436 Clinical Microbio logy and Infection, Volume 5 Number 7 , J u l y 1999

10. Havlir DV, Dube MP, Sattler FR, et al. Prophylaxis against disseminated Mycobacterium avium complex with weekly azithro- mycin, daily rifabutin or both. N Engl J Med 1996; 335: 392-8.

11. Horsburgh C R , Metchock B, Gordon SM, Havlik JA, McGowan JE, Thompson SE. Predictors of survival in patients with AIDS and disseminated Mycobacterium avium complex disease. J Infect Dis 1994; 170: 573-7.

12. Sathe SS, Gascone F', Lo W, Pinto R , Reichman LB. Severe anemia is an important negative predictor for survival with disseminated Mycobacterilrm avilrm in acquired immunodeficiency syndrome. Am Rev Respir Dis 1990; 142: 1306-12.

13. Singer J, Thorne A, Phlipps P, et al. Predictors of survival and eradication of Mycobacterium avium complex bacteremia (MAC) in AIDS patients in the Canadian randomized treatment trial. Canadian HIV Trials: Network Protocol 020 Study Group. AIDS

14. Benson CA. Disease due to the Mycobacterium avium complex in patients with AIDS: epidemiology and clinical syndrome. Clin Infect Dis 1994; 18(suppl 3): S218-22.

1999; 13: 575-82.

15. Hoover DR, Graham NMH, Bacellar H, et al. An epidemio- logic analysis of Mycobacterium avium complex disease in homosexual men infected by the human immunodeficiency virus type 1. Clin Infect Dis 1995; 20: 125G8.

16. Broggart CL, Louis TA, Hillman DW, et al. A randomized, placebo-controlled trial of the safety and efficacy of oral ganciclovir for prophylaxis of cytomegalovirus disease in HIV- infected individuals. Terry Beirn Community Programs for Clinical Research on AIDS. AIDS 1998; 12: 269-77.

17. Jouan M, McCutchan JA, Bartok A. Survival of AIDS patients coinfected with disseminated Mycobacterirtm avium complex and cytomegalovirus [abstract no. 1171. In: Program and abstracts of the 2nd Conference on Retroviruses and Opportunistic Infections, Washington, 1995.

18. Kemper CA, Havlir D, Bartok AE, et a1 Transient bacteremia due to Mycobacterium avium complex in patients with AIDS. J Infect Dis 1994; 170: 488-93.

Antimicrobial susceptibility of coagulase-negative staphylococci isolated between 1991 and 1996 from a German university hospital

Clin Microbiol Infect 1999; 5: 436-439

Franx-Josef Schnitx I,', Sylvia Theis', Ad C. Fluit', Jan Krhotf2, Hans-Peter Heinx' and Mark E. Jones3

Institute for Medical Microbiology and Virology, Heinrich-Heine-Universitat Diisseldorf, Germany; 2Eijkman-Winkler Institute for Medical Microbiology, University Hospital Utrecht, 3 M € U Pharmaceutical Services, Den Brielstraat 11, 3554XD, Utrecht, The Netherlands

Tel: +31 30 265 1794 Fax: +31 30 265 1784

E-mail: rnjonesQthetsn.com

Accepted 21 November 1998

Infections caused by coagulase-negative staphylococci (CNS) are commonly associated with implanted devices such as intravascular catheters, prosthetic heart valves and shunts [ 1,2]. Although several typing systems have helped to elucidate their epidemiology and mode of transmission, and to define outbreaks of nosocomial infections [3-6], little is known about the hospital epidemiology of CNS, including both methicdlin- resistant and methicillin-sensitive strains (MRCNS and MSCNS respectively). This is hrther complicated by the ubiquitous nature of CNS as human skin flora. CNS may also serve as a reservoir for antibiotic resistance genes that can be potentially transferred to other Gram-positive organisms, including Stuphylococcus aureus [7]. In order to better understand the epidemio- logy of antimicrobial resistance in CNS within the Heinrich-Heine University Hospital, Diisseldorf, routine susceptibility data for isolates derived from different

body sites in patients from intensive care units (ICUs), regular wards and the bone marrow transplantation unit were reviewed.

The Heinrich-Heine University Hospital Dussel- dorf is a 1500-bed hospital comprising regular care units and five units devoted to intensive medical care, namely the surgical ICU, pediatric ICU, neurologic ICU, internal medicine ICU and bone marrow trans- plant (BMT) unit. Between 1991 and 1996, all clinically significant CNS isolates &om blood cultures, catheter tips, liquors and aspirates, and all isolates &om immuno- compromised patients and other patients where CNS were suspected to be involved in a bacteremic episode, were referred for confirmation of identity and suscepti- bility testing.

Isolates were identified as CNS by demonstra- ting a positive catalase test, a negative coagulase test (Bacto Coagulase Plasma EDTA-Test, Difco, Augsburg,

Concise C o m m u n i c a t i o n s 437

Germany), a negative nuclease test (DNase agar, Oxoid, Wesel, Germany), and showing no anaerobic mannitol metabolism. ATCC reference strains 25923, 29213 and 33591 were used as controls.

Susceptibility to antibiotics was detected using agar diffusion tests and breakpoints according to the Deutsches Institut fur Normung (DIN) guidelines [8] and were reported as susceptible (S), intermediate (I) or resistant (R). Organisms were susceptibility tested to the following antibiotics, with breakpoint criteria defined as follows listed by drug, disk load (shown in parentheses as pg) and S, I and R category breakpoints shown in mm: oxacillin (5) 16, -, 15; penicillin G (6) 29, -, 28; gentamicin (10) 21, 15-20, 14; netilmicin (10) 22, 17-21, 16: tetracycline (30) 22, 17-21, 16; ciprofloxacin (5) 23, 19-22, 18; ofloxacin (5) 22, 18-21, 17; co-trimoxazole (1.25/23.75) 16, 11-15, 10; erythromycin (1 5) 2 1, 17-20, 16; clindamycin (1 0) 24, 19-23, 18; teicoplanin (30) 14, -, 10; chloramphenicol (30) 21, -, 20; vancomycin (30) 14, -, 10; fosfomycin (100) 20, 19-14, 13; fusidic acid (10) 20, -, 19; and rifampicin (30) 25, -, 24. A single dash indicates no intermediate class designated.

Between 1991 and 1996, a total of 6323 CNS considered to be clinically significant were isolated from different sample materials. Of these, 3 541 (56%) isolates were methicillin susceptible and 2 782 (44%) isolates were methicillin resistant. Of isolates from regular wards, 1126 of 3631 (31%) were MRCNS, while 1 138 of 1581 (72%) of isolates from ICUs and 5 18 of 1 11 1 (52%) of isolates from the BMT unit were MRCNS (p<0.05 using the chi-squared test). Despite the units being heterogeneous with respect to patient populations, prescribed antibiotic regimens and duration of hospital stay (data not shown), the proportion of MRCNS isolates from each of the four ICUs remained between 68% and 72% (Table 1). Between 70% and 87% of CNS from either blood cultures or from

Table 1 Numbers of CNS from body sites in different ICUs

catheters derived from each of the ICUs were methcillin resistant, while the rate of methicillin resistance among CNS from other sources (e.g. liquor, aspirates, wound swabs) varied from 55% to 65% (Table 1). The signifi- cantly higher prevalence of methicillin resistance in isolates derived from blood or catheter tips (p<0.05), as compared to isolates derived from other sites of infection, probably reflects exposure to higher concen- trations of antibiotics.

It is known that methicillin resistance in staphylo- cocci also predicts resistance to multiple classes of antibiotics besides p-lactams [7]. In accordance with this observation, the MRCNS isolates showed signi- ficantly increased resistance to other antimicrobials, especially to quinolones and aminoglycosides (Table 2). Methicillin resistance can be used as a fairly reliable marker for resistance to a range of different drugs. Therefore, it is clearly prudent to assume that MRCNS are multiresistant and to carefully tailor therapeutic regimens according to susceptibility test results.

A significant increase in the in vitro resistance to all drugs tested, with the exception of glycopeptides, was observed for isolates derived from ICUs as com- pared to those derived from regular wards. This was especially so for the prevalence of resistance to cipro- floxacin and ofloxacin, which increased in MSCNS from 9% to 34% between the two sample groups. Similarly, MRCNS isolates derived from ICUs were more resistant to each of the antibiotics tested, again with the exception of glycopeptides, than MRCNS isolated from the rest of the hospital, with particularly noticeable increases in resistance to ciprofloxacin and ofloxacin, from 42% and 41% to 73% and 76%, respec- tively (pC0.05). Similarly, resistance to netilmicin increased from 10% to 28% (pc0.05). Similar patterns of increasing resistance in the susceptibility profiles of CNS have been reported previously [2,7,9-121. The results of the EPIC Study focusing specifically on

Surgical ICU

Pedatric ICU

Internal Neurosurgical ICU ICU

BC"

CT

0 s '

67 (18%) (78% MR)

197 (53%) (87% MR)

107 (29%) (65% MR)

60 (20%) (70% MR)

151 (50%) (78% MR)

91 (30%) (55% MR)

50 (16%) (70% MR)

33 (15%) (80% MR)

170 (54%) 134 (61%) (74% MR) (76% MR)

95 (30%) (57% MR)

53 (23%) (57% MR)

Total 371 302 315 220 (72% MR) (70% MR) (68% MR) (72% MR)

"BC, CNS from blood cultures; bCT, CNS from catheter tips; 'OS, CNS from other sources (e.g. wound swabs, aspirates etc.); MR. methicillin resistant.

438 C l in ica l M i c r o b i o l o g y a n d I n f e c t i o n , V o l u m e 5 N u m b e r 7, J u l y 1999

Table 2 Resistance patterns of MSCNS and MRCNS from all hospital units. S, I and R values were calculated using DIN interpretative criteria [8]. Data points are shown as percentage values

MSCNS (n=3541) MRCNS (n=2782)

Antibiotics S I R S I R

Penicillin 28 - 72 0 - 100 (p<0.05)" Gentamicin 79 4 17 16 10 74 (p<0.05) Netilmicin 92 4 4 66 18 16 (p<0.05) Tetracycline 77 5 18 69 4 27 NSb Ciprofloxacin 80 4 16 40 5 55 (p<0.05) Ofloxacin 81 3 16 41 5 54 (p<0.05) Co-tr imoxazole 69 0 31 21 0 79 (p<0.05) Erythromycin 61 0 39 20 0 80 (pc0.05) Clindamycin 79 0 21 33 0 67 (p<0.05) Teicoplanin 100 0 0 100 0 0 NS Vancomycin 100 0 0 100 0 0 NS

Additional tested antibiotics Chloramphenicol Not tested 86 0 14

Fusidic acid Not tested 60 0 40 Rifampicin Not tested 71 0 29

Fosfomycin Not tested 55 0 45

"P-values were calculated using the chi-squared test. bNS, non-significant.

European ICU isolates showed that 70% of the CNS isolated were multiresistant MRCNS [2].

An analysis of methicillin resistance (and, by inference, multiple-drug resistance) in CNS isolates h m regular care units, from ICUs and from the BMT unit showed a significant yearly increase between 1991 and 1996. Whereas in 1991 only 27% of CNS from regular care units were resistant to methicillin, by 1996 this had increased to 36% (pCO.05). During this same period, methicillin resistance in CNS isolates from the ICUs and the BMT unit increased from 62% and 44% in 1991 to 78% and 58%, respectively, in 1996 (pCO.05). In a similar study during a 6-year period, Durand-Gasselin et al showed an increase in MRCNS from 33% to 56% on a hematology unit [l 11.

It is important to note that methicdhn resistance is much more frequently encountered in CNS than in isolates of S. aureus [13]. This is apparent in the Heinrich-Heine University Hospital, where about 6% of S. aureus isolates are methicillin resistant [14]. As antibiotics used to treat infections with both organisms are often the same, the selective pressure of antibiotic use alone cannot explain the selection or frequency of MRCNS, especially in ICUs. As CNS are ubiquitous skin organisms and can spread easily on skin particles or by air, hygiene measures implemented primarily to prevent S. aureus cross-infections may not be sufficient to control transmission of MRCNS.

It is clear that the prevalences of MRCNS in each of the four ICUs were similar, even though aspects of medical care and patient demography differed between them. In addition, a uniform increase of methicillin and associated multiresistance was noted in all units

during the study period. This may provide circum- stantial evidence to indicate continuous transmission of MRCNS between staff and patients in the ICUs.

Data from the BMT unit provided some evidence to support this hypothesis that MRCNS strains are continuously transmitted throughout the hospital between both staff and patients. The BMT unit has long-term patients in residence, a high consumption of broad- and narrow-spectrum antibiotics, and a rigorous standard of infection control and hygiene. Compared to the four ICUs, the BMT unit had a 20-25% lower incidence of methcillin resistance. Thls suggests that if the cycle of transmission is broken by extreme conditions of hygiene (such as those enforced within the BMT unit), it is possible to limit the extent of resistance. In contrast to MRSA [15], the transmission ofMRCNS is notoriously hard to demonstrate because of the ubiquitous nature of CNS.

In conclusion, it is clear that the prevalence of methicillin resistance and associated multiresistance in CNS has increased significantly during the 6-year study period. Differences can be seen between the suscepti- bility profiles of CNS from ICUs and regular hospital units. The reduced prevalence of MRCNS isolates from the BMT unit, compared to the ICUs, may provide circumstantial evidence that hygiene plays a key role in controlling the circulation of these organisms within the hospital environment.

References 1. moos WE, Bannerman TL. Update on clinical significance of

coagulase-negative staphylococci. Clin Microbiol Rev 1994; 7: 117-40.

Concise Communications 439

2. Spencer RC. Predominant pathogens found in the European prevalence of infection in intensive care study Eur J Clin Microbiol Infect Dis 1996; 15: 281-5.

3. John JF, Grieshop TJ, Atkins LM, Platt CG. Widespread colonization of personnel at a Veterans Mairs Medical Center by methicillin-resistant, coagulase-negative Stuphylococcus. J Clin Infect Dis 1993; 17: 38S8.

4. Kotilainen P, Houvinen S, Jarvinen H, Aro H, Houvinen P. Epidemiology of the colonization of inpatients and outpatients with ciprofloxacin-resistant coagulase-negative staphylococci. J Clin Infect Dis 1995; 21: 685-7.

5. Thurn JR, Crossly KB, Gerdts A, Baken L. Dynamics of coagulase-negative staphylococcal colonization in patients and employees in a surgical intensive care unit. J Hosp Infect 1991;

6. Patrick CH, John JE Levkoff AH, Atkins LM. Relatedness of strains of methicillin-resistant coagulase-negative Staphylococc~s colonizing hospital personnel and producing bacteriemias in a neonatal intensive care unit. J Pediatr Infect Dis 1991; 11: 93540.

7. Archer GL, Climo MW. Antimicrobial susceptibility of coagulase- negative staphylococci. Antimicrob Agents Chemother 1994;

8. Deutsches Institut fur Normung. Empfindlichkeitspriifung von Krankheitserregern gegen Chemotherapeutika. In: DIN- Taschenbuch medizinische mikrobiologie und immunologie. Berlin: Beuth, 1991; 331-406.

20: 247-55.

38: 2231-7.

9. Fomsgaard A, Hoiby N. Friis HM, et al. Prevalence and antibiotic sensitivity of Danish versus other European bacterial isolates from intensive care and hematology/oncology units. Eur J Clin Microbiol Infect Dis 1995; 14: 275-81.

10. Rello J, Ricart M, Mirelis B, et al. Nosocomial bacteremia in a mehcal-surgical intensive care unit: epidemiologic characteristics and factors influencing mortality in 111 episodes. Intens Care Med 1994; 20: 94-8.

11. Durand-Gasselin B, Leclercq R , Girard-Pipau F, et al. Evolution of bacterial susceptibllity to antibiotics during a six-year period in a hematology unit. J Hosp Infect 1995; 29: 19-33.

12. Tripoldi MF, Attanasio V, Alnolfi LE, et al. Prevalence of antibiotic resistance among clinical isolates of methicillin- resistant staphylococci. EurJ Clin Microbiol Infect Dis 1994; 13: 148-52.

13. Berger-Bachi B. Resistance not mediated by p-lactamase (methicillin resistance) In Crossley KB, Archer GL, eds. The staphylococci in human disease. Edmburgh Churchill Livingston,

14. Schmitz FJ, MacKenzie CR, Geisel R , et al. Methicillin resistant Staphylococcus aureus strains in the greater Dusseldorf area. Eur J Epidemiol 1997; 13: 709-17.

15. Allen KD, Anson JJ, Parsons LA, Frost NG. Staff carriage of methicillin-resistant Staphylococnrs aureuz (EMRSA 15) and the home environment: a case report. J Hosp Infect 1997; 35:

1997; 158-7'5.

307-1 1.

Interaction of insulin and Burkholderia cepacia

Clin Microbiol lnfect 1999; 5: 439-442

SarahJeromson I , Paula Keig' and Kevin Kerr2*

'Department of Chemical and Biological Sciences, University of Huddersfield, Huddersfield, "Department of Microbiology, University of Leeds, Leeds LS2 9JT, UK *Tel: +44 113 233 5617 Fax: f 4 4 113 233 5649 E-mail: mickgk@leeds.ac.uk

Accepted 10 November 1998

Key words: Cystic fibrosis, Burkholderia cepacia, insulin, cresol

Burkholderia cepacia has emerged as an important cause of pulmonary sepsis in individuals with cystic fibrosis (CF). Colonization of the airways with this bacterium, which often occurs in the second decade of life, is associated with three outcomes: chronic asymptomatic carriage, progressive deterioration over a prolonged period, or a rapidly fatal necrotizing pneumonia [l].

Putative virulence factors of B. cepacia remain ill-defined, although there is increasing evidence to suggest that the bacterium can invade respiratory epithelial cells and may thus behave as a facultative intracellular pathogen [2,3]. Another member of the genus Burkkolderia, B. pseudomallei-the causative agent of melioidosis-can also act in this manner [4].

Following observations that insulin-dependent (type I) diabetes is a risk factor for the development

of the septicemic form of melioidosis, Woods et a1 investigated the possibility that insulin may influence the growth of B. pseudomallei [5] and reported that recombinant human insulin inhibited the growth of this bacterium.

As diabetes is a well-recognized complication of CF, we have examined whether a similar interaction between B. cepacia and insulin might also occur. More- over, our findings, in conjunction with doubts raised over the conclusions ofWoods et a1 [6], suggest that the hypothesis that recombinant human insulin affects the growth of B. pseudomallei merits re-examination.

Two strains of B. cepacia were examined. C1359, isolated &om a CF patient, was a generous gift from Dr J. R. W Govan of the University of Edinburgh Medical School, and NCTC 10661, an environmental isolate,