antibiotic resistance (mar) locus in many Gram-negative bacteria, especially E. coli, is responsible for resistance to many unrelated antibiotic classes [12]. Mutations in the mar locus or outer-membrane protein alterations may be potential explanations for the ciprofloxacin resistance observed in ESBL-producing E. coli strains. It is generally considered that fluoroquinolone resistance is chromosomally-mediated [10]. However, plasmid-mediated ciprofloxacin resistance in E. coli and K. pneumoniae has now been reported [13]. The plasmid provided only low-

level ciprofloxacin resistance, but it facilitated high-level resistance when the organism possessed other properties such as porin deficiencies. In the present study, it was shown that all the ESBL-producing E. coli strains were also ciprofloxacin-resistant, and that this relationship was statistically significant. However, there was no such relationship between ESBL production and ciprofloxacin resistance in K. pneumoniae strains, although it should be noted that fewer Klebsiella strains than E. coli strains were studied. This could be a limitation to the detection of a statistically significant difference. There may be other possible explanations for the coexistence of resistance to b-lactams and quinolones in Gramnegative bacteria. For example, bacteria that are able to acquire the ability to produce ESBL can be selected by intensive quinolone use. Further epidemiological and molecular studies are needed for an understanding of the mechanism(s) underlying the selection of cross-resistance. REFERENCES 1. Bauernfeind A, Abele-Horn M, Emmerling P, Jungwirth R. Multiclonal emergence of ciprofloxacin-resistant clinical isolates of Escherichia coli and Klebsiella pneumoniae. J Antimicrob Chemother 1994; 34: 1074–1076. 2. Richard P, Delangle MH, Raffi F, Espaze E, Richet H. Impact of fluoroquinolone administration on the emergence of fluoroquinolone-resistant gram-negative bacilli from gastrointestinal flora. Clin Infect Dis 2001; 32: 162–166. 3. Hooper DC. Emerging mechanisms of fluoroquinolone resistance. Emerg Infect Dis 2001; 7: 337–341. 4. Paterson DL, Mulazimoglu L, Casellas JM et al. Epidemiology of ciprofloxacin resistance and its relationship to extended-spectrum beta-lactamase production in Klebsiella pneumoniae isolates causing bacteremia. Clin Infect Dis 2000; 30: 473–478. 5. National Committee for Laboratory Standards. Performance standards for antimicrobial susceptibility testing; Ninth Informational Supplement. NCCLS document M100-S9. Wayne, Pa: NCCLS, 1999. 6. Jarlier V, Nicolas MH, Fournier G, Philippon A. Extended broad-spectrum beta-lactamases conferring transferable resistance to newer beta-lactam agents in Enterobacteriaceae: hospital prevalence and susceptibility patterns. Rev Infect Dis 1988; 10: 867–878. 7. Babini GS, Livermore DM. Antimicrobial resistance amongst Klebsiella spp. collected from intensive care units in Southern and Western Europe in 1997–1998. J Antimicrob Chemother 2000; 45: 183–189. 8. Gales AC, Bolmstrom A, Sampaio J, Jones RN, Sader HS. Antimicrobial susceptibility of Klebsiella pneumoniae producing extended-spectrum beta-lactamase (ESBL) isolated in hospitals in Brazil. Braz J Infect Dis 1997; 1: 196– 203.

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

Research Note 75

9. Gonullu N, Aktas Z, Salcioglu M, Bal C, Ang O. Comparative in vitro activities of five quinolone antibiotics, including gemifloxacin, against clinical isolates. Clin Microbiol Infect 2001; 7: 499–503. 10. Deguchi T, Fukuoka A, Yasuda M et al. Alterations in the GyrA subunit of DNA gyrase and the ParC subunit of topoisomerase IV in quinolone-resistant clinical isolates of Klebsiella pneumoniae. Antimicrob Agents Chemother 1997; 41: 699–701. 11. Sanders CC, Sanders WE, Goering RV, Werner V. Selection of multiple antibiotic resistance by quinolones, beta-lactams, and aminoglycosides with special reference to crossresistance between unrelated drug classes. Antimicrob Agents Chemother 1984; 26: 797–801. 12. George AM, Levy SB. Amplifiable resistance to tetracycline, chloramphenicol, and other antibiotics in Escherichia coli: involvement of a non-plasmid-determined efflux of tetracycline. J Bacteriol 1983; 155: 531–540. 13. Martinez-Martinez L, Pascual A, Jacoby GA. Quinolone resistance from a transferable plasmid. Lancet 1998; 351: 797–799.

RESEARCH NOTE Quinolone resistance among Escherichia coli strains from community-acquired urinary tract infections in Greece S. Chaniotaki1,2, P. Giakouppi1, L. S. Tzouvelekis3, D. Panagiotakos1 M. Kozanitou1, G. Petrikkos4, A. Avlami2, the WHONET Study Group and A. C. Vatopoulos1 1

Department of Microbiology, National School of Public Health, 2Department of Microbiology, ‘Laikon’ General Hospital, 3Department of Microbiology, Medical School, University of Athens and 4Infectious Diseases and Antimicrobial Chemotherapy Research Laboratory ‘G. K. Daikos’, University of Athens, Athens, Greece

ABSTRACT Susceptibility data for 10 049 Escherichia coli isolates derived from community-acquired urinary tract infections in Greece during the period January Corresponding author and reprint requests: A. Vatopoulos, Department of Microbiology, National School of Public Health, 196 Alexandras Avenue, Athens 115 21, Greece Tel: + 30 2106422278 Fax: + 30.2106743294 E-mail: [email protected]

2000 to June 2002 indicated 8.1% resistance to nalidixic acid and 36% resistance to ciprofloxacin. In a sample of 170 E. coli isolates, mutations in gyrA (25 isolates) and parC (15 isolates) were consistent with the levels of resistance to quinolones. Previous exposure to quinolones and underlying chronic disease were independent risk factors for infection by quinolone-resistant E. coli strains. Community-acquired, Escherichia coli, quinolones, resistance, urinary tract infection

Keywords

Original Submission: 22 April 2003; Revised Submission: 3 July 2003; Accepted: 29 July 2003

Clin Microbiol Infect 2004; 10: 75–78 Intensive use of quinolones in the treatment of common infections in humans and animals has led to the spread of resistant microorganisms. Quinolone resistance in Escherichia coli is mainly caused by point mutations in the gyrA and parC genes [1,2]. Such mutants are frequently encountered in hospitals worldwide. There are also studies reporting the spread of quinolone-resistant E. coli (QREC) strains in the community [3,4]. Susceptibility data from Greece have indicated a relatively high incidence of quinolone resistance among E. coli strains derived from hospitalacquired urinary tract infections (UTIs) [5]. In this study, the frequency of QREC isolates causing community-acquired UTIs in Greece was estimated. The chromosomal mutations associated with quinolone resistance in a sample of such isolates and the risk factors associated with infection by QREC were also studied. The frequency of QREC strains isolated from community-acquired UTIs in 34 hospitals in Greece during January 2000 to August 2002 was calculated from the data collected by the National Surveillance System for Antimicrobial Resistance with the aid of WHONET software [6]. In order to further study the epidemiological and biological characteristics of resistance to quinolones, E. coli strains were collected consecutively (March to December 2000) from UTI outpatients at Laikon teaching hospital in Athens. This is one of the largest institutions participating in the National Surveillance System for Antimicrobial Resistance, with an active internal medicine outpatient clinic serving a large catchment area that could be considered representative of Athens. Patient data, including demographics, underlying disease,

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

76 Clinical Microbiology and Infection, Volume 10 Number 1, January 2004

immunological status, hospitalisations, invasive device use, and use of antibiotics during the last 12 months, were also obtained. Patients admitted to a hospital during the previous month were excluded. Associations between QREC isolation and various patient characteristics were assessed by contingency tables and chi-square tests with continuity corrections. Relative risks were estimated by calculating odds ratios and their corresponding 95% confidence intervals through backward multiple logistic regression analysis using Stata 6 software (StataCorp, College Station, TX, USA). Diagnosis of UTI was based on the presence of at least one of the following criteria: urgency, frequency, dysuria, or suprapubic tenderness and the isolation of > 105 CFU ⁄ mL of no more than two species of microorganisms from urine. Identification of microorganisms was performed with the API 20E System (bioMe´rieux, Marcy-l’Etoile, France). MICs of nalidixic acid and ciprofloxacin were determined by the Etest method (AB Biodisk, Solna, Sweden). Susceptibility to other antimicrobials was determined according to the disk diffusion breakpoints proposed by the National Committee for Clinical Laboratory Standards [7,8]. PCR-based typing of selected E. coli isolates was carried out with ERIC2 primer (5¢-AAGTAAGTGACTGGGGTGAGCG) as described previously [9]. Amplicons resulting from PCR amplification of the quinolone resistance-determining regions of the gyrA and parC genes [2,10] were purified with a QiaQuick spin PCR purification kit (Qiagen, Hilden, Germany), and the nucleotide sequences were determined with an automatic sequencer (ABI Prism; Applied Biosystems, Warrington, UK). During January 2000 to June 2002, 10 049 E. coli isolates were derived from community-acquired UTIs in the 34 hospitals of the National Surveillance System for Antimicrobial Resistance. Of the 10 049 isolates, 813 (8.1%) were resistant to nalidixic acid, and, of these, 366 (45.1%) were also resistant to ciprofloxacin. Resistance to quinolones was significantly associated with resistance to the four non-quinolones routinely tested in all participating hospitals (p < 0.01) (Table 1). For further study, 170 E. coli isolates were collected from UTI patients visiting the outpatient clinic of the Laikon teaching hospital in Athens. Of 25 (14.7%) nalidixic acid-resistant isolates, 15 (60%) were also resistant to ciprofloxacin.

Table 1. Percentage frequencies of resistance to ciprofloxacin and non-quinolone antibiotics among nalidixic acid-susceptible (Na-S) and -resistant (Na-R) isolates derived from outpatients during January 2000 to August 2002 from 34 hospitals in Greece Antibiotic

Nal-S (n = 9236)

Nal-R (n = 813)

Total (n = 10 049)

Ciprofloxacin Ampicillin Co-trimoxazole Gentamicin Tetracycline

0 33.6 31.4 5.4 44.6

45.1 76.1 72.6 17.4 84.7

3.6 37.1 34.7 6.4 47.8

Twenty-one (84%) QREC isolates were resistant to ampicillin and 17 (68%) to co-trimoxazole (Table 2), compared with 25% and 12.6%, respectively, for quinolone-susceptible isolates. The QREC isolates were also more frequently resistant to gentamicin (28%) and tetracycline (68%) than were the quinolonone-susceptible isolates (0.7% and 17.4%, respectively). All of these differences were statistically significant (p < 0.01). Mutations in the quinolone resistance-determining region of the gyrA gene were detected in all 25 nalidixic acid-resistant isolates. Additional mutations in the quinolone resistance-determining region of the parC gene were found in 15 of the QREC isolates. The most frequent mutations were those resulting in the substitutions Leu for Ser-83 of gyrA (22 isolates), Asn for Asp-87 of gyrA (15 isolates), and Ile for Ser-80 of parC (14 isolates). High-level resistance to ciprofloxacin was associated with mutations at both codons 83 and 87 of the gyrA gene and one or two additional mutations in parC (codons 80 and 84) (Table 2). ERIC2 PCR typing of the 25 QREC isolates showed 14 distinct types, each represented by between one and three isolates (data not shown). The patient data did not indicate epidemiological associations between these isolates. Univariate analysis showed that isolation of QREC strains was positively associated with age. In comparison with younger patients, older (‡ 65 years) UTI patients were at increased risk of infection by a QREC strain than a quinolonesusceptible strain (p ¼ 0.01). Isolation of QREC strains was also strongly associated with the presence of underlying chronic disease (p < 0.001), complicated UTIs (p ¼ 0.005), previous UTI episodes (p < 0.001), and recent exposure to quinolones (p < 0.001). There was no significant association with gender or the use of non-

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

Research Note 77

Table 2. Mutations in the quinolone resistance-determining regions and antibiotic resistance phenotypes of 25 QREC isolates derived from outpatients in Laikon hospital during March to December 2000 Amino-acid change gyrA

MICs (mg ⁄ L)

Other resistance markers

Glu-84

Nal

Cip

Amp

Sxt

Gm

Tet

– – – – – – – – – – – – – – – – – – – – – Lys Val Gly Lys Glu Glu

128 64 128 512 512 64 512 128 64 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 4 4

0.25 0.25 0.25 0.25 0.5 0.125 0.5 0.125 0.125 1 64 8 64 16 16 8 64 16 64 64 64 8 64 64 64 0.032 0.016

R R S R S R R R S S R R R R R R R R R R R R R R R R R

R R S R S R R S S S S R R S R R R R R R R S R R R S S

S S S S S S S S S S R S S R R R S S R S S S R R S S S

R S R R R R S S S S R R R S R R R R S R R S R R R S S

parC

Isolate no.

Ser-83

Asp-87

Ser-80

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 S1a S2a

Leu Leu Leu Leu Leu Leu Ile – – Leu Leu Leu Leu Leu Leu Leu Leu Leu Leu Leu Leu Leu Leu Leu Leu Ser Ser

– – – – – – – Gly Tyr Gly Asn Asn Asn Asn Asn Asn Asn Asn Asn Asn Asn Asn Asn Asn Asn Asp Asp

– – – – – – – – – – Ile Ile Ile Ile Ile Ile Ile Ile Ile Ile Ile – Ile Ile Ile Ser Ser

a

Quinolone-susceptible strains from the same setting. Amp, ampicillin; Cip, ciprofloxacin; Gen, gentamicin; Nal, nalidixic acid; Sxt, co-trimoxazole; Tet, tetracycline.

quinolone antibiotics. Multivariate logistic regression analysis indicated that only chronic disease (odds ratio 22.3; confidence interval 3–128; p ¼ 0.03) and use of fluorinated quinolones (odds ratio 80.7; confidence interval 11–613; p < 0.001) were independent risk factors for infection by a QREC strain in this population. The surveillance data indicated a significant rate of quinolone resistance among E. coli strains from community-acquired UTIs throughout Greece. Most of the QREC isolates examined in this study carried multiple chromosomal mutations known to confer resistance to quinolones. These mutations were consistent with the respective resistance levels to nalidixic acid and ciprofloxacin, although the simultaneous presence of other resistance mechanisms cannot be excluded.

The epidemiological and typing data indicated that such resistant strains probably emerge independently, possibly following selection during treatment with quinolones, or are acquired from exogenous sources [3,11–13]. The former possibility is supported by the strong association with previous use of fluoroquinolones. Indeed, all but two of the QREC isolates were from patients with recent exposure to these agents. Chronic disease also appeared to be an independent risk factor, but this might be a surrogate marker for antibiotic use before the period covered by the study. Notably, most QREC strains were also resistant to other drug classes. Multiresistance may be an important factor for the establishment of QREC strains in chronically ill individuals requiring frequent antimicrobial chemotherapy.

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

78 Clinical Microbiology and Infection, Volume 10 Number 1, January 2004

ACKNOWLEDGEMENTS The National Surveillance System for Antimicrobial Resistance is sponsored by the Hellenic Centre for Infectious Disease Control (KEEL), Ministry of Health. We thank Mrs Argiro Meni for excellent technical assistance. The members of the WHONET study group can be found at http://www.mednet.gr/whonet.

13. Horcajada JP, Vila J, Moreno-Martinez A et al. Molecular epidemiology and evolution of resistance to quinolones in Escherichia coli after prolonged administration of ciprofloxacin in patients with prostatitis. J Antimicrob Chemother 2002; 49: 55–59.

RESEARCH NOTE

REFERENCES 1. Piddock LJ. Mechanisms of fluoroquinolone resistance: an update 1994–1998. Drugs 1999; 58(suppl 2): S11–S18. 2. Weigel LM, Steward CD, Tenover FC. gyrA mutations associated with fluoroquinolone resistance in eight species of Enterobacteriaceae. Antimicrob Agents Chemother 1998; 42: 2661–2667. 3. Garau J, Xercavins M, Rodriguez-Carballeira M et al. Emergence and dissemination of quinolone-resistant Escherichia coli in the community. Antimicrob Agents Chemother 1999; 43: 2736–2741. 4. McDonald LC, Chen FJ, Lo HJ et al. Emergence of reduced susceptibility and resistance to fluoroquinolones in Escherichia coli in Taiwan and contributions of distinct selective pressures. Antimicrob Agents Chemother 2001, 45: 3084–3091. 5. Vatopoulos AC, Kalapothaki V, the Greek Network for the Surveillance of Antimicrobial Resistance, Legakis NJ. Trends in bacterial resistance to ciprofloxacin in Greece. Results from the Greek National Electronic Surveillance System. Emerg Infect Dis 1999; 5: 471–476. 6. Vatopoulos AC, Kalapothaki V, Legakis NJ, the Greek Network for the Surveillance of Antimicrobial Resistance. An electronic network for the surveillance of antimicrobial resistance in bacterial nosocomial isolates in Greece. Bull WHO 1999; 77: 595–601. 7. National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial disk susceptibility testing, 7th edn. Document M2-A7. Wayne, PA: NCCLS, 2000. 8. National Committee for Clinical Laboratory Standards. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 5th edn. Approved standard M7-A5. Wayne, PA: NCCLS, 2000. 9. Johnson JR, O’Bryan TT. Improved repetitive-element PCR fingerprinting for resolving pathogenic and non-pathogenic phylogenetic groups within Escherichia coli. Clin Diagn Lab Immunol 2000; 7: 265–273. 10. Everett MJ, Jin YF, Ricci V, Piddock LJ. Contributions of individual mechanisms to fluoroquinolone resistance in 36 Escherichia coli strains isolated from humans and animals. Antimicrob Agents Chemother 1996; 40: 2380– 2386. 11. Carratala J, Fernandez-Sevilla A, Tubau F, Callis M, Gudiol F. Emergence of quinolone-resistant Escherichia coli bacteremia in neutropenic patients with cancer who have received prophylactic norfloxacin. Clin Infect Dis 1995; 20: 557–560. 12. Dupeyron C, Mangeney N, Sedrati L, Campillo B, Fouet P, Leluan G. Rapid emergence of quinolone resistance in cirrhotic patients treated with norfloxacin to prevent spontaneous bacterial peritonitis. Antimicrob Agents Chemother 1994; 38: 340–344.

Preparation of stock solutions of macrolide and ketolide compounds for antimicrobial susceptibility tests A. Barry1, A. Bryskier2, M. Traczewski1 and S. Brown1 1

The Clinical Microbiology Institute, Wilsonville, Oregon, USA and 2Aventis Pharmaceuticals, Romainville, France

ABSTRACT Stock solutions of telithromycin, ABT-773, azithromycin, clarithromycin, erythromycin, roxithromycin and dirithromycin were each prepared with eight different combinations of solvents and diluents. Broth microdilution trays were then prepared and frozen at ) 60 C. Standard quality control strains were evaluated periodically during a 12-week storage time. There were no significant changes in MICs with different solvents and diluents. It was concluded that the easiest approach was to dissolve each compound in water with a small volume (< 2.5 lL ⁄ mL) of glacial acetic acid added in a dropwise fashion, followed by further dilutions in deionised water. Keywords Ketolide, macrolide, tions, susceptibility tests

stock

solu-

Original Submission: 22 October 2002; Revised Submission: 5 February 2003; Accepted: 12

February 2003 Clin Microbiol Infect 2004; 10; 78–83 As part of an effort to standardise antimicrobial susceptibility testing procedures, the National Corresponding author and reprint requests: A. Barry, 9725 SW Commerce Circle, Suite A-1, Wilsonville, OR 97070, USA Tel: + 1 503 682 3232 Fax: + 1 503 682 2065 E-mail: [email protected]

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

Research Note 79

Committee for Clinical Laboratory Standards (NCCLS) has issued a table specifying the solvent and diluent to be used for preparing stock solutions of antimicrobial agents [1]. For macrolide, azalide and ketolide compounds, the solvent is either ethanol or methanol, and the diluent is either water, a phosphate buffer, or a nutrient broth. The macrolides occasionally require a fairly large volume of alcohol in order to keep the drug in solution, which can affect the ability of stock solutions to be frozen for long-term storage. The potentially toxic solvent dimethylsulphoxide is an alternative solvent that might be considered. In the search for a more reliable method for dissolving and diluting macrolide and ketolide compounds, it was found that they would dissolve in water after addition of a very small amount of glacial acetic acid. A series of experiments was then performed to determine whether this method would affect the end result of microdilution tests and, specifically, whether quality control guidelines would need to be changed if a new solvent or diluent was used. Seven different compounds were studied: telithromycin, ABT-773, azithromycin, clarithromycin, erythromycin, roxithromycin, and dirithromycin (provided by their respective manufacturers). Stock solutions were prepared with four different solvents: ethanol, methanol, dimethylsulphoxide, or acetic acid. In each case, the solvent was added slowly until a clear solution was achieved, which was then further diluted with either water, a phosphate buffer, or Mueller–Hinton broth (Table 1). For dissolving with acetic acid, the compound was added to a small volume of water (half the total volume needed), and glacial acetic acid was then added dropwise until the material dissolved completely. This took < 2.5 lL of acetic acid ⁄ mL. The solution was then further diluted with deionised water. Working solutions were prepared to contain 320 mg ⁄ L or (for Haemophilus influenzae) 1280 mg ⁄ L. Subsequent dilutions were prepared with the broth medium appropriate for the species being tested: cation-adjusted Mueller– Hinton broth (CAMHB) for Staphylococcus aureus ATCC 29213; CAMHB with lysed horse blood 2–3% v ⁄ v for Streptococcus pneumoniae ATCC 49619; and Haemophilus Test Medium (HTM) for H. influenzae ATCC 49247. The range of concentrations in each microdilution panel was varied for each control strain in order to ensure

on-scale endpoints throughout the study. When the Gram-positive control strains were tested, the working solutions were further diluted in CAMHB to provide 16 mg ⁄ L and 1.0 mg ⁄ L, and these were then serially diluted to provide 12 doubling concentrations. For testing the H. influenzae control strain, the working solutions were diluted 1 : 5 in HTM broth to provide an initial concentration of 256 mg ⁄ L (64 mg ⁄ L for azithromycin or telithromycin). Serial dilutions were then prepared in HTM broth and dispensed into wells in microdilution trays. On the day of preparation, the appropriate control strain was tested in triplicate, using the methods defined by the NCCLS [1]. The remaining trays were stored at ) 60 C, and triplicate tests were repeated every 3 weeks for a total of 12 weeks. Broth from uninoculated microdilution panels was aspirated from wells containing the highest and lowest concentrations tested, and the pH of each was recorded. These pH values were compared with the pH of broth in the growth control well without antibiotics. Table 1 shows the median of 15 MIC values obtained in each solution. In all cases, the 15 MICs were no more than one doubling concentration from the median value. Median MICs from panels prepared with different types of working solution were essentially the same (± 1 doubling concentration) for each of the seven study agents. The solvent and diluent did not appear to influence the final result with quality control strains. Table 2 shows the overall range of all 120 MICs recorded for each control strain. This range of observed MICs is contrasted with the quality control ranges that have been defined in NCCLS documents [1,2]. Since roxithromycin and ABT773 are not included in the NCCLS tables, control limits were those proposed elsewhere [3] (data on file). Only ten of the 3024 MIC values were outside the quality control ranges, and none of those involved solutions prepared with acetic acid. The pH values in test wells prepared in CAMHB were essentially the same (± 0.1 unit) as that of the control broth without antibiotic. However, when trays were prepared with HTM broth, the working solutions were diluted only 1 : 5 to provide 256 mg ⁄ L or 1 : 20 for 64 mg ⁄ L (telithromycin and azithromycin). Wells prepared to contain 256 mg ⁄ L contained enough solvent or diluent to lower the pH by approximately 0.2 units, especially when acetic acid was the

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

80 Clinical Microbiology and Infection, Volume 10 Number 1, January 2004

Median of 15 MICs (mg ⁄ L)c Antimicrobial agent, stock solution solventsa and working solution diluentsb

Streptococcus Staphylococcus Haemophilus pneumoniae aureus influenzae ATCC 49619 ATCC 29213 ATCC 49247

Telithromycin Ethanol (9% v ⁄ v) in pH 7.2 buffer Water diluent 0.008 Buffer diluent 0.008 Methanol (9% v ⁄ v) in pH 7.2 buffer Water diluent 0.016 Buffer diluent 0.016 Acetic acid (2.5 lL ⁄ mL) in water Water diluent 0.008 Buffer diluent 0.008 DMSOd (9% v ⁄ v) in pH 7.2 buffer Water diluent 0.008 Buffer diluent 0.008 ABT-773 Ethanol (9% v ⁄ v) in pH 6.5 buffer Water diluent 0.016 Buffer diluent 0.016 Methanol (9% v ⁄ v) in pH 6.5 buffer Water diluent 0.016 Buffer diluent 0.008 Acetic acid (2.0 lL ⁄ mL) in water Water diluent 0.016 Buffer diluent 0.016 DMSO (13% v ⁄ v) in pH 6.5 buffer Water diluent 0.016 Buffer diluent 0.016 Azithromycin Ethanol (6% v ⁄ v) in CAMHB CAMHB diluent 0.12 Buffer diluent 0.12 Methanol (6% v ⁄ v) in CAMHB CAMHB diluent 0.12 Buffer diluent 0.12 Acetic acid (1.8 lL ⁄ mL) in water Water diluent 0.12 Buffer diluent 0.12 DMSO (6% v ⁄ v) in CAMHB CAMHB diluent 0.12 Buffer diluent 0.12 Clarithromycin Ethanol (40% v ⁄ v) in pH 6.5 buffer Water diluent 0.03 Buffer diluent 0.03 Methanol (42% v ⁄ v) in pH 6.5 buffer Water diluent 0.03 Buffer diluent 0.03 Acetic acid (1.2 lL ⁄ mL) in water Water diluent 0.03 Buffer diluent 0.03 DMSO (20% v ⁄ v) in pH 6.5 buffere Water diluent 0.03 Buffer diluent 0.03 Erythromycin Ethanol (10% v ⁄ v) in water Water diluent 0.06 Buffer diluent 0.06

0.12 0.12

2.0 1.0

0.12 0.12

2.0 1.0

0.06 0.06

1.0 2.0

0.06 0.06

1.0 2.0

0.06 0.06

2.0 2.0

0.06 0.06

2.0 2.0

0.12 0.12

2.0 2.0

0.12 0.06

2.0 2.0

1.0 1.0

1.0 1.0

2.0 2.0

2.0 2.0

1.0 1.0

2.0 2.0

1.0 1.0

2.0 2.0

0.5 0.5

8.0 8.0

0.5 0.5

8.0 8.0

0.25 0.25

4.0 4.0

0.5 0.5

8.0 NDf

0.5 0.5

4.0 4.0

Table 1. Performance of two ketolides, an azalide and four macrolides when microdilution panels were prepared with eight different types of working solution of each antimicrobial agent

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

Research Note 81

Table 1. continued

Median of 15 MICs (mg ⁄ L)c Antimicrobial agent, stock solution solventsa and working solution diluentsb Methanol (10% v ⁄ v) in water Water diluent Buffer diluent Acetic acid (1.2 lL ⁄ mL) in water Water diluent Buffer diluent DMSO (10% v ⁄ v) in water Water diluent Buffer diluent Roxithromycin Ethanol (10% v ⁄ v) in water Water diluent Buffer diluent Methanol (11% v ⁄ v) in water Water diluent Buffer diluent Acetic acid (1.3 lL ⁄ mL) in water Water diluent Buffer diluent DMSO (50% v ⁄ v) in watere Water diluent Buffer diluent Dirithromycin Ethanol (5% v ⁄ v) in water Water diluent Buffer diluent Methanol (5% v ⁄ v) in water Water diluent Buffer diluent Acetic acid (2.5 lL ⁄ mL) in water Water diluent Buffer diluent DMSO (15% v ⁄ v) in watere Water diluent Buffer diluent

Streptococcus Staphylococcus Haemophilus pneumoniae aureus influenzae ATCC 49619 ATCC 29213 ATCC 49247 0.06 0.06

0.5 0.5

8.0 8.0

0.06 0.06

1.0 0.5

8.0 8.0

0.06 0.06

0.5 0.5

4.0 4.0

0.12 0.12

1.0 1.0

8.0 16

0.12 0.12

1.0 1.0

8.0 16

0.12 0.12

1.0 1.0

8.0 8.0

0.12 0.06

1.0 1.0

16 NDf

0.12 0.12

2.0 2.0

16 16

0.12 0.12

2.0 2.0

16 16

0.12 0.12

2.0 2.0

16 16

0.25 0.25

4.0 4.0

16 NDf

a

Stock solutions with 2560 mg ⁄ L were prepared by dissolving antibiotic powder in a minimal volume of solvent and then adjusting by adding the diluent recommended by the manufacturer of each agent. b Working solutions with 320 mg ⁄ L (1280 mg ⁄ L when testing H. influenzae) were prepared by diluting each stock solution with deionised water and with a pH 7.2 phosphate buffer. c Serial dilutions of each working solution were prepared by diluting in cationadjusted Mueller–Hinton broth (CAMBH) with or without lysed horse blood or in Haemophilus Test Medium (HTM broth), and microdilution test panels were prepared and stored at ) 70 C until needed. Each control strain was tested in triplicate every 3 weeks to document short-term stability (12 weeks at ) 70 C). There was no consistent trend for a change in MICs during storage, and all 15 MICs differed by no more than one doubling concentration from the median. d DMSO, dimethylsulphoxide (potentially toxic solvent). e Clarithromycin, roxithromycin and dirithromycin were not very soluble in DMSO, and thus the stock solution contained only 1280 mg ⁄ L. f ND, not done; because of the limited solubility of two macrolides in DMSO, only one working solution was evaluated in final concentrations great enough to obtain on-scale endpoints with the H. influenzae control strain.

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

82 Clinical Microbiology and Infection, Volume 10 Number 1, January 2004

MIC (mg ⁄ L) ranges and quality control limits

Antimicrobial agent and MIC ranges Telithromycin Overall range of 120 MICs Quality control limits ABT-773 Overall range of 120 MICs Quality control limits Azithromycin Overall range of 120 MICs Quality control limits Clarithromycin Overall range of 120 MICs Quality control limits Erythromycin Overall range of 120 MICs Quality control limits Roxithromycin Overall range of 120 MICs Quality control limits Dirithromycin Overall range of 120 MICs Quality control limits

Streptococcus pneumoniae ATCC 49619

Staphylococcus aureus ATCC 29213

Haemophilus influenzae ATCC 49247

0.004–0.016 0.004–0.03

0.06–0.12 0.06–0.25

1.0–2.0 1.0–4.0

0.008–0.03a 0.002–0.016

0.06–0.12 0.06–0.25

1.0–2.0 1.0–4.0

0.06–0.25 None

1.0–4.0b 0.5–2.0

1.0–4.0 1.0–4.0

0.03–0.03 0.03–0.12

0.25–0.5 0.12–0.5

4.0–8.0d 4.0–16

0.03–0.06 0.03–0.25

0.5–1.0 0.25–1.0

2.0–16 None

0.06–0.12 0.06–0.25

1.0–2.0 0.5–2.0

8.0–16d 8.0–32

0.06–0.25 0.06–0.25

2.0–8.0c 1.0–4.0

16–32d 8.0–32

Table 2. Overall range of MICs described in Table 1 compared to quality control limits anticipated for each of three control strains

a

One of 144 MICs was outside the quality control range (stock dissolved in ethanol). b Two of 144 MICs were outside the quality control range (stock dissolved in methanol). c Seven of 144 MICs were outside the quality control range (stock dissolved in dimethylsulphoxide). d Only 75 MICs were available for analysis, because three drugs were not tested with a buffer diluent.

solvent. The pH values of lower concentrations did not differ from that of the control. When a 20-fold dilution was prepared to achieve an initial concentration of 64 mg ⁄ L, the pH values of all eight preparations were essentially identical to that of the broth control. The slight decrease in pH involved only extremely high concentrations of the study compounds and was well above the range of MICs expected for the quality control strains. However, it was concluded that working solutions should contain enough drug to require a 1 : 10 or 1 : 20 dilution with a broth medium before serial dilutions are carried out. To further evaluate the effect of diluents carried over in the first well, a set of ten strains was selected with elevated macrolide MICs (six S. pneumoniae, three Streptococcus pyogenes, and one Staph. aureus). Susceptibility tests were performed with the different types of working solutions defined by the NCCLS [1,2] and with a weak solution of acetic acid diluted in water or in

a phosphate buffer. MICs recorded for this challenge set of resistant strains were not affected by the solvent or diluent. MICs of all seven drugs were essentially the same, even when they were > 256 mg ⁄ L (data not shown). As ethanol, methanol and dimethylsulphoxide are not entirely satisfactory solvents for the macrolides, it was concluded that a dilute solution of glacial acetic acid is the preferred solvent for all seven study drugs. To prepare a stock solution, the compound should be added to deionised water (approximately half the total volume desired) and the glacial acetic acid then added in a dropwise fashion until the compound is completely dissolved (this should require < 2.5 lL ⁄ mL). Further dilution can then be carried out with deionised water. The working solution should contain at least 10–20-fold more antibiotic than the highest concentration that will be tested. This procedure has been recognised by the NCCLS subcommittee on susceptibility test-

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83

Research Note 83

ing and is acknowledged as an accepted alternative approach for preparing stock solutions of four study drugs. Solvents for ABT-773, dirithromycin and roxithromycin have not yet been presented to the subcommittee for inclusion in this document. ACKNOWLEDGEMENTS These studies were made possible by grants from Aventis Pharmaceuticals, Romainville, France and from Abbott Laboratories, Abbott Park, Illinois.

REFERENCES 1. National Committee for Clinical Laboratory Standards. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 5th edn. Approved standard M7-A5. Wayne, PA: NCCLS, 2000. 2. National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial susceptibility testing: Twelfth Informational Supplement. M100-S12. Wayne, PA: NCCLS, 2002. 3. Barry AL, Brown SD. Parameters for quality control of antimicrobial susceptibility tests of roxithromycin. Clin Microbiol Infect 1999; 5: 233–234.

 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 70–83